1 INFO-VAX	Thu, 29 Nov 2001	Volume 2001 : Issue 664       Contents:; Re: "Why Great Companies Fail": Includes DEC (perhaps HP ?) ; Re: "Why Great Companies Fail": Includes DEC (perhaps HP ?) * Re: Andrew Harrison : Alive & Still At SUN* Re: Andrew Harrison : Alive & Still At SUN& Re: Can We Turn Down the Flames a Bit? Re: Careful with Mozilla 0.9.69 Re: Compaqs VMS plans for IPF port ... any doubters left?  Re: Complie Error % DCL minute of the day: PCSI_MENU v1.1 " Re: DCPS stopped, inactive problem" Re: DCPS stopped, inactive problem Re: DEC C Error on GS v7.03 1 DECEMBER SPECIAL: DS20 1GB $3299 IN NEW Condition  DECNET Phase IV saves the day  Re: DECnet timeout? , Re: DSSI VAXcluster manual on line anywhere?/ FC  configurations with only one HSG controller 3 RE: FC  configurations with only one HSG controller 3 Re: FC  configurations with only one HSG controller  ftp performance  Re: ftp performance  Re: ftp performance  Re: Future of VMS ?  Re: Future of VMS ?  Re: Future of VMS ?  Re: Future of VMS ?  Re: Future of VMS ?  Re: Future of VMS ?  Re: Future of VMS ?  Re: Future of VMS ?  Re: Future of VMS ?  Re: Future of VMS ?  Re: Future of VMS ?  Re: Future of VMS ?  Re: Future of VMS ?  Re: Future of VMS ?  Re: Future of VMS ?  Re: Future of VMS ?  Re: Future of VMS ? 4 Re: Gartner and IDC say HP will effectively kill DLT4 Re: Gartner and IDC say HP will effectively kill DLT( Installing Icon permanently to CDE menu?, Re: Installing Icon permanently to CDE menu?, Re: Installing Icon permanently to CDE menu? LNC02 cont.  Re: LNC02 cont.  Re: LNC02 cont.  Re: LNC02 cont.  Re: LNC02 cont.  Re: LNC02 cont. ' Re: LNC02 problem with DCPS  was: LNC02 ! Re: logical names novice question   Mozilla DECWserver crash problem6 Re: New product allows VMS to become part of a PC Lan!6 Re: New product allows VMS to become part of a PC Lan!6 Re: New product allows VMS to become part of a PC Lan!6 Re: New product allows VMS to become part of a PC Lan!6 Re: New product allows VMS to become part of a PC Lan!6 Re: New product allows VMS to become part of a PC Lan! new to VMS cxx question...! NT to Solaris Migration Promotion % Re: NT to Solaris Migration Promotion % Re: NT to Solaris Migration Promotion / Re: Obtaining System Serial Number via Software 8 Re: OpenVMS VAXSHAD03_072 VAX V7.2 SHADOWING ECO Summary$ Oracle 8i OCI problem with VAX BASIC( Re: Oracle 8i OCI problem with VAX BASIC Oracle RDB announcment/letter ! Re: Oracle RDB announcment/letter ! Re: Oracle RDB announcment/letter . Re: OT: Andrew Harrison : Alive & Still At SUN. Re: OT: Andrew Harrison : Alive & Still At SUN  patch downloads - 3rd time lucky$ Re: patch downloads - 3rd time lucky2 Re: patch downloads - 3rd time lucky & compression- Re: Problem fseeking a 4.2 million block file  RECALL does not work Re: RECALL does not work Re: RECALL does not work Re: RECALL does not work? Re: SKC Musings on "Recent IPF Unpleasantness" at www.tru64.org E Re: Special IPF-Inside Issue of Shannon Knows Compaq at www.tru64.org E Re: Special IPF-Inside Issue of Shannon Knows Compaq at www.tru64.org E Re: Special IPF-Inside Issue of Shannon Knows Compaq at www.tru64.org E Re: Special IPF-Inside Issue of Shannon Knows Compaq at www.tru64.org 4 Re: Time for the Encompass US Inc CHAD-FREE ElectionE Re: Tru64 .vs. HP-UX  (was: Compaq's Secret VMS Plans (The Inquirer)) E Re: Tru64 .vs. HP-UX  (was: Compaq's Secret VMS Plans (The Inquirer)) E Re: Tru64 .vs. HP-UX  (was: Compaq's Secret VMS Plans (The Inquirer)) E Re: Tru64 .vs. HP-UX  (was: Compaq's Secret VMS Plans (The Inquirer)) E Re: Tru64 .vs. HP-UX  (was: Compaq's Secret VMS Plans (The Inquirer)) E RE: Tru64 .vs. HP-UX  (was: Compaq's Secret VMS Plans (The Inquirer)) E Re: Tru64 .vs. HP-UX  (was: Compaq's Secret VMS Plans (The Inquirer)) . vms7.2-1 tape initialize 30 minute timeout bug Wall Street Journal on Carly  Re: Wall Street Journal on Carly  Re: Wall Street Journal on Carly  Re: Wall Street Journal on Carly4 Re: why not a communityDeveloped[tm] version of VMS?4 Re: why not a communityDeveloped[tm] version of VMS?4 Re: why not a communityDeveloped[tm] version of VMS?1 Xbox (was: re: Compaqs VMS plans for IPF port ... 5 Re: Xbox (was: re: Compaqs VMS plans for IPF port ... 	 Re: xdmcp 5 Re: Yahoo - IBM unseated in new supercomputer ranking 5 Re: Yahoo - IBM unseated in new supercomputer ranking 5 Re: Yahoo - IBM unseated in new supercomputer ranking   F ----------------------------------------------------------------------  # Date: Thu, 29 Nov 2001 15:06:42 GMT & From: "Jeff Killeen" <Jeff@IDM-IO.com>D Subject: Re: "Why Great Companies Fail": Includes DEC (perhaps HP ?)8 Message-ID: <6asN7.1191$h24.103758@typhoon1.gnilink.net>  H Leaving aside fraud great companies almost always fail for one reason...  ;     A fundamental paradigm shift occurs in the market which @     the company fails to detect or fails to react to.  In effect:     the business that made the company great first becomes5     ineffective and then finally becomes a liability.   K The classic example is the USA railroads who failed to realize they were in K the shipping business and didn't react to the paradigm shift in the way the  country was shipping goods.   I I have seen posted here that Digital for example stopped listening to its J customers.  In fact like most companies that go through this Digital spentK too much time listening to their _current_ customers and in effect inhaling I Digital's own exhaust the current customers were feeding back to Digital. H What Digital failed to do is listen to its potential customers.  This isL classic behavior for a company caught by a hard shift in the market.  When aI company inhales its own exhaust it spends its time trying to make the old F business model work rather than abandoning it and finding a new viableJ business model.  That failure to let go of the current business model that' is failing is why great companies fail.   L Sun is an example of a company that succeeded making the shift when it movedL from workstations to servers.  It saw itself in the Unix market and followed Unix.   J Digital failed to realize that its business was a combination of being theL price/technology/service leader for ISV's producing applications that ran onK departmental/workgroup class systems.  When Digital's business model became J broken on the price side rather than fixing the model Digital attempted toL retool the model to work for enterprise customers - in other words chasing aG market that would support its growing product margin requirements.  The K effect was that ISV's could no longer compete with Digital systems and left H Digital in droves.  That failure to understand the market Digital was inE combined with the failure to fix the Digital business model when that G technology paradigm shift occurred in the market is why Digital failed.   D The fundamental mistake in Digital's strategy that resulted from notH understanding its market is not realizing that operating systems are notJ just a series of technical tradeoffs.  That operating systems are a series5 of price, time to market, and technology tradeoffs...    ------------------------------  # Date: Thu, 29 Nov 2001 15:07:14 GMT & From: "Jeff Killeen" <Jeff@IDM-IO.com>D Subject: Re: "Why Great Companies Fail": Includes DEC (perhaps HP ?)8 Message-ID: <CasN7.1192$h24.103018@typhoon1.gnilink.net>  H Leaving aside fraud great companies almost always fail for one reason...  ;     A fundamental paradigm shift occurs in the market which @     the company fails to detect or fails to react to.  In effect:     the business that made the company great first becomes5     ineffective and then finally becomes a liability.   K The classic example is the USA railroads who failed to realize they were in K the shipping business and didn't react to the paradigm shift in the way the  country was shipping goods.   I I have seen posted here that Digital for example stopped listening to its J customers.  In fact like most companies that go through this Digital spentK too much time listening to their _current_ customers and in effect inhaling I Digital's own exhaust the current customers were feeding back to Digital. H What Digital failed to do is listen to its potential customers.  This isL classic behavior for a company caught by a hard shift in the market.  When aI company inhales its own exhaust it spends its time trying to make the old F business model work rather than abandoning it and finding a new viableJ business model.  That failure to let go of the current business model that' is failing is why great companies fail.   L Sun is an example of a company that succeeded making the shift when it movedL from workstations to servers.  It saw itself in the Unix market and followed Unix.   J Digital failed to realize that its business was a combination of being theL price/technology/service leader for ISV's producing applications that ran onK departmental/workgroup class systems.  When Digital's business model became J broken on the price side rather than fixing the model Digital attempted toL retool the model to work for enterprise customers - in other words chasing aG market that would support its growing product margin requirements.  The K effect was that ISV's could no longer compete with Digital systems and left H Digital in droves.  That failure to understand the market Digital was inE combined with the failure to fix the Digital business model when that G technology paradigm shift occurred in the market is why Digital failed.   D The fundamental mistake in Digital's strategy that resulted from notH understanding its market is not realizing that operating systems are notJ just a series of technical tradeoffs.  That operating systems are a series5 of price, time to market, and technology tradeoffs...    ------------------------------  % Date: Thu, 29 Nov 2001 08:07:52 -0500 0 From: "Mike Zarudzki" <mike.zarudzki@compaq.com>3 Subject: Re: Andrew Harrison : Alive & Still At SUN 3 Message-ID: <KqqN7.2096$RL6.63115@news.cpqcorp.net>   ' "doesn't have access to a NNTP server."  New cost savings approach.  7 All opinions matter, even if they are "skewed". '>) '>)  -Mike Z.  6 "Jerry Leslie" <leslie@clio.rice.edu> wrote in message! news:9u3bgu$9v7$4@joe.rice.edu... E > I got a reply from Andrew from an email to Andrew.Harrison@Sun.COM. A > He's switched offices and doesn't have access to a NNTP server.  > ? > He was surprised that Fred Kleinsorge really asked about him.  >  > --Jerry Leslie   ------------------------------  % Date: Thu, 29 Nov 2001 10:44:56 -0500 5 From: "Fred Kleinsorge" <kleinsorge@star.zko.dec.com> 3 Subject: Re: Andrew Harrison : Alive & Still At SUN 3 Message-ID: <GMsN7.2108$RL6.63209@news.cpqcorp.net>    > 7 >"Jerry Leslie" <leslie@clio.rice.edu> wrote in message " >news:9u3bgu$9v7$4@joe.rice.edu...F >> I got a reply from Andrew from an email to Andrew.Harrison@Sun.COM.B >> He's switched offices and doesn't have access to a NNTP server. >>@ >> He was surprised that Fred Kleinsorge really asked about him. >> >   H While I can't say that I miss his "input" in this and other forums, I amI glad to hear that there is nothing wrong.  Looks like a win-win situation  ;-)    _Fred    ------------------------------  % Date: Thu, 29 Nov 2001 13:44:52 -0500 ; From: "Brian Tillman" <tillman_brian@notnoone.notnohow.com> / Subject: Re: Can We Turn Down the Flames a Bit? $ Message-ID: <3c06822b$1@news.si.com>  % >Hell, Indiana froze over circa 1954.   " Hell, Michigan freezes every year. --  A Brian Tillman                   Internet: tillman_brian at si.com A Smiths Aerospace                          tillman at swdev.si.com = 3290 Patterson Ave. SE, MS      Addresses modified to prevent < Grand Rapids, MI 49512-1991     SPAM.  Replace "at" with "@"8        This opinion doesn't represent that of my company   ------------------------------    Date: 29 Nov 2001 05:28:43 -0800) From: P.Young@unsw.EDU.AU (Patrick Young) ' Subject: Re: Careful with Mozilla 0.9.6 = Message-ID: <55f85d77.0111290528.14327848@posting.google.com>   [ John Reagan <john.reagan@compaq.com> wrote in message news:<3C04F75D.8030203@compaq.com>...  > Vance R. Haemmerle wrote:  > J > >   I just wanted to warn people that if they install Mozilla 0.9.6 theyG > > might want to copy their bookmarks.html file to a safe place first. L > > Mozilla ate mine and I had to get mine from a backup.  There's also some8 > > other strange problems, like window focus problems.  > >  > K > That is odd.  I updated from 0.9.3 to 0.9.6 and didn't have any problems  N > with my bookmarks.  I also haven't seen any problems with window focus, etc.  G Works OK for me too - bookmarks are OK and nothing seems to have broken G since the previous release. Looking good! - as someone who does not use I a PC this is my main browser (actually has been for a while). Using 0.9.6 K here at home (to type this now) under OpenVMS 7.3 and at work under OpenVMS  7.2-1.  E The annoying problem that appeared in 0.9.5 where you could not "back 8 button" from the output of a CGI program has been fixed.  B www.westpac.COM.AU still (as always, and also under other Mozilla @ supported platforms) has a broken main menu so I have to use the> specific URL for my online banking. Netscape 3.03 displays the menu.   I The file download name problem (where file names with multiple dots, etc, A are not saved preserving the name to ODS-5 disks) is still there.    ------------------------------    Date: 29 Nov 2001 04:07:11 -0800) From: P.Young@unsw.EDU.AU (Patrick Young) B Subject: Re: Compaqs VMS plans for IPF port ... any doubters left?= Message-ID: <55f85d77.0111290407.74521ea7@posting.google.com>   1 "David J. Dachtera" <djesys.nospam@fsi.net> wrote  > E > Isn't that *EXACTLY* what I've been doing since I got laid off from + > Advocate almost *TWO* *YEARS* *AGO*???!!!   @ Yup, I did notice. With all due respect you are not the OpenVMS B development team to whose _lack of_ "spit the dummy" conduct I did originally refer to.   > I > *WE* - You, me and many of the others here - *ARE* in their place, just  > somewhat less directly.  >   D Hoff's reply in this thread was interesting and very well put. UnderE the "new toy" perspective the paranoid side of me took it badly (well C this part opens and shuts, it smells a bit like a computer, not the @ way we would have done it, parts of this project are going to be) "interesting" in the worst possible way).   B Due to M$ and "Noname PC brand X, Y and Z Pty. Ltd." I don't thinkB Compaq (or HP) are in a financial position to throw a _great deal_@ of money away on a "grand conspiracy" to keep existing customersB happy. Especially since those existing customers in most cases are. not of the "average M$ PC consumer" mentality.  C In addition, Compaq were not listed as a government department last  time I checked :-)   E > Time to face facts, man - once they find a way to weasel out of the J > gov't contracts (probably just lie their way out, like they've done with6 > all their other "commitments"), VMS is history, man!  H I don't like to be that negative. Yes certainly the current situation is% a worry. Anything non M$ is a worry.     H > I'm training to teach Solaris Administration, and come Springtime, I'mI > looking at training to be an Oracle DBA. I still have 18 years before I D > can even think about retiring, I'm dead broke (but my net worth is  G Solaris is the UNIX for people who need tech support and vendor back up F and are willing to pay for it. I can't help but wonder that going downE the Solaris track will in the not so long term end you up in the same 2 "fast shrinking" position OpenVMS was/is enjoying.  E LINUX does more and is free. Get any number of well established LINUX + support companies out there and who knows??    >  > Get real, man!  E No, I prefer not to and live in a hazy .EDU fuzzy type place - you do A ("actually did" before M$) gain a huge range of skills in such an C environment. Gee - my *STRONG* dislike of M$ and partial dislike of   Solaris came from such a place.   D PS: on the Paul verses Patrick thing.... no worries... any number ofA people over the years have called me Paul. At first I suspected I A looked like someone they knew called Paul. I understand the first B two characters are the same, have not often been called Peter, but still I really don't get it -:)i   ------------------------------  # Date: Thu, 29 Nov 2001 17:22:52 GMTn2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman) Subject: Re: Complie Error3 Message-ID: <M9uN7.2113$RL6.63261@news.cpqcorp.net>o  K In article <9u46ec$kud$1@news.nuri.net>, "ȼ" <syahn@icols.com> writes:  :when C lib link timer :Occured following error.  :I don't know why....:  E   You will want to read the OpenVMS Frequently Asked Questions (FAQ).nE   Specifically, please review the section of the OpenVMS FAQ entitledBC   "How to resolve LINK-I-DATMISCH errors?".  You can find a copy oftB   the FAQ at the URL listed below -- I would encourage downloadingC   the current edition of the the text-format FAQ, and searching it.y   :Please E-mail and Answer....f     Ask here, get an answer here.l   ..5 :$ LINK FBI_SST, GR2D, KEYINPUT, SST_ENDCUT, SST_SUB,a% :STR_LST,STR_SHEET,STR_TOKEN, FB_DLG,o+ :SYS$COMMON:[000000.SYSLIB]DECWINDOWS/LIBRA  :RY I :%LINK-I-DATMISMCH, creation date of 21-JAN-1998 11:08 in shareable imageu- :SYS$COMMON:[000000.SYSLIB]DECW$XLIBSHR.EXE;2oJ :        differs from date of 21-MAR-1996 21:41 in shareable image library+ :SYS$COMMON:[000000.SYSLIB]DECWINDOWS.OLB;1e :$ SET NOVER    N  ---------------------------- #include <rtfaq.h> -----------------------------N       For additional, please see the OpenVMS FAQ -- www.openvms.compaq.com    N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------  % Date: Thu, 29 Nov 2001 17:19:12 +0100D, From: Didier Morandi <Didier.Morandi@gmx.ch>. Subject: DCL minute of the day: PCSI_MENU v1.1& Message-ID: <3C065FFF.1F4A9071@gmx.ch>   $!+! $! PCSI_MENU.COM $!J $! This procedure will build a PCSI kit from answers to questions during aM $! menu driven session. On-line help will also provide some training on PCSI.  $!H $! The PCSI Developper's Guide is available from the COMPAQ WEB site at:J $! www.openvms.compaq.com:8000/72final/5952/5952pro.html (for OpenVMS 7.2) $! ortJ $! www.openvms.compaq.com:8000/73final/5952/5952pro.html (for OpenVMS 7.3) $!G $! There is also an ADOBE PDF version available from the following URL:AL $! www.openvms.compaq.com:8000/73final/documentation/pdf/OVMS_73_PCSI_GD.pdf $!
 $! <input> $! First name, last name( $! help text option (Verbose/Brief/None) $! favourite editore $! product name, version, $! automatic IVP proc YES/NO		(next release)/ $! configuration template YES/NO	(next release)  $! pre-conf proc YES/NOo0 $! post-install proc YES/NO	      (next release) $! $! <output>c $! pre-conf procedurel $! <product>.pcsi$desc filee $! <product>.pcsi$text filek* $! post-install procedure			(next release) $! IVP						(next release)" $! release notes				(next release)' $! installation guide				(next release)n$ $! users guide (?)				(next release) $! PCSI kit file $! $! <side effects> M $! None. But note that an incomplete kit file will be deleted when restarting6; $! the same kit build if the version number did not change.a $!M $! LEGAL: This procedure is NothingWare and may be copied, given and modified.H $!        as wished, but without any support whatsoever from the author.N $!        To know more on PCSI, visit http://groups.yahoo.com/group/vmsinstal.G $!	    To know more on the author, visit http://Didier.Morandi.Free.fr.w $! $! Revision historyl $!( $! Version Date        Author     actionO $! ------- ----------- ---------- ---------------------------------------------q* $! v1.0-0  16-may-2001 D. Morandi creationP $v="1.1-0"!29-nov-2001 DMo	       add list of resulting files at end of process.5 $!				            fix bug in default directory values  $!-2 $ on error then goto EXITe $ set control=(T,Y)r $ on control then goto EXITl $ set symbol/scope=noglobal  $ say = "write sys$output"
 $ bell[0,8]=7e
 $ esc[0,8]=27r $ b1 = esc + "[1m" $ b0 = esc + "[m"] $ r1 = esc + "[7m"	 $ r0 = b0  $ bold = "write sys$output b1" $ rev  = "write sys$output r1" $ off  = "write sys$output r0" $ help_on = "call give_help"( $ explain = "call give_help HELP_SETUP " $ help_setup = "VERBOSE"" $ user := 'f$getjpi(0,"username")'& $ node  =  f$trnlnm("sys$node") - "::"1 $ debug =  (f$trnlnm("pcsi_menu$debug") .nes. "")n $ if p4 .nes. "" then debug = 1e $! $ first=f$trnlnm("first$name") $ default=""- $ if first .nes. "" then default = "''first'"-J $ if p1 .eqs. "" then inq p1 "* Please enter your first name [''default']"4 $ if p1 .eqs. "" .and. first .eqs. "" then goto EXIT $ if p1 .nes. "" $ then3 $    first = f$edit(f$extract(0,1,p1),"upcase") + - 3              f$edit(f$extract(1,99,p1),"lowercase")a' $    define/nolog first$name "''first'" C $    if debug then say "(first name default value set to ''first')"d $ endiff $! $ last=f$trnlnm("last$name") $ default=""+ $ if last .nes. "" then default = "''last'" I $ if p2 .eqs. "" then inq p2 "* Please enter your last name [''default']"s3 $ if p2 .eqs. "" .and. last .eqs. "" then goto EXIT  $ if p2 .nes. "" $ then $    last = f$edit(p2,"upcase")i" $    define/nolog last$name 'last'A $    if debug then say "(last name default value set to ''last')"o $ endife $!$ $ producer=f$trnlnm("producer$name") $ default=""3 $ if producer .nes. "" then default = "''producer'"n $ASK_PRODUCER: $ if p3 .eqs. "" then inq p3 -9          "* Enter Company three digits code [''default']"f7 $ if p3 .eqs. "" .and. producer .eqs. "" then goto EXITs $ if p3 .nes. "" $ then $    if f$len(p3) .ne. 3	 $    thena $	type sys$input $DECK   K Please enter a three digits code to identify the producer of this PCSI kit.e  J This information is mandatory as this code wil be part of different stringK    identifiers required to build the kit then install/remove your software.s  G See section 5952PRO_001.HTML#SUBFIELD_PDF_FORMAT_KIT in the HTML doc atmM    http://www.openvms.compaq.com:8000/72final/5952/5952pro_001.html for more.n  J Should you have a direct access to the Internet, select the above URL thenM    cut/paste it into your favourite WEB browser, then select the whole stringhM    starting with the # and cut/paste/append it to reach the section directly.l   $EOD $	p3=""s $	goto ASK_PRODUCER'
 $    endif# $    producer = f$edit(p3,"upcase")e* $    define/nolog producer$name 'producer'I $    if debug then say "(producer name default value set to ''producer')"c $ endif  $! $ type/page nl:p $ say ""I $ say "Hello ",first,", Welcome to PCSI Menu, version ",v," on ",f$time()n $ say "" $ editor_cmd = "edit/edt"l& $ edtini_file = "SYS$LOGIN:EDTINI.EDT"* $ if f$search(edtini_file) .nes. "" then -3      editor_cmd = "edit/edt/command=''edtini_file'"e $! $ASK_CUST_MENU:oD $ inq ask "* Do you wish to access the customization menu (Y/[N]/?)" $ ask = f$edit(ask,"upcase") $ if ask .eqs. "?" $ then $    help_on CUST_MENU $    goto ASK_CUST_MENUn $ endife" $ h_setup = f$trnlnm("help$setup")O $ if h_setup .eqs. ""  then define/nolog help$setup "V"	    !Verbose by default 2 $ if h_setup .eqs. "V" then help_setup = "VERBOSE"0 $ if h_setup .eqs. "B" then help_setup = "BRIEF"/ $ if h_setup .eqs. "N" then help_setup = "NONE"  $ say ""- $ say "(current HELP mode is ",help_setup,")"r $ say ""; $ if f$extr(0,1,ask) .nes. "Y" then goto SKIP_CUSTOMIZATION 2 $ if debug then say "[Start customization phase ]"2 $ if debug then say "[1. HELP mode preference   ]"" $ if h_setup .eqs. "N" then say "" $! $ explain HELP_MODEt $! $ASK_HELP_MODE: ? $ inq ask "* Select Verbose, Brief or No help mode (V/[B]/N/?)"t $ ask =f$edit(ask,"upcase")  $ if ask .eqs. "?" $ then $    help_on HELP_MODE $    goto ASK_HELP_MODEs $ endife  $ if ask .eqs. "" then ask = "B"L $ if ask.nes."V" .and. ask.nes."B" .and. ask.nes."N" then goto ASK_HELP_MODE $ define/nolog help$setup 'ask's $ h_setup = askh $ say ""2 $ if debug then say "[2. Editor preference      ]"" $ if h_setup .eqs. "N" then say "" $! $ explain DEFAULT_EDITOR $! $ASK_DEFAULT_EDITOR:@ $ inq ask "* Select your favourite editor, EDT or TPU ([E]/T/?)" $ if ask .eqs. "?" $ then $    help_on DEFAULT_EDITOR  $    goto ASK_DEFAULT_EDITOR $ endifo $ say ""+ $ ask = f$extract(0,1,f$edit(ask,"upcase"))r $ if ask .eqs. "T" $ then $    editor_cmd = "edit/tpu", $    say "(default editor is         : TPU)"- $    section = "SYS$LOGIN:TPUINI.TPU$SECTION"r" $    if f$search(section) .nes. ""	 $    theni2 $       editor_cmd = "edit/tpu/section=''section'"6 $       say "(default TPU section is    : ''section')"
 $    endif $ else, $    say "(default editor is         : EDT)") $    edtini_file = "SYS$LOGIN:EDTINI.EDT"i& $    if f$search(edtini_file) .nes. ""	 $    thenn6 $       editor_cmd = "edit/edt/command=''edtini_file'": $       say "(default EDTINI file is    : ''edtini_file')"
 $    endif $ endif'2 $ if debug then say "[end of customization phase]"" $ if h_setup .eqs. "N" then say "" $! $SKIP_CUSTOMIZATION: $! $ explain PRODUCT_NAMES $!www.openvms.compaq.com:8000/72final/5952/5952pro_001.html#subfield_pdf_format_kitr $! $ASK_PRODUCT_NAME:4 $ prod_name = f$edit(f$trnlnm("prod$name"),"upcase") $ default=""5 $ if prod_name .nes. "" then default = "''prod_name'"l $ read/end=EXIT -D@       /prompt="* Enter product name (9c. maxi) [''default']: " -        sys$command ask $ ask =f$edit(ask,"upcase")a $ if ask .eqs. "?" $ then n $    help_on PRODUCT_NAMEo $    goto ASK_PRODUCT_NAME $ else- $    if ask .eqs. "" .and. prod_name .eqs. ""a	 $    thennA $       say bell,b1,"You must enter a name here or ^Z to exit",b0t $       goto ASK_PRODUCT_NAMEg
 $    endif $ endift $ if ask .nes. ""c $ then $    prod_name = ask' $    define/nolog prod$name 'prod_name'aI $    if debug then say "(product name default value set to ''prod_name')"e $ endif  $! $ explain PRODUCT_VERSION I $! www.openvms.compaq.com:8000/72final/5952/5952pro_001.html#tmmnn_formatA $! $ASK_PRODUCT_VERSION:g! $ prod_ver = f$trnlnm("prod$ver")q $ default=""3 $ if prod_ver .nes. "" then default = "''prod_ver'"vC $ inq ask "* Enter version number (format xx.yy-zzzz) [''default']"  $ ask =f$edit(ask,"upcase")i $ if ask .eqs. "?" $ then t $    help_on PRODUCT_VERSION $    goto ASK_PRODUCT_VERSIONe $ else, $    if ask .eqs. "" .and. prod_ver .eqs. ""	 $    thens= $       say bell,b1,"You must enter a version number here",b0   $       goto ASK_PRODUCT_VERSION
 $    endif $ endifa $ if prod_ver .eqs. "" $ then $    prod_ver = ask-% $    define/nolog prod$ver 'prod_ver'1K $    if debug then say "(product version default value set to ''prod_ver')"c $ endifd $! $ explain MATERIAL_LOCL $! www.openvms.compaq.com:8000/72final/5952/5952pro_005.html#package_kit_nowM $! (the /MATERIAL value for the PACKAGE command, where the appli. is located)d $! $ASK_MATERIAL_LOC:" $ source_kit = f$trnlnm("src$kit") $ default=""7 $ if source_kit .nes. "" then default = "''source_kit'"  $ read/end=EXIT ->>       /prompt="* Enter kit material location [''default']: " -        sys$command ask $ ask =f$edit(ask,"upcase")n $ if ask .eqs. "?" $ then   $    help_on MATERIAL_LOCn $    goto ASK_MATERIAL_LOC $ else. $    if ask .eqs. "" .and. source_kit .eqs. ""	 $    then-E $       say bell,b1,"You must enter a location here or ^Z to exit",b0m $       goto ASK_MATERIAL_LOC 
 $    endif $ endif  $ if ask .eqs. ""s $ then $    source_kit = defaulte $ else $    source_kit = asko $ endifi8 $ source_kit = source_kit - ":*.*" - "*.*"			!in case...# $ define/nolog src$kit 'source_kit'tE $ if debug then say "(source kit default value set to ''source_kit')"iA $ logname_value = f$trnlnm("source_kit")		!in case it's a lognameo; $ if logname_value .nes. "" then source_kit = logname_valuep $ root_kit = source_kit > $ if f$locate("...]",source_kit) .ne. f$len(source_kit) then -*      root_kit  = root_kit  - "..." - "*.*"* $ if f$search("''source_kit'*.*") .eqs. "" $ then7 $    say bell,b1,"There are no files in ",source_kit,b0D $    goto ASK_MATERIAL_LOC $ endifn $! $ explain DEST_KIT_LOCI $! www.openvms.compaq.com:8000/72final/6258/6048pro_054.html#index_x_1662o3 $! (the /DESTINATION value for the PACKAGE command)t $! $ASK_DEST_KIT_LOC:! $ dest_kit = f$trnlnm("dest$kit")h $ default="" $ if dest_kit .nes. "" $ then $    default = "''dest_kit'" $ else $    default = "SYS$LOGIN" $ endiff9 $ inq ask "* Enter destination for kit file [''default']"s $ ask =f$edit(ask,"upcase")  $ if ask .eqs. "?" $ then   $    help_on DEST_KIT_LOC  $    goto ASK_DEST_KIT_LOC $ endifh $ if ask .nes. ""l $ then $    dest_kit = askw $ else $    dest_kit = defaultK $ endifo" $ define/nolog dest$kit 'dest_kit'J $ if debug then say "(dest kit location default value set to ''dest_kit')" $ define sys$output nl:i $ define sys$error  nl:u $ dir dest$kit $ status = $status $ deassign sys$outputt $ deassign sys$error= $ if status .eqs. "%X1001C04A" then create/dir/log 'dest_kit'o $ if status .eqs. "X00038030"r $ then $    say f$mess(status)r $    goto ASK_DEST_KIT_LOC $ endife $ set on $! $ explain PCSI_SRC_LOCN $! www.openvms.compaq.com:8000/72final/5952/5952pro_005.html#package_files_nowM $! (the /SOURCE value for the PACKAGE command, where the PDF/PTF are located)i $! $ASK_PCSI_SRC_LOC:) $ pcsi_src_dir = f$trnlnm("pcsi_src$dir")m $ default="" $ if pcsi_src_dir .nes. "" $ then  $    default = "''pcsi_src_dir'" $ else* $    default = "''dest_kit'"					!why not? $ endifv9 $ inq ask "* Enter PCSI material directory  [''default']"u $ if ask .eqs. "?" $ then S $    help_on PCSI_SRC_LOC  $    goto ASK_PCSI_SRC_LOC $ endif. $ if ask .nes. ""@ $ then $    pcsi_src_dir = ask  $ else $    pcsi_src_dir = dest_kit $ endif * $ define/nolog pcsi_src$dir 'pcsi_src_dir'P $ if debug then say "(source kit location default value set to ''pcsi_src_dir')" $! $ explain PRE_CONF_PROCcE $! www.openvms.compaq.com:8000/72final/5952/5952pro_007.html#exec_preh $! $ASK_PRE_CONF_PROC:r= $ inq ask "* Do you want a pre-installation procedure? Y/[N]"r $ if ask .eqs. "?" $ then , $    help_on PRE_CONF_PROC $    goto ASK_PRE_CONF_PROC  $ endifa $ ask = f$edit(ask,"upcase")! $ there_is_a_pre_install_proc = 0t  $ if ask .eqs. "" then ask = "N" $ if f$extr(0,1,ask) .eqs. "Y" $ then$ $    there_is_a_pre_install_proc = 1N $    inq ask "* Enter filename specification [''prod_name'$PRE_CONFIGURE.COM]" $    if ask .nes. ""	 $    then)" $	pre_install_file = f$search(ask) $	if pre_install_file .eqs. "" $	then# $	   say b1,ask," was not found",b0i $	   goto ASK_PRE_CONF_PROCf $	endifa	 $    elsen5 $	pre_install_file = "''prod_name'$PRE_CONFIGURE.COM"a $	boldE $	say "  Entering editing session for the PCSI pre-installation file" P $  say "  with a default name of ",pre_install_file,". Please do not change it."O $  say "  You can enter now any DCL commands or include any procedure in here."w $	say ""> $	say "  Note: This procedure will NOT be parsed by PCSI  and"@ $	say "        elements created here cannot be removed by PCSI." $	offo% $	inq dummy "Press <RETURN> to start"aG $	if f$search(pre_install_file) .eqs. "" then gosub CREATE_PRECONF_FILE # $	assign/user sys$command sys$inputc! $	'editor_cmd' 'pre_install_file'S
 $    endif $ endifkP $! Now we create an empty PTF (not necessarily required but useful for training)% $ ptf_name = "''prod_name'.PCSI$TEXT"t $! $!!! explain PTFC $! www.openvms.compaq.com:8000/72final/5952/5952pro_004.html#ptf_ch  $! $ if debug then say ""? $ if debug then say "[creating Product Text File ",ptf_name,"]"i $ close/nolog ptfn $ open/write ptf 'ptf_name'  $! >>> start update PTF <<<p $ gosub CREATE_EMPTY_PTF $! >>> end   update PTF <<<G $!/ $! Now we create the PDF (fasten you seat belt)p% $ pdf_name = "''prod_name'.PCSI$DESC"  $! $!!! explain PDFE $! www.openvms.compaq.com:8000/72final/5952/5952pro_002.html#spia_def" $! $ if debug then say ""F $ if debug then say "[creating Product Description File ",pdf_name,"]" $ close/nolog pdfi $ open/write pdf 'pdf_name'f $! >>> start update PDF <<<t $ gosub BUILD_HEADER $! >>> end   update PDF <<<n $ if debug then say ""0 $ if debug then say "[version checking section]" $! $ explain VMS_VER_CHKyJ $! www.openvms.compaq.com:8000/72final/5952/5952pro.html#pdf_file_example  $! $ASK_VMS_VER_CHK:O; $ inq ask "* Product needs minimum OpenVMS version? [Y]/N "  $ if ask .eqs. "?" $ then a $    help_on VMS_VER_CHK $    goto ASK_VMS_VER_CHK  $ endif" $ ask = f$edit(ask,"upcase")0 $ if ask .eqs. "" .or. f$extr(0,1,ask) .eqs. "Y" $ then $    explain MIN_VMS_VERK $!   www.openvms.compaq.com:8000/72final/5952/5952pro_001.html#tmmnn_formatn $! $ASK_MIN_VMS_VER:l9 $    inq ask "* Enter minimum OpenVMS version [V7.2-1H1]"  $    if ask .eqs. "?"s
 $    then  $       help_on MIN_VMS_VER$ $      goto ASK_MIN_VMS_VER$
 $    endif $    min_vms_version = askB $    if min_vms_version .eqs. "" then min_vms_version = "V7.2-1H1" $! >>> start update PDF <<<  $    if debug then say ""d% $    if debug then say "[update PDF]"  $    write pdf -
      "--!" $    write pdf -H      "if (<software DEC AXPVMS VMS version below ",min_vms_version,">);" $    write pdf -&      "    error UNSUPP_VMS_VER abort;" $    write pdf -      "end if;" $! >>> end   update PDF <<<d $!+t- $! Now we create the matching text in the PTFf $!-_ $! >>> start update PTF <<<"% $    if debug then say "[update PTF]"  $    write ptf -      "1 UNSUPP_VMS_VER"[ $    write ptf -<      "=prompt OpenVMS ",min_vms_version," minimum required." $    write ptf -N      " The version of ",prod_name," you are installing runs only with OpenVMS" $    write ptf -.      " version ",min_vms_version," or higher." $    write ptf -      " " $    write ptf -J      " Please check whether you are installing this software on the right" $    write ptf -P   " system, or do upgrade this system to OpenVMS ",min_vms_version," or higher." $! >>> end   update PTF <<<  $ endif_ $!+)H $! If there is a pre-installation procedure, let's declare it in the PDF $!-f  $ if there_is_a_pre_install_proc $ then $! >>> start update PDF <<< % $    if debug then say "[update PDF]"a $    write pdf -
      "--!" $    write pdf -G      "execute preconfigure ""@PCSI$SOURCE:[TEMP]",pre_install_file,""""f $    write pdf -      "        interactive" $    write pdf -/      "        uses [TEMP]",pre_install_file,";"_ $! >>> end   update PDF <<<  $ endife $ if debug then say ""+ $ if debug then say "[identifiers section]"m $! n $ explain IDENTIFIERSrB $! www.openvms.compaq.com:8000/72final/5952/5952pro_011.html#right $! _ $ASK_IDENTIFIERS:n= $ inq ask "* Do you want to create rights identifiers? Y/[N]"  $ if ask .eqs. "?" $ then u $    help_on IDENTIFIERS $    goto ASK_IDENTIFIERSi $ endif  $ ask = f$edit(ask,"upcase")  $ if ask .eqs. "" then ask = "N" $ if f$extr(0,1,ask) .eqs. "Y" $ then $ANOTHER_ID:6 $    inq ask "* Enter identifier name (<RETURN>=done)" $    if ask .nes. ""	 $    theno" $	rights_id = f$edit(ask,"upcase") $	inq ask -fD 	   "* Do you want the RESOURCE attribute for ''rights_id'? (Y/[N])" $	ask = f$edit(ask,"upcase")  $	if ask .eqs. "" then ask = "N" $	attr_string = ""S $	if f$extr(0,1,ask) .eqs. "Y" then attr_string = " with ""/attributes=resource"";"R $! >>> start update PDF <<<I" $	if debug then say "[update PDF]"
 $	write pdf -i  	"--!""
 $	write pdf - + 	"rights identifier ",rights_id,attr_stringh $! >>> end   update PDF <<<  $	goto ANOTHER_ID_
 $    endif $ endifn $ if debug then say ""+ $ if debug then say "[directories section]"  $! s $ explain DIRECTORIESiA $! www.openvms.compaq.com:8000/72final/5952/5952pro_006.html#direi $! $ASK_DIRECTORIES:e= $ inq ask "* Do you want to create directories?        Y/[N]"' $ if ask .eqs. "?" $ then u $    help_on DIRECTORIES $    goto ASK_DIRECTORIES  $ endifa $ ask = f$edit(ask,"upcase")  $ if ask .eqs. "" then ask = "N" $ if f$extr(0,1,ask) .eqs. "Y" $ then
 $ANOTHER_DIR:e> $    inq ask "* Enter directory specification (<RETURN>=done)" $    if ask .nes. ""	 $    then"- $	dir_spec = f$edit(ask,"upcase") - "[" - "]"  $! >>> start update PDF <<< " $	if debug then say "[update PDF]"
 $	write pdf -   	"--!"L
 $	write pdf -v6         "directory [",dir_spec,"] protection private;" $! >>> end   update PDF <<<v $	goto ANOTHER_DIR
 $    endif $ endifD $ if debug then say ""( $ if debug then say "[accounts section]" $! s $ explain ACCOUNTS@ $! www.openvms.compaq.com:8000/72final/5952/5952pro_006.html#acc $! $ASK_ACCOUNTS:= $ inq ask "* Do you want to create user accounts?      Y/[N]"e $ if ask .eqs. "?" $ then . $    help_on ACCOUNTS  $    goto ASK_ACCOUNTS $ endif  $ ask = f$edit(ask,"upcase")  $ if ask .eqs. "" then ask = "N" $ if f$extr(0,1,ask) .eqs. "Y" $ then $ANOTHER_ACCOUNT: 3 $    inq ask "* Enter account name (<RETURN>=done)"  $    if ask .nes. ""	 $    then ! $	username = f$edit(ask,"upcase")o	 $ASK_UIC: = $	inq uic_value "* Enter UIC for ''username' (ggg,mmm octal)" ) $	if uic_value .eqs. "" then goto ASK_UIC / $	uic_value = uic_value - "(" - ")" - "[" - "]"g# $	uic_value = "[" + uic_value + "]"3 $ASK_DEVICE:' $	inq dev_name "* Enter default device"_+ $	if dev_name .eqs. "" then goto ASK_DEVICEi! $	dev_name = dev_name - ":" - "_"a $	dev_name = dev_name + ":"r	 $ASK_DIR:p* $	inq dir_name "* Enter default directory"( $	if dir_name .eqs. "" then goto ASK_DIR! $	dir_name = dir_name - "[" - "]" ! $	dir_name = "[" + dir_name + "]"$ $ASK_ID:% $	inq identifier "* Enter identifier" ) $	if identifier .eqs. "" then goto ASK_ID  $!& $	inq pwd "* Enter password [WELCOME]"& $	if pwd .eqs. "" then pwd = "WELCOME" $!4 $	inq pwd_life "* Enter password lifetime [30 days]", $	if pwd_life .eqs. "" then pwd_life = "30-") $	say "(password is created pre-expired)"" $	say "" $! $! >>> start update PDF <<<e" $	if debug then say "[update PDF]"
 $	write pdf -n  	"--!"_
 $	write pdf -r 	"account ",username," with ("
 $	write pdf -oE 	"""/uic=",uic_value,"/device=",dev_name,"/directory=",dir_name,"""," 
 $	write pdf -c 	"""",identifier,""","
 $	write pdf -.: 	"""/account=",username,"/pwdlife=""""",pwd_life,""""""","
 $	write pdf -s% 	"""/password=",pwd,"/pwdexpired"");". $! >>> end   update PDF <<<_ 	goto ANOTHER_ACCOUNTg
 $    endif $ endif  $ type sys$input $DECKt  0 Now building a list of your application files...B (this may take a while if you have many files in your application)   $EOD< $ files_list = "sys$scratch:dir_''f$getjpi(0,""pid"")'.temp"C $ dir/col=1/notrail/nohead/excl=*.dir/out='files_list' src$kit:*.*;s $ gosub GET_NR_OF_RECORDSt4 $ say nr_of_rec," files found for your application." $ say "" $ close/nolog tmp_file! $ open/read tmp_file 'files_list'O" $ if debug then say "[update PDF]"
 $ write pdf -=   "--!"n $LOOP:& $ read/end=NO_MORE_FILES tmp_file line$ $ type      = f$parse(line,,,"type")B $ if type .eqs. ".PCSI"      then goto LOOP	!no need for this one,< $ if type .eqs. ".PCSI$DESC" then goto LOOP	!   or this one,B $ if type .eqs. ".PCSI$TEXT" then goto LOOP	!   or this other one.& $ device    = f$parse(line,,,"device")' $ version   = f$parse(line,,,"version")t% $ file_spec = line - device - versionI $! >>> start update PDF <<<s! $ write pdf "file ",file_spec,";"( $! >>> end   update PDF <<<e $ goto LOOPE $! $NO_MORE_FILES:i4 $ if debug then say "[end of building list process]" $ close tmp_file $ delete 'files_list';*w $! $! >>> start update PDF <<<a" $ if debug then say "[update PDF]"
 $ write pdf -=   "--!"n
 $ write pdf -t  "end product;"n
 $ write pdf -e   "--! End of ",pdf_name," file." $! >>> end   update PDF <<<  $ close pdfe $ close ptff
 $ say "Done."  $ copy = "copy"t $ delete = "delete",
 $ qualif = ""o $ if debug then qualif="/log"_/ $ if debug then say "[copy PDF to ''dest_kit']"A# $ copy'qualif' 'pdf_name' dest$kit:  $ status = $status+ $ if status then delete'qualif' 'pdf_name';_/ $ if debug then say "[copy PTF to ''dest_kit']"n$ $ copy'qualif' 'ptf_name' dest$kit:  $ status = $status+ $ if status then delete'qualif' 'ptf_name';"  $ if there_is_a_pre_install_proc $ thenD $    if debug then say "[copy pre installation file to ''dest_kit']". $    copy'qualif' 'pre_install_file' dest$kit: $    status = $statuse6 $    if status then delete'qualif' 'pre_install_file'; $ endif  $ if debug then say ""L $ say "The following elements required to build your kit have been created." $ say ""! $ dir = "directory/date/size=all"n  $ if there_is_a_pre_install_proc $ then= $    dir dest$kit:'pdf_name';,'ptf_name';,'pre_install_file';o $ else) $    dir dest$kit:'pdf_name';,'ptf_name';n $ endiff $ say ""$ $ node = f$trnlnm("sys$node") - "::" $PLEASE_ANSWER:L9 $ inq build_it "* Do you want to build the kit NOW (Y/N)". $ if build_it .eqs. "" .or. -sG      f$locate(build_it,"YN") .eq. 2 then goto PLEASE_ANSWER	!no defaultn
 $ if build_it: $ then $    gosub PCSI_BUILD  $PLEASE_ANSWER2: $    inq inst_it -O      "* Do you want to install ''prod_name' on this system (''node') NOW (Y/N)"H $    if inst_it .eqs. "" .or. -oG      f$locate(inst_it,"YN") .eq. 2 then goto PLEASE_ANSWER2	!no defaultu $    if debug then set verify 0 $    if inst_it then product install 'prod_name' $    if debug then set noverifyd $ endif. $EXIT: $ say "Goodbye." $ set symbol/scope=global  $ exitP $!------------------------------------------------------------------------------ $GIVE_HELP: SUBROUTINE
 $ esc[0,8]=27R $ bold=esc+"[1m" $ off =esc+"[m"i% $ help_setup = f$trnlnm("help$setup")l $ item = p1S $ if p2 .nes. "" then item = p2H $!+;G $! Let's see what kind of help we want (help mode = Verbose/Brief/None)bK $! HELP_SETUP in p1 forces a help text to be displayed *before* a question.-< $! p1 = "" when the user entered a "?" to a question prompt. $!-- $! here we want no help.? $ if p1 .eqs. "HELP_SETUP" .and. help_setup .eqs. "N" then exit-O $ if f$sea("sys$scratch:help.temp") .nes. "" then dele_ sys$scratch:help.temp;*SO $ help/libr=sys$login:pcsi.hlb/noprompt/nopage/out=sys$scratch:help.temp 'item'- $ close/nolog ch$ $ open/read ch sys$scratch:help.temp $ read ch dummyr $ read ch dummyM $ read ch dummyu $ write sys$output ""e $HELP_LOOP:N $ read/end=HELP_EOF ch line.O $ if line .eqs. "" then goto HELP_EOF !>>> blank lines should contain one spaceS  $ write sys$output bold,line,off1 $! here we want the brief version, one line only.lH $ if p1 .eqs. "HELP_SETUP" .and. help_setup .eqs. "B" then goto HELP_EOF $ goto HELP_LOOP
 $HELP_EOF:
 $ close ch $ write sys$output "": $ return $ ENDSUBROUTINEtO $!-----------------------------------------------------------------------------. $CREATE_PRECONF_FILE:  $ close/nolog ch" $ open/write ch 'pre_install_file' $ write ch "$!+"! $ write ch "$! ",pre_install_file0 $ write ch "$!"aM $ write ch "$! This file is the pre-installation procedure to be executed by" P $ write ch "$! the PCSI utility prior to the installation. It was automatically"B $ write ch "$! generated via PCSI_MENU version ",v," on ",f$time()K $ write ch "$! by ",first," ",last," (VMS user ",user,") on node ",node,"."S $ write ch "$!"e $ write ch "$! <input>"  $ write ch "$! None."  $ write ch "$!"$ $ write ch "$! <output>"N $ write ch "$! Any error messages will be displayed by PCSI during execution." $ write ch "$!"e $ write ch "$! <side effects>"3 $ write ch "$! (add here what this procedure does)") $ write ch "$!")  $ write ch "$! Revision history" $ write ch "$!".5 $ write ch "$! Version Date        Author     action"h $ write ch -P "$! ------- ----------- ---------- --------------------------------------------"! $ date = f$extract(0,11,f$time())e6 $ write ch "$! V1.0-0  ",date," ",user,"     creation" $ write ch "$!-"
 $ close ch $ return $! $BUILD_HEADER: $ write pdf "--!+" $ write pdf "--! ",pdf_name/ $ write pdf "--!"!
 $ write pdf -rP "--! This file is the Product Description File (PDF) for application ",prod_nameJ $ write pdf "--! automatically generated by PCSI_MENU v",v," on ",f$time()M $ write pdf "--! by ",first," ",last," (VMS user ",user,") on node ",node,"."o $ write pdf "--!""" $ write pdf "--! Revision history" $ write pdf "--!"e7 $ write pdf "--! Version Date        Author     action"t
 $ write pdf -eP "--! ------- ----------- ---------- -------------------------------------------"! $ date = f$extract(0,11,f$time()) O $ write pdf "--! V1.0-0  ",date," ",user,"     creation (via PCSI_MENU v",v,")"  $ write pdf "--!-"K $ write pdf "product ",producer," AXPVMS ",prod_name," ",prod_ver," full ;"t $ return $! $CREATE_EMPTY_PTF: $!+uK $! Note the = sign as the 1st character, and no ";" at the end of the line.P $!- 
 $ write ptf -E@   "=product ",producer," AXPVMS ",prod_name," ",prod_ver," full" $!+dF $! Unfortunately, comments are NOT allowed (at all) within PTFs... :-(8 $! So, we bypass this (non) feature with a tricky trick: $!- 
 $ write ptf -f   "1 'PRODUCT"
 $ write ptf -"   "=prompt ",prod_name
 $ write ptf -1N   " (this is a comment section written into the ",prod_name,".PCSI$TEXT file)"
 $ write ptf - H   " This file is the Product Text File (PTF) for application ",prod_name
 $ write ptf - =   " automatically generated by PCSI_MENU v",v," on ",f$time()o
 $ write ptf -"@   " by ",first," ",last," (VMS user ",user,") on node ",node,"."
 $ write ptf -    " "f
 $ write ptf -t   " Revision history"e
 $ write ptf -l   " "s
 $ write ptf -l*   " Version Date        Author     action"
 $ write ptf -rO   " ------- ----------- ---------- -------------------------------------------"I! $ date = f$extract(0,11,f$time())l
 $ write ptf -,B   " V1.0-0  ",date," ",user,"     creation (via PCSI_MENU v",v,")"
 $ write ptf -d   " "v $ return $! $GET_NR_OF_RECORDS:!) $ ver=f$ver(0)						!or the proc fails...e $ define sys$output x.xn $ copy/log 'files_list' nl:u $ deassign sys$output  $ close/nolog xt $ open/read x x.x 
 $ read x linep	 $ close xiM $ nr_of_rec = f$element(0," ",f$extract('f$locate("NL: (",line)'+5,999,line))I $ delete x.x;* $ ver=f$ver(ver) $ returnP $!------------------------------------------------------------------------------ $PCSI_BUILD:! $ on warning then goto FATAL_EXIT ! $ src = f$trnlnm("''source_kit'")i' $ if src .eqs. "" then src = source_kit   $ dest = f$trnlnm("''dest_kit'")' $ if dest .eqs. "" then dest = dest_kit  $ if debug then set verify $!# $ product package 'prod_name'   	- c      /base       =AXPVMS			- p       /producer   ='producer'		- B      /source     ='dest''prod_name'	-	! where to find PDF and PTF 7      /destination='dest'			-	! where to put .PCSI file a<      /material   ='src'			-	! where to find product material      /format     =sequential $! $ if debug then set noverify" $ dir 'producer'*'prod_name'*.pcsi $ say "" $ return $! $FATAL_EXIT: $ if debug then set noverify, $ say "The PCSI build procedure has failed." $ return   Enjoy.   D. -- fG   ---------------------------------------------------------------------"E MORANDI Consulting.  WEB: http://Didier.Morandi.Free.fr/index_us.htmlmE Pflanzschulstrasse 53, 8004 Zurich, Switzerland. GSM: +41 79 705 4670E/ 19, chemin de la Butte, 31400 Toulouse, France.h  H Disaster Recovery Plans, Computer Security Audits, DEC OpenVMS ExpertiseH On parle franais, Man spricht Deutsch, Habla Castellano, English spoken   ------------------------------  % Date: Thu, 29 Nov 2001 09:06:32 +0100S% From: "Fred Zwarts" <F.Zwarts@KVI.nl>i+ Subject: Re: DCPS stopped, inactive problemh. Message-ID: <9u4qa8$eoa$1@info.service.rug.nl>  4 <paddy.o'brien@zzz.tg.nsw.gov.au> wrote in message =/ news:01KBA1VTCE9E000WCE@tgmail.tg.nsw.gov.au...eJ > Since our contractors have forced boots on DCPS printers, I have been=20G > having problems on my machines.  The contractors have absolutely no =e idea=20sD > what to do.  These queues have been working with no problems for =	 years,=20tG > until our new contractors did something on their machine from which =s the=20 > printers boot. >=20H > The boot machine is a VAX running VMS 6.2.  My machines are a VAX at = 6.2=20 > and two Alphas at 7.2. >=20C > I can start the DCPS queue and it will happily sit there idle.  =L
 Submitting=20tH > a print job (ordinary text) via a generic queue, the queue goes busy = for a=20J > while and then is shown as stopped with autostart inactive.  The print = job=20* > is back on the generic queue as pending. >=20G > The operator log shows the following sequence of events at this time:! >=20: > %%%%%%%%%%%  OPCOM  27-NOV-2001 09:55:11.86  %%%%%%%%%%%# > Message from user SYSTEM on GECKO C > Process SYMBIONT_53: %SYSTEM-F-ACCVIO, access violation, reason =O
 mask=3D00,=20P; > virtual address=3D7FF60A00, PC=3D81F4B595, PSL=3D03C00004. >=20: > %%%%%%%%%%%  OPCOM  27-NOV-2001 09:55:12.05  %%%%%%%%%%%) > Message from user QUEUE_MANAGE on GECKO 9 > %QMAN-E-SYMDEL, unexpected symbiont process termination  >=20: > %%%%%%%%%%%  OPCOM  27-NOV-2001 09:55:12.05  %%%%%%%%%%%) > Message from user QUEUE_MANAGE on GECKOKB > -SYSTEM-F-ACCVIO, access violation, reason mask=3D00, virtual=203 > address=3D00000000, PC=3D0001DC6C, PSL=3D03C00000e >=20: > %%%%%%%%%%%  OPCOM  27-NOV-2001 09:55:12.05  %%%%%%%%%%%) > Message from user QUEUE_MANAGE on GECKOsA > %QMAN-I-QUEAUTOOFF, queue DCPS$P1201L is now autostart inactive  >=20I > Can anyone suggest anything that I can do on my machines, or anything =  I can=20 > tell the contractors to fix? >=20 > TIAs >=20 > Regards, Paddy >=20  ? We need more information about your queue setup. Could you postN+ the output of SHOW QUEUE /FULL DCPS$P1201L?"/ (Assuming that DCPS$P1201L is your DCPS queue.)N   ------------------------------  % Date: Thu, 29 Nov 2001 10:38:35 -0500<0 From: Paul Anderson <paul.r.anderson@compaq.com>+ Subject: Re: DCPS stopped, inactive problemi; Message-ID: <291120011038352962%paul.r.anderson@compaq.com>e  5 In article <01KBA1VTCE9E000WCE@tgmail.tg.nsw.gov.au>,n( <paddy.o'brien@zzz.tg.nsw.gov.au> wrote:  H > Since our contractors have forced boots on DCPS printers, I have been K > having problems on my machines.  The contractors have absolutely no idea rJ > what to do.  These queues have been working with no problems for years, J > until our new contractors did something on their machine from which the  > printers boot.  > Are these PrintServer printers that boot from an OpenVMS host?  K > The boot machine is a VAX running VMS 6.2.  My machines are a VAX at 6.2 > > and two Alphas at 7.2.  C What version of DCPS are you running?  If you are using TCP/IP as a G transport, what version of TCP/IP Services or other IP software are youl running?   Paul   -- p  Paul Anderson   OpenVMS Engineeringw   Compaq Computer Corporation[   ------------------------------  % Date: Thu, 29 Nov 2001 10:42:41 +0100d- From: Jouk Jansen <joukj@hrem.stm.tudelft.nl>i$ Subject: Re: DEC C Error on GS v7.032 Message-ID: <3C061121.7A21EF1@hrem.stm.tudelft.nl>   Richard L. Dyson wrote:_ > H > I am working on getting GS v7.03 working for my OpenVMS systems.  I amI > using Compaq C v6.4-008 on an OpenVMS/Alpha v7.2-1 system.  I get quiteSL > a way into the build when I get the following errors that are entirely new > to me: >  > [...]( > $sg > CC/NODEBUG/OPTIMIZE/DECC/PREFIX=ALL/NESTED_INCLUDE=PRIMARY/NAME=(AS_IS,SHORT)/DEFINE=("HAVE_MKSTEMP")n > -"H > /INCLUDE=([.obj] ,[.src]) /OBJECT=[.obj]gdevpdfm.obj  [.src]gdevpdfm.c > $ag > CC/NODEBUG/OPTIMIZE/DECC/PREFIX=ALL/NESTED_INCLUDE=PRIMARY/NAME=(AS_IS,SHORT)/DEFINE=("HAVE_MKSTEMP")n > -=H > /INCLUDE=([.obj] ,[.src]) /OBJECT=[.obj]gdevpdfo.obj  [.src]gdevpdfo.cE > Assertion failure:  Deleting instruction with DefinesRoutineCtx setf > Instruction:M > COD INSTRUCTION  BNE  : NEXT=0109B848, PREV=0109B498, LOCATOR={17224:1-30},, > S...F >          DEFINES_ROUTINE_CTX, OPCODE=66, NOT_IN_CURRENT_RTN, OP1=R1,
 > OP2=Targ... C > %GEM-F-ASSERTION, Deleting instruction with DefinesRoutineCtx set 1 > %TRACE-F-TRACEBACK, symbolic stack dump follows.L >   image    module    routine             line      rel PC           abs PC, >  DECC$COMPILER  GEM_DB  GEM__DB_ASSERT_END@ >                                           844 0000000000000314 > 00000000005AE0B48 >  DECC$COMPILER  GEM_CODE_CELL  gem_cn_delete_code_cell@ >                                        109577 0000000000004934 > 0000000000474934+ >  DECC$COMPILER  GEM_FI_PEEP  MOVE_SECTIONU@ >                                          5685 0000000000007D34 > 000000000063C1841 >  DECC$COMPILER  GEM_FI_PEEP  REVERSE_CROSS_JUMPe@ >                                          7842 000000000000A968 > 000000000063EDB8< >  DECC$COMPILER  GEM_FI_PEEP  GEM_FI_PEEP_BRANCH_PROCESSING@ >                                          3004 0000000000002F14 > 0000000000637364? >  DECC$COMPILER  GEM_FI_PEEP_ALPHA  GEM_FI_PEEP_APPLY_PEEPHOLEc@ >                                           733 00000000000001C4 > 000000000063F8044 >   DECC$COMPILER  GEM_FI_PEEP  PROCESS_PATTERN_LIST@ >                                          7461 0000000000009FF8 > 000000000063E448* >  DECC$COMPILER  GEM_FI_PEEP  GEM_FI_PEEP@ >                                          1528 0000000000000638 > 0000000000634A88@ >  DECC$COMPILER  GEM_FI  rtn_phase      134194 0000000000000F64 > 000000000047D1C4@ >  DECC$COMPILER  GEM_FI  gem_fi         134376 0000000000001ACC > 000000000047DD2C/ >  DECC$COMPILER  GEM_CO  GEM_CO_COMPILE_MODULEt@ >                                          3428 00000000000010F4 > 0000000000550684) >  DECC$COMPILER  COMPILE  gemc_be_master"@ >                                         85852 0000000000000E38 > 00000000001B0E38) >  DECC$COMPILER  COMPILE  gem_xx_compile @ >                                         85071 0000000000000000 > 0000000000000000) >  DECC$COMPILER  GEM_CP_VMS  GEM_CP_MAIN @ >                                          2551 0000000000001CA8 > 0000000000526F88@ >                                             0 FFFFFFFF94CF1414 > FFFFFFFF94CF1414F > %MMK-F-ERRUPD, error status %X19DB800C occurred when updating target > [.OBJ]GDEVPDFO.OBJL > %MMK-F-ERRUPD, error status %X1C14803C occurred when updating target MACRO > + >         Does anyone have any suggestions?> > M >         Has anyone else successfully compiled GhostScript v7.03 on OpenVMS?_ >   E This is due to a compiler bug. I reported it some time ago to Compaq.uA I got a patch to solve the problem. Unfortunately I have not seen , this patch in the public FTP-area of Compaq.  E  But I noticed that on earlier compiler versions although it seems to H compile correctly, it generates corrupt code for this file when compiled with all optimizations on.  C Solution : turn off the optimizations or get the patch from Compaq.u                   Jouk   ------------------------------  % Date: Thu, 29 Nov 2001 09:10:34 -0500 - From: "www.islandco.com" <sales@islandco.com>p: Subject: DECEMBER SPECIAL: DS20 1GB $3299 IN NEW Condition/ Message-ID: <u0cfjrnnau4n85@news.supernews.com>t   Alphserver DS20  EV6 500Mhz 21264 CPU
 1.44MB FLoppy  32x SCSI CD-ROMo 1GB Memory (from Camintonn)_   -- David B Turner	 Sales Dpte Island Computers US Corporation  2700 Gregory Street 	 Suite 180N Savannah GA 31404l Tel: 912 447 6622  Fax: 912 201 0096y sales@islandco.com www.islandco.com' http://www.islandco.com/legal-email.htmt   We sell Alpha's !"   ------------------------------  % Date: Thu, 29 Nov 2001 15:51:24 +0000O% From: Alan Greig <a.greig@virgin.net> & Subject: DECNET Phase IV saves the day8 Message-ID: <sblc0u0a68km04olakm5j0vcfqioabpp8f@4ax.com>  F On Tuesday evening an extremely well known organization who manage ourC corporate WAN installed new routing equipment locally. On Wednesday B morning the kit failed taking down all site external IP networking9 plus ISDN backup systems -  a should not happen scenario.a  C During various planning sessions I have argued for the retention of E DECNET Phase IV to certain key sites worldwide as opposed to shuttinge- down or transition to DECNET Phase V over IP.   C You guessed it DECNET connectivity remained up between Scotland and E France (which still had IP connectivity) allowing us to rapidly setupyD captive TCPIP/DECNET relay logons on a VMS box in France to SET HOSTC into our European MANMAN systems in Scotland. Additionally VMS mailt? over DECNET (once the corporate standard email) again became ofe critical importance.  ( Glad I didn't cave in and drop Phase IV.  F Yes I know there are serious issues we have to resolve with this "well known organization" as well... -- Alan   ------------------------------  % Date: Thu, 29 Nov 2001 11:43:03 +0100e= From: Oswald Knoppers <Oswald.Knoppers@contrastmediagroep.nl>i Subject: Re: DECnet timeout?5 Message-ID: <3C061137.343DA608@contrastmediagroep.nl>h   Didier Morandi wrote:   Q > Do you know how to enable TCP/IP RPC proxy? I had a look at the doc and did not 
 > find it.   Something like:n  E TCPIP> add proxy localuser/remote_user=username/host=remote_host_namem  F But remember that if the remote and local usernames are different, theC username must be specified in the R command. This is very different  compared to DECnet proxies.    Regards,   Oswald   ------------------------------  % Date: Thu, 29 Nov 2001 17:04:45 +0100 & From: "B.Eckstein" <B.Eckstein@cli.de>5 Subject: Re: DSSI VAXcluster manual on line anywhere?p% Message-ID: <3C065C9D.6070805@cli.de>(   Robert DiRosario wrote:s  F > The DSSI cable doesn't go directly to the system board, it goes to aE > small card that plugs into the system board, under the disk drives.u    ? Yes, according to the "The Design of the VAX 4000 Model 100 andE@ MicroVAX 3100 Model 90 Desktop Systems" the DSSI-Controller is a, expansion-Card to the KA50/KA52 Motherboard.    -C > I can open my system and look for a part numbers on the board andd > cable if you want.    8 Thanks for this offer, but I am able to get such a board   from a broken 4100 for my 3190.!   ------------------------------    Date: 29 Nov 2001 01:57:10 -0800- From: peter@kaj-und-pedro.org (Peter Flunger)t8 Subject: FC  configurations with only one HSG controller= Message-ID: <29f941b3.0111290157.7a247c9b@posting.google.com>    Hi all, C has anybody got an explamation for this, any inout would be highly f0 welcome, I cannot find anything in the manuals..  B We have got pretty much the same configuration running in several  sites across Europe:" ( listing nly the relevant parts )  0 dual redundand HSG80 im Multipath failover mode,
 two switches,m- two controllers in each Alpha OpenVMS server.H; Each HSG has one port connected to one of the two switches,t%   the other port to the other switch.l& Each AS is connected to both switches. Result ( as expected ) e= There are four paths to each of the DGA ( and of course GGA )E	 devices .h  A But there is one site where we are running a very similar config  A except for the fact that we are only running HSG60's and only one- HSG60 for each DiskSubsystem.-@ And instead of seeing two paths ( which would be logical to me ) i can see only one path.   Any ideas ?e     Thanks
 Peter Flunger"   ------------------------------  % Date: Thu, 29 Nov 2001 14:57:38 +0100T From: zessin@decus.det< Subject: RE: FC  configurations with only one HSG controller* Message-ID: <00A05C6E.A687FE0D.9@decus.de>   Peter Flunger wrote: [...]g2 > dual redundand HSG80 im Multipath failover mode, > two switches, / > two controllers in each Alpha OpenVMS server.n= > Each HSG has one port connected to one of the two switches,h& >  the other port to the other switch.( > Each AS is connected to both switches. > Result ( as expected )? > There are four paths to each of the DGA ( and of course GGA )i > devices .   B > But there is one site where we are running a very similar configC > except for the fact that we are only running HSG60's and only oneV > HSG60 for each DiskSubsystem.iB > And instead of seeing two paths ( which would be logical to me ) > i can see only one path.  / Oh, that's easy and is the same for HSG60 + 80:h  D When the controller pair is in multi-bus failover mode units D0..199D are presented through all 4 ports and all connections have a default offset of 0.  > In single-controller or dual-redundancy mode, units D0..99 areB presented through each controller's port 1 and units D100..199 are+ presented through each controller's port 2.o@ The connection for PORT_1 has an offset of 0, but the connectionC for PORT_2 has a default offset of 100 so that D100..199 are mappeds  to the LUN range of 0..99 again.6 In those cases you have a single path to your disk(s).   -- t
 Uwe Zessin   ------------------------------  % Date: Thu, 29 Nov 2001 16:12:53 +0100-* From: "peter flunger /news" <p-i-b@gmx.at>< Subject: Re: FC  configurations with only one HSG controller0 Message-ID: <9u5j86$3bv$1@newsreader1.netway.at>  ( <zessin@decus.de> schrieb im Newsbeitrag$ news:00A05C6E.A687FE0D.9@decus.de... > Peter Flunger wrote:1 > Oh, that's easy and is the same for HSG60 + 80:t >e@ > In single-controller or dual-redundancy mode, units D0..99 areD > presented through each controller's port 1 and units D100..199 are- > presented through each controller's port 2.(B > The connection for PORT_1 has an offset of 0, but the connectionE > for PORT_2 has a default offset of 100 so that D100..199 are mapped " > to the LUN range of 0..99 again.8 > In those cases you have a single path to your disk(s).  > Thank you indeed. The manual was not so specific about single-. controller mode, they only mentioned d-r mode.# But I am afraid that you are right,,D this really is a pain [.. omitted ..], especially since we canot buy4 two more HSG60's 'cause they have been discontinued.I OK, I'm off to check with our accountants, maybe they have got some money  left
 somewhere. :)   
 Cheers again,a Peter    ------------------------------  % Date: Thu, 29 Nov 2001 17:07:47 -0000-4 From: "Chris Sharman" <chris.sharman@ccagroup.co.uk> Subject: ftp performanceA Message-ID: <1007053505.2905.0.nnrp-02.9e989e7e@news.demon.co.uk>t  J We've got a ds20e connected via both a 100M card & a 1Gb card & a Mac file server (100M only). 2 We can transfer files Mac to Mac at around 600k/s./ Mac to ie5.0/win2000 seems to be around 220k/s.n4 Mac to ftpnetdrive/win2000 seems to be around 800k/s& Mac to VMS is consistently 170-180k/s.K This is using vms 7.3, tcp/ip 5.1-3, native ftp, mgftp2.6-5 (with & without-; sysprv, and with madgoat_ftp_window_size 64000 and 131072).-" All in binary mode where possible.A Nothing seems to have significant impact on the throughput rates.f! Anything else to try or look at ?    Suggestions anyone ? Hunter ?)   ------------------------------    Date: 29 Nov 2001 11:13:28 -0600 From: briggs@encompasserve.org Subject: Re: ftp performance3 Message-ID: <MOoAQYEDZSh2@eisner.encompasserve.org>/  x In article <1007053505.2905.0.nnrp-02.9e989e7e@news.demon.co.uk>, "Chris Sharman" <chris.sharman@ccagroup.co.uk> writes:L > We've got a ds20e connected via both a 100M card & a 1Gb card & a Mac file > server (100M only).g4 > We can transfer files Mac to Mac at around 600k/s.1 > Mac to ie5.0/win2000 seems to be around 220k/s.t6 > Mac to ftpnetdrive/win2000 seems to be around 800k/s( > Mac to VMS is consistently 170-180k/s.M > This is using vms 7.3, tcp/ip 5.1-3, native ftp, mgftp2.6-5 (with & without = > sysprv, and with madgoat_ftp_window_size 64000 and 131072). $ > All in binary mode where possible.C > Nothing seems to have significant impact on the throughput rates.u# > Anything else to try or look at ?n >  > Suggestions anyone ? Hunter ?V  D Default extend quantity?  I sure wouldn't want to be in the positionC of an FTP server process continually tacking 11 blocks onto the end  of a 50 megabyte download file.   
 Just a guess.@   	John Briggs   ------------------------------  % Date: Thu, 29 Nov 2001 10:46:16 -0700i( From: emanuel stiebler <emu@ecubics.com> Subject: Re: ftp performance+ Message-ID: <3C067468.448F2E17@ecubics.com>    Chris Sharman wrote: > L > We've got a ds20e connected via both a 100M card & a 1Gb card & a Mac file > server (100M only).v > [...]e > Suggestions anyone ? Hunter ?   D Because none of the transfer rates exceeds 800 KB/s, I guess, you're
 just sitting t: on a 10Mb network, regardles of the cards you're using ...   ------------------------------  # Date: Thu, 29 Nov 2001 07:41:09 GMTy* From: "Bill Todd" <billtodd@metrocast.net> Subject: Re: Future of VMS ?< Message-ID: <pElN7.64786$YD.5774359@news2.aus1.giganews.com>  8 "Rob Young" <young_r@encompasserve.org> wrote in message- news:9pueAaha4WC3@eisner.encompasserve.org... J > In article <IUjN7.168687$dk.12068132@bin1.nnrp.aus1.giganews.com>, "Bill& Todd" <billtodd@metrocast.net> writes:   ...0  K > > While the inability of Compaq to compete effectively with Dell on price0 isG > > indeed relevant to the question of whether Compaq could make Itanicu boxes toL > > sell for less than Alpha boxes, a second issue was that the Dell product was K > > in no way a match for the ES40.  And AFAIK you can still buy a DS10 for> wellE > > under the price of even the cheapest (Dell) Itanic box.  Finally,b
 additionalI > > discussion elicited the information that a large portion of the pricedK > > difference between an ES40 and a Dell Itanic box with considerably lesseG > > performance was attributable to Compaq's significant mark-up on the0 Alpha4K > > memory (a mark-up which would apply equally to a Compaq Itanic box, onetL > > would assume - unless Alpha's significantly superior performance allowedF > > them to soak Alpha customers more, as seems to be the case today). > >d >t+ > Signifcantly higher memory and cpu costs:s  J The latter is precisely what I referred to above.  And the availability ofH equivalent memory for Alphas from other sources at *far* more reasonableH prices indicates that this cost difference is purely artificial:  CompaqB simply chooses to charge several times as much as Dell charges forI comparable memory, because Alpha is a good enough product for them to geth
 away with it.m  I As for any difference in CPU costs, let's see if Aaron Spink will be morevL forthcoming in that area:  it's certainly true that Itanic was introduced atH a minimum price of over $1K (up to over  $4K for the vaunted 800 MHz pigK with more cache, IIRC), while Compaq's *internal* cost for an Alpha suffersaL at worst from fab costs plus at most $1K (since Terry has seemed inclined toL up his previous $500 estimate) to cover the on-going development costs (thatJ are now no longer on-going).  So once again Compaq is simply charging whatL the market will bear for a superb platform, while Dell is charging as littleI as possible and *still* having conspicuously little success in moving thew= product that you'd have us consider comparable, last I heard.s   >eB > Itanium servers are quite a bit cheaper than AlphaServers today, > never mind the future.  K Not at the DS10 level, nor at higher levels if you compare servers of equalvK capability rather than consider each Itanic pig a full equal to each Alpha.dK But I understand that the capabilities that seemed so important to you when L you were an Alpha supporter became an embarrassment when you switched sides.   - bill   ------------------------------   Date: 29 Nov 2001 08:53:11 GMTB From: bertrand@perceval.cnam.fr (BERTRAND =?iso-8859-1?Q?Jo=EBl?=) Subject: Re: Future of VMS ?6 Message-ID: <slrna0btrj.c5o.bertrand@perceval.cnam.fr>  " Le Wed, 28 Nov 2001 19:03:38 +01006 John McLean <mcleanj@swissonline.delete.ch> crivait : > G >My recent postings about Compaq have been quite negative and there's arH >very good reason for this;  personally I believe that the best prospectE >of a future for VMS lies with killing the merger with HP dead in itsd >tracks. >r >Why do I think this ? > G >1.  There is no current indication that the merger will do anything at E >all to improve the situation for VMS.  At best it looks like will be B >more of the same (niche markets, no promotion, no press releases,G >retiring products ...etc.).  At worst it could be another e3000, maybe < >not right away but some time in the not too distant future. >6E >2.  The current management of Compaq - the same individuals who willsD >have a strong influence on the merged company - are people who haveC >directed billions of dollars into the product that makes the leasteB >profit, and neglected to make any serious efforts to market theirD >best-earners.  To do this for one year would be unfortunate, but toF >persist in doing this year after year really makes one question theirE >competence.  It looks like the stockmarket is also losing confidencerI >because when Capellas took over the stock price was about $28, then rose.G >slowly to $35 (up 20%) but is now down at less than $10 (down 70% fromo >that $35).5 >0H >3.  The current roadmap expects the IPF processor to be on schedule andE >meet performance specifications but this must be regarded as wishfuldA >thinking when delivery is several years away.   Right now VMS isAC >slipping into "coasting" mode, no foot on the pedal, just driftingsE >along.  If IPF slips by even a few months (as is Intel's habit) thenrH >even more VMS users will desert it and quite possibly this will reach aG >point where it casts doubt upon the viability of the porting exercise.2  M         Don't forget that OpenVMS is only avalaible on VAX and APX. I believeTJ that this system can be more used if it is avalaible on more architecture.G Thus, I have restarted the old FreeVMS project. First, I was alone, butkF we are currenty about 30. We are working on kernel, libraries and DCL,2 and we have a SRM-like console which runs on i386.  ;         All informations about this project can be found att% http://freshmeat.net/projects/freevms            Regards,           JKBe   -- oO La guerre des sexes n'est pas prs de s'teindre, il y a trop de fraternisatione avec l'ennemi. -- Georges Bernard Shawa   ------------------------------  % Date: Thu, 29 Nov 2001 10:35:31 +00004% From: Alan Greig <a.greig@virgin.net>s Subject: Re: Future of VMS ?8 Message-ID: <ee3c0usfqsrc71ng62uppshfda8qr2bcpi@4ax.com>  D On 29 Nov 2001 00:11:26 -0600, young_r@encompasserve.org (Rob Young) wrote:     >bF >	Have you priced out 16 GBytes of memory for an AlphaServer lately?  G >	The memory is considerably more expensive for an AlphaServer.  I was aJ >	looking at 16 GBytes of AlphaServer memory... around $48000* U.S.  That E >	same 16 GBytes of memory for a 4-way Itanium server is $25000 U.S.:.  F I've just had  a quote for ES40 Compaq memory (new) 4 GB module $6000.= So 16 GB would be $24,000. The supplier is well known to thishD newsgroup. Official Compaq  list price is far higher. Also CamintonnF memory by same supplier approx $3000 per 4GB module. A UK supplier hasE quoted the lowest at $2000 for 4GB module (Dataram I think) or approxr $8000 for 16 GB.   You need to shop about more!     >Ea >http://configure.us.dell.com/dellstore/config.asp?customer_id=04&order_code=pe7150&keycode=6w300> >CB >	What about quad processor?  If you load that Itanium box up with? >	CPUs (4MByte L2 - top of the line at 800 MHz) you add a total F >	of $20K to the cost.  That puts a quad Itanium server with 16 GBytesE >	of RAM, 4 CPUs at $63K U.S.  The same config for an ES40 would mostdE >	likely be in the neighborhood of $115K U.S.  ($48K* for memory, 27K0E >	base system, each processor add 13K.  I may be $10K low on the base  >	system, but what's $10K ?).  >hC >	The cost of making that Dell system a quad-processor is $19000 , e8 >	quite a bit cheaper than the cost to uplift an ES40 toF >	quad status.  I don't expect that to change much when McKinley shows  E Hmm I've also had quotes very much below list for new ES40 processors + which are close to the Dell cost you quote.p  8 	up, should actually go down as McKinley is in a smaller process andl? >	L2 comes on CPU.  That separate L2 is very costly stuff.  You-' >	are off the mark on McKinley pricing.: >1A >	Itanium servers will continue to be substantially cheaper than pE >	AlphaServers and CPU cost is - and will continue to be - a sizable r	 >	factor.   # Depends on who you shop with again.o   >1 -- Alan   ------------------------------  % Date: Thu, 29 Nov 2001 10:38:20 +0000n% From: Alan Greig <a.greig@virgin.net>  Subject: Re: Future of VMS ?8 Message-ID: <as3c0ukakndlb9o8ohnvvf0r43780sd5dt@4ax.com>  F On Thu, 29 Nov 2001 07:41:09 GMT, "Bill Todd" <billtodd@metrocast.net> wrote:  J >As for any difference in CPU costs, let's see if Aaron Spink will be moreM >forthcoming in that area:  it's certainly true that Itanic was introduced atwI >a minimum price of over $1K (up to over  $4K for the vaunted 800 MHz piguL >with more cache, IIRC), while Compaq's *internal* cost for an Alpha suffersM >at worst from fab costs plus at most $1K (since Terry has seemed inclined to M >up his previous $500 estimate) to cover the on-going development costs (thatr  D The loss of Alpha CPU sales to NetApp won't have helped. To the bestD of my knowledge the person who wrote the Compaq literature attackingD NetApp for using Alpha processors has not ben sacked despite costing< the company millions or perhaps tens of millions of dollars.       -- Alan   ------------------------------  # Date: Thu, 29 Nov 2001 13:37:13 GMT & From: "Jeff Killeen" <Jeff@IDM-IO.com> Subject: Re: Future of VMS ?8 Message-ID: <dSqN7.1170$h24.101002@typhoon1.gnilink.net>  8 "Rob Young" <young_r@encompasserve.org> wrote in message- news:9pueAaha4WC3@eisner.encompasserve.org...   @ > Itanium servers will continue to be substantially cheaper thanD > AlphaServers and CPU cost is - and will continue to be - a sizable	 > factor.A  G What I suspect you find is there won't be a noticeable price differencetL between IA64 servers delivering the same price/performance as Alpha servers.L The primary reason being that R&D costs really drive the cost of mid to highH end servers.  The manufacturing costs (parts) are only a fraction of theH price to the customer.  With a secondary reason being the need for added IA64 processors...   ------------------------------  # Date: Thu, 29 Nov 2001 14:14:55 GMT14 From: "Terry C. Shannon" <terryshannon@mediaone.net> Subject: Re: Future of VMS ?; Message-ID: <zprN7.491$zX1.1434991@typhoon.ne.mediaone.net>a  7 > In article <3C05647D.E2DA8D82@videotron.ca>, JF MezeiE& <jfmezei.spamnot@videotron.ca> writes: > > Rob Young wrote:K > >>         But as discussed before, Itanium servers today are quite a bite( > >>         cheaper than Alpha servers. > >r > > Today ? Are you sure ? > > J > > Is it a case of Alpha servers being artificially jacked up in price to preventu! > > their competing against PCs ?t  E Why on God's green earth would CPQ artificially jack up the prices ofr
 AlphaServers?t  1 To deter competition against peecees? Get real!!!   J Even a peecee Conspiracy Theorist would have to realize that inflating theI prices of AlphaServers would render them less competitive with RISC boxesp from SUNW, HWP, and IBM.  H The last time I looked, CPQ was competing pretty aggressively with those guys... and gals.n   ------------------------------  # Date: Thu, 29 Nov 2001 14:16:02 GMT-4 From: "Terry C. Shannon" <terryshannon@mediaone.net> Subject: Re: Future of VMS ?; Message-ID: <CqrN7.492$zX1.1435328@typhoon.ne.mediaone.net>3  2 "Keith Brown" <kbrown780@isd.net> wrote in message) news:u0b7ogbhdmfs52@corp.supernews.com...S > JF Mezei wrote:m >p > > John McLean wrote:I > >> along.  If IPF slips by even a few months (as is Intel's habit) then L > >> even more VMS users will desert it and quite possibly this will reach aK > >> point where it casts doubt upon the viability of the porting exercise.U > >IL > > On a certain angle, I disagree. The more IA64 is delayed, the longer theL > > remaining VMS customers still stay on Alpha. Customers will only migrateJ > > to IA64 if/when they are forced to or if the performance difference is! > > really worth all the trouble., > >]L > > The longer it takes for IA64 to materialise with some competitive speed,> > > the longer Compaq will have to keep Alpha on life support. > >UG > > I dare Compaq to come out in public and announce VMS , like MPE  is. being K > > terminated. Is it really worth the aggravation of witnessing VMS wither. > > slowly day after day ? > >iK > > The longer the slow agony of VMS lasts, the longer people who loved VMSd# > > will be mad , saddened etc etc.r > >eG > > Might as well get it over with ASAP, and perhaps 5-6 years down the  road,_I > > former VMS customers will have new decision makers who would not have4 any00 > > grudges against Compaq and return to Compaq. >c >eG > If Compaq kill VMS then I will never buy another product from Compaq.0D > There will be no returning to patronize the vendor who screwed me. >   E Keith, that says it all! It would make a fine statement at the annual0 stockholders meeting!P   ------------------------------  # Date: Thu, 29 Nov 2001 14:17:31 GMT 4 From: "Terry C. Shannon" <terryshannon@mediaone.net> Subject: Re: Future of VMS ?; Message-ID: <%rrN7.494$zX1.1435623@typhoon.ne.mediaone.net>   9 "Didier Morandi" <Didier.Morandi@gmx.ch> wrote in message$  news:3C05C7D9.5ADD5328@gmx.ch... > "Terry C. Shannon" wrote:  > > D > > Well, he's welcome to start when I launch "Shannon KNEW Compaq!" >  > Hi Terry,I > ) > well, you may start to prepare SKHP :-)  >   
 Hi Didier,  J That's a distinct possibility, although I haven't a clue as to whether the acquisition will happen or not.   H But I *did* take the first step... I joined INTEREX last week. Not a bad deal for $30USD per year.    ------------------------------    Date: 29 Nov 2001 09:31:19 -0600+ From: young_r@encompasserve.org (Rob Young)0 Subject: Re: Future of VMS ?3 Message-ID: <FPkJf26G+Ej+@eisner.encompasserve.org>   ` In article <ee3c0usfqsrc71ng62uppshfda8qr2bcpi@4ax.com>, Alan Greig <a.greig@virgin.net> writes:F > On 29 Nov 2001 00:11:26 -0600, young_r@encompasserve.org (Rob Young) > wrote: >  >  >>G >>	Have you priced out 16 GBytes of memory for an AlphaServer lately?   H >>	The memory is considerably more expensive for an AlphaServer.  I was K >>	looking at 16 GBytes of AlphaServer memory... around $48000* U.S.  That  F >>	same 16 GBytes of memory for a 4-way Itanium server is $25000 U.S.: > H > I've just had  a quote for ES40 Compaq memory (new) 4 GB module $6000.? > So 16 GB would be $24,000. The supplier is well known to thisBF > newsgroup. Official Compaq  list price is far higher. Also CamintonnH > memory by same supplier approx $3000 per 4GB module. A UK supplier hasG > quoted the lowest at $2000 for 4GB module (Dataram I think) or approxe > $8000 for 16 GB. >  > You need to shop about more! >   C 	Sure... I can do the same for memory.  Just looking at large OEMs,t 	not aftermarket prices.   				Robh   ------------------------------  % Date: Thu, 29 Nov 2001 17:27:28 +0100s% From: "Dr. Dweeb." <Dweeb@nospam.com>r Subject: Re: Future of VMS ?/ Message-ID: <CltN7.94$WK5.4569@news.get2net.dk>t   Fred,   
 Dja vu ??  J "While I cannot say anything specific about the, I will say that I support/ it 100%.  I think it will be good for OpenVMS.",  F I remember hearing/reading these exact same words (multiple times from. multiple sources) before CompaQ swallowed DEC.   Just an observation.   Dweeb.@ "Fred Kleinsorge" <kleinsorge@star.zko.dec.com> wrote in message- news:zPbN7.2067$RL6.62887@news.cpqcorp.net...e >yK > John McLean wrote in message <3C0526FA.7A739645@swissonline.delete.ch>...p > >oI > >My recent postings about Compaq have been quite negative and there's a J > >very good reason for this;  personally I believe that the best prospectG > >of a future for VMS lies with killing the merger with HP dead in its-
 > >tracks. > >h >2 >oJ > While I cannot say anything specific about the merger, I will say that I8 > support it 100%.  I think it will be good for OpenVMS. >jH > The only thing I will comment on in your post is the IPF reference.  IE > believe merger or not, the Alpha/IPF decision will not be revisted.o   ------------------------------  % Date: Thu, 29 Nov 2001 17:32:15 +0100I% From: "Dr. Dweeb." <Dweeb@nospam.com>- Subject: Re: Future of VMS ?/ Message-ID: <4qtN7.95$WK5.4728@news.get2net.dk>u  : "JF Mezei" <jfmezei.spamnot@videotron.ca> wrote in message& news:3C05A498.4997E7B1@videotron.ca... > Keith Brown wrote:I > > If Compaq kill VMS then I will never buy another product from Compaq.eF > > There will be no returning to patronize the vendor who screwed me. >aJ > But as you change jobs, the guy who will replace you at your old job may notTJ > have that anti-Compaq attitude and Compaq might be able to wine and dine himE$ > to become a Compaq customer again.  J Assuming of course that CompaQ actually exists 5-6 years down the line.  A bold prediction at best.  B A beauty of capitalism is that we have the ability to "punish" theG incompetant by spending our $ somewhere else.  Should enough of CompaQs2L customers move their $ somewhere else, then there will be no more CompaQ (or HP for that matter).   Dweebe   ------------------------------  % Date: Thu, 29 Nov 2001 17:36:06 +0100r% From: "Dr. Dweeb." <Dweeb@nospam.com>t Subject: Re: Future of VMS ?/ Message-ID: <HttN7.97$WK5.4649@news.get2net.dk>c  5 "Bill Todd" <billtodd@metrocast.net> wrote in messagee6 news:N5eN7.61962$YD.5501533@news2.aus1.giganews.com... > @ > "John McLean" <mcleanj@swissonline.delete.ch> wrote in message1 > news:3C0526FA.7A739645@swissonline.delete.ch...t > >Clip ...  >sL > To address Rob's comment about making the Itanic pig fly:  of course IntelH > will.  An old aviation adage is that you can make a barn door fly withK > enough power, and Itanic uses power like an electric furnace (thus makingi5 > the adage apply both metaphorically and literally).s >r  # I guess that was the F4-Phantom :-)d   ------------------------------  % Date: Thu, 29 Nov 2001 12:27:02 -0500l5 From: "Fred Kleinsorge" <kleinsorge@star.zko.dec.com>q Subject: Re: Future of VMS ?3 Message-ID: <oguN7.2114$RL6.63331@news.cpqcorp.net>c  K No question.  And in fact, since Compaq bought DEC, VMS had managed to turn L around inside the company, and had managed some growth.  That is, instead ofK a prevailing attitude of "riding VMS down", things changed to a "how can wed sustain this business".e  F In the current case, I believe that HP brings to the table a number of8 things that VMS itself may be able to take advantage of.     Dr. Dweeb. wrote in message ...v >Fred, >M >Dja vu ??m > K >"While I cannot say anything specific about the, I will say that I supportf0 >it 100%.  I think it will be good for OpenVMS." >dG >I remember hearing/reading these exact same words (multiple times fromp/ >multiple sources) before CompaQ swallowed DEC.o >c >Just an observation.y >a >Dweeb.tA >"Fred Kleinsorge" <kleinsorge@star.zko.dec.com> wrote in messageg. >news:zPbN7.2067$RL6.62887@news.cpqcorp.net... >>L >> John McLean wrote in message <3C0526FA.7A739645@swissonline.delete.ch>... >> >J >> >My recent postings about Compaq have been quite negative and there's aK >> >very good reason for this;  personally I believe that the best prospectrH >> >of a future for VMS lies with killing the merger with HP dead in its >> >tracks.i >> > >> >>K >> While I cannot say anything specific about the merger, I will say that It9 >> support it 100%.  I think it will be good for OpenVMS.a >>I >> The only thing I will comment on in your post is the IPF reference.  I F >> believe merger or not, the Alpha/IPF decision will not be revisted. >  >s >8   ------------------------------  % Date: Thu, 29 Nov 2001 18:34:44 +0100.% From: "Dr. Dweeb." <Dweeb@nospam.com>8 Subject: Re: Future of VMS ?0 Message-ID: <EkuN7.115$WK5.4919@news.get2net.dk>  L So in the "hope springs eternal" frame of mind, after HP take over, we mightD expect to see a "how can we grow this business" corporate attitude ?  G I hope so, but as you may have noted, I am not too confident right now.    Dweeb.@ "Fred Kleinsorge" <kleinsorge@star.zko.dec.com> wrote in message- news:oguN7.2114$RL6.63331@news.cpqcorp.net... H > No question.  And in fact, since Compaq bought DEC, VMS had managed to turnK > around inside the company, and had managed some growth.  That is, instead. ofJ > a prevailing attitude of "riding VMS down", things changed to a "how can we > sustain this business".6 >2H > In the current case, I believe that HP brings to the table a number of: > things that VMS itself may be able to take advantage of. >e >d! > Dr. Dweeb. wrote in message ...e > >Fred, > >a
 > >Dja vu ??  > >iE > >"While I cannot say anything specific about the, I will say that In supportd2 > >it 100%.  I think it will be good for OpenVMS." > >tI > >I remember hearing/reading these exact same words (multiple times from.1 > >multiple sources) before CompaQ swallowed DEC.i > >r > >Just an observation.t > >v	 > >Dweeb.kC > >"Fred Kleinsorge" <kleinsorge@star.zko.dec.com> wrote in messaged0 > >news:zPbN7.2067$RL6.62887@news.cpqcorp.net... > >>! > >> John McLean wrote in messagee, <3C0526FA.7A739645@swissonline.delete.ch>... > >> >L > >> >My recent postings about Compaq have been quite negative and there's aD > >> >very good reason for this;  personally I believe that the best prospectJ > >> >of a future for VMS lies with killing the merger with HP dead in its
 > >> >tracks.' > >> > > >> > >>K > >> While I cannot say anything specific about the merger, I will say thatw Ie; > >> support it 100%.  I think it will be good for OpenVMS.  > >>K > >> The only thing I will comment on in your post is the IPF reference.  IhH > >> believe merger or not, the Alpha/IPF decision will not be revisted. > >  > >e > >a >d >e   ------------------------------  % Date: Thu, 29 Nov 2001 12:42:30 -0500e) From: "Mike Foley" <mikiefoley@yahoo.com>  Subject: Re: Future of VMS ?/ Message-ID: <u0csmj79lsfi45@corp.supernews.com>a  C I think cost (of a new EV6 design of the NetApp box) was probably amH greater factor than anything anyone said in public or private. Remember,@ NetApp boxes were EV5x based. x86 designs have rapidly caught up9 with 1995-1997 Alpha designs, at a MUCH lower entry cost.-  = FWIW, I'd love to see HP/Q spin off VMS as a seperate company<L after the Itanium port. That's just a personal preference. :) I think in the. long run, the port to IPF is the wiser choice.     mikeH                                         Former system manager in the VMS GroupuH                                         Former marketing engineer at API  2 "Alan Greig" <a.greig@virgin.net> wrote in message2 news:as3c0ukakndlb9o8ohnvvf0r43780sd5dt@4ax.com...H > On Thu, 29 Nov 2001 07:41:09 GMT, "Bill Todd" <billtodd@metrocast.net> > wrote: >.L > >As for any difference in CPU costs, let's see if Aaron Spink will be moreL > >forthcoming in that area:  it's certainly true that Itanic was introduced atK > >a minimum price of over $1K (up to over  $4K for the vaunted 800 MHz pig F > >with more cache, IIRC), while Compaq's *internal* cost for an Alpha suffersBL > >at worst from fab costs plus at most $1K (since Terry has seemed inclined toI > >up his previous $500 estimate) to cover the on-going development costsd (thate >oF > The loss of Alpha CPU sales to NetApp won't have helped. To the bestF > of my knowledge the person who wrote the Compaq literature attackingF > NetApp for using Alpha processors has not ben sacked despite costing> > the company millions or perhaps tens of millions of dollars. >f >a >S > -- > Alan   ------------------------------  % Date: Thu, 29 Nov 2001 12:44:18 -0500	5 From: "Fred Kleinsorge" <kleinsorge@star.zko.dec.com>  Subject: Re: Future of VMS ?3 Message-ID: <zwuN7.2115$RL6.63338@news.cpqcorp.net>s   Dr. Dweeb. wrote in message ... G >So in the "hope springs eternal" frame of mind, after HP take over, we  might E >expect to see a "how can we grow this business" corporate attitude ?t >v     Not at all.s  G VMS itself was already in a "how can we grow these segments" mode.  The J Itanium port may cause a brief shift of focus, but it remains the same.  IJ cannot make any predictions about what current or future senior management may do.c  J I'm am really thinking of software and hardware capabilities that we might leverage from HP's UNIX base.C  H I am optimistic about the future of VMS regardless of the outcome of the merger.s   _Freda  H >I hope so, but as you may have noted, I am not too confident right now. >0 >Dweeb.2A >"Fred Kleinsorge" <kleinsorge@star.zko.dec.com> wrote in messager. >news:oguN7.2114$RL6.63331@news.cpqcorp.net...I >> No question.  And in fact, since Compaq bought DEC, VMS had managed tot >turnrL >> around inside the company, and had managed some growth.  That is, instead >of K >> a prevailing attitude of "riding VMS down", things changed to a "how can  >wed >> sustain this business". >>I >> In the current case, I believe that HP brings to the table a number ofC; >> things that VMS itself may be able to take advantage of.o >> >>" >> Dr. Dweeb. wrote in message ...	 >> >Fred,e >> > >> >Dja vu ?? >> >F >> >"While I cannot say anything specific about the, I will say that I >support3 >> >it 100%.  I think it will be good for OpenVMS."  >> >J >> >I remember hearing/reading these exact same words (multiple times from2 >> >multiple sources) before CompaQ swallowed DEC. >> > >> >Just an observation. >> >
 >> >Dweeb.D >> >"Fred Kleinsorge" <kleinsorge@star.zko.dec.com> wrote in message1 >> >news:zPbN7.2067$RL6.62887@news.cpqcorp.net...J >> >>l" >> >> John McLean wrote in message- ><3C0526FA.7A739645@swissonline.delete.ch>...0 >> >> >oK >> >> >My recent postings about Compaq have been quite negative and there's  asE >> >> >very good reason for this;  personally I believe that the bestn	 >prospect K >> >> >of a future for VMS lies with killing the merger with HP dead in itse >> >> >tracks. >> >> >t >> >>  >> >>sL >> >> While I cannot say anything specific about the merger, I will say that >I< >> >> support it 100%.  I think it will be good for OpenVMS. >> >>mL >> >> The only thing I will comment on in your post is the IPF reference.  II >> >> believe merger or not, the Alpha/IPF decision will not be revisted.t >> > >> > >> > >> >> >y >a   ------------------------------  % Date: Thu, 29 Nov 2001 19:08:27 +0100-1 From: John McLean <mcleanj@swissonline.delete.ch>m Subject: Re: Future of VMS ?5 Message-ID: <3C06799B.E4B2085A@swissonline.delete.ch>g   Fred Kleinsorge wrote: >   J > While I cannot say anything specific about the merger, I will say that I8 > support it 100%.  I think it will be good for OpenVMS. > H > The only thing I will comment on in your post is the IPF reference.  IE > believe merger or not, the Alpha/IPF decision will not be revisted.   E And I believe that the only truly encouraging sign to come out of allyE this deal with Intel is the fact that as the porting takes place, the / hardware dependencies in VMS are being reduced.a  H This at least will permit a hasty "Plan-B" if things go to mush, becauseA it means a port to yet another processor would be a simpler task.e    H I also want to point out that current porting is to IA64 and is _not_ toH the target IPF platform which is still 2, 3, 4 .. ? years away.  It mustD be reasonable to assume that some degree of change would be requiredH between IA64 and (whatever), but perhaps that change will be minor.  (IfH this was all plain sailing than there seems no good reason why VMS could: not be released on IA64 prior to original scheduled date.)  G Oh dear, so you cannot comment on the merger.  If you had read the finebG print of the merger document that HP submitted to the SEC you would seeaH that Compaq's board and its employees are supposed to endorse the merger8 at every reasonable opportunity or HP may pull the plug.  G Now maybe it is communication problem within Compaq that this has neveraE filtered down the tree.  Or maybe it is not meeting with overwhelmingrC approval in the corridors of Compaq.  Neither of these alternativesmD reflects well on Compaq's management.  Either they can't communicateH within their own organisation or management are making decisions that do not impress the workers.  H I do realise that I may be putting you in an difficult position here.  IE really don't mind if you post something saying that the merger is theUD best thing since sliced bread (or whatever preceeded sliced bread). H Alternatively I don't mind if you decide to say nothing, in which case I3 will say "Say no more. Nudge, nudge.  Wink, wink !"      John McLeans   ------------------------------  % Date: Thu, 29 Nov 2001 09:38:23 +0000 % From: Alan Greig <a.greig@virgin.net>w= Subject: Re: Gartner and IDC say HP will effectively kill DLTd8 Message-ID: <6cvb0us1lt7qqsjqghfqb5db54i5fbli22@4ax.com>  B On 28 Nov 2001 14:12:21 GMT, Chris Petersen <havoc@apk.net> wrote:   >. >cL >Heck, I spec'ed out some new HP 9000 A-class hardware for them a few monthsL >back and HP wouldn't even offer me an Ultrium drive in their "Unix" storageJ >lineup.  All they have for Unix & HP 3000-class machines are DATs and DLTL >drives.  If this is going to happen HP's really going to have to get behindK >Ultrium and push it, something they haven't done since they introduced the: >format 2-3 years ago...  F Hmm HP have no problem offering us Ultrium drives on HP-UX systems and6 I know of someone with an Ultrium on an HP3000 system.  	 And go toga http://www.products.storage.hp.com/eprise/main/storage/tapebackup/selectionPages/tapeTechComp.htmn6 then tell me that HP aren't favouring Ultrium over DLT  B Plus I can find no reference to SuperDLT products on HP's web site under tape storage.e  B Finally Ultrium has only been around a few months - not 2-3 years.  A In summary I cannot make sense of your experience or post at all.     f -- Alan   ------------------------------   Date: 29 Nov 2001 15:42:40 GMT$ From: Chris Petersen <havoc@apk.net>= Subject: Re: Gartner and IDC say HP will effectively kill DLT ( Message-ID: <9u5l1g$dmk$1@plonk.apk.net>  6 In comp.sys.dec Alan Greig <a.greig@virgin.net> wrote:D : On 28 Nov 2001 14:12:21 GMT, Chris Petersen <havoc@apk.net> wrote:   :> :>M :>Heck, I spec'ed out some new HP 9000 A-class hardware for them a few months1M :>back and HP wouldn't even offer me an Ultrium drive in their "Unix" storage K :>lineup.  All they have for Unix & HP 3000-class machines are DATs and DLTeM :>drives.  If this is going to happen HP's really going to have to get behind-L :>Ultrium and push it, something they haven't done since they introduced the :>format 2-3 years ago...   H : Hmm HP have no problem offering us Ultrium drives on HP-UX systems and8 : I know of someone with an Ultrium on an HP3000 system.  K And yet the sales staff in Cleveland, OH, won't offer such a combination atrF all.  Furthermore, HP's own online HP9000 configuration & price quoter@ didn't offer the combinations when I used it last 3 weeks ago...  B Finally, the Ultrium drive is only available in the "Surestore" HPB packaging, which is primarily targeted at their PC & non-HP serverK customers.  The preferred storage line for the "Enterprise" product line isoK the SMART Storage system.  Look at the DLT8000 drive page for an example of. what I'm talking about:   b (http://www.products.storage.hp.com/eprise/main/storage/DisplayPages/summary.htm?DataPage=dlt8000)  G "    * HP SMART Storage  products are optimized for customers with UNIXaE computing environments, including HP-UX and Solaris, and who purchasel products directly from HP."t  I SMART storage is all your average HP9000/3000 sales rep will push at you,aJ from what my experience has been.  And it should be noted that my employerH (EDS PLM Solutions, formerly Unigraphics Solutions) has been one of HP'sJ larger VARs for years, so I would feel confident that our customer base isK respective of the larger HP9000 base.  That said, HP will sell you whateveruI you're willing to buy, and if Ultrium is what the customer wants, they'll L get you Ultrium.  It just takes the dumb sales rep another 2 weeks to figureL out how to order a new server from the factory that way (personal experience4 here, on Surestore vs. SMART storage product lines).   : And go to c : http://www.products.storage.hp.com/eprise/main/storage/tapebackup/selectionPages/tapeTechComp.htm-8 : then tell me that HP aren't favouring Ultrium over DLT  H No, it looks to me like a simple page showing the plusses of the variousK formats they offer.  For example, they have some glowing things to say eventG about 9840, a format that doesn't see a whole lot of use outside of IBM:H environments (last I checked).  On the other hand, the specs for Ultrium look better than before...  D : Plus I can find no reference to SuperDLT products on HP's web site : under tape storage.r  D Last word I heard on this one was that Quantum (or whomever owns theG mechanism rights now) was charging an arm & a leg for SuperDLT.  Last IoK checked, none of the major system vendors were offering OEM-versions of thehK SuperDLT drives yet, and I believe it was for just such a reason.  If I caneH find some reference to this for you I'll post it later, but that was the word through the grapevine.   D : Finally Ultrium has only been around a few months - not 2-3 years.  A Ultrium is one of 2 implementations of LTO technology, or "Linear-I Tape-Open".  LTO was/is a consortium of IBM, Seagate, and HP to build neweI tape formats based on a common set of open standards.  As you'll see fromaH the link below, Ultrium & Accelis (a dual-reel LTO format) licenses haveJ been available to manufacturers since April of 1998!  I've been personally: following the LTO formats since about September of 1998.    ) http://www.ltotechnology.com/license.htmlo   Licensing participants include:o  5 http://www.ltotechnology.com/licensing/lic_parts.htmlr  J Ultrium is a DLT-style implementation, as favored by HP & Seagate, whereasJ Accelis is a nod to IBM's Magstar line, where the tape format is dual-reelK and the "home" position of the tape is in the middle of the reel (i.e. withc0 both reels containing half of the overall tape).  J HP has only this year finally introduced Ultrium drives, so that may tell L you something about the hurdles the format has faced.  I'm not sure when theH first drives were introduced or who made them, but I seem to recall that: Ultrium drives were available since at least fall of 2000.  K However, I did note that when I reviewed the Ultrium drives @ HP's web sitelF just now, they are offering drives specifically for HP 9000 servers.  L Somebody should clue in their sales reps on that fact!  The problem is that J HP's "Enterprise" sales reps (9000 series HP-UX boxen) have had the SMART K storage product line all to themselves - it's built using HP components by  O their division, thus they see the profits on their bottom line.  The Surestore oO stuff used to be a seperate division (might not anymore), hence when they sold  N Surestore stuff they were passing profit out of their bottom line to the other: division, which didn't really help them directly enough...  C : In summary I cannot make sense of your experience or post at all.u  K Hopefully this helps comfort your sense.  I've been working with enterpriselJ server installations for about 7 years now, which is a short time comparedG with many here but quite some time nonetheless.  Hopefully I've clearedaL things up and given you some reading for more research.  Personally, UltriumJ looks like an interesting and potentially promising format.  However, I'veL always been a big fan of DLT (I've got an older DLT-IIIXT drive here at homeE for my own personal backups), and if SuperDLT can be competitive withtF Ultrium but provide some level of backward compatibility, then all the better.s   Christ   -- E Chris Petersen? Systems Engineer, Industry Services, Unigraphics Solutions Inc.h% Co-founder, auctions.workstations.orgt Email: havoc@apk.net       :    : -- : Alan   ------------------------------   Date: 29 Nov 2001 09:24:43 GMT3 From: gartmann@immunbio.mpg.de (Christoph Gartmann)m1 Subject: Installing Icon permanently to CDE menu? 0 Message-ID: <9u4usr$jph$1@n.ruf.uni-freiburg.de>   Hello,  H dragging a command-procedure or a program from the file manager over theI "install icon" item in a CDE workspace adds the icon to the correspondingaL menu. But as soon as I log out and log in again the icon has disappeared andJ I have to reinstall it again. I tried to save my current session, but thisO didn't change anything. Thus, how can I make my change permanent? OpenVMS 7.1-2e and Decwindows-Motif 1.2-5.    Regards,    Christoph Gartmanne  H -- --------------------------------------------------------------------+H | Max-Planck-Institut fuer      Phone   : +49-761-5108-464   Fax: -452 |H | Immunbiologie                                                        |H | Postfach 1169                 Internet: gartmann@immunbio.mpg.de     |H | D-79011  Freiburg, FRG                                               |H +--------- http://www.immunbio.mpg.de/home/english/menue.html ---------+   ------------------------------   Date: 29 Nov 2001 10:30:50 GMT3 From: vance@alumni.caltech.edu (Vance R. Haemmerle)m5 Subject: Re: Installing Icon permanently to CDE menu? , Message-ID: <9u52oq$jeg@gap.cco.caltech.edu>  0 In article <9u4usr$jph$1@n.ruf.uni-freiburg.de>,4 Christoph Gartmann <gartmann@immunbio.mpg.de> wrote: >Hello,h > I >dragging a command-procedure or a program from the file manager over thenJ >"install icon" item in a CDE workspace adds the icon to the correspondingM >menu. But as soon as I log out and log in again the icon has disappeared andhK >I have to reinstall it again. I tried to save my current session, but thisfP >didn't change anything. Thus, how can I make my change permanent? OpenVMS 7.1-2 >and Decwindows-Motif 1.2-5.  F   Doing this should change the file [.DT.SESSIONS]DTWMFP.SESSION underM your login directory.  Is this directory or file not writable and/or not have L delete prot?  I seem to remember a patch to Motif (earlier version 1.2-3 andF 1.2-4 or so) that fixed a session update problem because of directory  protection..   -- Vance Haemmerle  vance@alumni.caltech.edu   ------------------------------   Date: 29 Nov 2001 13:38:24 GMT3 From: gartmann@immunbio.mpg.de (Christoph Gartmann)n5 Subject: Re: Installing Icon permanently to CDE menu?a0 Message-ID: <9u5dog$od0$1@n.ruf.uni-freiburg.de>  b In article <9u52oq$jeg@gap.cco.caltech.edu>, vance@alumni.caltech.edu (Vance R. Haemmerle) writes:1 >In article <9u4usr$jph$1@n.ruf.uni-freiburg.de>, 5 >Christoph Gartmann <gartmann@immunbio.mpg.de> wrote:m >>Hello, >>J >>dragging a command-procedure or a program from the file manager over theK >>"install icon" item in a CDE workspace adds the icon to the corresponding N >>menu. But as soon as I log out and log in again the icon has disappeared andL >>I have to reinstall it again. I tried to save my current session, but thisQ >>didn't change anything. Thus, how can I make my change permanent? OpenVMS 7.1-2t >>and Decwindows-Motif 1.2-5., >eG >  Doing this should change the file [.DT.SESSIONS]DTWMFP.SESSION underiN >your login directory.  Is this directory or file not writable and/or not haveM >delete prot?  I seem to remember a patch to Motif (earlier version 1.2-3 and G >1.2-4 or so) that fixed a session update problem because of directory   >protection.  O The file and the directories are owned by me and the protections are all right.)N Nevertheless the contents of the file doesn't change although the modificationM date gets modified. But it contains only four lines with a single digit each.d   Regards,    Christoph Gartmann   H -- --------------------------------------------------------------------+H | Max-Planck-Institut fuer      Phone   : +49-761-5108-464   Fax: -452 |H | Immunbiologie                                                        |H | Postfach 1169                 Internet: gartmann@immunbio.mpg.de     |H | D-79011  Freiburg, FRG                                               |H +--------- http://www.immunbio.mpg.de/home/english/menue.html ---------+   ------------------------------  % Date: Thu, 29 Nov 2001 14:10:19 +0100u- From: Serge ZANGHERI <serge.zangheri@sema.fr>a Subject: LNC02 cont.' Message-ID: <3C0633BB.879F773F@sema.fr>e   ok brian, ok paul, I have VMS 6.1, DCPS-PLUS 1.0c,.D I installed a DCPS-OPEN license. I don  t have the DCPS-OPEN license! required message anymore. Only ateD the first job (with a reply/enable=printer) the unrecognized printer  message is returned immediatly..G I don t know how to avoid to get the DCPS-OPEN license required withouty this license, anyway.o' Now, I got a problem of delay to print.o  G The first job is printing very fast until the page previous of the lasts one (the last page of the docbG or of trailer). I mean The last page of the job is printing about aftere a few minutes.H The next job is printing about 10 minutes after !!!!!!!Actually when the printer say ok, I m ready.  E I tried the no_sync param as I saw in this forum, but nothing happen. " I tried interrupt_when_busy -> no." I didn t try the nofeed param yet.5 I putted the 2 LNC02 modules from the DCPS 2.0 -> no.n   Well do you help me again ?? Thanx a lot. Merci beaucoup Serge     ) "Brian Schenkenberger, VAXman-" a crit :   8 > In article <3C039534.C727528B@sema.fr>, Serge ZANGHERI  <serge.zangheri@sema.fr> writes:% > >Well, my problem is very weird....LD > >There is no Hubs to connect this printer directly on the network. >eB > Then how is it connected?????  Help us to better understand your environment.< > and we will be able to better help you address your issue. >s) > >This printer replace the dec2100 plus.1 > >nB > >So, the first job is printing well but just after the following message come back : "DCPS - W -3E > >unrecognize printer" and the others job never print until reset orm stop/que/manager..G > >It seems that after the first print job the printer give his name tot DCPS . And this name is not F > >recognize, so the followings jobs never print and the message about license required on DCPS-Openo > >appears.a@ > >Is there a way to disable the printer LNC02 to say his name ? >- > Perhaps... >-A > >Is there a param to configure in DCPS to avoid DCPS to see thel
 printer name?n@ > >May be this is a DCPS version problem, what is your version ? >oH > I am running V1.7.  Paul Anderson will likely recall the version which dida: > support the LNC02.  My guest, IIRC, is that is was V1.6. >n4 > >May be we need just the library for this printer.% > >Do you use the DCPS open product ?w >  > Yes. >g > --4 > VAXman- OpenVMS APE certification number: AAA-0001 VAXman(at)TMESIS(dot)COM >bE >   "And of course, I'm a genius, so people are naturally drawn to mya fierynD >   intellect.  Their admiration overwhelms their envy!" -- Calvin & Hobbes   ------------------------------  % Date: Thu, 29 Nov 2001 15:18:20 +0100f- From: Serge ZANGHERI <serge.zangheri@sema.fr>S Subject: Re: LNC02 cont.' Message-ID: <3C0643AC.D34BC4A5@sema.fr>e  J Well, if I ask again it s because I thought that worked before, but may be. not and I know that I got a version "problem".F So I just  would like to know what the jogging parameter do and if theF nofeed parameter could interact with my problem. For the moment all my3 prints are ansi, I did not try with PS file yet.... J Is there a parameter to force DCPS to cut connection after the job print ?G May be one solution is to declare in the print parameter a timeout withfC about 4 minutes and the user just have to wait to get the followingr print.........* to be continued but thanx to your support. Serge    Serge ZANGHERI a crit :   > ok brian, ok paul,! > I have VMS 6.1, DCPS-PLUS 1.0c,SF > I installed a DCPS-OPEN license. I don  t have the DCPS-OPEN license# > required message anymore. Only at F > the first job (with a reply/enable=printer) the unrecognized printer" > message is returned immediatly..I > I don t know how to avoid to get the DCPS-OPEN license required withoutR > this license, anyway.g) > Now, I got a problem of delay to print.  >oI > The first job is printing very fast until the page previous of the last  > one (the last page of the doctI > or of trailer). I mean The last page of the job is printing about after  > a few minutes.J > The next job is printing about 10 minutes after !!!!!!!Actually when the > printer say ok, I m ready. >$G > I tried the no_sync param as I saw in this forum, but nothing happen. $ > I tried interrupt_when_busy -> no.$ > I didn t try the nofeed param yet.7 > I putted the 2 LNC02 modules from the DCPS 2.0 -> no.g >t > Well do you help me again ?? > Thanx a lot. > Merci beaucoup > Serges >s+ > "Brian Schenkenberger, VAXman-" a crit :e >a: > > In article <3C039534.C727528B@sema.fr>, Serge ZANGHERI" > <serge.zangheri@sema.fr> writes:' > > >Well, my problem is very weird.....F > > >There is no Hubs to connect this printer directly on the network. > >cD > > Then how is it connected?????  Help us to better understand your
 > environmentr> > > and we will be able to better help you address your issue. > >s+ > > >This printer replace the dec2100 plus.b > > >oD > > >So, the first job is printing well but just after the following! > message come back : "DCPS - W -sG > > >unrecognize printer" and the others job never print until reset orc > stop/que/manager..I > > >It seems that after the first print job the printer give his name to. > DCPS . And this name is notnH > > >recognize, so the followings jobs never print and the message about > license required on DCPS-Opene
 > > >appears.tB > > >Is there a way to disable the printer LNC02 to say his name ? > >P > > Perhaps... > >rC > > >Is there a param to configure in DCPS to avoid DCPS to see the  > printer name? B > > >May be this is a DCPS version problem, what is your version ? > >OJ > > I am running V1.7.  Paul Anderson will likely recall the version which > dide< > > support the LNC02.  My guest, IIRC, is that is was V1.6. > >o6 > > >May be we need just the library for this printer.' > > >Do you use the DCPS open product ?0 > >: > > Yes. > >r > > --6 > > VAXman- OpenVMS APE certification number: AAA-0001 > VAXman(at)TMESIS(dot)COM > >4G > >   "And of course, I'm a genius, so people are naturally drawn to my  > fierypF > >   intellect.  Their admiration overwhelms their envy!" -- Calvin & > Hobbes   ------------------------------  % Date: Thu, 29 Nov 2001 10:55:40 -050010 From: Paul Anderson <paul.r.anderson@compaq.com> Subject: Re: LNC02 cont.; Message-ID: <291120011055404468%paul.r.anderson@compaq.com>   6 In article <3C0643AC.D34BC4A5@sema.fr>, Serge ZANGHERI <serge.zangheri@sema.fr> wrote:e  L > Well, if I ask again it s because I thought that worked before, but may be0 > not and I know that I got a version "problem".  # > > I have VMS 6.1, DCPS-PLUS 1.0c,nH > > I installed a DCPS-OPEN license. I don  t have the DCPS-OPEN license% > > required message anymore. Only at.H > > the first job (with a reply/enable=printer) the unrecognized printer$ > > message is returned immediatly..  7 > I putted the 2 LNC02 modules from the DCPS 2.0 -> no.   D You certainly do have a version problem.  DCPS V1.0C is very old.  IG understand you have the kit for DCPS V2.0, supplied by a support person0E at Compaq.  Please install it, not just put the LNC02 modules in youro device control library.a  F You get the "unrecognized printer" message because DCPS V1.0C does not support the LNC02 printer.  G > I tried the no_sync param as I saw in this forum, but nothing happen.o  8 The NO_SYNC logical name is not needed for this printer.  F What is the connection type you are using to this printer?  It must beE either AppleTalk or serial, as IP connections to this type of printer # were not supported until DCPS V1.4.o  H > So I just  would like to know what the jogging parameter do and if theH > nofeed parameter could interact with my problem. For the moment all my5 > prints are ansi, I did not try with PS file yet....V  @ Are you referring to the printer's jog setting?  That should notF produce the symptoms you are seeing.  And what nofeed parameter do you mean?t  L > Is there a parameter to force DCPS to cut connection after the job print ?I > May be one solution is to declare in the print parameter a timeout withhE > about 4 minutes and the user just have to wait to get the followingi > print.........  H DCPS normally closes the connection to the printer after the job prints.  C It is possible this is a printer or network card firmware problem. t2 What versions display on the printer startup page.   Paul   -- I  Paul Anderson   OpenVMS Engineeringu   Compaq Computer Corporationu   ------------------------------  % Date: Thu, 29 Nov 2001 17:46:50 +0100I- From: Serge ZANGHERI <serge.zangheri@sema.fr>  Subject: Re: LNC02 cont.' Message-ID: <3C06667A.2F07AF1E@sema.fr>    Paul Anderson a crit :n  8 > In article <3C0643AC.D34BC4A5@sema.fr>, Serge ZANGHERI! > <serge.zangheri@sema.fr> wrote:  >rN > > Well, if I ask again it s because I thought that worked before, but may be2 > > not and I know that I got a version "problem". >t% > > > I have VMS 6.1, DCPS-PLUS 1.0c,sJ > > > I installed a DCPS-OPEN license. I don  t have the DCPS-OPEN license' > > > required message anymore. Only attJ > > > the first job (with a reply/enable=printer) the unrecognized printer& > > > message is returned immediatly.. >v9 > > I putted the 2 LNC02 modules from the DCPS 2.0 -> no.  >,F > You certainly do have a version problem.  DCPS V1.0C is very old.  II > understand you have the kit for DCPS V2.0, supplied by a support person G > at Compaq.  Please install it, not just put the LNC02 modules in your' > device control library.w >l  K ok,  yes I m in contact with a support person at Compaq. The change of DCPSaN version is what I m going to suggest to the client, because I know about that.E About the LNC02 modules, I can undestand the version problem as well.h   > H > You get the "unrecognized printer" message because DCPS V1.0C does not > support the LNC02 printer.   I undestand too.   >tI > > I tried the no_sync param as I saw in this forum, but nothing happen.  >e: > The NO_SYNC logical name is not needed for this printer.   Ok, thanks.s   >h >iH > What is the connection type you are using to this printer?  It must beG > either AppleTalk or serial, as IP connections to this type of printerm% > were not supported until DCPS V1.4.e  K It is serial of course and I know about raw/tcpip and this old dcps version  (does not work together).o   >l >hJ > > So I just  would like to know what the jogging parameter do and if theJ > > nofeed parameter could interact with my problem. For the moment all my7 > > prints are ansi, I did not try with PS file yet....r  = I always did print /que=lnc02 login.com     (never a PS file)    >e >sB > Are you referring to the printer's jog setting?  That should notH > produce the symptoms you are seeing.  And what nofeed parameter do you > mean?e  M The DCPS$startup P5 parameter : /default=nofeed or feed.  (chapter 3.5 in ther4 MGUIDE). I really don t know if i have to change it.   >n >rN > > Is there a parameter to force DCPS to cut connection after the job print ?K > > May be one solution is to declare in the print parameter a timeout witheG > > about 4 minutes and the user just have to wait to get the followingt > > print......... >sJ > DCPS normally closes the connection to the printer after the job prints. >tD > It is possible this is a printer or network card firmware problem.4 > What versions display on the printer startup page. >a  O I already check with this contact at compaq : the lnc02 system release is 1.1.1 L rev.:3.11, the  version : 2011.26 rev:.2, the test build : 2.1. Should be ok right ?   O My new investigation : It seems there is a parameter in DCPS which delay job ofa 4 minutes by default.r' refer to D.13 Stall Timeouts of MGUIDE. G and 3.4.4 Purging the Symbiont Process's Working Set of MGUIDE as well.hC               The DCPS symbiont purges its working set after it hasiH               been idle for a period of time in order to conserve systemG               resources. The time delay is intended to help prevent the F               system from thrashing by keeping the program in physical6               memory while more work is apt to arrive.D               By default, DCPS waits ten (10) minutes after becomingH               idle before purging its working set. You can increase thisC               value, if desired, by defining a system-wide logical:sG               $ DEFINE/SYSTEM/EXECUTIVE DCPS$PURGE_TIME "0 hh:mm:ss.00"s  C and 5-6 Manipulating Print Jobs and Print Queues of MGUIDE as well.sF                 ________________________Note  ________________________@                 When communications problems arise with a serialB                 or raw TCP/IP interconnect, the symbiont will keepB                 listening for up to 4 minutes before disconnectingC                 from the printer. Therefore, under some conditions, C                 the device is not released immediately after a STOP /                 /QUEUE/RESET command is issued.   L Can I decrease these values ? It is exactly the times I have to wait betweenF jobs. 4 minutes to the end of the fisrt one and 10 before begin of the+ following. May be there is no link, but....2  P I would like to do some test again..... and suggest to change version anyway, ok ?s% So if you can help me one more time ? 	 Merci :-D    >o > Paul >c > -- >  Paul Anderson >   OpenVMS Engineeringu >   Compaq Computer Corporationa   ------------------------------  # Date: Thu, 29 Nov 2001 17:45:15 GMTr= From: system@SendSpamHere.ORG (Brian Schenkenberger, VAXman-)  Subject: Re: LNC02 cont.0 Message-ID: <00A05C5C.286A9F1C@SendSpamHere.ORG>  n In article <291120011055404468%paul.r.anderson@compaq.com>, Paul Anderson <paul.r.anderson@compaq.com> writes:7 >In article <3C0643AC.D34BC4A5@sema.fr>, Serge ZANGHERI.  ><serge.zangheri@sema.fr> wrote:
 >{...snip...}SG >What is the connection type you are using to this printer?  It must belF >either AppleTalk or serial, as IP connections to this type of printer$ >were not supported until DCPS V1.4.  F I asked this guy to explain this after his first posting.  Unless I'veE missed the posting, I do not believe he has compied with the request.    --O VAXman- OpenVMS APE certification number: AAA-0001     VAXman(at)TMESIS(dot)COM-            -J   "And of course, I'm a genius, so people are naturally drawn to my fiery I   intellect.  Their admiration overwhelms their envy!" -- Calvin & Hobbes    ------------------------------  % Date: Thu, 29 Nov 2001 13:43:47 -0500g0 From: Paul Anderson <paul.r.anderson@compaq.com> Subject: Re: LNC02 cont.; Message-ID: <291120011343479785%paul.r.anderson@compaq.com>h  6 In article <3C06667A.2F07AF1E@sema.fr>, Serge ZANGHERI <serge.zangheri@sema.fr> wrote:o  ? > I always did print /que=lnc02 login.com     (never a PS file)n  G By the time your LOGIN.COM gets to the printer, it's PostScript becauses& the DCPS ANSI translator converted it.  O > The DCPS$startup P5 parameter : /default=nofeed or feed.  (chapter 3.5 in then6 > MGUIDE). I really don t know if i have to change it.  ) I've never had to fiddle with this value.   K > I already check with this contact at compaq : the lnc02 system release isSG > 1.1.1 rev.:3.11, the  version : 2011.26 rev:.2, the test build : 2.1.   C You are running the latest printer code.  Look for the network code A version.  The latest two versions are 5.23 and 5.49 (test build).s  ; > $ DEFINE/SYSTEM/EXECUTIVE DCPS$PURGE_TIME "0 hh:mm:ss.00"   B This is a possible performance issue and appears unrelated to your problem.  2 > When communications problems arise with a serial4 > or raw TCP/IP interconnect, the symbiont will keep4 > listening for up to 4 minutes before disconnecting5 > from the printer. Therefore, under some conditions,-5 > the device is not released immediately after a STOPa! > /QUEUE/RESET command is issued.  >4 > Can I decrease these values ?m  G I am intimately familiar with this timeout due to recent work I've beentD doing.  This is hardcoded and, incidentally, will be lowered in DCPS V2.1.   C This would not affect your situation unless you were issuing a STOP-% /QUEUE /RESET command after each job.i   Paul   -- -  Paul Anderson   OpenVMS Engineering-   Compaq Computer Corporation:   ------------------------------  % Date: Thu, 29 Nov 2001 12:56:32 +01002- From: Serge ZANGHERI <serge.zangheri@sema.fr>e0 Subject: Re: LNC02 problem with DCPS  was: LNC02' Message-ID: <3C062270.64F481C8@sema.fr>I   ok brian, ok paul, I have VMS 6.1, DCPS-PLUS 1.0c,yf I installed a DCPS-OPEN license. I don  t have the DCPS-OPEN license required message anymore. Only ate the first job (with a reply/enable=printer) the unrecognized printer message is returned immediatly.. ] I don t know how to avoid to get the DCPS-OPEN license required without this license, anyway. ' Now, I got a problem of delay to print.l  e The first job is printing very fast until the page previous of the last one (the last page of the docuV or of trailer). I mean The last page of the job is printing about after a few minutes.c The next job is printing about 10 minutes after !!!!!!!Actually when the printer say ok, I m ready.   E I tried the no_sync param as I saw in this forum, but nothing happen.t" I tried interrupt_when_busy -> no." I didn t try the nofeed param yet.5 I putted the 2 LNC02 modules from the DCPS 2.0 -> no.S   Well do you help me again ?? Thanx a lot. Merci beaucoup Serge@    ) "Brian Schenkenberger, VAXman-" a crit :f  Y > In article <3C039534.C727528B@sema.fr>, Serge ZANGHERI <serge.zangheri@sema.fr> writes:e% > >Well, my problem is very weird....hD > >There is no Hubs to connect this printer directly on the network. > N > Then how is it connected?????  Help us to better understand your environment< > and we will be able to better help you address your issue. >i) > >This printer replace the dec2100 plus.g > >eb > >So, the first job is printing well but just after the following message come back : "DCPS - W -X > >unrecognize printer" and the others job never print until reset or stop/que/manager..c > >It seems that after the first print job the printer give his name to DCPS . And this name is notid > >recognize, so the followings jobs never print and the message about license required on DCPS-Open > >appears. @ > >Is there a way to disable the printer LNC02 to say his name ? >S > Perhaps... >nO > >Is there a param to configure in DCPS to avoid DCPS to see the printer name? @ > >May be this is a DCPS version problem, what is your version ? >iL > I am running V1.7.  Paul Anderson will likely recall the version which did: > support the LNC02.  My guest, IIRC, is that is was V1.6. > 4 > >May be we need just the library for this printer.% > >Do you use the DCPS open product ?A >0 > Yes. >a > --Q > VAXman- OpenVMS APE certification number: AAA-0001     VAXman(at)TMESIS(dot)COMm > K >   "And of course, I'm a genius, so people are naturally drawn to my fiery K >   intellect.  Their admiration overwhelms their envy!" -- Calvin & Hobbes0   ------------------------------  % Date: Thu, 29 Nov 2001 09:57:29 +0100n( From: "Pino Gargiulo" <jk3380@naida.org>* Subject: Re: logical names novice question2 Message-ID: <9u4rkn$5pn$1@fe1.cs.interbusiness.it>  - "jlsue" <jlsuexxxz@home.com> wrote in messaged2 news:0ada0ug386b0fbif2q68s7cckbjpaavnv5@4ax.com...D > If logicals are used correctly in your application, you can alwaysG > check-for, and remove any process-level logicals that try to overrider > system-level logicals. >oD > However, I've found this process-level override to be very useful.= > Taking a cue from the JCL DD statements, I began developingrF > applications such that all data files are opened via a logical name.? > Now, for most users this logical name is in the system table. : > However, when there's something I need to do in testing,H > troubleshooting, etc., I would define my process-level logical name toH > point to a non-production data file instead.  This eliminated the needH > for commenting code between development->test->production (and all theD > inherent possibilities of forgetting to comment important lines ofD > code).  The code would remain *exactly* the same between the three  I Thanks for all the replies! Logicals are even more powerfully than what Is
 understoodK from the manual. Rats, I only use VMS for fun and so getting something REALs done# will full power logical is not easy    /PGt   ------------------------------  # Date: Thu, 29 Nov 2001 12:04:41 GMTe= From: system@SendSpamHere.ORG (Brian Schenkenberger, VAXman-)i) Subject: Mozilla DECWserver crash probleme0 Message-ID: <00A05C2C.9450F207@SendSpamHere.ORG>  G Just wanted to let everyone know that I can now run Mozilla on my Alpha G without it crashing the DECWserver.  The Alpha has a ZLXp-E3 card whicheH is 24 bits.  There was an override, DECW$SERVER_DEFAULT_VISUAL_CLASS, inG the DECW$PRIVATE_SERVER_SETUP.COM to cause this card to use PseudoColorwH or 8 bits.  I've removed this and Mozilla now runs -- it's a slow, pain- fully slow, pig but it runs.  G Thanks to George Watson who sent me an email describing his pains usingrF the DECWindows server with the same consequences when running Mozilla.  H Now Mozilla needs only a little^Jlarge amount of performance tweaking or
 optimization.S --O VAXman- OpenVMS APE certification number: AAA-0001     VAXman(at)TMESIS(dot)COM              J   "And of course, I'm a genius, so people are naturally drawn to my fiery I   intellect.  Their admiration overwhelms their envy!" -- Calvin & Hobbesr   ------------------------------  % Date: Thu, 29 Nov 2001 09:56:22 +0000f% From: Alan Greig <a.greig@virgin.net>a? Subject: Re: New product allows VMS to become part of a PC Lan!I8 Message-ID: <td1c0ugl6075tq6p0tqjp2hhm8g7b393ed@4ax.com>  F On 28 Nov 2001 08:36:22 -0600, koehler@encompasserve.org (Bob Koehler) wrote:  i >In article <d7791aa1.0111271509.4f1bf490@posting.google.com>, bob@instantwhip.com (Bob Ceculski) writes:O > F >> Just ran across this amazing product as we were looking to make our >> AlphaB >> VMS system part of our local PC Lan and ran across this amazing >> product ...H >> they also have a product that allows VMS pc disk access ... amazing! 
 >> here is' >> the link if anyone is interested ...n > I >   Nothing like being 15 years behind the times, and 10 years behind the 
 >   freeware.   F I thought at first surely this is just Pathworks or Samba but actuallyF the product Bob describes allows VMS systems to access NT print queues@ and shares which is the reverse of the functionality provided byC Pathworks and Samba. Although I have heard that Samba may have some:5 ability to allow a user to access NT shares from VMS.d -- Alan   ------------------------------    Date: 29 Nov 2001 05:52:31 -0800( From: bob@instantwhip.com (Bob Ceculski)? Subject: Re: New product allows VMS to become part of a PC Lan! = Message-ID: <d7791aa1.0111290552.616d67b3@posting.google.com>a  h Kilgallen@SpamCop.net (Larry Kilgallen) wrote in message news:<Mb4QyUKM4nWY@eisner.encompasserve.org>...e > In article <LKB0LPX64wO1@eisner.encompasserve.org>, koehler@encompasserve.org (Bob Koehler) writes:il > > In article <d7791aa1.0111271509.4f1bf490@posting.google.com>, bob@instantwhip.com (Bob Ceculski) writes: > > H > >> Just ran across this amazing product as we were looking to make our
 > >> AlphaD > >> VMS system part of our local PC Lan and ran across this amazing > >> product ...J > >> they also have a product that allows VMS pc disk access ... amazing!  > >> here is) > >> the link if anyone is interested ...u > > L > >    Nothing like being 15 years behind the times, and 10 years behind the > >    freeware. > @ > It looks suspiciously like the first message I ever saw postedC > by Bob Ceculski (a few months ago, before he was in my killfile).r > @ > Perhaps he is repeating the information about his product just2 > to let us know he has learned how to capitalize.  M I guess you don't read your posts then, because the first product I mentioneddH was a completely different one and a different company than this one ...   ------------------------------  % Date: Thu, 29 Nov 2001 07:56:17 -0800t' From: David Mathog <mathog@caltech.edu>o? Subject: Re: New product allows VMS to become part of a PC Lan! + Message-ID: <3C065AA1.36AC48AB@caltech.edu>n   Alan Greig wrote:p > H > On 28 Nov 2001 08:36:22 -0600, koehler@encompasserve.org (Bob Koehler) > wrote: > k > >In article <d7791aa1.0111271509.4f1bf490@posting.google.com>, bob@instantwhip.com (Bob Ceculski) writes:c > >tH > >> Just ran across this amazing product as we were looking to make our
 > >> AlphaD > >> VMS system part of our local PC Lan and ran across this amazing > >> product ...I > >> they also have a product that allows VMS pc disk access ... amazing!o > >> here is) > >> the link if anyone is interested ...  > >1K > >   Nothing like being 15 years behind the times, and 10 years behind theL > >   freeware.g > H > I thought at first surely this is just Pathworks or Samba but actuallyH > the product Bob describes allows VMS systems to access NT print queuesB > and shares which is the reverse of the functionality provided byE > Pathworks and Samba. Although I have heard that Samba may have somem7 > ability to allow a user to access NT shares from VMS.n  H Samba has smbclient allows you to do a lot of things from the VMS side.  For fileD access/transfer it's not nearly as handy as some sort of pseudo-disk driver wouldG be.  Anyway, here's some code for queueing print jobs through smbcliente
 to windows	 printers:h  E   ftp://saf.bio.caltech.edu/pub/software/openvms/print_to_windows.zipV   Regards,   David Mathog mathog@caltech.edu   ------------------------------   Date: 29 Nov 2001 16:55:55 GMT1 From: bill@triangle.cs.uofs.edu (Bill Gunshannon)u? Subject: Re: New product allows VMS to become part of a PC Lan!h, Message-ID: <9u5par$19so$1@info.cs.uofs.edu>  8 In article <td1c0ugl6075tq6p0tqjp2hhm8g7b393ed@4ax.com>,(  Alan Greig <a.greig@virgin.net> writes: |>I |> I thought at first surely this is just Pathworks or Samba but actuallycI |> the product Bob describes allows VMS systems to access NT print queuesrC |> and shares which is the reverse of the functionality provided bysF |> Pathworks and Samba. Although I have heard that Samba may have some8 |> ability to allow a user to access NT shares from VMS. |>  D smbclient (the client side of SAMBA) does this. I don't know if thatC was implemented on VMS as I have never looked at VMS SAMBA.  We usem+ NFS and LPR for printing.  Works just fine.t   bill   -- oJ Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesD bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. University of Scranton   |A Scranton, Pennsylvania   |         #include <std.disclaimer.h>       ------------------------------    Date: 29 Nov 2001 09:19:24 -0800/ From: jeremy.lakey@ndchealth.com (jeremy lakey)a? Subject: Re: New product allows VMS to become part of a PC Lan!t= Message-ID: <70a82a12.0111290919.273b03ce@posting.google.com>   e Alan Greig <a.greig@virgin.net> wrote in message news:<td1c0ugl6075tq6p0tqjp2hhm8g7b393ed@4ax.com>...cH > I thought at first surely this is just Pathworks or Samba but actuallyH > the product Bob describes allows VMS systems to access NT print queuesB > and shares which is the reverse of the functionality provided byE > Pathworks and Samba. Although I have heard that Samba may have some 7 > ability to allow a user to access NT shares from VMS.S  F samba works both ways, with both a client (accessing NT shares & printF queues from VMS) and server (accessing VMS directories from NT) and it has for a long time now...   ------------------------------    Date: 29 Nov 2001 09:36:42 -07001 From: nothome@spammers.are.scum (Malcolm Dunnett),? Subject: Re: New product allows VMS to become part of a PC Lan!k, Message-ID: <UyNhASZJxsIM@malvm6.mala.bc.ca>  9 In article <td1c0ugl6075tq6p0tqjp2hhm8g7b393ed@4ax.com>, y,      Alan Greig <a.greig@virgin.net> writes:  H > On 28 Nov 2001 08:36:22 -0600, koehler@encompasserve.org (Bob Koehler) > wrote: > j >>In article <d7791aa1.0111271509.4f1bf490@posting.google.com>, bob@instantwhip.com (Bob Ceculski) writes: >>G >>> Just ran across this amazing product as we were looking to make oure	 >>> AlpharC >>> VMS system part of our local PC Lan and ran across this amazingm >>> product ... I >>> they also have a product that allows VMS pc disk access ... amazing! t >>> here ist( >>> the link if anyone is interested ... >>J >>   Nothing like being 15 years behind the times, and 10 years behind the >>   freeware. > H > I thought at first surely this is just Pathworks or Samba but actuallyH > the product Bob describes allows VMS systems to access NT print queuesB > and shares which is the reverse of the functionality provided byE > Pathworks and Samba. Although I have heard that Samba may have some 7 > ability to allow a user to access NT shares from VMS.I  F     Multiware used to have such a feature for Novell print queues, butJ I've not seen anything else that does this for SMB print queues. PathworksJ certainly doesn't have this capability. I recall Samba having an SMBCLIENTE program ( which I'd call the equivalent of "FTP for SMB networks", it J provided an interactive utility to copy files but didn't allow transparentH access to SMB shares ). I don't believe SMBCLIENT did anything for print7 queues, but I didn't spend too much time looking at it.   H     Most of our printers are network connected and we print to them fromE VMS via IP, but it seems this product could be very useful if you had F a bunch of printers connected directly to servers or workstations that! you wanted VMS to have access to.-  G    If there are other commercial products or freeware that provide this.6 functionality I'd be interested in hearing about them.   ------------------------------    Date: 29 Nov 2001 08:55:02 -0800/ From: jeremy.lakey@ndchealth.com (jeremy lakey) # Subject: new to VMS cxx question....= Message-ID: <70a82a12.0111290855.2f59ef57@posting.google.com>N  8 I've got the compaq xml libraries loaded...   they're at $1$dga41:[ebus.xml]   & how do i include that path as headers?  @ cxx /include_directory=$1$DGA41:[EBUS.XML.XERCES-C-SRC1_3_0.SRC] testxml.cppw  = does not work... i've tried to define it as a system include:  $show logical cxx*E    "CXX$SYSTEM_INCLUDE" = "$1$DGA41:[EBUS.XML.XERCES-C-SRC1_3_0.SRC]"p  F that doesn't work..  i'm new to cxx and vms (as in just this week new)C but i've read all the documentation at www.openvms.compaq.com i can ; find and other than using /include_directory i'm still lostn  & any help would be greatly appreciated.   Thanks   Jeremy Lakey Software Architect
 NDC Health   ------------------------------   Date: 29 Nov 2001 09:00:51 GMT) From: leslie@clio.rice.edu (Jerry Leslie)L* Subject: NT to Solaris Migration Promotion' Message-ID: <9u4tg3$lh8$1@joe.rice.edu>eG Keywords: sun,tradein,for,Windows,servers,marketing,marketing,MARKETINGo  E A Sun person posted the attached article announcing Sun's trade-in of ) Wintel Cartel boxes on Sun Netra servers.i  B Obviously Sun knows how to make something from the Gartner Group'sD suggestion about leaving IIS as a webserver. From the first web page in the attached article:  <   "Compared to Wintel systems, Sun brings the advantages of:J    Improved security with minimal reports of virus attacks on Sun Solaris"   To HWP  & CPQ:  E    This is called "marketing". It CAN be used for non-Wintel systems.b  A The odds are better on a winning lottery ticket than there is fore  such a marketing effort for VMS.    4 --Jerry Leslie     (my opinions are strictly my own)N ==============================================================================% From: drew.wrenn@sun.com (Drew Wrenn) A Newsgroups: comp.unix.solaris,alt.solaris.x86,comp.sys.sun.admin,l             comp.sys.sun.misc * Subject: NT to Solaris Migration Promotion  Date: 28 Nov 2001 14:57:52 -0800' Organization: http://groups.google.com/ 	 Lines: 15-= Message-ID: <4ebe311b.0111281457.51b4cf78@posting.google.com>E NNTP-Posting-Host: 192.18.243.4 , Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bitd7 X-Trace: posting.google.com 1006988272 15407 127.0.0.1 n"         (28 Nov 2001 22:57:52 GMT)( X-Complaints-To: groups-abuse@google.com+ NNTP-Posting-Date: 28 Nov 2001 22:57:52 GMTQ   Trade up and save up to $3,000!i  3 For a limited time only, developers can save up to -5 $3,000 for a reliable web server solution.  Sun will -3 allow trade-ins of current servers for a Sun Netra  . T1 or Netra X1 rack-optimized thin(1U) server.  / For more information on this promotion, visit: e  % http://www.sun.com/netra/tradeuppromo   5 To learn more about Sun's FREE tools and resources ton8 facilitate migration from NT to the Solaris OE, visit:     http://www.sun.com/migration/ntl   ------------------------------  # Date: Thu, 29 Nov 2001 12:59:25 GMTrG From: Simon Clubley <simon_clubley@remove_me.altavista.co.uk-Earth.UFP>D. Subject: Re: NT to Solaris Migration Promotion6 Message-ID: <NiqN7.42081$xS6.70602@www.newsranger.com>  J On 29 Nov 2001 09:00:51 GMT, in article <9u4tg3$lh8$1@joe.rice.edu>, Jerry
 Leslie wrote:h >oF >   This is called "marketing". It CAN be used for non-Wintel systems. > B >The odds are better on a winning lottery ticket than there is for! >such a marketing effort for VMS., >m  @ That's too optimistic. :-) To tie this to another c.o.v. thread:  K General VMS marketing is as likely as an insurance company agreeing to sella life insurance to Bin Laden.   Simon.   -- C@ Simon Clubley, simon_clubley@remove_me.altavista.co.uk-Earth.UFPK In the task of removing Microsoft from the marketplace, I have discovered atE truly remarkable plan, but this signature is too small to contain it.r   ------------------------------  # Date: Thu, 29 Nov 2001 14:10:44 GMTo3 From: "Tom Wade" <t.wade@vms.eurokom.ie.removespam> . Subject: Re: NT to Solaris Migration Promotion, Message-ID: <ElrN7.1413$_3.6464@news.iol.ie>  6 "Jerry Leslie" <leslie@clio.rice.edu> wrote in message! news:9u4tg3$lh8$1@joe.rice.edu...  > ... > >   "Compared to Wintel systems, Sun brings the advantages of:L >    Improved security with minimal reports of virus attacks on Sun Solaris" >r > To HWP  & CPQ: > G >    This is called "marketing". It CAN be used for non-Wintel systems.r >vC > The odds are better on a winning lottery ticket than there is foro" > such a marketing effort for VMS.  K Which is sadly ironic given that most of the world's lottery systems run onb OpenVMS.  L ---------------------------------------------------------------------------- --A Tom Wade    | EMail: T.Wade@vms.eurokom.ie  (all domain mailers). G EuroKom     | X400:  g=tom;s=wade;o=eurokom;p=eurokom;a=eirmail400;c=ie & 30, Dale Rd | Tel:   +353 (1) 278-7878& Stillorgan  | Fax:   +353 (1) 278-78793 Co Dublin   | Disclaimer:  This is not a disclaimern@ Ireland     | Tip:         "Friends don't let friends do Unix !"   ------------------------------    Date: 29 Nov 2001 06:02:35 -0800) From: P.Young@unsw.EDU.AU (Patrick Young))8 Subject: Re: Obtaining System Serial Number via Software= Message-ID: <55f85d77.0111290602.599e6998@posting.google.com>o  h "Mark Buda" <buda@tabasco.zko.dec.no.spam.com> wrote in message news:<9u1g84$9bam$1@lead.zk3.dec.com>...F > This does not go back very far, so has limited ability.  It works onJ > wildfire systems with all patches and V7.3 on, from what I can remember. > 2 > star> write sys$output f$getsyi("serial_number")" > 00000000000000000000000000000000 >   F That is sooooo weird! I am probably missing something here (there mustH be a good reason for it :-). serial_number will work on Itanic systems??   We now have:  / 1. f$getsyi("sid") is disabled on Alpha systemss@ 2. Alpha systems may not have a serial number anyway (well the 4&    different models I've tried didn't)E 3. Another f$getsyi item is introduced sometime between 7.2-1 and 7.3,F    (but not documented under 7.3) to read the (missing) serial number   A $ write sys$output f$getsyi("version"), f$getsyi("serial_number")c( V7.3    00000000000000000000000000000000  & H write sys$output f$getsyi("version") 7.2-1n, H write sys$output f$getsyi("serial_number")A %DCL-W-IVKEYW, unrecognized keyword - check validity and spellinge  \SERIAL_NUMBER\   ------------------------------  % Date: Thu, 29 Nov 2001 11:31:42 -0500i  From: norm.raphael@jamesbury.comA Subject: Re: OpenVMS VAXSHAD03_072 VAX V7.2 SHADOWING ECO Summaryl4 Message-ID: <C2256B13.005A7483.00@jklh21.valmet.com>  3 I just found out why I couldn't download this patche' (I'm sorry if mail wraps this for you).   ' Tracking down this stuff is enervating,a to say the least. -Norm4  N The five lines below should read as follows (without the included "/rtr/4.1"):  N <a href=/patches/public/vms/vax/v7.2/vaxshad03_072.README>vaxshad03_072.README </a>N <a href=/patches/public/vms/vax/v7.2/vaxshad03_072.CHKSUM>vaxshad03_072.CHKSUM </a>> <a href=/patches/public/vms/vax/v7.2/vaxshad03_072.CVRLET_TXT> vaxshad03_072.CVRLET_TXT</a>@ <a href=/patches/public/vms/vax/v7.2/vaxshad03_072.a-dcx_vaxexe> vaxshad03_072.a-dcx_vaxexe</a>> <a href=/patches/public/vms/vax/v7.2/vaxshad03_072.CVRLET_TXT> vaxshad03_072.CVRLET_TXT</a>   in the following block insiden <URL:eO http://ftp.support.compaq.com/patches/public/vms/vax/v7.2/vaxshad03_072.README>eI (and as an aside, it seems the <pre></pre> is unneeded, though harmless):y   <pre>^B <a href=/patches/public/vms/vax/v7.2/rtr/4.1/vaxshad03_072.README> vaxshad03_072.README</a>B <a href=/patches/public/vms/vax/v7.2/rtr/4.1/vaxshad03_072.CHKSUM> vaxshad03_072.CHKSUM</a>F <a href=/patches/public/vms/vax/v7.2/rtr/4.1/vaxshad03_072.CVRLET_TXT> vaxshad03_072.CVRLET_TXT</a>H <a href=/patches/public/vms/vax/v7.2/rtr/4.1/vaxshad03_072.a-dcx_vaxexe> vaxshad03_072.a-dcx_vaxexe</a>F <a href=/patches/public/vms/vax/v7.2/rtr/4.1/vaxshad03_072.CVRLET_TXT> vaxshad03_072.CVRLET_TXT</a> </pre>  , <!-- INSERT YOUR CONTENT ABOVE THIS POINT-->  
         </td>6
         </tr>l      <tr><td><br></td></tr>h         </table>         </div>  
      </td>       <!-- End content column -->! <!--End main body content area-->t      </TR> </TABLE> <!-- End body table -->c   ------------------------------  % Date: Thu, 29 Nov 2001 11:42:33 -0500 $ From: "John" <jhook@regenstrief.org>- Subject: Oracle 8i OCI problem with VAX BASICa- Message-ID: <9u5ogg$8vs$1@hercules.iupui.edu>    Hi,t  C I have a machine that has been running a VAX BASIC Oracle OCI basedrL application successfully for several years. Recently, we've installed OracleJ 8i on our development machine to begin migrating the application to OracleG 8. Unfortunately, I can't even get the application to login to Oracle 8  without crashing.f   Production machine -----------------------AE OpenVMS V7.2-1, Oracle7 Server Release 7.3.2.2.0 - Production Release    Development machinem ------------------------J OpenVMS V7.2-1, Oracle8i Enterprise Edition Release 8.1.7.0.0 - Production    G I've created the following small program that demonstrates the problem.t   PROGRAM TESTORALOGIN              DECLARE STRING UNAMEo            DECLARE STRING PWD@>            DECLARE WORD LDA(1 TO 32)   !ORACLE LOGON DATA AREA7            DECLARE WORD HDA(1 TO 128)   !HOST DATA AREA1B            DECLARE WORD CDA1(1 TO 32)   !DEFINE A CURSOR DATA AREA           INPUT "USR:"; UNAMEa         INPUT "PWD: "; PWD  9         PRINT "CALLING ORLON UID: "; UNAME; " PWD: "; PWD J         CALL ORLON(LDA(1) BY REF, HDA(1), UNAME BY REF, LEN(UNAME), PWD BY REF, LEN(PWD), 0)V         IF (LDA(7) <> 0) THENL/                 PRINT "ERROR LDA(7) = "; LDA(7)          ELSE(                 PRINT "LOGIN SUCCESSFUL"         END IF           PRINT "CALLING OLOGOF""         CALL OLOGOF(LDA(1) BY REF)   ENDt  B This program can be compiled and linked with the following command
 procedure.   $ basic testoralogin.bas6 $ @ora_rdbms:lnoci testoralogin.exe testoralogin.obj T    E When the program is linked on the machine with Oracle 7.3 the programeG executes as expected. If I link the program on the Oracle 8 machine the 0 program terminates with the following exception.     $ run testoralogin USR:? scotti PWD: ? tiger# CALLING ORLON UID: scott PWD: tigert- %BAS-F-MEMMANVIO, Memory management violation 3 -BAS-I-USEPC_PSL, at user PC=005289A0, PSL=0000001B ; -SYSTEM-F-ACCVIO, access violation, reason mask=00, virtualo: address=000000007B65BAAB, PC=00000000005289A0, PS=0000001B% -BAS-I-FROMOD, In module TESTORALOGINc/ %TRACE-F-TRACEBACK, symbolic stack dump followsoJ   image    module    routine             line      rel PC           abs PC>                                             0 FFFFFFFF8045EF44 FFFFFFFF8045EF44>  DEC$BASRTL                                 0 000000000000DC0C 000000007BFC5C0C= ----- above condition handler called with exception 0000000C: ; %SYSTEM-F-ACCVIO, access violation, reason mask=00, virtual.: address=000000007B65BAAB, PC=00000000005289A0, PS=0000001B ----- end of exception message>                                             0 FFFFFFFF80085C3C FFFFFFFF80085C3C>  ORACLIENT_V817  NIGCON  osncon        129799 0000000000000780 00000000005289A0>  ORACLIENT_V817  KPUADEF  kpuadef      159234 000000000000006C 0000000000158A1C>  ORACLIENT_V817  UPIPRV  upiini        261280 000000000000156C 0000000000197C6C>  ORACLIENT_V817  UPICDC  upiah0        227967 00000000000000C8 00000000001984D8>  ORACLIENT_V817  UPICDC  upiahm        228019 0000000000000150 0000000000198560>  ORACLIENT_V817  UPILOG  upilgn       1410103 0000000000000520 00000000001A8960>  ORACLIENT_V817  UPILOG  upilon       1409844 0000000000000048 00000000001A8488>  ORACLIENT_V817  OCI  ocirlo           175588 0000000000002B08 00000000001904F8>                                             0 0000000000000000 FFFFFFFF80068EAC>  ORACLIENT_V817  OCICEE  ORLON                               ? ? )  TESTORALOGIN  TESTORALOGIN  TESTORALOGINN>                                            14 0000000000000258 0000000000020258>  PTHREAD$RTL                                0 00000000000312FC 000000007BB712FC>  PTHREAD$RTL                                0 0000000000012B48 000000007BB52B48>                                             0 FFFFFFFF8ECA53F4 FFFFFFFF8ECA53F4    L Obviously, there is a memory management problem. A coworker suggested that I7 create a linker option file and add the following line:t  " CLUSTER=SGAPAD,512,,ORA_OLB:SGAPAD  D I did this on the Oracle 7.3 machine and the program still worked asH intended. However, the ORA_OLB:SGAPAD object module doesn't exist on theK Oracle 8 box so I couldn't try this technique. In fact the following tidbitmB from the installation manual implies that the SGAPAD isn't needed.  H "Starting with release 7.3.2.3.2, Alpha OpenVMS supported the Very LargeJ Memory (VLM) 64-bit feature. The Alpha OpenVMS operating system has nativeG 64-bit memory addressing, which allows Oracle8i to implement the 64-bit|E feature. The need to estimate the maximum System Global Area (SGA) at I installation time is now eliminated and process startup times are faster.e) The Server no longer includes an SGAPAD."u  - Any suggestions would be greatly appreciated.o   Thanks,t John   ------------------------------   Date: 29 Nov 2001 18:12:59 GMT' From: "Jim Strehlow" <jims@data911.com>e1 Subject: Re: Oracle 8i OCI problem with VAX BASIC./ Message-ID: <9u5trb$7e@dispatch.concentric.net>w  7 We opened Oracle Trouble Assistance Request 1758826.999o! in August regarding that problem.u@ Months later, the "T.A.R." exceeded 32K and it is now 13486009.6  G We are still running Oracle 8.0.5.0.1 in production because we got thatm same error on our test machine.a  1 We sent to Oracle all of our SGA, UAF values etc.w  0 Oracle claims we need to reprogram our VAX BASIC? Oracle OCI code that has been working for a year in production.t  = Oracle was able to reproduce the crash on multiple platforms; ( so it is not just restricted to OpenVMS.  1 Do you have access to http://metalink.oracle.com/R in order to view the TARs?  7 We feel that Oracle must have changed an API somewhere;6< but we have not yet been able to change our test program and4 production programs based upon their latest findings7 (Oracle updated the TAR with a "solution" on Nov. 27th)s  due to other project priorities.  > Jim Strehlow, Data911, Systems Manager, DBA Assistant, et. al. Alameda, CAw  / "John" <jhook@regenstrief.org> wrote in messager' news:9u5ogg$8vs$1@hercules.iupui.edu...n > Hi,k >uE > I have a machine that has been running a VAX BASIC Oracle OCI basedtG > application successfully for several years. Recently, we've installedV OracleL > 8i on our development machine to begin migrating the application to OracleI > 8. Unfortunately, I can't even get the application to login to Oracle 8i > without crashing.. >  > Production machine > -----------------------rG > OpenVMS V7.2-1, Oracle7 Server Release 7.3.2.2.0 - Production Releaset >c > Development machine  > -----------------------eL > OpenVMS V7.2-1, Oracle8i Enterprise Edition Release 8.1.7.0.0 - Production >d >aI > I've created the following small program that demonstrates the problem.V >I > PROGRAM TESTORALOGIN >a! >            DECLARE STRING UNAMEq >            DECLARE STRING PWDg@ >            DECLARE WORD LDA(1 TO 32)   !ORACLE LOGON DATA AREA9 >            DECLARE WORD HDA(1 TO 128)   !HOST DATA AREAED >            DECLARE WORD CDA1(1 TO 32)   !DEFINE A CURSOR DATA AREA >  >         INPUT "USR:"; UNAMEt >         INPUT "PWD: "; PWD >t; >         PRINT "CALLING ORLON UID: "; UNAME; " PWD: "; PWD L >         CALL ORLON(LDA(1) BY REF, HDA(1), UNAME BY REF, LEN(UNAME), PWD BY > REF, LEN(PWD), 0)  >         IF (LDA(7) <> 0) THENd1 >                 PRINT "ERROR LDA(7) = "; LDA(7)i >         ELSE* >                 PRINT "LOGIN SUCCESSFUL" >         END IF >y  >         PRINT "CALLING OLOGOF"$ >         CALL OLOGOF(LDA(1) BY REF) >I > END  > D > This program can be compiled and linked with the following command > procedure. >e > $ basic testoralogin.bas8 > $ @ora_rdbms:lnoci testoralogin.exe testoralogin.obj T >- >0G > When the program is linked on the machine with Oracle 7.3 the programxI > executes as expected. If I link the program on the Oracle 8 machine them2 > program terminates with the following exception. >y >  > $ run testoralogin
 > USR:? scotti > PWD: ? tiger% > CALLING ORLON UID: scott PWD: tiger$/ > %BAS-F-MEMMANVIO, Memory management violation.5 > -BAS-I-USEPC_PSL, at user PC=005289A0, PSL=0000001By= > -SYSTEM-F-ACCVIO, access violation, reason mask=00, virtual$< > address=000000007B65BAAB, PC=00000000005289A0, PS=0000001B' > -BAS-I-FROMOD, In module TESTORALOGINh1 > %TRACE-F-TRACEBACK, symbolic stack dump followstL >   image    module    routine             line      rel PC           abs PC@ >                                             0 FFFFFFFF8045EF44 > FFFFFFFF8045EF44@ >  DEC$BASRTL                                 0 000000000000DC0C > 000000007BFC5C0C? > ----- above condition handler called with exception 0000000C:N= > %SYSTEM-F-ACCVIO, access violation, reason mask=00, virtual.< > address=000000007B65BAAB, PC=00000000005289A0, PS=0000001B  > ----- end of exception message@ >                                             0 FFFFFFFF80085C3C > FFFFFFFF80085C3C@ >  ORACLIENT_V817  NIGCON  osncon        129799 0000000000000780 > 00000000005289A0@ >  ORACLIENT_V817  KPUADEF  kpuadef      159234 000000000000006C > 0000000000158A1C@ >  ORACLIENT_V817  UPIPRV  upiini        261280 000000000000156C > 0000000000197C6C@ >  ORACLIENT_V817  UPICDC  upiah0        227967 00000000000000C8 > 00000000001984D8@ >  ORACLIENT_V817  UPICDC  upiahm        228019 0000000000000150 > 0000000000198560@ >  ORACLIENT_V817  UPILOG  upilgn       1410103 0000000000000520 > 00000000001A8960@ >  ORACLIENT_V817  UPILOG  upilon       1409844 0000000000000048 > 00000000001A8488@ >  ORACLIENT_V817  OCI  ocirlo           175588 0000000000002B08 > 00000000001904F8@ >                                             0 0000000000000000 > FFFFFFFF80068EAC@ >  ORACLIENT_V817  OCICEE  ORLON                               ? > ? + >  TESTORALOGIN  TESTORALOGIN  TESTORALOGINn@ >                                            14 0000000000000258 > 0000000000020258@ >  PTHREAD$RTL                                0 00000000000312FC > 000000007BB712FC@ >  PTHREAD$RTL                                0 0000000000012B48 > 000000007BB52B48@ >                                             0 FFFFFFFF8ECA53F4 > FFFFFFFF8ECA53F4 >g >cL > Obviously, there is a memory management problem. A coworker suggested that I09 > create a linker option file and add the following line:  >0$ > CLUSTER=SGAPAD,512,,ORA_OLB:SGAPAD >uF > I did this on the Oracle 7.3 machine and the program still worked asJ > intended. However, the ORA_OLB:SGAPAD object module doesn't exist on theF > Oracle 8 box so I couldn't try this technique. In fact the following tidbitD > from the installation manual implies that the SGAPAD isn't needed. >oJ > "Starting with release 7.3.2.3.2, Alpha OpenVMS supported the Very LargeL > Memory (VLM) 64-bit feature. The Alpha OpenVMS operating system has nativeI > 64-bit memory addressing, which allows Oracle8i to implement the 64-bitsG > feature. The need to estimate the maximum System Global Area (SGA) atoK > installation time is now eliminated and process startup times are faster.r+ > The Server no longer includes an SGAPAD."a >t/ > Any suggestions would be greatly appreciated.  >e	 > Thanks,e > John   ------------------------------  % Date: Thu, 29 Nov 2001 08:11:51 -0500S2 From: "Sue Skonetski" <susan.skonetski@compaq.com>& Subject: Oracle RDB announcment/letter3 Message-ID: <qtqN7.2097$RL6.63071@news.cpqcorp.net>    got this last night.   sue   4 ____________________________________________________   Dear Rdb Customer:  ? On June 29, shortly after Compaq announced plans to consolidatee  E Compaq's entire 64-bit server family on the Itanium architecture froms  C Intel, I wrote to you regarding Oracle Rdb, Oracle CODASYL DBMS and   @ the Itanium Processor Family. At that time, I said we would work  @ closely with Compaq and our customers to determine the best path  : forward for Rdb. I am writing to you now with an update on  9 developments regarding a port of Rdb and DBMS to Itanium.u  F From our discussions with you, our customers, over the past months, it  C is clear that many of you are very interested in seeing Oracle Rdb,   A Oracle CODASYL DBMS and Oracle9i available on OpenVMS on Itanium.I  E As a result of this interest, Juan Jones, Oracle's Vice President forn  @ the Systems Platforms Division, recently, released the following  
 statement:  7 Oracle and Compaq have a long and successful history ofs  < delivering enterprise solutions to our OpenVMS customers. In  8 July, Oracle released Rdb 7.1 for OpenVMS. In September,  8 Oracle9i for OpenVMS was released. Given Compaq's recent  = announcement to consolidate its 64-bit servers on the Itanium   < Processor Family (tm), Oracle's current plan is to team with  < Compaq and work toward a delivery of Oracle Rdb and Oracle9i  : for OpenVMS on IPF based upon Compaq's current engineering   roadmap.  F Consistent with this direction we have developed a preliminary project  C plan for our porting activities for both Rdb and CODASYL DBMS. Thisi  A preliminary plan was developed with the assistance of the OpenVMSt  > Engineering Group at Compaq. This plan is based on the current  ( delivery dates provided to us by Compaq.   Highlights of the plan include:b  D * By early next year, Compaq will provide us with cross compilers to  < allow us to begin initial testing of our code in the Itanium   environment.  @ * In 2003, Compaq will deliver to us and their other partners an  C early version of OpenVMS for Itanium that will allow us to complete   : the port and begin testing of Rdb in this new environment.  > * In 2004, Compaq will deliver the first production release of   OpenVMS on Itanium.A  @ Oracle normally ships supported versions of Rdb and CODASYL DBMS  D within 90 days of Compaq's production release date for new operating  A systems and processors. Assuming Compaq delivers Itanium softwareo  E according to their announced schedule and that this software performsn  D as expected, we will deliver the first production release of Rdb and  F CODASYL DBMS for Itanium within 90 days of Compaq's production release   date for OpenVMS on Itanium.  E These are still early days and Compaq's plans and road maps are stillE  D evolving. We expect to learn more as we and Compaq progress with our  ? respective porting efforts. In the meantime we will continue toH  0 enhance and improve Rdb on the OpenVMS platform.  C We will provide an updated report on our progress on Itanium at theu  E Rdb Forum at Oracle Headquarters near San Francisco on December 1 andr  E 2 as well as at other Rdb Forums to be scheduled throughout the worldi  A in 2002. I hope to see you at one of these events. (You can still"  D register for the Forum at our web site, http://www.oracle.com/rdb/.)  
 Best regards,>   Kevin Duffyh   Oracle Rdb Development Director2  F ----------------------------------------------------------------------  F I sent you this mail because of your business relationship with Oracle  A Rdb. If you no longer wish to receive mailings from us about Rdb,a  B please reply to this message with the word "Remove" in the subject  C line of your message. We will immediately remove your name from ourL  
 mailing list.p   ------------------------------  # Date: Thu, 29 Nov 2001 14:21:59 GMTa4 From: "Terry C. Shannon" <terryshannon@mediaone.net>* Subject: Re: Oracle RDB announcment/letter; Message-ID: <bwrN7.495$zX1.1437893@typhoon.ne.mediaone.net>7  = "Sue Skonetski" <susan.skonetski@compaq.com> wrote in messaged- news:qtqN7.2097$RL6.63071@news.cpqcorp.net...e > got this last night.  L This is very good news for the IPF port of VMS. Oracle was one of the gating4 factors, looks like they're completely on board now!   ------------------------------  % Date: Thu, 29 Nov 2001 16:11:35 +0000/% From: Alan Greig <a.greig@virgin.net> * Subject: Re: Oracle RDB announcment/letter8 Message-ID: <1gmc0u0q9u6qnnahcmsuk7i9inqausriut@4ax.com>  4 On Thu, 29 Nov 2001 14:21:59 GMT, "Terry C. Shannon"" <terryshannon@mediaone.net> wrote:   > > >"Sue Skonetski" <susan.skonetski@compaq.com> wrote in message. >news:qtqN7.2097$RL6.63071@news.cpqcorp.net... >> got this last night.- >-M >This is very good news for the IPF port of VMS. Oracle was one of the gatinge5 >factors, looks like they're completely on board now!   B Excellent news and quite clearly worded as well. Says plainly that@ they *will* release RDB, DBMS and 9i (or 10i by then) as long as Compaq completes the port.  > Just in time as well for me to forward on to US colleagues forA inclusion in some more board level powerpoint slides. We now haveyB enough statements from key suppliers of our software for me to sayD that I am confident we can run on VMS/Itanium assuming all suppliers keep their word.  ; I have also included a copy of your latest SKC as posted tomD www.openvms.org for their perusal. Yes the board did come back afterE earlier presentations and ask for more clarification on why we shouldp: buy new Alphaservers after "Compaq had just dropped Alpha"  F Maybe they'll pick up a few subscriptions directly themselves. And canE I congratulate you on, if not totally doing an about turn,  returning ' more to what I remember as the old SKC.A   Still like a SAP port though!1 -- Alan   ------------------------------  % Date: Thu, 29 Nov 2001 12:40:43 +0000 % From: Alan Greig <a.greig@virgin.net>O7 Subject: Re: OT: Andrew Harrison : Alive & Still At SUNA8 Message-ID: <81bc0u0h7ij31dqaolktm8dfct3gh33cle@4ax.com>  @ On 28 Nov 2001 18:47:58 GMT, leslie@clio.rice.edu (Jerry Leslie) wrote:  D >I got a reply from Andrew from an email to Andrew.Harrison@Sun.COM.@ >He's switched offices and doesn't have access to a NNTP server.  E And presumably email which somehow blocks info-vax and browsers whiche somehow block deja/google.  A Switched offices could be code for "no longer employed monitoringo VMS".   > >He was surprised that Fred Kleinsorge really asked about him. >h >--Jerry Leslier   -- Alan   ------------------------------   Date: 29 Nov 2001 17:06:25 GMT1 From: bill@triangle.cs.uofs.edu (Bill Gunshannon) 7 Subject: Re: OT: Andrew Harrison : Alive & Still At SUNN, Message-ID: <9u5puh$19so$2@info.cs.uofs.edu>  8 In article <81bc0u0h7ij31dqaolktm8dfct3gh33cle@4ax.com>,(  Alan Greig <a.greig@virgin.net> writes:C |> On 28 Nov 2001 18:47:58 GMT, leslie@clio.rice.edu (Jerry Leslie)A	 |> wrote:0 |> 5G |> >I got a reply from Andrew from an email to Andrew.Harrison@Sun.COM.EC |> >He's switched offices and doesn't have access to a NNTP server.  |>  H |> And presumably email which somehow blocks info-vax and browsers which |> somehow block deja/google.F |> 5  D Actually, many offices prohibit the use of these facilities from the@ office LAN using social pressures rather than technical methods.@   "Any non-company use of the corporate LAN will get you fired!"  - My sister works in just such an organization.5   bill   -- 0J Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesD bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. University of Scranton   |A Scranton, Pennsylvania   |         #include <std.disclaimer.h>   9   ------------------------------  % Date: Thu, 29 Nov 2001 12:51:03 -0000 4 From: "Chris Sharman" <chris.sharman@ccagroup.co.uk>) Subject: patch downloads - 3rd time lucky0A Message-ID: <1007038099.2915.0.nnrp-14.9e989e7e@news.demon.co.uk>   ! What does everyone use for this ?9I We've got an ISDN line, and I tried to fetch the dce 3.0 u1 eco yesterday0H evening (via ie5.0 on my pc) - this morning it had given up at 4.9M. I'mJ trying again now with copy/ftp/bin from vms (7.3, tcpip 5.1-3) -  it seemsL to have stalled at 8-8.5M (the file's not grown, and the process has done noL I/O in half an hour), so I presume it's going to give up again. Perhaps I'll7 get lucky at my third attempt - a bargain at 25M total.O  I The patch is 12.4M (with the dismal DCX compression). I've seen kits with0H SFX compression - much better and therefore much more likely to download7 first time - why are Compaq not using it all the time ?    Thanks,0 ChrisB   ------------------------------  % Date: Thu, 29 Nov 2001 14:50:27 +0000F% From: Alan Greig <a.greig@virgin.net>l- Subject: Re: patch downloads - 3rd time luckyr8 Message-ID: <ckic0ughi9cu9m76s2al5bgues26klu591@4ax.com>  3 On Thu, 29 Nov 2001 12:51:03 -0000, "Chris Sharman"P% <chris.sharman@ccagroup.co.uk> wrote:i  " >What does everyone use for this ?J >We've got an ISDN line, and I tried to fetch the dce 3.0 u1 eco yesterdayI >evening (via ie5.0 on my pc) - this morning it had given up at 4.9M. I'mr  D If you download from a recent browser or PC ftp download manager viaB HTTP (or FTP with restart extensions) the download should continue@ where it left off. I'm fairly sure Compaq's path download server supports both.  K >trying again now with copy/ftp/bin from vms (7.3, tcpip 5.1-3) -  it seemsTM >to have stalled at 8-8.5M (the file's not grown, and the process has done no M >I/O in half an hour), so I presume it's going to give up again. Perhaps I'lli8 >get lucky at my third attempt - a bargain at 25M total. >eJ >The patch is 12.4M (with the dismal DCX compression). I've seen kits withI >SFX compression - much better and therefore much more likely to downloadO8 >first time - why are Compaq not using it all the time ? >d >Thanks, >Chris >t   -- Alan   ------------------------------  % Date: Thu, 29 Nov 2001 14:25:20 -0000t4 From: "Chris Sharman" <chris.sharman@ccagroup.co.uk>; Subject: Re: patch downloads - 3rd time lucky & compression.B Message-ID: <1007043932.17647.0.nnrp-12.9e989e7e@news.demon.co.uk>  # > What does everyone use for this ? K > We've got an ISDN line, and I tried to fetch the dce 3.0 u1 eco yesterdayOJ > evening (via ie5.0 on my pc) - this morning it had given up at 4.9M. I'mL > trying again now with copy/ftp/bin from vms (7.3, tcpip 5.1-3) -  it seemsK > to have stalled at 8-8.5M (the file's not grown, and the process has done  noI > I/O in half an hour), so I presume it's going to give up again. Perhaps  I'll9 > get lucky at my third attempt - a bargain at 25M total.i  K Finally, on my 3rd attempt (7.3, tcpip 5.1-3, mgftp 2.6-5) I got it in justo over 40 minutes.K mgftp has a nice ^T response, but apart from that I don't know whether it'sg faster or better or whether I was just lucky.  K > The patch is 12.4M (with the dismal DCX compression). I've seen kits witheJ > SFX compression - much better and therefore much more likely to download9 > first time - why are Compaq not using it all the time ?s   Uncompressed: 42720 blocks dcx_axpexe: 25501 
 zip/gz: 16902h  - Seems like a large & worthwhile saving to me. J It usually is. Indeed, this file seems to have compressed better than mostH with DCX, but it still doesn't compare with zip/gzip (which is I believe similar to the sfx algorithm).   Thanks,7 Chris0   ------------------------------  % Date: Wed, 28 Nov 2001 15:07:57 -0500- From: William_Bochnik@acml.com6 Subject: Re: Problem fseeking a 4.2 million block file> Message-ID: <OFC15A5167.E1A1EA58-ON85256B12.006E3CFC@acml.com>  ; I get the impression he was asking if he would have to messA5 around like that, rather than asking for workarounds.     s                                                                                                                     s                     Andy Burns                                                                                     Ns                     <andy@burns.n                To:  Info-VAX@Mvb.Saic.Com                                        Fs                     et>                          cc:                                                                s                                          Subject:     Re: Problem fseeking a 4.2 million block file                 s                     11/28/2001                                                                                     as                     02:01 PM                                                                                       .s                     Please                                                                                         as                     respond to                                                                                     ys                     Andy Burns                                                                                     Os                     <andy@burns.n                                                                                  os                     et>                                                                                            os                                                                                                                    ns                                                                                                                           A On 28 Nov 2001 10:44:53 -0800, steve_p@trigent.com (Steve Porter)0 wrote:  % >Since fseek expects a signed int ford? >the offset, will there be a problem if the offset passed in is  soA >large it 'overwrites' the sign bit?  Will this affect subsequentA< >fseeks, assuming all fseeks start from the beginning of the file?   < rather than doing absolute seeks from the start of the file,	 could youEA try relative seeks for a point already seeked to halfway into the0; file? or pehaps absolute seeks from the end rather than the0 start?   messy I know ... --
 Andy Burns          F ______________________________________________________________________;  The information contained in this transmission may contain0@ privileged and confidential information and is intended only forA the use of the person(s) name above.  If you are not the intendedg= recipient, or an employee or agent responsible for deliveringE3 this message to the intended recipient, any review,8@ dissemination, distribution or duplication of this communication? is strictly prohibited.  If you are not the intended recipient, A please contact the sender immediately by reply e-mail and destroyA# all copies of the original message.    ------------------------------  % Date: Thu, 29 Nov 2001 10:20:45 +0100 , From: "Ferry Bolhar" <bol@adv.magwien.gv.at> Subject: RECALL does not work28 Message-ID: <1007025593.127337@mozart.adv.magwien.gv.at>  	 Hi @ all,   J In a procedure, I do not want to save user input (requested by INQUIRE) in  the DCL recall buffer. So I do a   $ RECALL/OUTPUT=X.X    at the procedure start and a   $ RECALL/ERASE/INPUT=X.X  K at the end. However, this does not work when invoked from the procedure. NotF File X.X is created and user input is still in the recall buffer. When> entering these commands interactively, it behaves as expected.  $ Does someone know what happens here?  ; BTW: This occurs on OpenVMS V6.2 and V7.1 on VAX and Alpha.o  $ MTIA and kind greetings from Vienna,   Ferryt   -- Ing. Ferry Bolhart% Municipality of Vienna, Department 14a A-1010 Vienna / AUSTRIAa E-mail: bol@adv.magwien.gv.at    ------------------------------  + Date: Thu, 29 Nov 2001 10:47:15 +0100 (MET)o9 From: Phillip Helbig <HELBPHI@sysdev.deutsche-boerse.com> ! Subject: Re: RECALL does not worki; Message-ID: <01KB9U2GYXHS90YLDV@sysdev.deutsche-boerse.com>.   > In a procedure,  >  > $ RECALL/OUTPUT=X.Xr > A > However, this does not work when invoked from the procedure. No H > File X.X is created and user input is still in the recall buffer. When@ > entering these commands interactively, it behaves as expected. > & > Does someone know what happens here?  D It is documented that RECALL only works interactively, not in a DCL 
 procedure.   ------------------------------  % Date: Thu, 29 Nov 2001 10:50:49 +00007( From: Martyn <mpattKNICKERS@bigfoot.com>! Subject: Re: RECALL does not work_* Message-ID: <3C061309.8040206@bigfoot.com>  C You could replace the "Inquire"s with "Read SYS$INPUT"s,  then the d* responses don't go into the recall buffer.  # Maybe you already know that though.n     Ferry Bolhar wrote:d  
 >Hi @ all, >rK >In a procedure, I do not want to save user input (requested by INQUIRE) inw! >the DCL recall buffer. So I do a  >  >$ RECALL/OUTPUT=X.X >e >at the procedure start and ad >I >$ RECALL/ERASE/INPUT=X.X  >uL >at the end. However, this does not work when invoked from the procedure. NoG >File X.X is created and user input is still in the recall buffer. Whenc? >entering these commands interactively, it behaves as expected.  >r% >Does someone know what happens here?e > < >BTW: This occurs on OpenVMS V6.2 and V7.1 on VAX and Alpha. > % >MTIA and kind greetings from Vienna,t >  >Ferry >i >--n >Ing. Ferry Bolhar& >Municipality of Vienna, Department 14 >A-1010 Vienna / AUSTRIA >E-mail: bol@adv.magwien.gv.at >g >t >r   -- u, Remove KNICKERS before replying by Email ;-)   ------------------------------  % Date: Thu, 29 Nov 2001 14:47:57 +0100s, From: "Ferry Bolhar" <bol@adv.magwien.gv.at>! Subject: Re: RECALL does not works8 Message-ID: <1007041623.962597@mozart.adv.magwien.gv.at>   > > In a procedure,  > >p > > $ RECALL/OUTPUT=X.Xv > >fC > > However, this does not work when invoked from the procedure. NorJ > > File X.X is created and user input is still in the recall buffer. WhenB > > entering these commands interactively, it behaves as expected. > >b( > > Does someone know what happens here? >wE > It is documented that RECALL only works interactively, not in a DCLt > procedure.  I Where? I didn't find this information, neither in the online help, nor ini the DCL dictionary.i  L And this restriction isn't a useful one, IMHO. RECALL with /ERASE, /INPUT orL /OUTPUT should be allowed in DCL procedures (eg., the recall buffer could be  pre-initialized from LOGIN.COM).   Greetings, Ferry ---p   Ing. Ferry Bolhare% Municipality of Vienna, Department 14a A-1010 Vienna / AUSTRIAh E-mail: bol@adv.magwien.gv.at    ------------------------------  % Date: Thu, 29 Nov 2001 18:54:04 +0100t1 From: John McLean <mcleanj@swissonline.delete.ch>rH Subject: Re: SKC Musings on "Recent IPF Unpleasantness" at www.tru64.org5 Message-ID: <3C06763C.F67C548F@swissonline.delete.ch>    "Terry C. Shannon" wrote:i >  > Folks, > B > Given the controversy that my special SKC issue entitled "CompaqH > Cartographers Drew the IPF Roadmap" (.html | .pdf) precipitated when IM > posted the piece, it seemed like the best thing to do was wite a follow-up.rN > If you haven't read the original document, feel free to do so. Draw your ownL > conclusions, but please take a moment to review the follow-up as per below > (Nov 10, 2001) >  >       Shannon knows Compaq( >       An Outpouring of Alphacide Angst >       (Nov 28, 2001)    B The problem with Compaq is more than just poor communication, poorH marketing, poor press-releases, poor understanding of the user community/ ... etc, etc.  The problem is their competence.l  E There have been too many errors of judgement (eg. extreme emphasis on.H the PC side of the business, roadmaps and commitments that are cancelledG before the ink is dry) and their credibility now rates somewhere in the>
 negatives.  G Once customers lose confidence in a supplier they rapidly start to look-B elsewhere; those customers who stay with the product, for whateverE reason, become extremely skeptical of any statement from that companyi7 and will examine every statement in microscopic detail.y  G This is what has happened with Compaq and all too often that detail hasaA been found to be wanting, thus diminishing their credibility evenp further.  C Communication is a small part of their problem.  Credibility is the 
 bigger issue.o     John McLeanm   ------------------------------  % Date: Thu, 29 Nov 2001 16:26:28 +01007% From: "Dr. Dweeb." <Dweeb@nospam.com>tN Subject: Re: Special IPF-Inside Issue of Shannon Knows Compaq at www.tru64.org/ Message-ID: <rssN7.89$WK5.3550@news.get2net.dk>l  	 Question.   H How many Fortune 100 CIOs do you personally know, and from whom you have gleaned this information ?  # Enquiring minds would like to know.-   Dweeb.1 "Jeff Killeen" <Jeff@IDM-IO.com> wrote in message/0 news:YX7N7.608$h24.58835@typhoon1.gnilink.net... >:< > "JF Mezei" <jfmezei.spamnot@videotron.ca> wrote in message( > news:3C0503C9.5488110C@videotron.ca...L > > In the past, corporations viewed computing in the same way as building aJ > > tower. You know the building will last a long time so you need to have > veryL > > strong foundations and a very good design that will keep the tower up noJ > > matter what weather comes, and you will also have to have the means toH > > renovate floors and perhaps add a few more floors later on. And then once > all G > > of the structure is done and proven solid, you add the decorations.b >hL > Let me share with you what the typical USA Fortune 100 CIO would say about	 > this...o > G > 1) Strong foundations were viable when technology lifecycles were 3-5n years.* > Today they can be as short as 18 months. >o> > 2) Applications, and not base technology, are the key today. > A > 3) If delaying deploying an application, that gives my businessc competitiveaA > advantage, is delayed by preserving the design of my current ITbI > infrastructure I have cost my business major lost opportunity costs andiF > potentially market share.  The cost of that far out weighs any other issue. > K > The important part isn't whether you agree with the above.  The importantgL > part is know the senior management of IBM, Compaq, HP, and SUN are gettingF > their market input from IT professionals who think like the above... >  >-J > > But guess what, even if they must rebuild their buildings every couple of
 > > weeks, >@I > And if you were an IT supplier would you view that as a good thing or as badD > thing? >j >q   ------------------------------    Date: 29 Nov 2001 17:36:31 +0100- From: Robert Harley <harley@estephe.inria.fr>rN Subject: Re: Special IPF-Inside Issue of Shannon Knows Compaq at www.tru64.org. Message-ID: <rz7adx5pm9c.fsf@estephe.inria.fr>  ( peter@abbnm.com (Peter da Silva) writes:I > I think it's obvious that the gap in performance between Alpha and IA64c; > is closing... given the current performance if IA64 [...]H   Wot, like this?t  3     CPU           Clock   SPECint SPECfp (Int+FP)/2L6   ----------------------------------------------------/     Power 4       1300      808    1169     988l/     Alpha EV6     1001      679     960     820:/     Sparc III     1050      610     827     718-/     Pentium 4     2000      664     734     700e/     Alpha EV6      833      571     784     678A/     Athlon XP     1600      701     634     668N/     Pentium 4     1900      634     696     665 /     Athlon XP     1533      671     615     643m/     Sparc III      900      530     730     630E/     PA-8700        750      604     581     592s/     Itanium        800      379     701     540i/     MIPS R14000    500      427     463     445h    ? Then again, last I heard, Itanium is too buggy for Compaq to bei willing ship it...     RobeI     .-.                                                               .-..J    /   \           .-.        http://argote.ch/        .-.           /   \K   /     \         /   \       .-.     _     .-.       /   \         /     \ L  /       \       /     \     /   \   / \   /   \     /     \       /       \M /         \     /       \   /     `-'   `-'     \   /       \     /         \aB            \   /         `-'                     `-'         \   /A             `-'                                               `-'0   ------------------------------  % Date: Thu, 29 Nov 2001 18:15:43 +0100w% From: "Dr. Dweeb." <Dweeb@nospam.com>1N Subject: Re: Special IPF-Inside Issue of Shannon Knows Compaq at www.tru64.org0 Message-ID: <S2uN7.108$WK5.4900@news.get2net.dk>  H It appears that Jeff concurs with this sentiment (even though it was not mine).   Dweebn@ "Fred Kleinsorge" <kleinsorge@star.zko.dec.com> wrote in message- news:ukTM7.2011$RL6.62015@news.cpqcorp.net...8 > Ah, but not fair -   cut ...t  K > So far your argument is based on your belief that Jeff isn't competent top, > argue with you on some intellectual level. >p cut ...g   ------------------------------  # Date: Thu, 29 Nov 2001 15:43:47 GMT)& From: "Jeff Killeen" <Jeff@IDM-IO.com>N Subject: Re: Special IPF-Inside Issue of Shannon Knows Compaq at www.tru64.org8 Message-ID: <TIsN7.1201$h24.103953@typhoon1.gnilink.net>  5 "Bill Todd" <billtodd@metrocast.net> wrote in message16 news:2mXM7.54560$YD.4934129@news2.aus1.giganews.com... >j3 > "Jeff Killeen" <Jeff@IDM-IO.com> wrote in messageD2 > news:k6VM7.140$h24.23231@typhoon1.gnilink.net... > > Read carefully folks...0 > >C9 > > "Bill Todd" <billtodd@metrocast.net> wrote in messageW? > > news:2kSM7.76249$8q.10317432@bin2.nnrp.aus1.giganews.com...t > > > 7 > > > "Jeff Killeen" <Jeff@IDM-IO.com> wrote in messagen > >cL > > > > Even Todd at this point will acknowledge that a multi-IA64 processor	 > > couldrE > > > > achieve the same performance level as a multi-Alpha processor  system.s > > >oJ > > > As 'could' a multi-IA32-processor, or even a multi-PDP-11-processor,G > > > configuration - for *some* but definitely not *all* uses.  You've5	 > ignored  > > myH > > > very explicit clarification of this, with is either incompetent or > > > dishonest. > >oI > > The reality is that the number of applications this applies to is the J > > minority, and not the majority, in the typical commercial environment.D > > Being one who saw the magic of clusters when they first came out
 > understandseJ > > what they can do for a typical commercial mixed workload requirements. > @ > If you believe that cluster behavior is directly applicable toJ > tightly-coupled MP systems, you're even more incompetent than I thought.  B No that is you putting words in my mouth - I believe the number ofL applications that would benefit from tightly coupled MP of EV8 class are farL more limited than you are portraying here.  That you are using the exceptionJ to justify the rule.  Note I was very careful in the QuickBlade postings IJ made to clarify the processors where group and not all tightly coupled.  IL never said that tightly couple MP and clusters were the same - but for given, work loads there are two ways to skin a cat.  B > As far as characterizing the conditions under which more, slower
 processorsH > cannot effectively substitute for fewer, faster processors, to a firstJ > approximation it's simple:  if the typical number of runnable threads in thenF > system is less than the number of processors in the system, you'd be better$ > off with fewer, faster processors.  E We actually agree on that point.  Here is where I think we disagree - L typical commercial interactive job mixes have lots of threads to be run.  MyK point is the reader here should think about their actual production mix and I that the IA64 processors we are talking about will at least equal today'szK Alpha processor.  The make up there own mind about the relevancy of the EV8d4 to IA64 gap as it applies to them in the real world.  % >  After that, you start getting intoeI > second-order effects that tend to make processing power scale less thansK > linearly as the number of processors increases, which again favors fewer,5J > faster processors (in all cases, not just special ones) but usually only > slightly.c   I don't disagree  L > > When Todd is giving his examples sanity check them against the work load > mixgF > > in your environment.  Remember also the examples are based on much fasterJ > > Alpha processors.  When you are thinking about your applications todayJ > > remember the IA64 processors 2-3 generations out will be _faster_ thanL > > today's Alpha processors.  So the applications that fit Todd's exceptionH > > would be ones that require more processor power than you have today. HowbE > > many of your application fit that model if we buy his assumption?  > D > More to the point, how many will fit that model *then*?  While VMSL > applications may not suck up as many more processor cycles each release asG > MS bloatware, there's still a noticeable tendency to use more as they  become > available.  K I think in a Alpha -> IA64 context we should stay focused on OVMS and Tru64   H > > > > commercial me wrap-up with this  - assume Compaq delivers on its	 > promise  > > that
 > > > IA64K > > > > is only a compile away and delivers IA64 based servers running OVMS  or > > > Tru64 @ > > > > that is equal to or better than the competition in price performance. > > > Grant H > > > > the assumption for the moment.  Do you think many will care that OVMS > is	 > > > now K > > > > running on IA64 versus Alpha.  If it is only a  recompile away, andO > the K > > > > price performance is there, what rational business reason would any F > > > > unemotional person have for caring it is not running on Alpha? > > > H > > > 1.  They will care because *verifying* that the recompile of theirL > > > applications (and any recompiled components they obtained from others) > was . > > > all it took takes non-negligible effort. > > > K > > > 2.  They will care because Compaq will have spent 3 years preoccupied  > withI > > > the porting effort rather than adding *real* value to VMS (and ISVsr will3 > > > have been diverted by such activity as well).o > >,I > > No matter how you attempt to cut it the pain of doing this will be ati > leastsH > > a magnitude less than (a likely many magnitude less than) porting to	 > anothere
 > > platform.  >eL > Don't change the question you posed and I answered:  you asked for reasonsH > people would care, not reasons why staying with VMS on Itanic would be  > easier than porting elsewhere. >nF > But to take up your new implied question:  reasons people would port include_F > a) they believed that some time soon they could no longer obtain theI > performance they needed on the platforms (stagnant EV7, relatively sloweK > Itanic) VMS ran on,  b) they had no confidence that Compaq would continuev toF > provide the support or new development they needed, or  c) they wereH > sufficiently disgusted with Compaq's corporate behavior that it seemed worthm > the pain.r  2 B and C are business decisions and not a technical  # > > The decision on Alpha is final.e >dL > Only death is final, and aren't you one of those whe's been asserting that > Alpha isn't dead?   F Not me - I believe on June 25th Alpha was irrevocably destroyed in theI marketplace.  The reason why is that if the decision was reversed the ISV:K community would never trust that it would stick.  It is the same reason whyaI I feel Tru64 suffered an irrevocable death in mid September.  Even if thesK merger never happens the ISV community would never trust in the marketplace C viability of Tru64 again after the way Compaq justified the merger.e  + >   Even if Compaq changed its mind at thisoL > > point the tenuous credibility that Alpha had in the marketplace is gone. >uB > Your opinion, of course.  The predominant reaction here suggests
 otherwise,J > though in terms of *momentum* (as distinct from credibility) it would ofC > course have been far better had Compaq not screwed up so royally.r  J Perception is reality and the perception of the ISV's is what drives theseL markets.  Few ISV's will bet major dollars on Alpha and Tru64 after what has	 happened.e  1 > > The best course of action is to move forward.i >eI > Your opinion again, at least from what I'd infer you mean by 'forward'.rF > Others believe that the best course of action is to replace Compaq'sI > incompetent management (without which action *no* course of action willr haveI > much chance of success) and 'move forward' in the direction that shouldt have  > been taken in the first place.  K I believe whatever anyone feels about replacing Compaq's management at thisoI point is a side issue. The damage to marketplace credibility of Alpha haslL been done.  ISV's will say yea right - what happens the next time management changes.  H > > The alternative is you can take the Todd approach which is to try toG > > discredit his ex-employer and try to destroy the viability of OVMS.l >eF > You're a real shithead, Jeff.  1.  I've never worked for Compaq.  2. CompaqI > has discredited itself to a degree that would be extremely difficult tohD > increase.  3.  VMS's viability has already been destroyed, and I'mG > advocating the only course that shows any promise of resurrecting it.   1 More unemotional professionalism from our friend.    Did you not work for Digital?=   >   OfL > > course it would be a little hard to claim one is trying to save OVMS but< > > trashing it to the point where everyone migrates off it. > >WK > > > 3.  They will care because the risk that the transition will *not* beD asK > > > smooth as you posit and/or that Itanic will continue to suck so badly7 > thatE > > > cost-effective platforms simply cannot be based on it will haveiH > > > significantly reduced VMS's market (and ISV base) over the 3 years > beforeC > > > such risks have been proved not to exist (*if* we accept yourK
 > assumptionsr > > toF > > > this effect) and therefore VMS will have a significantly smaller
 > customerD > > > base to fund its future development (if there is such) or even > existence. > >eH > > Again Todd stating his opinion as fact.  IA64 will suck badly 2 to 3I > > generations out - how do we know this as fact?  Because of course the  sage > > Todd has said so.... >rH > And cited the evidence to support it.  One generation out is McKinley, which H > Intel is already suggesting will be about 1.6 times the performance ofL > Merced (i.e., pretty much the same improvement one would expect from otherH > architectures given the year between their introductions, and thus notG > noticeably improving Itanic's relative position in the speed sweeps).i Intel H > has also indicated that Madison (two generations out from now) will beL > McKinley with a process-shrink and more cache, which also promises only toI > follow the normal speed-improvement curve and thus again not noticeablylL > improve Itanic's industry position.  And either the generation beyond that@ > is not yet even publicly defined or it's Deerfield, which is a cost-reducedH > Madison with less cache (and hence won't improve things at all, unless& > another process-shrink is involved). >lI > By contrast, EV7 will not only complete the Alpha move to a 0.18 micronlI > process but add dramatic improvements in memory latency, bandwidth, andoG > on-chip MP support - thus increasing performance faster than industrye norms.C > And EV8 would have added SMT and other tweaks to, again, increaseeI > performance (certainly in server-style use) faster than industry norms.hF > Which means that 2 - 3 generations out, Alpha would be significantly faster& > relative to Itanic than it is today. >pJ > Just because you're abysmally ignorant of the available data surrounding0 > this discussion doesn't mean that everyone is.  L There is an interesting POV you keep ignoring.  Your arguments are all basedG on IA64 performance relative to Alpha - I am far more concerned at thistI point whether IA64 can do the job for me instead of theoretical argumentsr about the gap.  L > > > And, of course, for some non-zero number of installations, recompiling > isC > > > not an option in any event - and even if running via a binarys
 translator > > isJ > > > effective that too will require verification effort.  And some otherJ > > > non-zero number of installations will have to choose between runningG > > > unsupported cluster mixes that include VAXen and reconfiguring to  avoidlF > > > them.  And I'm sure there are some other minorities that will be
 > affected > > inL > > > other ways that don't pop into my mind (but likely can be filled in byH > > > others here):  for a system with as small a user population as VMS	 > alreadyCJ > > > has, it doesn't take too many such minorities to add up to something > > > significant. > >2L > > Folks neither need to take my opinion or your opinion on this.  They canK > > contact Compaq and start getting "Golden Blanket" written guarantees on- > thes > > conversion.w >eJ > You're trying to change the question you asked again:  people don't want toL > have to exercise such 'guarantees', they want smooth evolution - which had; > the Alpha development path continued they would have had.i  C No I am facing reality - you believe that the Alpha decision can be H reversed - I do not believe so.  If you don't believe it can be reversed: than the options you are placing on the table don't exist.  L > > > Those are the down-sides when we *grant* your assumptions.  The actualI > > > potential down-sides are far worse.  And there's *no* real up-side:v at > > bestJ > > > (and again granting your assumptions rather than using a less biased > > crystaloG > > > ball), by 2005 - 6 people will be able to buy hardware to run VMS  that'sK > > > somewhat less expensive for the same performance than would otherwisea > haveE > > > been available, but hardware cost is not (at least according to1 severaluL > > > surveys that used to be on the Compaq Web site) a dominant concern for > VMSeH > > > customers anyway (very-low-end applications possibly excepted, but > > *that's* > > > a whole 'nother subject).  > >e: > > The upside is not doing a far more painful conversion.  J > No, you idiot:  you asked why people would care that they were no longerG > running on Alpha, not whether that would make them mad enough to movel > elsewhere.  H I was responding to your claim there is no upside.  Once again Bill yourH calling someone an idiot because you believe your assumption to be fact.K You assume, as proven by the statements below, that Alpha business was just,K humming along fine.  I do not believe that was the case.  Most, if not all,kL of the major industry analyst were questioning how long Compaq could justifyI its investment in Alpha given the potential return.  I know to you we arei: all idiots - I feel in good company with my fellow idiots.  I > I'll spell out the current problem once more, because you still seem to0 have4 > no grasp of it (let alone the associated details): >SI > It's the cancellation of Alpha.  Period.  This was an incredibly stupidDL > move, because it  a) immediately turned one of the most profitable productF > lines Compaq has into yet another struggling one,  b) in the process likelyF > cost more money than eliminating development will save over the same periodH > (let alone the cost in future sales),   c) cut VMS's its nascent salesL > up-turn off at the knees (thus putting yet another stake into the heart ofJ > its already-fragile long-term viability)  d) did similar damage to Tru64K > (though the merger now shows signs of turning that lingering illness into    aeD > more precipitous demise), and  e) highlighted the fact that CompaqI > management are incompetent weasels with whom at least many people would  not 9 > choose to do business given any reasonable alternative.   J "A" above is flat out not true.  I have seen the internal numbers and whatI the Alpha servers have been contributing in profits over the last several0K years is significantly below two other groups.  Sorry but the information It saw I can't release.  8 "B" is an assumption reasonable people will disagree on.  K "C" is likely a 12 month window unless Compaq fails to develop strong proof = points about the conversion - yes it is a risk but not a fact   . "D" - Tru64 has been given a serious 1-2 punch   "E" is opinion  L Re: "E" BTW if you want to see what a real weasel manager is look at this...  F http://computerworld.com/nlt/0%2C3590%2CNAV47_STO66102_NLTPM%2C00.html  I > The future 'blade' architecture you've described does not change any of  the I > above in the slightest - in fact, it supports it.  Because Compaq *has*-H > elected to make the blades support both Alpha and Itanic engines, they willK > indeed benefit from any volume market that Itanic might provide, and thuswH > the only difference in cost between an Alpha-based blade server and anD > Itanic-based blade server will be the per-processor cost, which weK > apparently both agree is not a major contributor to overall server price.   I But the flip side is required the extra processors shouldn't have a major.E effect on the overall cost of a IA64 system either - Acknowledging wem1 disagree about the value of the extra processors.e  H > (I said 5 months ago when this brouhaha began that there was no reason that: > Alpha-based platforms couldn't take advantage of all the 'industry-standard'dF > components that comparable Itanic platforms used, and this bi-sexual 'blade'9G > design just takes that to its logical extreme.  As long as the design H > doesn't cripple Alpha's strengths, it's win-win for both, and the onlyK > reason I suggested earlier that it *would* cripple Alpha was because thattL > was the only explanation for your assertion that somehow for a given levelJ > of performance the Itanic version would be less expensive than the Alpha > version.).  D Actually no - for the record I always felt there wouldn't be a priceH difference.  Meaning that it was going to require extra IA64 processors.C What my contention has been is Compaq should be able to deliver theeK competitive price/performance while avoiding the R&D costs of Alpha.  NeveriJ was that meant to imply same performance for less dollars.  The issue withJ avoiding the R&D costs is NOT to lower the price but to mitigate the risk.  K > Let's go over that once again, so that there's at least a chance you willnE > understand it:  Compaq has ensured that the only difference in costl between J > an Alpha blade system and an Itanic blade system will be the cost of theD > processors (or even less, if the Itanic system requires extra glue elementsJ > that the Alpha system by virtue of its on-chip features does not).  ThatE > difference will be only a small percentage of overall server cost -i assuming7 > that both systems have the same number of processors.t  J Lets try it again to see if we can get you to see the business dimensions.I Assume that with the added processors there is no cost advantage.  I knowa= you claim the added IA64 processors won't achieve competitiveuL price/performance but humor me.  What the issue here is risk.  If Compaq canL achieve the same competitive price performance without the R&D risk of AlphaH the text book business decision to make is to avoid the risk because theL added reward is not there.  I suspect you believe people who see things thisI way are idiots but it is the way the investment world works.  I also know J you believe that they have taken a far greater risk with the loss of theirD customer base - we choose to disagree on that point - I suspect your$ response will be well your an idiot.   > But in that case theA > Alpha-based system will have about twice the performance of thee Itanic-basedL > system - or, if you prefer, will require only about half the blades of theL > Itanic system and therefore cost considerably *less* than the Itanic-based# > system of comparable performance.o  K We will disagree on that - I don't claim you are wrong - what I do claim isq< you don't have the data to prove Compaq wrong at this point.  I > I have no problem if Compaq can sell zillions of Itanic systems of thato typeJ > (though strongly doubt that it will, because I doubt that they will turn outeB > to be competitive with the other options that will be out there, principallysH > POWER4 and Hammer).  What I have a problem with is not maintaining the AlphaaI > development that would have allowed them to offer right alongside those F > Itanic systems the option of buying more cost-effective (though lessK > 'industry-standard') Alpha systems - *even though each Alpha processor inrH > those systems carried the full burden of its development cost and thusK > funded it fully*.  *That's* what's so stupid - not only because it's lostnH > opportunity for Compaq, but because the other effects of killing AlphaJ > *will* be felt by customers (as described previously) and the incrediblyJ > sleazy and perfidious manner in which the murder was handled *will* have- > consequences on Compaq's future in general.t  H I suspect no matter how the interment of Alpha was handled it would have" caused a strong negative reaction.  K Understand Bill I see your judgement call on this that Compaq won't be able C to compete.  I see the judgement call of the chief architect of theuH TurboLaser and the Wildfire coming to a different conclusion.  I _known_F that the conclusion he came to originated in his group and effectivelyI killed his own project.  If you can objectively look at it for a moment - I and understanding that we are all idiots who see legitimacy in the CompaqeJ POV - if you were a bystander who's judgement would you tend to place moreL credibility in?  Or to put it another way, not to be nasty, as bright as youH are how many TurboLaser and Wildfire class systems have you delivered to market?s  K Bill I respect your opinion but I find the way you treat it as fact and thevL way you treat people who challenge you truly distastefully.  I have actuallyD used some of your POV's in discussing issue with Compaq on behalf of; Encompass but I took great care to remove the bitterness...   I > None of those are marginal calls.  Alpha's per-processor performance isaL > superior and would have only become more so over the time period for whichJ > any real information is available.  Significant percentages of customers areeI > already significantly upset at  a) the disruption they foresee in their A > operations and  b) the way Compaq went about this.  And Alpha'stI > profitability was well-established and relatively constant, despite the J > neglect of its owner (so this argument doesn't depend on an expansion of itsdG > markets, though such expansion was without doubt eminently possible).a > J > Compaq was stupid.  Compaq was sleazy.  Compaq was wrong.  And it's time to	 > fix it.   K Can't be changed - when it comes to marketplace perception you can't unringe	 a bell...l   ------------------------------    Date: 29 Nov 2001 08:06:07 -06009 From: kaplow_r@eisner.encompasserve.org.mars (Bob Kaplow)n= Subject: Re: Time for the Encompass US Inc CHAD-FREE Electionp3 Message-ID: <dQJUTk0bEiJm@eisner.encompasserve.org>   r In article <R_gN7.1641$29.1642086@typhoon.ne.mediaone.net>, "Terry C. Shannon" <terryshannon@mediaone.net> writes:L > Well, the Encompass BoD ballot is available in PDF format at the following; > URL. Print it out, fill it out, and mail it back in ASAP.s > I > You can vote for up to four of the eight candidates on the ballot, or a ! > lesser number if you so choose.t > M > Please take the time to download the PDF, cast your vote(s), and return thel6 > ballot by fax or snailmail. Deadline is December 12. >  > Thank you for your support!r > 2 > http://www.encompassus.org/whatsnew/bodElec.html  F Does this mean that Encompas will not be mailing out ballots? How willG members who don't have valid email addresses on file and don't read thev' newsgroups be notified of the election?   G And IIRC I've got 2 or more unique numbers. Do I get to vote "early and  often"? I'm from Chicago....    1 	26-October, 2001: A day that will live in infamy 4 	Support Freedom: http://www.indefenseoffreedom.org/   ------------------------------  % Date: Thu, 29 Nov 2001 02:54:08 -0500 - From: Jack Patteeuw <jjpatteeuw@peoplepc.com>eN Subject: Re: Tru64 .vs. HP-UX  (was: Compaq's Secret VMS Plans (The Inquirer)), Message-ID: <3C05E9A0.4EBE3132@peoplepc.com>   "Terry C. Shannon" wrote:   M > The HP way will more likely involve grafting the goodness of Tru64 onto theiJ > long-in-tooth HP-UX kernel. This will give Sun, IBM, et al a competitive/ > knockoff: nice clustering, antiquated kernel.t  I Terry, thanks for stating the obvious !  If there is a way that Carly andc( Curly can mess something up they will !!  I This is why we are (probably) going to Solaris another antiquated kernel,yM carrying much to much baggage, and certainly less robust and harder to manageT than Tru64.m   -- a  
 Jack Patteeuw    ------------------------------  % Date: Thu, 29 Nov 2001 19:31:51 +0010m' From: <paddy.o'brien@zzz.tg.nsw.gov.au>tN Subject: Re: Tru64 .vs. HP-UX  (was: Compaq's Secret VMS Plans (The Inquirer))5 Message-ID: <01KBAD80O81E000W44@tgmail.tg.nsw.gov.au>   - >In article <9u4fnr$cau@gap.cco.caltech.edu>,o5 >Vance R. Haemmerle <vance@alumni.caltech.edu> wrote:t >>H >>   Do I have to start this?  Little-endian is "industry standard"! It  >>must be because PCs are. >, >  I forgot this:  ;-> >o >--o >Vance Haemmerle  . From you, we guessed, and before I forget ;-).   Regards, Paddy   ------------------------------  % Date: Thu, 29 Nov 2001 10:54:31 +0100 5 From: Martin Knoblauch <Martin.Knoblauch@TeraPort.de>VN Subject: Re: Tru64 .vs. HP-UX  (was: Compaq's Secret VMS Plans (The Inquirer))+ Message-ID: <3C0605D7.5B093182@TeraPort.de>e   Duane Sand wrote:* > > > What I should have said, is that any automatic byte-swapping? > service supplied by the OS or runtime library, is going to be ; > useless for automatically converting files whose internal = > layouts are unknown to the system.  So the system generallyu  F  true for the OS, but not [neccessarily] true for runtime support. TheH OS of course does not know the internals of a binary input file. The RTLC may very well know. The DEC/COMPAQ Frotran compiler has support forhH reading/writing binary files both ways. I think it can even convert moreE exotic stuff like the old VAX formats. But of course, the RTL "knows"w. what to expect when it comes to a binary read.  A > can't help a big-endian version of an application work together = > with a little-endian version of that same application.  They> > problem is that when items of different sizes are intermixedA > in a file or struct, the two-byte numbers must be byte-reversedr: > on a two-byte basis, and four-byte numbers must be byte-9 > reversed on a four-byte basis, etc.  The system doesn'ts6 > know enough about each application's declarations or5 > methods to know how to parse and reverse each part.  >c  ?  The system doesn't, but the RTL may very well know. See above.l  o< > Some cases of file interchange can work via automatic byte5 > swapping.  This includes data bases (whose internalA< > structures are self-described somewhere), and simple files8 > where all items are the same size (eg text files), and< > Fortran binary files (where the read/write statements show; > how to parse each record into bigger and smaller fields).n >s  &  shoot. I should have read further :-)  ; > The cases that don't work, include C programs reading andn; > writing structs with scalar fields of different sizes, or < > worse yet, unions of those. (And that's most every program > I deal with.)d >T  D  But in that case, the program was not [trivially] portable from theG beginning. I mean, next you will ask that you can read absolute pointerr values from a binary file :-))  a Martin --  B ------------------------------------------------------------------B Martin Knoblauch         |    email:  Martin.Knoblauch@TeraPort.de7 TeraPort GmbH            |    Phone:  +49-89-510857-309n7 C+ITS                    |    Fax:    +49-89-510857-111n5 http://www.teraport.de   |    Mobile: +49-170-4904759c   ------------------------------    Date: 29 Nov 2001 05:55:39 -0600- From: Kilgallen@SpamCop.net (Larry Kilgallen)nN Subject: Re: Tru64 .vs. HP-UX  (was: Compaq's Secret VMS Plans (The Inquirer))3 Message-ID: <r$dg9OSITIa8@eisner.encompasserve.org>a  p In article <dGjN7.54809$RG1.29981861@news1.rdc1.sfba.home.com>, "Duane Sand" <duane.sand@mindspring.com> writes:  : > And I believe the OVMS group is aiming to supply one for > Alpha-to-IPF.p  E VMS Development says their translator will also translate VAX-to-IPF.    ------------------------------  % Date: Thu, 29 Nov 2001 12:59:50 +0100t/ From: Nigel Arnot <sysmgr@maxwell.ph.kcl.ac.uk>mN Subject: Re: Tru64 .vs. HP-UX  (was: Compaq's Secret VMS Plans (The Inquirer))7 Message-ID: <00A05C5E.31AC51CA.16@maxwell.ph.kcl.ac.uk>i   >  > The starting question is,tE > Will the merged Unix be big-endian, like PA-RISC and IPF HP-UX are,  > or little-endian, like Tru64?pJ > The IPF hardware can do both, even at same time for different processes,E > but no practical OS will ever be bi-endian in its raw file support.n  E Surely must be little-endian, because of various influences from bothtB the Linux/free software (mostly x86) and Microsoft desktop worlds. > I > One camp of users or the other will be losers, in having to fix endian-gH > sensitive nonportable source code before doing the "just a recompile",  C Badly-written code is always hard to port. Endian-ness problems aresG only one of many classes of latent bug: changes in call frame structuredF often cause latent bugs to emerge on any architecture migration (which> is one reason why there are still a lot of VAXen in use). And F unlike endian-ness problems, stack- and heap-structure ones tend to beJ "Heisenbugs" which may disappear if you try to compile with debug support,J or which appear at "random" (ie depend on the exact sequence of many prior+ operations, or even on the timing thereof).e  J Even the best-written code can have latent bugs, so you have to re-qualifyA everything even if it did recompile and run without hassle. WhichnG can be expensive. And which in the Unix world may open a door that endseG up with the customer migrating to his local PHM-approved vendor rather hC than staying in the fold. Or even falling victim to the Borg. Sigh.o   	Yours,e
 		Nigel Arnotu- 		NRA@MAXWELL.PH.KCL.AC.UK                   A  7 		"In the beginning there was nothing, which exploded."    ------------------------------  % Date: Thu, 29 Nov 2001 14:12:08 +0100a7 From: "Dijk, Jeroen van" <Jeroen.vanDijk@Getronics.com>eN Subject: RE: Tru64 .vs. HP-UX  (was: Compaq's Secret VMS Plans (The Inquirer))O Message-ID: <2795B75EF003D311801A00A0C906B511011C6AF4@cucexec.gbc.getronics.nl>s  > > > Q3: Obviously you can't go to market with two versions of  > UNIX--HP-UX andq& > > Compaq Tru64 UNIX. Which one wins? > > < > > A: HP has a great general-purpose data center UNIX with  > broad ISV coverage" > > and increasing share momentum. > > E > > Compaq has great UNIX clustering, RAS (reliability, availability,a1 > > scalability) and performance characteristics,  True.e  : Tru64 UNIX and HP-UX will became one OS on a new platform.E The 'new' name will be HP-UX, but it will be the best of both worlds.e  3 The 'news' comes from an Compaq storage specialist.g   ------------------------------  # Date: Thu, 29 Nov 2001 14:09:02 GMTl* From: "Bill Todd" <billtodd@metrocast.net>N Subject: Re: Tru64 .vs. HP-UX  (was: Compaq's Secret VMS Plans (The Inquirer))B Message-ID: <2krN7.144469$qx2.9013007@bin5.nnrp.aus1.giganews.com>  B "Dijk, Jeroen van" <Jeroen.vanDijk@Getronics.com> wrote in messageI news:2795B75EF003D311801A00A0C906B511011C6AF4@cucexec.gbc.getronics.nl...s   ...   < > Tru64 UNIX and HP-UX will became one OS on a new platform.G > The 'new' name will be HP-UX, but it will be the best of both worlds.p >l5 > The 'news' comes from an Compaq storage specialist.n  I I hope you're sufficiently well-acquainted with this situation to realizea? the necessity for placing 'news' within quotations marks above.r  K 'Spin' would be equally appropriate.  Duane's original comment was correct:oD some users (likely HP's) will win, some (likely Compaq's) will lose.   - bill   ------------------------------   Date: 29 Nov 2001 18:27:22 GMT' From: "Jim Strehlow" <jims@data911.com> 7 Subject: vms7.2-1 tape initialize 30 minute timeout bugf0 Message-ID: <9u5uma$clt@dispatch.concentric.net>  8 Compaq has been able to reproduce my tape drive problem.9 OpenVMS v7.2-1 from the CD-ROM does not have the problem.e? They are researching which patch creates my tape drive problem.   7 Jim Strehlow, Data911 Systems Manager, jimS@Data911.come Alameda, CA           Posted 11/19/2001 to comp.os.vms  ; Before loading the latest set of patches for OpenVMS v7.2-1sA a nightly backup process would either initialize or mount a tape.t If found, run the tape backup.# If not found, skip the tape backup.n  ' If there was no tape in the tape drive,d either initialize or a mount  would timeout within 30 seconds.6 I would test the error code and handle it accordingly.  E After the v7.2-1 patches, initialize times out in thirty (30) minutesr' and mount times out in six (6) minutes.v  : Is there a new undocumented initialize or sysgen parameter- that I must now set to reduce the timeout ...oA or is there a bug with one of the latest patches (e.g. MOUNT96) ?h   OpenVMS AlphaServer DS-10s   $ product show historyL ----------------------------------- ----------- ----------- ----------------( PRODUCT KIT TYPE OPERATION DATE AND TIMEL ----------------------------------- ----------- ----------- ----------------G DEC AXPVMS VMS721_DRIVER V4.0       Patch       Install     14-NOV-2001  07:43:07G DEC AXPVMS VMS721_CLIUTL V3.0       Patch       Install     13-NOV-2001w 07:45:18G DEC AXPVMS VMS721_SYS V11.0         Patch       Install     11-NOV-2001g 18:22:31G DEC AXPVMS VMS721_RMS V3.0          Patch       Install     11-NOV-2001p 18:08:05G DEC AXPVMS VMS721_MOUNT96 V3.0      Patch       Install     11-NOV-2001p 18:05:56G DEC AXPVMS VMS721_MAIL V1.0         Patch       Install     11-NOV-2001t 18:04:12G DEC AXPVMS VMS721_LIBRTL V3.0       Patch       Install     11-NOV-2001/ 17:47:26G DEC AXPVMS VMS721_LAT V1.0          Patch       Install     11-NOV-2001  17:43:35G DEC AXPVMS VMS721_FIBRE_SCSI V5.0   Patch       Install     11-NOV-2001o 17:39:04G DEC AXPVMS VMS721_DW_MOT_MUP V1.0   Patch       Install     11-NOV-2001h 17:34:15G DEC AXPVMS VMS721_DRIVER V3.0       Patch       Install     11-NOV-2001h 17:18:34G DEC AXPVMS VMS721_AUDSRV V2.0       Patch       Install     11-NOV-2001y 17:11:03G DEC AXPVMS TCPIP_ECO V5.1-153       Patch       Install     09-NOV-2001i 12:13:32G DEC AXPVMS VMS721_SYS V10.0         Patch       Install     07-JUL-2001  11:53:12G DEC AXPVMS VMS721_RMS V2.0          Patch       Install     07-JUL-2001s 11:45:21G DEC AXPVMS VMS721_LIBRTL V3.0       Patch       Install     06-JUL-2001y 09:03:18G DEC AXPVMS VMS721_UPDATE V3.0       Patch       Install     06-JUL-2001h 08:31:35G DEC AXPVMS VMS721_SYS V9.0          Patch       Install     22-JUN-20014 22:25:27G DEC AXPVMS VMS721_QMAN V1.0         Patch       Install     25-MAY-2001s 08:52:27G DEC AXPVMS VMS721_PTHREAD V3.0      Patch       Install     25-MAY-2001t 08:51:45G DEC AXPVMS VMS721_PPPD V1.0         Patch       Install     25-MAY-2001  08:39:48G DEC AXPVMS VMS721_LIBRTL V1.0       Patch       Install     25-MAY-2001t 08:29:41G DEC AXPVMS VMS721_DRIVER V2.0       Patch       Install     25-MAY-2001o 08:15:37G DEC AXPVMS VMS721_DELETE V1.0       Patch       Install     25-MAY-2001t 08:14:00G DEC AXPVMS VMS721_DEBUG V1.0        Patch       Install     25-MAY-2001w 08:06:54G DEC AXPVMS VMS721_CLIUTL V1.0       Patch       Install     25-MAY-2001t 08:04:16G DEC AXPVMS VMS721_BACKUP V1.0       Patch       Install     25-MAY-2001i 07:54:49G DEC AXPVMS VMS721_ACRTL V3.0        Patch       Install     25-MAY-2001l 07:33:35G DEC AXPVMS VMS721_LAN V2.0          Patch       Install     25-MAY-2001a 07:15:26G DEC AXPVMS VMS721_PCSI V1.0         Patch       Install     25-MAY-2001e 07:13:06G DEC AXPVMS VMS721_UPDATE V1.0       Patch       Install     25-MAY-2001l 06:37:06G DEC AXPVMS DECNET_PHASE_IV V7.2-1   Full LP     Install     22-MAY-2001b 18:25:57G DEC AXPVMS OPENVMS V7.2-1           Platform    Install     22-MAY-2001s 18:25:57G DEC AXPVMS VMS V7.2-1               Oper System Install     22-MAY-2001m 18:25:57  % Jim Strehlow, Data911 Systems Manager  Alameda, CAv   ------------------------------  + Date: Thu, 29 Nov 2001 09:11:30 -0700 (MST)y" From: John Nebel <nebel@csdco.com>% Subject: Wall Street Journal on Carly-G Message-ID: <Pine.OSF.4.21.0111290908540.22153-100000@athena.csdco.com>b  ? The article is in the 11/26 edition on the editorial page, A18.s   "Beware the Self-promoting CEO"s   ------------------------------  # Date: Thu, 29 Nov 2001 17:41:02 GMTk& From: "Jeff Killeen" <Jeff@IDM-IO.com>) Subject: Re: Wall Street Journal on Carlyd8 Message-ID: <OquN7.1232$h24.106865@typhoon1.gnilink.net>  I An article I hope no one takes the article to heart all the way through -u+ not because of this (which I agree with)...s  L "The really striking point is that Ms. Fiorina had been on the job less thanJ six months before she posed for the cover of Forbes as a superhero savior.I She hadn't done anything significant. And now, two years later, she stilllG hasn't produced any significant results. During her watch, HP stock haseC fallen more than 50% -- substantially underperforming the Dow JoneseI Industrial Average and S&P 500, and falling nearly twice as far as Apple,h/ IBM and Sun Microsystems over the same period."m  G ...but because of the thinking below would not be good for the Compaq I  want...d  C "Charles R. Walgreen III, for example, transformed Walgreens from auJ mediocrity into a machine that produced cumulative stock returns that beatK Intel by nearly two to one, GE by nearly five to one, and the general stockcK market by over 15 times from 1975 to 2000. These results stem directly fromvI Mr. Walgreen's stoic resolve to make a series of tough decisions, such aslE jettisoning the entire food-service business (which dated back to his G grandfather and was quite profitable) because it was not an arena where 3 Walgreens could be the absolute best in the world."e   ------------------------------  % Date: Thu, 29 Nov 2001 19:14:17 +0100e1 From: John McLean <mcleanj@swissonline.delete.ch>l) Subject: Re: Wall Street Journal on Carlyi5 Message-ID: <3C067AF9.5F210265@swissonline.delete.ch>t   John Nebel wrote:  > A > The article is in the 11/26 edition on the editorial page, A18.h > ! > "Beware the Self-promoting CEO"u    
 Which one ???      John McLeanf   ------------------------------  % Date: Thu, 29 Nov 2001 19:16:13 +0100l1 From: John McLean <mcleanj@swissonline.delete.ch> ) Subject: Re: Wall Street Journal on Carly 5 Message-ID: <3C067B6D.42103C31@swissonline.delete.ch>    Jeff Killeen wrote:r > K > An article I hope no one takes the article to heart all the way through --- > not because of this (which I agree with)...a > N > "The really striking point is that Ms. Fiorina had been on the job less thanL > six months before she posed for the cover of Forbes as a superhero savior.K > She hadn't done anything significant. And now, two years later, she still.I > hasn't produced any significant results. During her watch, HP stock hastE > fallen more than 50% -- substantially underperforming the Dow JonesSK > Industrial Average and S&P 500, and falling nearly twice as far as Apple,,1 > IBM and Sun Microsystems over the same period."l  C Apple, IBM and Sun .... but no mention of Compaq.  I wonder why ???h     John McLeana   ------------------------------  + Date: Thu, 29 Nov 2001 15:10:00 +0100 (MET) 9 From: Phillip Helbig <HELBPHI@sysdev.deutsche-boerse.com>t= Subject: Re: why not a communityDeveloped[tm] version of VMS?l; Message-ID: <01KBA3YGVU7C90YLDV@sysdev.deutsche-boerse.com>n  B > Wouldn't the intervening period be good to either duplicate from	 > scratchm  I Calculate how many person-years have gone into VMS.  Do you really think .E that you can muster those resources?  The comparison with Linux is a dF no-go; that some students can throw something together in their spare C time which rivals commercial unix says something about the lack of tI quality of commercial unix, not about the ease of creating/duplicating a V real OS from scratch.)  D > Or, pigs flying now, maybe Compaq could kindly just throw open theG > source code, and let companies of greater insight provide support foru  E Why should they?  Having proprietary access to the source code makes l them money.e  C Personally, I would rather pay for good source code developed by a  C small, tightly-knit team of professionals than go with open-source e4 software, certainly for something the size of an OS.   ------------------------------  % Date: Thu, 29 Nov 2001 16:56:02 +0100 % From: "Dr. Dweeb." <Dweeb@nospam.com>h= Subject: Re: why not a communityDeveloped[tm] version of VMS?-/ Message-ID: <8UsN7.91$WK5.4468@news.get2net.dk>0   Abso-fucking-lutely !   F "Phillip Helbig" <HELBPHI@sysdev.deutsche-boerse.com> wrote in message5 news:01KBA3YGVU7C90YLDV@sysdev.deutsche-boerse.com...0D > > Wouldn't the intervening period be good to either duplicate from > > scratch. >>J > Calculate how many person-years have gone into VMS.  Do you really thinkF > that you can muster those resources?  The comparison with Linux is aG > no-go; that some students can throw something together in their spare D > time which rivals commercial unix says something about the lack ofJ > quality of commercial unix, not about the ease of creating/duplicating a > real OS from scratch.t >KF > > Or, pigs flying now, maybe Compaq could kindly just throw open theI > > source code, and let companies of greater insight provide support fore > F > Why should they?  Having proprietary access to the source code makes
 > them money.  >eD > Personally, I would rather pay for good source code developed by aD > small, tightly-knit team of professionals than go with open-source6 > software, certainly for something the size of an OS.   ------------------------------   Date: 29 Nov 2001 16:04:05 GMTB From: bertrand@perceval.cnam.fr (BERTRAND =?iso-8859-1?Q?Jo=EBl?=)= Subject: Re: why not a communityDeveloped[tm] version of VMS?p6 Message-ID: <slrna0cn3h.c5o.bertrand@perceval.cnam.fr>  " Le Thu, 29 Nov 2001 16:56:02 +0100( Dr. Dweeb. <Dweeb@nospam.com> crivait : >Abso-fucking-lutely ! > G >"Phillip Helbig" <HELBPHI@sysdev.deutsche-boerse.com> wrote in messageo6 >news:01KBA3YGVU7C90YLDV@sysdev.deutsche-boerse.com...E >> > Wouldn't the intervening period be good to either duplicate froms >> > scratch >>K >> Calculate how many person-years have gone into VMS.  Do you really thinkvG >> that you can muster those resources?  The comparison with Linux is aeH >> no-go; that some students can throw something together in their spareE >> time which rivals commercial unix says something about the lack ofrK >> quality of commercial unix, not about the ease of creating/duplicating a. >> real OS from scratch. >>G >> > Or, pigs flying now, maybe Compaq could kindly just throw open thedJ >> > source code, and let companies of greater insight provide support for >>G >> Why should they?  Having proprietary access to the source code makes  >> them money. >>E >> Personally, I would rather pay for good source code developed by a E >> small, tightly-knit team of professionals than go with open-source:7 >> software, certainly for something the size of an OS.d  C 	I have restarted the old FreeVMS project (http://freevms.free.fr). E At this time, we have a kernel for i386 in prealpha stage, a SRM-likewF console which can boot this kernel or another operating system and I'mF working on the DCL interpreter. We have a CVS tree and a mailing list.A You can find all informations about this project on our web page.-  C 	We need some contributers to (re)write some libraries (STR$, RTL$,KN SMG$...) and the basic functions (COPY, PURGE, SET...) in the DCL interpreter.  	 	Regards,n   	JKB   ------------------------------  % Date: Thu, 29 Nov 2001 12:32:53 +0100t/ From: Nigel Arnot <sysmgr@maxwell.ph.kcl.ac.uk>5: Subject: Xbox (was: re: Compaqs VMS plans for IPF port ...7 Message-ID: <00A05C5A.6DC22300.28@maxwell.ph.kcl.ac.uk>n   > > ' > > The Xbox uses a 733MHz Pentium III.  > > C > > As far as I can tell (I havn't actually exampined one) the onlysK > > ports on the thing are the 4 game controller ports (it should be fairly K > > easy for them to make a keyboard and mouse that plugs into one of them)gG > > and a 10/100 ethernet port. Yep, the Xbox comes with ethernet - notsI > > that there is anything that will currently run on it that uses it for-L > > anything, but the intended use is to plug it into a broadband connection > > to the internet.  4 There's a detailed analysis at www.tomshardware.com.    > Will it boot off the ethernet? >   M This, and suchlike questions, depend on Microsoft and on the law. It would be I very hard to prevent at the hardware level, but its BIOS is quite anotherl matter.i  H The key issue is the degree of control that the vendor (MS) has over theK platform. The X-box is designed to prevent you, the end user, from using iteL to run anything other than software sanctioned by MS. Reverse-engineering itH (ie finding out how to make it run *your* code off the network) may wellI be illegal under the DMCA (and might end up with you facing several yearsiM in jail, should you live in or even later visit the USA). I suspect, from the H presence of the hard disk and Ethernet, that Microsoft's idea is that itG will be a low-cost alternative to a PC, if you're willing to sign up tofG Microsoft's forthcoming(?) X-box subscription services (which IMO will oM start dirt cheap and increase in cost at a rate well above inflation, once a pA lot of moron^H^H^H^H^Hcustomers have entangled themselves in thisi7 devilish web. I feel slightly ill thinking about this.),  E It's worth noting that Sony sell a linux-on-PS2 kit in Japan and I'veiJ heard that they intend to market it in the rest of the world soon. I thinkJ this must mean that the boot details for the PS2 are "open" and if so it'sA clearly the cosole of choice for hackers and anyone who wants to hL resist "the Borg". Will be interesting to see whether open games and/or use I of PS2s for "serious" computing emerge, and if so whether they help Sony eH sell enough to counterbalance the resulting loss of licensing royalties.   	Yours,K
 		Nigel Arnot - 		NRA@MAXWELL.PH.KCL.AC.UK                   o  7 		"In the beginning there was nothing, which exploded."    ------------------------------  % Date: Thu, 29 Nov 2001 14:47:09 +0000v% From: Alan Greig <a.greig@virgin.net>u> Subject: Re: Xbox (was: re: Compaqs VMS plans for IPF port ...8 Message-ID: <m7ic0uohi7c86623d445o71phdjl7t7tot@4ax.com>  / On Thu, 29 Nov 2001 12:32:53 +0100, Nigel Arnot'$ <sysmgr@maxwell.ph.kcl.ac.uk> wrote:    F >It's worth noting that Sony sell a linux-on-PS2 kit in Japan and I'veK >heard that they intend to market it in the rest of the world soon. I think   @ A quick search of XBOX chat reveals that various Linux types areC confident they can get Linux up on the XBOX although Microsoft willm
 not be happy.s  = Also I see someone has already reverse engineered some of thewC networking code and has released Linux source code to act as a gamebF network server for the XBOX allowing you to play multi-user across the< internet. Microsoft intend to release their own subscriptionF controlled servers for mult-user XBOX next year so likely will try and4 take legal action to shut down the free server code.  F Allegedly MS have also twiddled round a few connecting cables here and+ there to disguise the standards underneath.i   -- Alan   ------------------------------  % Date: Thu, 29 Nov 2001 03:10:08 -0500r- From: Jack Patteeuw <jjpatteeuw@peoplepc.com>w Subject: Re: xdmcp, Message-ID: <3C05ED60.BB120D2E@peoplepc.com>   Hoff Hoffman wrote:b > H >   XDM is supported on current TCP/IP Services versions.  Specifically,; >   on TCP/IP Services V5.1 and later.  (DHCP client, too.)t >   ) Yep, its there and it works ... mostly !!w  L Right now there is a P*SSING contest going on between different parts of VMSL Engineering on who will fix a couple of (submitted, documented and verified)H bugs related to XDMCP.  The biggest one is that logins done via an XDMCPE connection are **NOT** registered as an "INTERACTIVE" login in SYSUAF 	 database.v  M It seems that no one can get the "children" to play nice together and resolvec
 this problem.   K Not a bug, but would sure be nice if we could put up a better splash screen 7 than "Username: Password:" with no graphic background !i     -- p  
 Jack Patteeuwa   ------------------------------    Date: 29 Nov 2001 06:02:14 -0800( From: bob@instantwhip.com (Bob Ceculski)> Subject: Re: Yahoo - IBM unseated in new supercomputer ranking= Message-ID: <d7791aa1.0111290602.4925344f@posting.google.com>n  m "Ken Farmer" <kfarmer@tru64.org> wrote in message news:<qUeN7.92719$fm5.17025390@typhoon.southeast.rr.com>...nD > "...At the top of the new list, created by IDC in cooperation withJ > supercomputer users, is a 3,024-processor Compaq Computer machine called> > Terascale based at the Pittsburgh Supercomputing Center...." > N > http://dailynews.yahoo.com/h/cn/20011128/tc/ibm_unseated_in_new_supercompute > r_ranking_1.html  M I thought Alpha couldn't keep up to Itanium?  By the way, where is Itanium onh	 the list?    ------------------------------  # Date: Thu, 29 Nov 2001 14:23:39 GMTc4 From: "Terry C. Shannon" <terryshannon@mediaone.net>> Subject: Re: Yahoo - IBM unseated in new supercomputer ranking; Message-ID: <LxrN7.496$zX1.1438268@typhoon.ne.mediaone.net>-  5 "Bob Ceculski" <bob@instantwhip.com> wrote in messageR7 news:d7791aa1.0111290602.4925344f@posting.google.com...73 > "Ken Farmer" <kfarmer@tru64.org> wrote in messaget; news:<qUeN7.92719$fm5.17025390@typhoon.southeast.rr.com>... F > > "...At the top of the new list, created by IDC in cooperation withL > > supercomputer users, is a 3,024-processor Compaq Computer machine called@ > > Terascale based at the Pittsburgh Supercomputing Center...." > >  > >XL http://dailynews.yahoo.com/h/cn/20011128/tc/ibm_unseated_in_new_supercompute > > r_ranking_1.html >VL > I thought Alpha couldn't keep up to Itanium?  By the way, where is Itanium on > the list?a  I Itanium is nowhere on the list, it'll probably be several years before it.L makes a showing. Nobody ever said Alpha couldn't keep up with Itanium; CPQ'sL projections were that Alpha would lose its performance edge at or beyond the% EV8 timeframe. Your mileage may vary.    ------------------------------  % Date: Thu, 29 Nov 2001 18:21:02 +0100t9 From: Ralf Eichmann <ralf.eichmann@physik.uni-marburg.de>5> Subject: Re: Yahoo - IBM unseated in new supercomputer ranking5 Message-ID: <3C066E7E.6A1150BC@physik.uni-marburg.de>J   Bob Ceculski wrote:  > o > "Ken Farmer" <kfarmer@tru64.org> wrote in message news:<qUeN7.92719$fm5.17025390@typhoon.southeast.rr.com>...aF > > "...At the top of the new list, created by IDC in cooperation withL > > supercomputer users, is a 3,024-processor Compaq Computer machine called@ > > Terascale based at the Pittsburgh Supercomputing Center...." > >VP > > http://dailynews.yahoo.com/h/cn/20011128/tc/ibm_unseated_in_new_supercompute > > r_ranking_1.html > O > I thought Alpha couldn't keep up to Itanium?  By the way, where is Itanium on  > the list?   < Take a look at positions no. 34 and 132 in the 11/2001 list.   Best, 
 Ralf Eichmannn --  A Adr.: Universitaet Marburg     Fachbereich Physik, AG HL(Theorie)c#       Renthof 5     D-35032 Marburg1A Tel.: +49 (0)6421 28 24240              Fax: +49 (0)6421 28 270765   ------------------------------   End of INFO-VAX 2001.664 ************************