1 INFO-VAX	Sat, 02 Nov 2002	Volume 2002 : Issue 605       Contents:
 Re: <None>
 Re: <None>
 Re: <None>
 Re: <None> ANN: GRAB - a search utility+ Re: can vms C access the keyboard directly? + Re: can vms C access the keyboard directly? + Re: can vms C access the keyboard directly? + Re: can vms C access the keyboard directly?  Re: CHARON-ALPHA - soon ???? Re: CHARON-ALPHA - soon ???? Re: Configuring OpenVMS-6.2 4 CSWING bug squashed. (was: OpenVMS 7.3-1 and CSWING)' Re: Does anyone use P7 in DCPS$STARTUP? ' Re: Does anyone use P7 in DCPS$STARTUP? ' Re: Does anyone use P7 in DCPS$STARTUP? ' Re: Does anyone use P7 in DCPS$STARTUP? ' Re: Does anyone use P7 in DCPS$STARTUP? ' Re: Does anyone use P7 in DCPS$STARTUP? ' Re: Does anyone use P7 in DCPS$STARTUP?  Re: equivalent of touch in VMS Re: errno equivalent Re: FTP with SSL for VMS/Alpha?  Re: FTP with SSL for VMS/Alpha?  Re: FTP with SSL for VMS/Alpha?  HELP suggestion 1 How to do inverse video, bold, etc. in Vax Pro-C? 5 Re: How to do inverse video, bold, etc. in Vax Pro-C? 5 Re: How to do inverse video, bold, etc. in Vax Pro-C? 5 Re: How to do inverse video, bold, etc. in Vax Pro-C? 5 Re: How to do inverse video, bold, etc. in Vax Pro-C? 5 Re: How to do inverse video, bold, etc. in Vax Pro-C?  Re: HP Advocacy Site RE: HP Advocacy Site Re: HP Advocacy Site Re: HP User Advocacy ADE survey 7 Re: Intel Conceding 64-Bit Desktops to IBM, AMD - eWeek 7 RE: Intel Conceding 64-Bit Desktops to IBM, AMD - eWeek 7 RE: Intel Conceding 64-Bit Desktops to IBM, AMD - eWeek 7 Re: Intel Conceding 64-Bit Desktops to IBM, AMD - eWeek 7 RE: Intel Conceding 64-Bit Desktops to IBM, AMD - eWeek 7 RE: Intel Conceding 64-Bit Desktops to IBM, AMD - eWeek 7 RE: Intel Conceding 64-Bit Desktops to IBM, AMD - eWeek 7 Re: Intel Conceding 64-Bit Desktops to IBM, AMD - eWeek 7 Re: Intel Conceding 64-Bit Desktops to IBM, AMD - eWeek 7 Re: Intel Conceding 64-Bit Desktops to IBM, AMD - eWeek 7 Re: Intel Conceding 64-Bit Desktops to IBM, AMD - eWeek 7 Re: Intel Conceding 64-Bit Desktops to IBM, AMD - eWeek 7 Re: Intel Conceding 64-Bit Desktops to IBM, AMD - eWeek 7 Re: Intel Conceding 64-Bit Desktops to IBM, AMD - eWeek 7 Re: Intel Conceding 64-Bit Desktops to IBM, AMD - eWeek 5 Re: Itanic ordered to halt by court. Iceberge free... & Re: Looking for contract opportunities Netware and OpenVMS 7.3 7 Re: Not sure if anyone else I seen this from LinuxWorld 0 Re: OpenVMS printing to printer via windows 2000$ Re: Was OpenVMS left out on purpose?$ Re: Was OpenVMS left out on purpose?$ Re: Was OpenVMS left out on purpose?$ Re: Was OpenVMS left out on purpose?$ Re: Was OpenVMS left out on purpose?% Why Is ALTPRI Needed to Set Affinity? A [OT] - Sub-standard network guys (was: Re: Remote Console access)   F ----------------------------------------------------------------------   Date: 1 Nov 2002 15:35:46 -0800 7 From: jones.computer.srv@worldnet.att.net (Daryl Jones)  Subject: Re: <None> = Message-ID: <8a646952.0211011535.5cd3d36d@posting.google.com>   ` "David J. Dachtera" <djesys.nospam@fsi.net> wrote in message news:<3DBF478E.7073C390@fsi.net>... > Tom Linden wrote:  > > = > > FWIW, some years ago we used to figure that 2 VUP = 1 MIP  > $ > More like .9 VUP = roughly 1 MIP.  > H > There's a story there, but it's been too long and I don't remember the
 > details. > J > Nowadays, they talk about "transactions" per second, but I've yet to see: > a lucid explanation of what constitutes a "transaction".   Hello Everyone,   A Back in the 1980s, the VAX/VMS 11/780 was rated by DEC as a 1 MIP F machine. However, there was a brew-ha-ha with IBM over what was a MIP.F This caused DEC to rate the VAX 11/780 as a 1 VUP machine instead of aD 1 MIP machine. The origainal VAX processor had a 250,000 instructionC set. The MicroVAX chip with its 150,000 instruction set was used in F the MicroVAX I and II, then later in the VAX6000 and VAX7000 machines. The Alpha is a RISC chip.   B I did a sort test once on a VAX750 against a IBM 4381. The IBM wasF faster. However, IBM will use its IO processor for sorting rather than8 its main processor. Thereby, cpu run-time being shorter.   C I ran a Cobol program that used packed-decimal arithmetic once on a F VAX6420 and VAX8820. The program ran faster in the VAX8820 even thoughD the VAX6420 was 2 VUP faster. Pack-decimal caused the program to run; slower on the VAX6420 with fewer number of instruction set.    D I was recently involved in a VAX 7800 class machine to an Alpha 8400E machine. The VAX (6) had more processors than the Alpha (4), however, F the VAX was nailed to the wall. What I did notice was a VAX running atC 100% and after the conversion the Alpha was running 66% on average. A Even though the Alpha EV600 appeared to much much faster than the 
 VAX7800 chip.   C Comparison between IBM and HP/COMPAQ/DEC hardware has been going on E since the 1980s. You can use the comparison between IBM and Alpha via = Oracle. However, there is more than speed of a processor when C comparing IBM and Alpha. I do remember that changing from VM/DOS to E VM/VSE operating system on IBM was a $50k to $75k on the same machine A on the same operating system. VMS has always been reasonable when > converting from one VAX processor to the next generations thisD includes the VAX to Alpha upgrades. Once you have converted from VAXE to Alpha, then the convertion from Alpha to I64 will be less painfull = than VAX to I64. This excludes privilege code changes. Please F remember, cluster capablitlies on VMS, THERE IS NONE LIKE IT! However, there are cheap imitations.    I hope this helps.   Daryl    ------------------------------  % Date: Fri, 01 Nov 2002 22:07:11 -0600 1 From: "David J. Dachtera" <djesys.nospam@fsi.net>  Subject: Re: <None> ' Message-ID: <3DC34F6F.5C3FFCAD@fsi.net>    John Smith wrote:  > > > "David J. Dachtera" <djesys.nospam@fsi.net> wrote in message# > news:3DBF478E.7073C390@fsi.net...  > > Tom Linden wrote:  > > > ? > > > FWIW, some years ago we used to figure that 2 VUP = 1 MIP  > > % > > More like .9 VUP = roughly 1 MIP.  > > J > > There's a story there, but it's been too long and I don't remember the > > details. > > L > > Nowadays, they talk about "transactions" per second, but I've yet to see< > > a lucid explanation of what constitutes a "transaction". > G > transaction - everything the bank dings you with a service charge for   A You don't wanna know how I misread that the first time through...    --   David J. Dachtera  dba DJE Systems  http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/    ------------------------------  % Date: Fri, 01 Nov 2002 22:10:28 -0600 1 From: "David J. Dachtera" <djesys.nospam@fsi.net>  Subject: Re: <None> ' Message-ID: <3DC35034.C9F80240@fsi.net>    Daryl Jones wrote: > b > "David J. Dachtera" <djesys.nospam@fsi.net> wrote in message news:<3DBF478E.7073C390@fsi.net>... > > Tom Linden wrote:  > > > ? > > > FWIW, some years ago we used to figure that 2 VUP = 1 MIP  > > % > > More like .9 VUP = roughly 1 MIP.  > > J > > There's a story there, but it's been too long and I don't remember the > > details. > > L > > Nowadays, they talk about "transactions" per second, but I've yet to see< > > a lucid explanation of what constitutes a "transaction". >  > Hello Everyone,  > C > Back in the 1980s, the VAX/VMS 11/780 was rated by DEC as a 1 MIP H > machine. However, there was a brew-ha-ha with IBM over what was a MIP.H > This caused DEC to rate the VAX 11/780 as a 1 VUP machine instead of aF > 1 MIP machine. The origainal VAX processor had a 250,000 instructionE > set. The MicroVAX chip with its 150,000 instruction set was used in H > the MicroVAX I and II, then later in the VAX6000 and VAX7000 machines. > The Alpha is a RISC chip.  >  > [Story snipped]  >  > I hope this helps.  B Well, o.k., but what the hell is a "transaction", how many machineC cycles does it take to complete one, and how does that help measure + actual CPU performance as compared to MIPs?    --   David J. Dachtera  dba DJE Systems  http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/    ------------------------------  % Date: Fri, 01 Nov 2002 23:28:40 -0500 - From: JF Mezei <jfmezei.spamnot@videotron.ca>  Subject: Re: <None> , Message-ID: <3DC3546C.901BE158@videotron.ca>   "David J. Dachtera" wrote:I > > transaction - everything the bank dings you with a service charge for  > C > You don't wanna know how I misread that the first time through...    Yes, I want to know...   ------------------------------  % Date: Sat, 02 Nov 2002 00:24:27 +0000 . From: Graham Burley <100625.30@compuserve.com>% Subject: ANN: GRAB - a search utility . Message-ID: <3DC31B3B.17376347@compuserve.com>  D GRAB is a search utility for VMS, the main features of interest are:  A  o Search using wildcard search strings (* and %) using /WILDCARD E  o Extract windows or sections by search strings using /CUT=(s1[,s2]) 3  o Incremental processing using /CONTEXT and /LIMIT C  o Scripting support using /SAVE to save information in DCL symbols   B Source distribution including help file and example procedures can be downloaded from here:  2  http://www.encompasserve.org/~burley/grab_src.zip  ! Requires DEC C or VAX C to build.   A This is version 0.21, the first to be released into the wild, I'd  be grateful for any feedback.   . %HOBBYIST-?-CODE, coded by an OpenVMS hobbyist   ------------------------------  % Date: Fri, 01 Nov 2002 20:56:29 -0000 ! From: Z  <zarlenga@conan.ids.net> 4 Subject: Re: can vms C access the keyboard directly?/ Message-ID: <us5qjtlt6cul87@corp.supernews.com>   ! John Santos <JOHN@egh.com> wrote: N :> How does getchar handle PF keys that generate multiple characters on UNIX ?Q :> Does it feed it to to your application one at a time as if they had been typed P :> separately (forcing our app to have the logic to reconstitute multi-characterL :> escape sequences), or does it provide some method to get you a PFKEY as a :> single entity ?   :>  O :> On VMS, the $QIO service and terminal drivers do have the ability to process O :> escape sequences as a single IO and you are told how long the terminator is.   K : If you use SMG, it does all this for you, and more.  You get a terminator J : code such as SMG$K_TRM_UP or SMG$K_TRM_SELECT or SMG$K_TRM_CR, which youN : can directly dispatch on.  Your app doesn't need to know anything about ANSI  @ Can SMG$ distinguish between someone pressing ESC, [, and then A" and someone pressing the up arrow?     = : That being said, I can see where an exact emulation of Unix A : behaviour would make porting from Unix and maintaining portable 7 : apps easier, and I think we are all in favor of that.    Agreed.    ------------------------------  % Date: Fri, 01 Nov 2002 20:57:06 -0000 ! From: Z  <zarlenga@conan.ids.net> 4 Subject: Re: can vms C access the keyboard directly?/ Message-ID: <us5ql2257omma4@corp.supernews.com>   . JF Mezei <jfmezei.spamnot@videotron.ca> wrote:M : How does getchar handle PF keys that generate multiple characters on UNIX ?    One key at a time.    P : Does it feed it to to your application one at a time as if they had been typedO : separately (forcing our app to have the logic to reconstitute multi-character  : escape sequences),   Yes.    8 : or does it provide some method to get you a PFKEY as a : single entity ?    Not that I know of.    ------------------------------   Date: 1 Nov 2002 15:19:00 -0600 - From: koehler@encompasserve.org (Bob Koehler) 4 Subject: Re: can vms C access the keyboard directly?3 Message-ID: <6Dlp03TppIUe@eisner.encompasserve.org>   S In article <us5qjtlt6cul87@corp.supernews.com>, Z  <zarlenga@conan.ids.net> writes:  > B > Can SMG$ distinguish between someone pressing ESC, [, and then A$ > and someone pressing the up arrow?  D    No.  There is no distinction unless your are on a workstation andI    can access the scan codes.  IIRC VMS does not supply a documented API  F    to the scan code outside of X11.  There's a decw$util: program that    can access the scan codes.    ------------------------------  $ Date: Sat, 2 Nov 2002 00:42:01 -0500  From: John Santos <JOHN@egh.com>4 Subject: Re: can vms C access the keyboard directly?4 Message-ID: <1021102004125.400C-100000@Ives.egh.com>  ! On 1 Nov 2002, Bob Koehler wrote:   U > In article <us5qjtlt6cul87@corp.supernews.com>, Z  <zarlenga@conan.ids.net> writes:  > > D > > Can SMG$ distinguish between someone pressing ESC, [, and then A& > > and someone pressing the up arrow? > F >    No.  There is no distinction unless your are on a workstation andK >    can access the scan codes.  IIRC VMS does not supply a documented API  H >    to the scan code outside of X11.  There's a decw$util: program that >    can access the scan codes.   E You answered this better than I could have.  I just want to point out E that if you *want* to see each character separately, and not have the @ terminal driver recognize the escape sequences, and not have SMGA determine their meaning for you, then you can set the terminal to H NOESCAPE mode and process them your self (a la getchar().)  This defeats0 one of the main advantages of using SMG, though.   --   John Santos  Evans Griffiths & Hart, Inc. 781-861-0670 ext 539   ------------------------------  $ Date: Fri, 1 Nov 2002 14:00:59 -0500# From: "Tom Cole" <tom.cole@sas.com> % Subject: Re: CHARON-ALPHA - soon ???? / Message-ID: <apuj1b$fbb$1@license1.unx.sas.com>   I I bet as one of the authors of CHARON-VAX, he's not pulling it out of his 
 butt... :)  L If you think about the issues around implementing a CISC instruction set viaI emulation, each VAX instruction neatly translates into a reasonably small J but non-zero-sized amount of logic, expressable as a method switch case orI whatever invocation architecture you choose.  The cost of translating the C instruction is relatively low compared to the cost of executing it, G especially since many VAX instructions require *quite* a bit of code to H implement correctly - for example, D_FLOAT math isn't trivial on an IEEEJ architecture, and EMODD becomes more than a few lines of code to implement as a result... :)   J By comparison, each Alpha instruction is quite a bit more succinct in whatI it does - the cost of decoding the instruction could easily match or even K overwhelm the cost of executing the logic associated with that instruction. I In short, you don't get a pleasing tradeoff between virtual machine speed 8 and compiled methods like you do in a CISC architecture.  I I would love to see an Alpha emulator worth it's salt, and I believe that H SRI are the folks who can do it, but I do agree that a 1:4 ratio is veryJ unlikely between emulated and physical machine speeds - even with a wickedJ clever instruction decoder, it's hard to imagine anything faster than 1:20E or something like that, especially given that you pretty much have to K emulate the virtual memory processing no matter how clever your instruction 0 decoder is, and that's a bear in it's own right.  F In short, I don't think that chip complexity is the issue; it's more a' fundamental aspect of RISC versus CISC.     9 "Keith A. Lewis" <lewis@mazda.mitre.org> wrote in message ( news:apuc51$9mv$1@newslocal.mitre.org...@ > "Robert Boers" <robert.boers@softresint.com> writes in articleD <3dc24668$1@news.deckpoint.ch> dated Fri, 1 Nov 2002 10:16:21 +0100:G > >The ratio of 1.5 GHz to 400 MHz is far too optimistic, and mapping a  64-bitF > >CPU emulator on a 32-bit host system is * very * inefficient. Given Alpha's K > >excellent arithmetic units, a reasonable host would be a dual AMD Hammer G > >system. A dual 2 GHz Hammer system would provide the equivalent of a  100 - $ > >140 MHz Alpha, not very exciting. > K > I agree that Dean's ratio is over-optomistic, and the way he presented it J > made it sound like he pulled the numbers out of his butt.  Are you doing the J > same thing?  If there are ALU and transistor-count comparisons out there% > somewhere, I'd like to take a look.  > - > --Keith Lewis              klewis$mitre.org @ > The above may not (yet) represent the opinions of my employer.   ------------------------------  % Date: Fri, 01 Nov 2002 17:10:16 -0800 % From: Dean Woodward <deanw@rdrop.com> % Subject: Re: CHARON-ALPHA - soon ???? ( Message-ID: <3DC325F8.6060402@rdrop.com>   Keith A. Lewis wrote:  > "Robert Boers" <robert.boers@softresint.com> writes in article <3dc24668$1@news.deckpoint.ch> dated Fri, 1 Nov 2002 10:16:21 +0100:  > M >>The ratio of 1.5 GHz to 400 MHz is far too optimistic, and mapping a 64-bit M >>CPU emulator on a 32-bit host system is * very * inefficient. Given Alpha's J >>excellent arithmetic units, a reasonable host would be a dual AMD HammerL >>system. A dual 2 GHz Hammer system would provide the equivalent of a 100 -# >>140 MHz Alpha, not very exciting.   D I'll admit to knowing little to nothing of the issues involved, and A bow to those who have a clue where I merely have a desire- quite  D possibly, the same desire as many others- some sort of portable[1], 5 affordable[2], relatively up-to-date[3] VMS platform.   K > I agree that Dean's ratio is over-optomistic, and the way he presented it ; > made it sound like he pulled the numbers out of his butt.   < I'd prefer to pull them out of the air- much less offensive,
 thank you.   [1] i.e., a laptop. G [2] for "workstation" class hardware values of "affordable".  $10K for  @ a laptop is a bit extreme...  But a Wintel laptop with a couple E thousand of emulation software might be doable- especially if it can  ) do other things besides run the emulator. G [3] We've stopped developing for VAX; I expect our next production and  + development boxen will have "Intel inside".    ------------------------------  # Date: Fri, 01 Nov 2002 17:37:41 GMT 4 From: Tim Llewellyn <tim.llewellyn@blueyonder.co.uk>$ Subject: Re: Configuring OpenVMS-6.20 Message-ID: <3DC2BBC8.E3D8B767@blueyonder.co.uk>   Jurren Bouman wrote: >  > Dear All,  > G > We have moved a box with OpenVMS version 6.2 to another location. But F > now we need to change the IP-address, Gateway address and the SubnetF > mask address. I've never done and I'd like to know how this is done.' > Does anyone here know how to do this?  >   F If UCX/TCPIP Services, it can be done from the ucx$config/tcpip$config utility.    The restart ucx/tcpip or reboot.  ? I have done this with a remote site once or twice. Double check  everything you do.  H > Then I'd like to know the mac-address of the ethernet card. How can we > find this under OpenVMS 6.2? >    $ tcpip show interface/full    for the hp ip stack.   >    --   tim.llewellyn@blueyonder.co.uk    H * PLEASE NOTE tim.llewellyn@cableinet.co.uk address is NO LONGER VALID *   ------------------------------  # Date: Sat, 02 Nov 2002 02:13:51 GMT ! From: Carlos Costa <calo@shaw.ca> = Subject: CSWING bug squashed. (was: OpenVMS 7.3-1 and CSWING) ( Message-ID: <B9E874DE.2015%calo@shaw.ca>  I Hi, everyone. First I'd like to thank everyone who responded and the many I excellent tidbits of information and suggestions. I have learned a lot in G the last month. I have gotten CSWING working now under VMS 7.3-1. Craig J Berry hit the nail right on the head with his observation that CSWING doesJ not maintain the search context because it zeros the file_length parameterJ and sets the filename to null before each QIOW call. Moving the setting ofL these variables outside the "while(1)" loop maintains the search context and all is well:  + ORIGINAL CODE (from get_dirs() in SWING.C):   
     while(1)       { %         static char dir_template[20]; (         strcpy(dir_template, "*.dir;*");         filename[0] = 0;         file_length = 0;         uchar = 0;           status = sys$qiow(0,"                           channel,%                           IO$_ACCESS, #                           &iosb[0],                            0,                           0,$                           &fib_desc,-                           DSC1(dir_template), '                           &file_length, *                           DSCA1(filename),                           acb,                           0);         
 WORKING CODE:        filename[0] = 0;     file_length = 0;     
     while(1)       { %         static char dir_template[20]; (         strcpy(dir_template, "*.dir;*");         uchar = 0;           status = sys$qiow(0,"                           channel,%                           IO$_ACCESS, #                           &iosb[0],                            0,                           0,$                           &fib_desc,-                           DSC1(dir_template), '                           &file_length, *                           DSCA1(filename),                           acb,                           0);     A He, and others, are also correct that CSWING could use some major I overhauling. It seems that noone is maintaining the code anymore which is K too bad, but if those people that have gotten a version compiled could make K it available (I know at least one person, Grant, has done that) it would be  a start.   As well, Dan Allen's comments:   > K > See Craig Berry's post on the misuse of the context value and name string K > in the FIB in the CSWING source code. Most likely the problem.  Note also I > that this is what I would call a latent application bug, i.e. it is not I > providing the proper input to the QIO call.  The fact that it worked by H > chance in 7.3 but fails in 7.3-1 is not what I would call a bug in theF > OS and I doubt VMS engineering would be motivated to restore the old > behavior.   % proved to be right on the money also.   I I looked through he release notes again, though, and was not able to find K any indication that this area of the VMS code had been worked on and fixed. H I can appreciate that they can't put *every* change that was made in theK release notes, but maybe HP/Compaq could rethink their assertion that 7.3-1 E is "binary compatible" and "needs no testing", as several people have  already mentioned.   Again, thanks to everyone.   Carlos       On 10/26/02 7:56 AM, in article A 7f15589f.0210260656.2e947aa0@posting.google.com, "Craig A. Berry" , <craig.berry@SignalTreeSolutions.com> wrote:   > In article    B > <rdeininger-2510022234130001@1cust118.tnt2.nashua.nh.da.uu.net>,5 > rdeininger@mindspring.com (Robert Deininger) wrote:  >  > % >>> Searching, root = SHODAN:[000000] & >>> %SYSTEM-W-NOSUCHFILE, no such file >>    K >> Ah!  This is useful.  "No such file" means we asked for a file spec that  >> isn't there.    >>    J >> Might be worth looking at whether the program deals with this condition >> properly.   > 3 > From get_dirs in swing.c after an IO$_ACCESS qio:  >  >       /*  * >       **  file in dir but not indexf.sys >       */  & >       if (iosb[0] == SS$_NOSUCHFILE) >       {    >           fib.fib$l_wcc++; >           continue;  >       }    > D > Isn't the "wild card context" of the file information block one ofF > those opaque longword values that one should not depend on coming inG > sequence, much less increment oneself and hope that doing so will not G > get you in trouble?  I certainly don't see anything in the I/O User's H > Guide that suggests I could predict or should attempt to influence theH > value of the wcc.  I have a strong suspicion that the code I've quotedC > is the source of the looping problem.  Section 1.3.1.2 of the I/O  > User's Guide says: > F > "On the first lookup, FIB$L_WCC should contain zero entries. On eachE > lookup, the ACP returns a nonzero value in FIB$L_WCC, which must be E > passed back on the next lookup call. In addition, you must pass the D > resultant name string returned by the previous lookup using the P4D > result string buffer, and its length in the P3 result length word.F > This string is used together with FIB$L_WCC to continue the wildcard3 > search at the correct position in the directory."  > F > The code in question, in addition to depending on what appears to beG > undocumented (and apparently obsolete) behavior of the context value, E > also does not pass the previous result name on subsequent calls (it H > zeros the first byte of the name string before each QIO and also zeros > the length). > E > John Malmberg's approach of following the compiler warnings will no B > doubt bring great improvement to the package, but there are alsoF > things that should be rethought and possibly reimplemented.  I wouldE > expect that lib$find_file (or even readdir() for that matter) would G > offer a more maintainable method for getting a directory listing than  > the QIO method.    ------------------------------  % Date: Fri, 01 Nov 2002 13:49:37 -0500 & From: David M Smith <dsmit115@csc.com>0 Subject: Re: Does anyone use P7 in DCPS$STARTUP?8 Message-ID: <sli5su0mjqflgllcs2amlaehf2sim26ni6@4ax.com>  M On Fri, 01 Nov 2002 16:11:15 GMT, Paul Anderson <paul.anderson@hp.com> wrote:   G >In article <8b65su8tmv8smch8qik0ic0pue5dndnd48@4ax.com>, David M Smith  ><dsmit115@csc.com> wrote: > H >> I would love to see HP move to a "common print queue setup" structureF >> similar to ours, since it is common on our systems to have multipleF >> types of print queues, and having to go to multiple files to create7 >> and manage them is complex for many system managers.  > H >You mentioned DCPS and DQS queues.  Are there other types of queues you; >define in your data file, like LPR or Telnet print queues?  >  >Paul   N Yes. In our current implementation we support generic, LPS, DQS, LAT, MultiNetO TCP/IP (LPD and stream) and direct-connect (e.g., TTA2:) print queues. Our data P table includes the information required for each type of queue, all collected inK one place, and then our startup code "deduces" (using some simple rules and L mandatory conventions) the type of queue and connection and performs all theP steps required to properly instantiate that queue at startup time. This includesP such things as creating the LTA device and setting its characteristics, creatingL the queue (if not present), etc. Both autostart and non-autostart queues are handled.  O Of course, changes and enhancements keep occurring in both the operating system P and the print-related layered products and we have not been able to keep up withM them with our limited resource (me and a few folks). Some printer "connection O types" are not currently handled, for example TCP/IP SERVICES-based queues, due I to both resource and technical issues like "how do we create/maintain the P printcap file?". Both DCPS and TCP/IP SERVICES are examples of classes of queuesM which have "auxiliary information" (i.e., not a part of "normal" VMS queueing L mechanisms) that are required for the print symbiont to work, and which haveP widely different implementations (i.e. printcap file vs. "special" logical names and device strings).  K A single, unified, consistent approach to creating and maintaining all this 7 print-queue-related information from HP would be great. I ------------------------------------------------------------------------- I David M. Smith 302.391.8533                       dsmit115 at csc dot com I Computer Sciences Corporation     (Opinions are those of the writer only) I -------------------------------------------------------------------------    ------------------------------  # Date: Fri, 01 Nov 2002 19:44:45 GMT * From: Paul Anderson <paul.anderson@hp.com>0 Subject: Re: Does anyone use P7 in DCPS$STARTUP?5 Message-ID: <011120021444258689%paul.anderson@hp.com>   F In article <sli5su0mjqflgllcs2amlaehf2sim26ni6@4ax.com>, David M Smith <dsmit115@csc.com> wrote:   D > A single, unified, consistent approach to creating and maintainingB > all this print-queue-related information from HP would be great.  C We've often thought of a data-driven (vs. command procedure-driven) B method to define DCPS queues.  It's a good idea but would be a big project.   Paul   --    Paul Anderson   OpenVMS Engineering    Hewlett-Packard Company    ------------------------------  % Date: Fri, 01 Nov 2002 15:07:36 -0500 - From: JF Mezei <jfmezei.spamnot@videotron.ca> 0 Subject: Re: Does anyone use P7 in DCPS$STARTUP?, Message-ID: <3DC2DF07.293B86E7@videotron.ca>   Paul Anderson wrote:I > > I would love to see HP move to a "common print queue setup" structure   I > You mentioned DCPS and DQS queues.  Are there other types of queues you < > define in your data file, like LPR or Telnet print queues?    M But one may have a generalized "queue" procedure that include not only print, J but also batch, and "server" queues (such as SMTP, and even ALL-IN-1 had aN couple of server queues to execute scripts in background as well as backgroundL formatting of large documents). Message router had its own batch queues. AndJ its startup procedure would automatically start those queues (they are notJ autostart), but i had to fight it because I didn't want the MR maintenanceR stuff to begin until after the system was fully booted (so that it boots quicker).    K I have to admit that when I started with CPS/DCPS, I was happy to have that H "template" startup procedure. It was all very foreign to me and I didn'tN really know how the software was implemented. But once you've installed it andL seen how it is implemented, is there anything more than setting the terminalM characteristics and defining a few logicals and letting the autostart feature  do its magic ?  L Instead of a command procedure with a gazzilion arguments, I would suggest aM template command procedure that one can use to incorporate the right commands J into whatever local procedure one has to setup queues. This template wouldN include all the DCPS logical names with a short explanation of their function,# as well as queue specific commands.    ------------------------------  % Date: Fri, 01 Nov 2002 15:32:49 -0500 - From: JF Mezei <jfmezei.spamnot@videotron.ca> 0 Subject: Re: Does anyone use P7 in DCPS$STARTUP?, Message-ID: <3DC2E4EE.D525915F@videotron.ca>   Paul Anderson wrote:E > We've often thought of a data-driven (vs. command procedure-driven) D > method to define DCPS queues.  It's a good idea but would be a big
 > project.  4 I think that you are overthinking/engineering this.   M You might want to consider some configuration template procedure that creates % the queues with the right attributes.   J And then a much simpler template that is used during startup to define the? right logical names and let the queue manager do the autostart.    ------------------------------  # Date: Sat, 02 Nov 2002 02:54:08 GMT - From: "John E. Malmberg" <wb8tyw@qsl.network> 0 Subject: Re: Does anyone use P7 in DCPS$STARTUP?; Message-ID: <k7Hw9.4547$6g.1160888@news1.news.adelphia.net>    Paul Anderson wrote:H > In article <sli5su0mjqflgllcs2amlaehf2sim26ni6@4ax.com>, David M Smith > <dsmit115@csc.company> wrote:  > D >>A single, unified, consistent approach to creating and maintainingB >>all this print-queue-related information from HP would be great. > E > We've often thought of a data-driven (vs. command procedure-driven) D > method to define DCPS queues.  It's a good idea but would be a big
 > project.  F I found that a data file driven approach was too confusing to those I I tried to distribute system management tasks to.  Also maintenance on the  H command files was a pain, as a mis-edit in a data file would cause many 	 problems.   D What I finally settled on was having a common command procedure for G logical, serial, lat, and telnetsym queues, and another common command  0 procedure that called the DCPS / CPS procedures.  C These procedures set up the print queues, and also spooled devices  ? either with a serial port, or a dummy LAT port.  It also put a   description on the queue.   D These common procedures did everything needed to make sure that the C queue was properly set up.  They would sanity check and repair the   settings of an existing queue.    I These procedures were called by an outer command procedure, one for each     major queue.  G These outer procedures would just define a set of local symbols before   calling the common procedures.  I This turned out to be easy to manage, and if someone mangled the edit to  & a file, only that queue was messed up.   -John  wb8ytw@qsl.network Personal Opinion Only    ------------------------------  % Date: Fri, 01 Nov 2002 22:20:44 -0500 - From: JF Mezei <jfmezei.spamnot@videotron.ca> 0 Subject: Re: Does anyone use P7 in DCPS$STARTUP?, Message-ID: <3DC34485.1FA8A416@videotron.ca>  K What would be needed is an improvement in the VMS queue management to allowN/ DCPS information to be stored inside the queue.A  L Or perhaps add just one more field to the queue definition that could hold a5 filename which DCPS would use as a configuation file.t  K I.E. for a queue that has the DCPS symbiont, the symbiont would look ta theNO queue's "config" file to extract all the optiosn the user wants for that queue.S   ------------------------------  % Date: Fri, 01 Nov 2002 22:22:22 -0600i1 From: "David J. Dachtera" <djesys.nospam@fsi.net>A0 Subject: Re: Does anyone use P7 in DCPS$STARTUP?' Message-ID: <3DC352FE.3F29AA18@fsi.net>    "John E. Malmberg" wrote:S >  > Paul Anderson wrote:J > > In article <sli5su0mjqflgllcs2amlaehf2sim26ni6@4ax.com>, David M Smith! > > <dsmit115@csc.company> wrote:q > >eF > >>A single, unified, consistent approach to creating and maintainingD > >>all this print-queue-related information from HP would be great. > >,G > > We've often thought of a data-driven (vs. command procedure-driven)xF > > method to define DCPS queues.  It's a good idea but would be a big > > project. > G > I found that a data file driven approach was too confusing to those InJ > tried to distribute system management tasks to.  Also maintenance on theI > command files was a pain, as a mis-edit in a data file would cause manyo > problems.e > [snip]  H My experience has been quite the reverse. Clearing up the "confusion" isB usually a simple matter of training, and they usually end up beingE better programmers for it. "Cut-and-paste" code (where the parameters H stay the same and/or vary only slightly) and only the queue names changeE is a royal *BITCH* to maintain, especially under pressure, especially - when the queue counts get up over 400 or so.    F I have also found most such attempts at parameter-driven procedures toB be rather poorly designed and/or poorly executed. Even my own codeD rarely meets my standards in this arena. Some depend on fixed formatG data files, while others use F$LOCATE() to parse for delimiters insteadnC of F$ELEMENT(). Most fail to provide a logical name pointing to thenD target device of the queue, and most attempts at parameters to driveA such things as queue processor selection tend to demonstrate onlyeB single- or two-dimensional thinking. Most also fail to provide for< starting a single queue, as when one is added during uptime.  8 DCPS adds a whole new set of complications into the mix.   --   David J. Dachtera  dba DJE Systems  http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/s   ------------------------------  % Date: Fri, 01 Nov 2002 22:23:58 -0600i1 From: "David J. Dachtera" <djesys.nospam@fsi.net>m' Subject: Re: equivalent of touch in VMS-' Message-ID: <3DC3535E.6C05A69D@fsi.net>-   Peter LANGSTOEGER wrote: >  > In article <8EA11405E59BD611BA7100104B93C2607580BA@exdel01.del.mgsl.com>, Kesav Tadimeti <Kesav_Tadimeti@KeaneIndia.com> writes:N > >What is the equivalent command in VMS to create an empty file? In UNIX this > >would be touch. > . > How about       $ touch == "APPEND/NEW _NL:"' >                 $ touch file-name.exti   Thank you, Peter!I  @ I nominate that for "definitive answer" to the original poster's. question as well as to the purpose of "touch".   --   David J. Dachterao dba DJE Systems  http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/a   ------------------------------  % Date: Fri, 01 Nov 2002 13:57:59 -0500 2 From: Atlant Schmidt <atlantnospam@mindspring.com> Subject: Re: errno equivalenta. Message-ID: <3DC2CEB7.9592548D@mindspring.com>   Shane Smith wrote:  D > I'm with John Briggs though, I prefer my return statii as functionJ > return values, or write arguments. You /know/ where the value came from.  + Generally speaking, of course, the success/e, failure of a C'ish/Unix'ish system call *IS*. returned by the function in some way (although0 interpretation of the value unfortunately varies4 all over the map: is zero, is non-zero, is negative, etc.).  ) You only refer to errno after you know anh) error has occurred and you're looking for ) some sort of additional detail as to what2 really went wrong.  1 But I agree that the it_succeeded/it_failed+errno6. concept sure is different from the VMS mindset of a detailed status return.   Atlant   ------------------------------   Date: 1 Nov 2002 11:27:32 -0800V( From: bob@instantwhip.com (Bob Ceculski)( Subject: Re: FTP with SSL for VMS/Alpha?= Message-ID: <d7791aa1.0211011127.41ee3ef0@posting.google.com>t  d winston@SSRL.SLAC.STANFORD.EDU wrote in message news:<00A164D8.42E33057@SSRL04.SLAC.STANFORD.EDU>...f > In article <d0141774.0211010132.304407e2@posting.google.com>, issinoho@slayme.com (issinoho) writes: > ) > >Why not TCPware? I've always liked it.g > P > Sorry, I was unclear.  I meant "I don't know about TCPware, _I_ don't use it,"0 > rather than intending to recommend against it. > 	 > -- Alan  >   8 TCPware does everything multinet does except its command3 structure is more vms like and it is the fastest ofs6 all vms IP stacks because it is still based on the VMS/ kernel ... best IP stack for VMS hands down ...    ------------------------------  % Date: Fri, 01 Nov 2002 18:16:44 +0100 2 From: martin@radiogaga.harz.de (Martin Vorlaender)( Subject: Re: FTP with SSL for VMS/Alpha?; Message-ID: <3dc2b6fc.524144494f47414741@radiogaga.harz.de>n  % winston@SSRL.SLAC.STANFORD.EDU wrote:i( > issinoho@slayme.com (issinoho) writes:+ > > winston@SSRL.SLAC.STANFORD.EDU wrote... J > > > If you'd rather spend money than effort, Multinet (and maybe TCPware$ > > > - don't use it), supports SCP. >a* > > Why not TCPware? I've always liked it. > K > Sorry, I was unclear.  I meant "I don't know about TCPware, _I_ don't used5 > it," rather than intending to recommend against it.   L As of version 5.6, TCPware is on par with MultiNet 4.4's SSH implementation.   cu,y   Martin -- dF   OpenVMS:                | Martin Vorlaender  |  VMS & WNT programmer3    The operating system   | work: mv@pdv-systeme.detF    God runs the           |   http://www.pdv-systeme.de/users/martinv/:    earth simulation on.   | home: martin@radiogaga.harz.de   ------------------------------   Date: 1 Nov 2002 16:26:38 -0500r& From: fdc@columbia.edu (Frank da Cruz)( Subject: Re: FTP with SSL for VMS/Alpha?1 Message-ID: <apurie$e7t$1@watsol.cc.columbia.edu>h  5 In article <slrnas2p96.38v.WADE@DAX.ASUB.ARKNET.EDU>,/. Wade Fincher <WADE@DAX.ASUB.ARKNET.EDU> wrote:J : Does anyone know of such a beast?  Ckermit 8 has it for unix and windows : but not vms. : I Because no VMS programmers have ported the code.  The SSL/TLS part shouldtI be easy, since (reportedly) OpenSSL has been ported to VMS; the semi-hard @ part is adapting the C-Kermit FTP to VMS, which is beyond my VMSE capabilities.  To do this right requires RMS, ASTs, and all the rest.eE But it would be worth doing, since the Kermit FTP client, unlike most  others, is fully scriptable:  0   http://www.columbia.edu/kermit/ftpscripts.html  G Actually, this could be done in steps.  First, somebody who has OpenSSL I installed, as well as DEC C (or maybe GCC), can try building C-Kermit 8.0 F as it is, without including FTP, but adding SSL/TLS support.  This wayE you get a form of secure (and optionally scriptable) Telnet.  If thisCE goes smoothly, FTP could be adapted to use it, perhaps incrementally;wD a first pass would use only C library calls; subsequent passes couldG convert to native VMS APIs for file i/o and terminal interruption. Plus0C I imagine some special measures will be needed to handle non-streamo& files in the FTP protocol data stream.  G There's an HTTP 1.1 client in there too that hasn't been adapted to VMS  yet.  H Anybody who's interested in tackling any of this, let me know.  C-Kermit 8.0 is here:  -   http://www.columbia.edu/kermit/ckermit.htmlk   - Frankt   ------------------------------  % Date: Sat, 02 Nov 2002 00:12:54 -0500t- From: JF Mezei <jfmezei.spamnot@videotron.ca>y Subject: HELP suggestion, Message-ID: <3DC35EC7.2DCDF4DA@videotron.ca>  K I noticed that on an Alpha 7.3-1 system, the help contains much more stuff.   ? However, I feel that its structure leaves room for improvement.i  ? They've added lots of xxxx_routines in the main list of topics vL (DCX_routines, CLI_routines, CQUAL_routines, EDT_routines, MAIL_routines etc etc etc)  H Shouldn't all of those have been added under the "RTL" main help topic ?    1 For instance, because there is an entry for both: G MAIL_command and MAIL_routines, typing "HELP MAIL" doesn't get you to a ( "Subtopic?" prompt for the Mail command.  U Not sure if this is particular to that one site, but if not, it is really bad design..   ------------------------------   Date: 1 Nov 2002 11:13:41 -0800s  From: aronoffs@hotmail.com (stu): Subject: How to do inverse video, bold, etc. in Vax Pro-C?= Message-ID: <80d4088b.0211011113.56ead662@posting.google.com>e  M Is there a way in Pro-C to display text in bold, inverse, etc. like VMS does?    Thanks.a   ------------------------------   Date: 1 Nov 2002 15:16:19 -0600x- From: koehler@encompasserve.org (Bob Koehler) > Subject: Re: How to do inverse video, bold, etc. in Vax Pro-C?3 Message-ID: <nZirAIo2gVi+@eisner.encompasserve.org>h  ` In article <80d4088b.0211011113.56ead662@posting.google.com>, aronoffs@hotmail.com (stu) writes:O > Is there a way in Pro-C to display text in bold, inverse, etc. like VMS does?0 >   C    Changes to text on ANSI CRTs are done via ANSI escape sequences.sD    DEC terminals may have a few extensions beyond ANSI< but the idea    is the same.o  G    The only problem sending some of the escape sequences from C is thateB    some of them may have embedded null characters in certain uses.   ------------------------------   Date: 1 Nov 2002 15:51:40 -0800f7 From: jones.computer.srv@worldnet.att.net (Daryl Jones)a> Subject: Re: How to do inverse video, bold, etc. in Vax Pro-C?= Message-ID: <8a646952.0211011551.2821ede9@posting.google.com>e  e aronoffs@hotmail.com (stu) wrote in message news:<80d4088b.0211011113.56ead662@posting.google.com>...dO > Is there a way in Pro-C to display text in bold, inverse, etc. like VMS does?f > 	 > Thanks.e    D Since the 1980s, ANSI standard escape sequence code could be used toF perform bold, inverse, and etc. from any language to any ANSI standard	 terminal.n   Daryln   ------------------------------   Date: 1 Nov 2002 18:19:48 -0800 0 From: chris_doran@postmaster.co.uk (Chris Doran)> Subject: Re: How to do inverse video, bold, etc. in Vax Pro-C?= Message-ID: <948f0720.0211011819.570b062e@posting.google.com>r  e aronoffs@hotmail.com (stu) wrote in message news:<80d4088b.0211011113.56ead662@posting.google.com>...-O > Is there a way in Pro-C to display text in bold, inverse, etc. like VMS does?   * Outside the SQL bits, Pro*C is just C so:-  3 1) Use Screen Management Group functions (SMG$), orm 2) Use curses, oreF 3) Embed escape sequences in your printf/puts calls (see your terminal manual).  C (1) is the preferred way for VMS, is best for terminal-independencedB and complicated screen displaying, but (arguably) takes longest toC learn, (2) is more-or-less portable across operating systems, being-A the Un*x way, but some of the curses functions vary a bit betweensF implementations, and (3) is easiest to learn and is probably your bestB choice for simple programs which only need to work on one terminal type, e.g. ANSI.   Chriso   ------------------------------  $ Date: Fri, 1 Nov 2002 17:44:52 -0800* From: "Jack Peacock" <peacock@simconv.com>> Subject: Re: How to do inverse video, bold, etc. in Vax Pro-C?2 Message-ID: <OqCdnZ7xA7qIs16gXTWc2Q@mpowercom.net>  - "stu" <aronoffs@hotmail.com> wrote in message[7 news:80d4088b.0211011113.56ead662@posting.google.com...NI > Is there a way in Pro-C to display text in bold, inverse, etc. like VMSc does?  >nL You might want to look at the VMS SMG (Screen Management Guidelines??) callsJ if you have the time and a large enough project to justify the work.  It'sL great for things like popup windows on a VT screen.  Been very stable for meH over the last 12+ years too.  I use SMG for all screen controls in C and) Dibol, on both VAX and Alpha VMS systems.0   Jack Peacock   ------------------------------  % Date: Sat, 02 Nov 2002 04:36:22 -00007! From: Z  <zarlenga@conan.ids.net>c> Subject: Re: How to do inverse video, bold, etc. in Vax Pro-C?/ Message-ID: <us6li6pr2ddp7d@corp.supernews.com>n  ! stu <aronoffs@hotmail.com> wrote:.O : Is there a way in Pro-C to display text in bold, inverse, etc. like VMS does?h   As I recall ...   @ ESC = escape ( ASCII 1B(hex), 33(octal), 27(decimal) or ctrl+[ )  + Sending ESC[0m will turn off all attributes   Sending ESC[1m will turn on bold% Sending ESC[4m will turn on underlinea! Sending ESC[5m will turn on blinkt# Sending ESC[7m will turn on reverses  = And, IIRC, you can also use ESC[X;Ym ... such as ESC[0;4m to e: turn off all previous attributes and to turn on underline.  C Once you turn on attributes, all chars printed to the terminal haveu, those attributes on until you turn them off.  : eg: In C, you can use "\033[0m" in a string to get ESC[0m.   ------------------------------  $ Date: Fri, 1 Nov 2002 13:38:28 -0500* From: "Syltrem" <syltremzulu@videotron.ca> Subject: Re: HP Advocacy Siteo5 Message-ID: <mQzw9.17188$H67.76080@tor-nn1.netcom.ca>    It's already there!b  F TITLE: Market OpenVMS just like Tru64 - not only for so-called "niche" marketsu  " There's only one me too right now.   --   SyltremtI http://pages.infinit.net/syltrem (OpenVMS related web site - en franais)A8 To reply to myself directly, remove zulu from my address  A "Shane Smith" <ssmith@icius.com> a crit dans le message de news:s% 01C2800B.EF740970@sulfer.icius.com...yI > So, how about starting a VMS advertising issue, and dropping as many mep > too's as we can on it? > I > Only one problem. I enter my email address to log in so I can put in antF > issue, and it takes me straight back to the login page. This is withH > Mozilla and IE (hawk, spit) with and without cookies, with and without, > Javascript. How did you submit your issue? >  > Shanet >t > -----Original Message-----F > From: ed.stuart@austinenergy.com [mailto:ed.stuart@austinenergy.com], > Sent: Wednesday, October 30, 2002 11:49 AM > To: Info-VAX@Mvb.Saic.Comh > Subject: HP Advocacy Siteo >o > H > While c.o.v is a good place to exchange ideas, thoughts, questions andH > solutions it may not be the best place to get a response from HP.  TheD > HP Advocacy site at http://www.hpuseradvocacy.org/ allows folks toB > submit issues for HP to respond to. While most c.o.v readers areH > probably already aware of this site I just learned something new about@ > it today. I submitted an issue and the auto-generated responseH > informed me that "An issue that has collected at least 5 Me Too! votesC > is guaranteed to be brought to the attention of HP management andeB > responded to on the HP User Advocacy Program's web site." So, if? > there's an issue that the readers of c.o.v would like to have H > addressed then one reader could post it to the advocacy site, and thenF > ask other c.o.v readers who agree to go to the site and submit a "MeD > Too" vote. This has more of a potential to get a response from HP, > than posting to c.o.v. >  > EdG > **Please apply a generous amount of all the usual disclaimers here.**t   ------------------------------  $ Date: Fri, 1 Nov 2002 12:04:27 -0800$ From: Shane Smith <ssmith@icius.com> Subject: RE: HP Advocacy Siter0 Message-ID: <01C2819E.E6D9E250@sulfer.icius.com>  G Aha! A fresh install of Mozilla with all the security switched off gets  me in at last!  F Where is it? I only get offered 6 issues when I hit the "me too" link,F and it isn't one of them. Please give me the list of links to hit from the login page.    Shaned   -----Original Message-----/ From: Syltrem [mailto:syltremzulu@videotron.ca]t( Sent: Friday, November 01, 2002 10:38 AM To: Info-VAX@Mvb.Saic.Com  Subject: Re: HP Advocacy Site      It's already there!e  F TITLE: Market OpenVMS just like Tru64 - not only for so-called "niche" markets,  " There's only one me too right now.   --   Syltremw? http://pages.infinit.net/syltrem (OpenVMS related web site - en  fran=E7ais)08 To reply to myself directly, remove zulu from my address  C "Shane Smith" <ssmith@icius.com> a =E9crit dans le message de news:o% 01C2800B.EF740970@sulfer.icius.com...+H > So, how about starting a VMS advertising issue, and dropping as many = me > too's as we can on it? >oH > Only one problem. I enter my email address to log in so I can put in = anF > issue, and it takes me straight back to the login page. This is withH > Mozilla and IE (hawk, spit) with and without cookies, with and without, > Javascript. How did you submit your issue? >n > Shanet >t > -----Original Message-----F > From: ed.stuart@austinenergy.com [mailto:ed.stuart@austinenergy.com], > Sent: Wednesday, October 30, 2002 11:49 AM > To: Info-VAX@Mvb.Saic.Com  > Subject: HP Advocacy Sitet >a >3H > While c.o.v is a good place to exchange ideas, thoughts, questions andH > solutions it may not be the best place to get a response from HP.  TheD > HP Advocacy site at http://www.hpuseradvocacy.org/ allows folks toB > submit issues for HP to respond to. While most c.o.v readers areH > probably already aware of this site I just learned something new about@ > it today. I submitted an issue and the auto-generated responseH > informed me that "An issue that has collected at least 5 Me Too! votesC > is guaranteed to be brought to the attention of HP management andxB > responded to on the HP User Advocacy Program's web site." So, if? > there's an issue that the readers of c.o.v would like to have H > addressed then one reader could post it to the advocacy site, and thenF > ask other c.o.v readers who agree to go to the site and submit a "MeD > Too" vote. This has more of a potential to get a response from HP, > than posting to c.o.v. >n > EdG > **Please apply a generous amount of all the usual disclaimers here.**n   ------------------------------  % Date: Fri, 01 Nov 2002 22:31:54 -0600 1 From: "David J. Dachtera" <djesys.nospam@fsi.net>  Subject: Re: HP Advocacy Site ' Message-ID: <3DC3553A.41D33DE8@fsi.net>f   Syltrem wrote: >  > It's already there!o > H > TITLE: Market OpenVMS just like Tru64 - not only for so-called "niche"	 > markets. > $ > There's only one me too right now.  ( I don't see that one. The ones I see at O http://www.hpuseradvocacy.org/Advocacy/Issues/MeToo/Index.cfm?IssueCategoryID=0.   are:   Universal SYSLOG Capabilitiesc CISC COMPILER SUPPORT ! Documentation on Legacy Equipments: roadmap to migrate CA's Advantage:Gen to HP OpenVMS on IPF NAPSE North American Support Embeded Wireless LAN in Ipaq  8 What page are you looking at, and how did you get there?   --   David J. DachteraD dba DJE Systemsc http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/    ------------------------------  % Date: Fri, 01 Nov 2002 20:43:11 +0100m9 From: Jan-Erik =?iso-8859-1?Q?S=F6derholm?= <aaa@aaa.com>i( Subject: Re: HP User Advocacy ADE survey' Message-ID: <3DC2D94F.A419DAB7@aaa.com>   9 Hi. OK so here is my conversation regarding this survey :p" (Read from the bottom as usual...)2 (Note that Jessica says "platform" in singular...)    
 NonStop ?????d  7 And what exactly has Windows to do with "NonStop" ?????   ; In what way, exactly, is the listed OS'es *more* "non stop"  then OpenVMS ???   Please enlighten me !!   Best Regards Jan-Erik Sderholm.m       -----Original Message-----3 From: Moser, Jessica [mailto:Jessica_Moser@sba.com]  Sent: den 1 november 2002 16:27, To: 'Jan-erik Sderholm (QAC)' Subject: RE: VMS ?     Jan-Erik - C  C    This survey was sponsored and written by the NonStop Development G department within HP so it's scope was limited to the NonStop platform.  WeB apologize if this wasn't communicated properly. Thank you for your interest> in the advocacy program. Please continue to check our web siteB (www.hpuseradvocacy.org) for future advocacy related initiatives.    -----Original Message----- From: Jan-erik Sderholm (QAC)+ [mailto:Jan-erik.Soderholm@pac.ericsson.se]e) Sent: Thursday, October 31, 2002 11:20 AM" To: 'admin@hpuseradvocacy.org' Subject: VMS ?     Hi.o3 I saw the "Customer Advocacy Program Questionnaire" A on "http://www.hpuseradvocacy.com/Surveys/2002October/index.cfm".   ! But where is OpenVMS ????????????   7 You must have done a mistake when setting this site up. , When can I expect this mistake to be fixed ?   Best Regards Jan-Erik Sderholm Sweden.    ------------------------------   Date: 1 Nov 2002 12:15:08 -0800,7 From: jones.computer.srv@worldnet.att.net (Daryl Jones) @ Subject: Re: Intel Conceding 64-Bit Desktops to IBM, AMD - eWeek= Message-ID: <8a646952.0211011215.7e6a2bd5@posting.google.com>a  t "John Smith" <a@nonymous.com> wrote in message news:<UYgv9.86381$mxk1.78654@news04.bloor.is.net.cable.rogers.com>...8 > "Barry Treahy, Jr." <Treahy@MMaz.com> wrote in message# > news:3DBD8644.5070705@MMaz.com...r > >oL > > If Intel were to release an 64-bit X86, they would undermine their IA-64K > > efforts!  This could create serious problems for us VMS folks expecting K > > to run on IA-64 because what was the Compaq rational for killing Alpha?i; > >  Not enough volume to support the design and fab costs.i > > F > > So, if Intel does introduce a 64-bit X64 for the desktop, you knowL > > manufacturers will introduce servers based on that chip set, essentially- > > killing high volume demand for the IA-64.t > >hJ > > Am I the only one that sees the irony of this?  A major porting effortH > > of VMS to another 'low volume and essentially proprietary' chip set?J > >  Worse yet, a chip set that HP/Compaq doesn't directly control and may8 > > not even be as fast as the Alpha that was abandoned? >  > ? > No Barry, you aren't the only one who sees the irony of this.v > M > However all that do see the irony don't occupy executive level positions at: > HP.i  
 Gentlemen:  E Before the HP-COMPAQ merger, both HP and Compaq annouced that I64 wasbC to be their chip of the future. Compaq gave the Alpha technology toeE Intel. The Merced/Itanium chip performance would be camparable to thenB EV7. What made the Alpha performance was not only the chip but theA compilier technology. Intel in this area was ten years behing the F Alpha. By 2004, HP will have only one architecture to work with ratherC than two and it is out of the hardware business to boot. I wouldn'trD worry about VMS. There are about 25 programmers in NH working on theD conversion from Alpha to Intel. The jobs were posted in DICE several months ago.i   ------------------------------  $ Date: Fri, 1 Nov 2002 12:18:15 -0800$ From: Shane Smith <ssmith@icius.com>@ Subject: RE: Intel Conceding 64-Bit Desktops to IBM, AMD - eWeek0 Message-ID: <01C281A0.CD15E5B0@sulfer.icius.com>  @ How many people are working on the conversion is nowhere near asH important as whether the final product will be worth using. If Itanium'sH performance doesn't improve drastically the move will make VMS unviable.   Shane    -----Original Message-----) From: jones.computer.srv@worldnet.att.net , [mailto:jones.computer.srv@worldnet.att.net]( Sent: Friday, November 01, 2002 12:15 PM To: Info-VAX@Mvb.Saic.Comc@ Subject: Re: Intel Conceding 64-Bit Desktops to IBM, AMD - eWeek    . "John Smith" <a@nonymous.com> wrote in messageE news:<UYgv9.86381$mxk1.78654@news04.bloor.is.net.cable.rogers.com>...t8 > "Barry Treahy, Jr." <Treahy@MMaz.com> wrote in message# > news:3DBD8644.5070705@MMaz.com...s > > L > > If Intel were to release an 64-bit X86, they would undermine their IA-64K > > efforts!  This could create serious problems for us VMS folks expecting$K > > to run on IA-64 because what was the Compaq rational for killing Alpha?t; > >  Not enough volume to support the design and fab costs.o > >tF > > So, if Intel does introduce a 64-bit X64 for the desktop, you knowL > > manufacturers will introduce servers based on that chip set, essentially- > > killing high volume demand for the IA-64.s > > J > > Am I the only one that sees the irony of this?  A major porting effortH > > of VMS to another 'low volume and essentially proprietary' chip set?J > >  Worse yet, a chip set that HP/Compaq doesn't directly control and may8 > > not even be as fast as the Alpha that was abandoned? >  > ? > No Barry, you aren't the only one who sees the irony of this.i > M > However all that do see the irony don't occupy executive level positions at- > HP.-  
 Gentlemen:  E Before the HP-COMPAQ merger, both HP and Compaq annouced that I64 wassC to be their chip of the future. Compaq gave the Alpha technology to E Intel. The Merced/Itanium chip performance would be camparable to the1B EV7. What made the Alpha performance was not only the chip but theA compilier technology. Intel in this area was ten years behing thesF Alpha. By 2004, HP will have only one architecture to work with ratherC than two and it is out of the hardware business to boot. I wouldn'toD worry about VMS. There are about 25 programmers in NH working on theD conversion from Alpha to Intel. The jobs were posted in DICE several months ago.a   ------------------------------   Date: 1 Nov 2002 13:49:40 -0600s+ From: young_r@encompasserve.org (Rob Young)k@ Subject: RE: Intel Conceding 64-Bit Desktops to IBM, AMD - eWeek3 Message-ID: <FBZW$5W1DUhq@eisner.encompasserve.org>t  W In article <01C281A0.CD15E5B0@sulfer.icius.com>, Shane Smith <ssmith@icius.com> writes:lB > How many people are working on the conversion is nowhere near asJ > important as whether the final product will be worth using. If Itanium'sJ > performance doesn't improve drastically the move will make VMS unviable. >   ? 	Recent articles talk about Madison (Itanium 3) being a 30-50% tF 	performance improvement over Itanium 2.  If we chose 40% as a guess, 0 	that would put Madison at or near these values:   	1300 SpecInt2000  	1800 Specfp2000" 	115000 tpmC for a 4 processor box  > 	Maybe it doesn't show up until mid-2003, but I'm sure we will> 	hear a lot more about it.  It should be at or near the top of@ 	performance when it ships.  I suspect it will trail in SpecInt C 	performance to a 4 GHz Pentium 4 (Prescott) as a 2.8 GHz Pentium 4y 	does 1010 today:l  R http://www.specbench.org/osg/cpu2000/results/res2002q3/cpu2000-20020909-01639.html  G         (40% speed up should mean more than a 30% speed up in integer).a 	e 	By the way, when you say:  M "If Itanium's performance doesn't improve drastically the move will make VMS  
 unviable."  C 	Just what metric do you have in mind?  Of those metrics, is Power4yF 	or UltraSparc 3 any better at any of Itanium2's ?  If so, which ones?A 	I overlooked them if they are out there and directly comparable.t   					Rob   ------------------------------  % Date: Fri, 01 Nov 2002 13:54:50 -0700g+ From: "Barry Treahy, Jr." <Treahy@MMaz.com>-@ Subject: Re: Intel Conceding 64-Bit Desktops to IBM, AMD - eWeek' Message-ID: <3DC2EA1A.5060305@MMaz.com>    Daryl Jones wrote:  u >"John Smith" <a@nonymous.com> wrote in message news:<UYgv9.86381$mxk1.78654@news04.bloor.is.net.cable.rogers.com>...- >  - >-8 >>"Barry Treahy, Jr." <Treahy@MMaz.com> wrote in message# >>news:3DBD8644.5070705@MMaz.com...o >>     >>K >>>If Intel were to release an 64-bit X86, they would undermine their IA-640J >>>efforts!  This could create serious problems for us VMS folks expectingJ >>>to run on IA-64 because what was the Compaq rational for killing Alpha?: >>> Not enough volume to support the design and fab costs. >>> E >>>So, if Intel does introduce a 64-bit X64 for the desktop, you knowvK >>>manufacturers will introduce servers based on that chip set, essentiallye, >>>killing high volume demand for the IA-64. >>>bI >>>Am I the only one that sees the irony of this?  A major porting effortlG >>>of VMS to another 'low volume and essentially proprietary' chip set?oI >>> Worse yet, a chip set that HP/Compaq doesn't directly control and mayj7 >>>not even be as fast as the Alpha that was abandoned? 	 >>>      9 >>>7? >>No Barry, you aren't the only one who sees the irony of this.P >>M >>However all that do see the irony don't occupy executive level positions ate >>HP.r >>     >> >y >Gentlemen:  > F >Before the HP-COMPAQ merger, both HP and Compaq annouced that I64 wasD >to be their chip of the future. Compaq gave the Alpha technology toF >Intel. The Merced/Itanium chip performance would be camparable to theC >EV7. What made the Alpha performance was not only the chip but the B >compilier technology. Intel in this area was ten years behing theG >Alpha. By 2004, HP will have only one architecture to work with rathertD >than two and it is out of the hardware business to boot. I wouldn'tE >worry about VMS. There are about 25 programmers in NH working on thesE >conversion from Alpha to Intel. The jobs were posted in DICE severalh >months ago. >  i >C Daryl,  H you spent time with a nice history lesson but you neglected to focus on H the crux of my point - Alpha was abandoned because it was too expensive E to push to EV7 and beyond due to low chip volume.  There is no doubt yG that VMS is porting to IA-64, that was never the point of the comment. lI  The real question is that if AMD, and everyone one else, kick's Intel's cD butt hard enough with their X86 64-bit chipsets and forces Intel to F produce their own, the IA-64 running VMS will be left flapping in the F breeze by itself once again, an OS dependent on a low volume yet very  expensive chipset...   Barryc     -- w  @ Barry Treahy, Jr  *  Midwest Microwave  *  Vice President & CIO   A E-mail: Treahy@mmaz.com * Phone: 480/314-1320 * FAX: 480/661-70280   ------------------------------  $ Date: Fri, 1 Nov 2002 12:57:34 -0800$ From: Shane Smith <ssmith@icius.com>@ Subject: RE: Intel Conceding 64-Bit Desktops to IBM, AMD - eWeek0 Message-ID: <01C281A6.6053E3E0@sulfer.icius.com>  G So far we've seen two versions of Itanium not so much in the wild as indG wildlife parks. Both apparently lag behind contemporary chips in enough/G areas that they're pretty much laughing stocks. OK, yes, there are someaE things it's good at but that's not enough. It has to have enough of anH performance /advantage/ over its competitors in enough areas that peopleF sit up and take notice. It's got to knock people's socks off. RememberD when they were trying to sell the Alpha? It was significantly faster@ than IA32, but that wasn't enough to overcome the lack of native	 software.   E VMS is already hard enough to sell. If it's perceived as slow too, itnF won't matter if HP decide to start marketing it. It won't be sellable.   Shanex   -----Original Message-----B From: young_r@encompasserve.org [mailto:young_r@encompasserve.org]( Sent: Friday, November 01, 2002 11:50 AM To: Info-VAX@Mvb.Saic.Coma@ Subject: RE: Intel Conceding 64-Bit Desktops to IBM, AMD - eWeek    < In article <01C281A0.CD15E5B0@sulfer.icius.com>, Shane Smith <ssmith@icius.com> writes:B > How many people are working on the conversion is nowhere near asJ > important as whether the final product will be worth using. If Itanium'sJ > performance doesn't improve drastically the move will make VMS unviable. >   ? 	Recent articles talk about Madison (Itanium 3) being a 30-50%  F 	performance improvement over Itanium 2.  If we chose 40% as a guess, 0 	that would put Madison at or near these values:   	1300 SpecInt2000  	1800 Specfp2000" 	115000 tpmC for a 4 processor box  > 	Maybe it doesn't show up until mid-2003, but I'm sure we will> 	hear a lot more about it.  It should be at or near the top of@ 	performance when it ships.  I suspect it will trail in SpecInt C 	performance to a 4 GHz Pentium 4 (Prescott) as a 2.8 GHz Pentium 4e 	does 1010 today:s  H http://www.specbench.org/osg/cpu2000/results/res2002q3/cpu2000-20020909-
 01639.html  G         (40% speed up should mean more than a 30% speed up in integer).a 	t 	By the way, when you say:  H "If Itanium's performance doesn't improve drastically the move will make VMS 
 unviable."  C 	Just what metric do you have in mind?  Of those metrics, is Power4wF 	or UltraSparc 3 any better at any of Itanium2's ?  If so, which ones?A 	I overlooked them if they are out there and directly comparable.p   					Rob   ------------------------------   Date: 1 Nov 2002 14:48:44 -0600D+ From: young_r@encompasserve.org (Rob Young)u@ Subject: RE: Intel Conceding 64-Bit Desktops to IBM, AMD - eWeek3 Message-ID: <QnMC59o01$oy@eisner.encompasserve.org>   W In article <01C281A6.6053E3E0@sulfer.icius.com>, Shane Smith <ssmith@icius.com> writes:iI > So far we've seen two versions of Itanium not so much in the wild as in  > wildlife parks.s  * 	That isn't accurate, if accuracy matters:  ( http://www.theinquirer.net/?article=6025  K AN EARLIER REPORT here that HP was suffering a lack of availability for its2K Itanium 2 machines - the ZX 6000, and the ZX2000 - was scotched by the firmR today.N A glitch on the Web site meant that the ZX6000 wasn't easily visible while theL ZX 2000, listed as on back order, is actually available, said Larry Mahoney,E Itanium Product Manager at HP's workstation division in the US today.p  N In fact, said Mahoney, although the HP site showed the ZX2000 on backorder, itJ and the ZX8000 were readily available and could ship in four to five days.  & "Both machines are shipping," he said.  L Mahoney claimed that the HP machines, which use its own proprietary chipset,B were the fastest available Itanium 2 based machines on the market.  9 > Both apparently lag behind contemporary chips in enoughr1 > areas that they're pretty much laughing stocks.   ? 	Name an area, any area.  There may be a few but they aren't ofd> 	the greatest importance.  Again, any of these areas less than$ 	Power4 and UltraSparc3 performance?  G > things it's good at but that's not enough. It has to have enough of aaJ > performance /advantage/ over its competitors in enough areas that people? > sit up and take notice. It's got to knock people's socks off.h  @ 	It does and it will , especially in floating point and databaseA 	metrics.  A clean migration to Itanium for HP/UX and VMS and news@ 	Windows64 will be happening.  In other words, there is built-in 	customer base for Itanium.y  
 > RememberF > when they were trying to sell the Alpha? It was significantly fasterB > than IA32, but that wasn't enough to overcome the lack of native > software.r >   = 	This is a good point.  Part of Itanium's success will lie ingC 	the Itanium porting money available.  A fully funded program worthm6 	over >$150 milllion if I recall (maybe $100 million?)  G > VMS is already hard enough to sell. If it's perceived as slow too, it.H > won't matter if HP decide to start marketing it. It won't be sellable.  = 	But it will be a better platform.  It will be cheaper.  I'veiB 	rolled in several AlphaServers in several places.  It was/is goodB 	they are running VMS.  If they were running Windows 2000 (pretend@ 	for a second the world flipped upside down) , they would be farC 	less attractive than currently shipping (comparable) Windows 2000 .F 	Intel hardware in price and nearly as great in performance (certainlyG 	better in certain database metrics but hard to get an apples to appleseF 	comparison on that).  I expect that Itanium will follow a much nicer F 	price performance curve than Alpha was ever on, and I'm an Alpha fan.  @ 	There is a last hurrah on the upside... Marvel is Marvelous andB 	quite astounding in performance and VMS will fly on it, squashing  	a lot of those "slow" mumblers.   				Rob,  B Men with walkie-talkie                  I'm home again to you babeC Men with flashlights waving             You know it makes me wonder G Up upon the tower                       Sittin' in the quiet slipstreamr> The clock reads daylight savin'         Rollin' in the thunder  -                                 -- Neil Young    ------------------------------  $ Date: Fri, 1 Nov 2002 14:09:19 -0800$ From: Shane Smith <ssmith@icius.com>@ Subject: RE: Intel Conceding 64-Bit Desktops to IBM, AMD - eWeek0 Message-ID: <01C281B0.51AB6750@sulfer.icius.com>  G I didn't say it wasn't shipping, I was implying that it's only in a fewhH specialised places. A wildlife park being closer to the wild than a zoo.  C Any area where Itanium lags? 32 bit. It runs them, but so slow it'seH painful. They should have just left the 32bit compatability out, becauseB it's marring Itanium's image. Don't even /try/ to run any MadOnion? benchmarks on it, or winbench. A lot of the unwashed masses paykC attention to those, because they're so prevalent in the PC hobbyistnG community. SpecINT? What's that? Remember, it's perception that matters H in sales, not facts. (Windows anyone?) Itanium 1 was found to just aboutG match a 100mhz vanilla Pentium quite early on in 32 bit benchmarks, and  that's stuck in many minds.l  G Windows 64-bit isn't a captive market if W64 also runs on Hammer, which1B I hear it will. And of course Hammer can run the 32 bit stuff veryE quickly indeed. Any idea if 32 bit apps will run on W64 if the chip's F compatible? It would be a good move for MS if they will. And of courseF if so, Itanium's got a big disadvantage there in available software as well as speed.  G Don't get me wrong, since VMS is going there I want Itanium to be good."G I'm just not convinced yet it'll be good /enough/. Alpha led the world,  and look where that got us.n   Shanex   -----Original Message-----B From: young_r@encompasserve.org [mailto:young_r@encompasserve.org]( Sent: Friday, November 01, 2002 12:49 PM To: Info-VAX@Mvb.Saic.Comd@ Subject: RE: Intel Conceding 64-Bit Desktops to IBM, AMD - eWeek    < In article <01C281A6.6053E3E0@sulfer.icius.com>, Shane Smith <ssmith@icius.com> writes:I > So far we've seen two versions of Itanium not so much in the wild as in. > wildlife parks.L  * 	That isn't accurate, if accuracy matters:  ( http://www.theinquirer.net/?article=6025  G AN EARLIER REPORT here that HP was suffering a lack of availability forl its F Itanium 2 machines - the ZX 6000, and the ZX2000 - was scotched by the firm today.D A glitch on the Web site meant that the ZX6000 wasn't easily visible	 while theaC ZX 2000, listed as on back order, is actually available, said Larry  Mahoney,E Itanium Product Manager at HP's workstation division in the US today.o  @ In fact, said Mahoney, although the HP site showed the ZX2000 on
 backorder, it D and the ZX8000 were readily available and could ship in four to five days.r  & "Both machines are shipping," he said.  C Mahoney claimed that the HP machines, which use its own proprietarya chipset,B were the fastest available Itanium 2 based machines on the market.  9 > Both apparently lag behind contemporary chips in enought1 > areas that they're pretty much laughing stocks.s  ? 	Name an area, any area.  There may be a few but they aren't ofK> 	the greatest importance.  Again, any of these areas less than$ 	Power4 and UltraSparc3 performance?  G > things it's good at but that's not enough. It has to have enough of amJ > performance /advantage/ over its competitors in enough areas that people? > sit up and take notice. It's got to knock people's socks off.p  @ 	It does and it will , especially in floating point and databaseA 	metrics.  A clean migration to Itanium for HP/UX and VMS and newu@ 	Windows64 will be happening.  In other words, there is built-in 	customer base for Itanium.   
 > RememberF > when they were trying to sell the Alpha? It was significantly fasterB > than IA32, but that wasn't enough to overcome the lack of native > software.  >   = 	This is a good point.  Part of Itanium's success will lie inaC 	the Itanium porting money available.  A fully funded program worthn6 	over >$150 milllion if I recall (maybe $100 million?)  G > VMS is already hard enough to sell. If it's perceived as slow too, itoH > won't matter if HP decide to start marketing it. It won't be sellable.  = 	But it will be a better platform.  It will be cheaper.  I'veiB 	rolled in several AlphaServers in several places.  It was/is goodB 	they are running VMS.  If they were running Windows 2000 (pretend@ 	for a second the world flipped upside down) , they would be farC 	less attractive than currently shipping (comparable) Windows 2000 /F 	Intel hardware in price and nearly as great in performance (certainlyG 	better in certain database metrics but hard to get an apples to applesfF 	comparison on that).  I expect that Itanium will follow a much nicer F 	price performance curve than Alpha was ever on, and I'm an Alpha fan.  @ 	There is a last hurrah on the upside... Marvel is Marvelous andB 	quite astounding in performance and VMS will fly on it, squashing  	a lot of those "slow" mumblers.   				Robu  B Men with walkie-talkie                  I'm home again to you babeC Men with flashlights waving             You know it makes me wonderiG Up upon the tower                       Sittin' in the quiet slipstreamA> The clock reads daylight savin'         Rollin' in the thunder  -                                 -- Neil Young    ------------------------------  # Date: Fri, 01 Nov 2002 23:04:42 GMTl# From: "John Smith" <a@nonymous.com>-@ Subject: Re: Intel Conceding 64-Bit Desktops to IBM, AMD - eWeekI Message-ID: <eMDw9.143614$Q3S.48163@news01.bloor.is.net.cable.rogers.com>   8 "Rob Young" <young_r@encompasserve.org> wrote in message- news:QnMC59o01$oy@eisner.encompasserve.org...n >lA > There is a last hurrah on the upside... Marvel is Marvelous and-C > quite astounding in performance and VMS will fly on it, squashings! > a lot of those "slow" mumblers.R    L There are rumors circulating about that HP is looking to downplay the Marvel. numbers so it doesn't make Itanic look so bad.  I The old adage, "Where there's smoke, there's fire" probably holds true on  this.e   ------------------------------  # Date: Fri, 01 Nov 2002 23:16:12 GMT # From: "John Smith" <a@nonymous.com> @ Subject: Re: Intel Conceding 64-Bit Desktops to IBM, AMD - eWeekI Message-ID: <0XDw9.143659$Q3S.90346@news01.bloor.is.net.cable.rogers.com>r  1 "Shane Smith" <ssmith@icius.com> wrote in messaget* news:01C281B0.51AB6750@sulfer.icius.com...  J > And of course if so, Itanium's got a big disadvantage there in available > software as well as speed.    J If the worst happens to "The Gang That Couldn't Patent Infringe Straight",- Itanic won't have any speed at all...0.00GHz.8( http://www.theinquirer.net/?article=5989  5 Guess they'll just have to go back to developing EV8.s   ------------------------------  % Date: Fri, 01 Nov 2002 19:15:50 -0500 ( From: David Froble <davef@tsoft-inc.com>@ Subject: Re: Intel Conceding 64-Bit Desktops to IBM, AMD - eWeek, Message-ID: <3DC31936.6050104@tsoft-inc.com>   Daryl Jones wrote:  v > "John Smith" <a@nonymous.com> wrote in message news:<UYgv9.86381$mxk1.78654@news04.bloor.is.net.cable.rogers.com>... > 8 >>"Barry Treahy, Jr." <Treahy@MMaz.com> wrote in message# >>news:3DBD8644.5070705@MMaz.com...v >>K >>>If Intel were to release an 64-bit X86, they would undermine their IA-64.J >>>efforts!  This could create serious problems for us VMS folks expectingJ >>>to run on IA-64 because what was the Compaq rational for killing Alpha?: >>> Not enough volume to support the design and fab costs. >>>QE >>>So, if Intel does introduce a 64-bit X64 for the desktop, you knowsK >>>manufacturers will introduce servers based on that chip set, essentiallyt, >>>killing high volume demand for the IA-64. >>>pI >>>Am I the only one that sees the irony of this?  A major porting effortcG >>>of VMS to another 'low volume and essentially proprietary' chip set?eI >>> Worse yet, a chip set that HP/Compaq doesn't directly control and may 7 >>>not even be as fast as the Alpha that was abandoned?d >>>i >>? >>No Barry, you aren't the only one who sees the irony of this.a >>M >>However all that do see the irony don't occupy executive level positions at- >>HP.- >> >  > Gentlemen: > G > Before the HP-COMPAQ merger, both HP and Compaq annouced that I64 wasuE > to be their chip of the future. Compaq gave the Alpha technology toDG > Intel. The Merced/Itanium chip performance would be camparable to thew > EV7.    4 How much crippling of EV7 will be required for this?  ? > What made the Alpha performance was not only the chip but theeC > compilier technology. Intel in this area was ten years behing theS > Alpha.    L So what are you saying here?  10 years behind in compiler technology, or 10  years behind in CPU technology?h  A > By 2004, HP will have only one architecture to work with ratherrE > than two and it is out of the hardware business to boot. I wouldn'ti > worry about VMS.    Q The only people that should not worry about VMS are those that don't give a damn  M about VMS.  If you have a desire to see VMS survive, then you should be very t worried.  O When Hammer forces Intel to compete, (Yamhill), then IA-64 is D-E-A-D!  It may nM quiver for a while, but who other than HP will buy it.  Surely not Dell, who x" will go with the MS prefered chip.  5 > There are about 25 programmers in NH working on the F > conversion from Alpha to Intel. The jobs were posted in DICE several
 > months ago.d    L Fine.  I'm all for such a port.  I do wonder whether such would be occuring J without the Intel financing.  Who will finance the port to Yamhill/Hammer?     Dave   ------------------------------  $ Date: Fri, 1 Nov 2002 21:38:14 -0500* From: "Bill Todd" <billtodd@metrocast.net>@ Subject: Re: Intel Conceding 64-Bit Desktops to IBM, AMD - eWeek2 Message-ID: <NsWcnXAmzdELp16gXTWc3w@metrocast.net>  D "Daryl Jones" <jones.computer.srv@worldnet.att.net> wrote in message7 news:8a646952.0211011215.7e6a2bd5@posting.google.com...l   ...s  G > Before the HP-COMPAQ merger, both HP and Compaq annouced that I64 was9E > to be their chip of the future. Compaq gave the Alpha technology toMG > Intel. The Merced/Itanium chip performance would be camparable to the  > EV7.  G This post is threatening to set new standards of cluelessness in c.o.v.l$ Let's start with that last sentence.  H Merced truly stood alone in the commercial performance sweepstakes.  TheI first SPECint2K_base score released for it after it shipped was 314 (on a0I Dell box), though HP later managed to get that figure up first to 342 and  eventually to 379.  I This wasn't just a lot slower than Alpha (which scored 621 at that time -eL note that this was of course on EV6, not EV7, which should easily top 1000):K it was slower than *everyone*.  Even SPARC, which is often maligned here asoE something of a SPEC slug, posted a score - for systems it was already L shipping - of 439 prior to Merced's launch, and bumped that up to 537 only 3 months after Merced hit 379.  >  What made the Alpha performance was not only the chip but theC > compilier technology. Intel in this area was ten years behing the. > Alpha.  L That would have been difficult, since Alpha was only 9 years old when MercedJ launched.  And especially since EPIC had been in design/development almost: as long as Alpha had (HP started working on EPIC in 1989).  D As far as compiler technology goes, Itanic (unlike virtually all itsK competition) has been using feedback-directed optimization in its SPEC baseiC testing (because it pretty much has to to yield anything that's notaG completely laughable).  One can question whether such optimization willoI prove typical for application developers (the rationale behind the 'base'yG set of scores being that it should reflect common practices rather thanvJ pedal-to-the-metal quests for speed), but even if the answer is "yes" thenL the next question is how much other platforms' scores will increase once the competition does likewise.  @  By 2004, HP will have only one architecture to work with rather: > than two and it is out of the hardware business to boot.  F That's HP's problem.  What people here are trying to ensure it that it doesn't become theirs as well.   - bill   ------------------------------  $ Date: Fri, 1 Nov 2002 21:50:55 -0500* From: "Bill Todd" <billtodd@metrocast.net>@ Subject: Re: Intel Conceding 64-Bit Desktops to IBM, AMD - eWeek2 Message-ID: <YiGdnWcsi98SoF6gXTWcqQ@metrocast.net>  6 "Barry Treahy, Jr." <Treahy@MMaz.com> wrote in message! news:3DC2EA1A.5060305@MMaz.com...s   ...n  2 > Alpha was abandoned because it was too expensive3 > to push to EV7 and beyond due to low chip volume.@  J Hello?  Hello?  Have you been paying any attention whatsoever for the past
 16 months?  K Alpha was abandoned because Compaq really didn't want to be responsible foruL significant technology development (the reason VMS will likely be mothballedL as well as soon as it seems convenient), or because of a fit of pique in theL server group, or because Shane Robison is terminally incompetent, or becauseG Curly is, or as a requisite prelude to the merger (to avoid FTC hasslessG related to HP buying a viable competitive architecture), or all or sometI combination of the above.  Despite its relatively modest volume (which ofaL course could have been increased many-fold with any real effort at marketingK and visible commitment) Alpha was profitable:  the impact that declaring itiJ dead had on VMS and Tru64 profits alone would have paid for completing EV8L several times over - even if EV8 had then never generated a penny of revenue on its own.g   - bill   ------------------------------  % Date: Fri, 01 Nov 2002 20:45:06 -0500t- From: JF Mezei <jfmezei.spamnot@videotron.ca>n@ Subject: Re: Intel Conceding 64-Bit Desktops to IBM, AMD - eWeek, Message-ID: <3DC32E21.776B62F6@videotron.ca>   Rob Young wrote:G >         Recent articles talk about Madison (Itanium 3) being a 30-50%e2 >         performance improvement over Itanium 2.   
 Whoopty doo !o  N But by the time it comes out, what performance imporvements will Power , Sparc# and even the 8086 have been given ?n  M Remember the Digital propaganda that blasted IA64 and showed how Alpha was soeH superior ? Well, guess, what, the argument that IA64 was bloated, overlyK complex and very tie consuming and expensive to design still stick with me.   M Until someone convinces me otherwise, I will continue to believe that IA64 isiI trying desperatly to keep up with other architectures and has no hopes ofaG leading because they will be so busy just managing their overly complexrJ architecture that they won't have the time to incorate fancy new features.  N Everyone benefits from process shrinks. So if they hope to get 30% improvementF through process shrink and faster clock rate, then it means that otherC architectures will also get those improvements so IA64 won't gain a0 competitive edge.    ------------------------------  $ Date: Fri, 1 Nov 2002 21:13:18 -0500* From: "Bill Todd" <billtodd@metrocast.net>@ Subject: Re: Intel Conceding 64-Bit Desktops to IBM, AMD - eWeek2 Message-ID: <4HWdnahnFPAjqV6gXTWc3w@metrocast.net>  8 "Rob Young" <young_r@encompasserve.org> wrote in message- news:FBZW$5W1DUhq@eisner.encompasserve.org...h   ...v  ? > Recent articles talk about Madison (Itanium 3) being a 30-50%:F > performance improvement over Itanium 2.  If we chose 40% as a guess,1 > that would put Madison at or near these values:a >g > 1300 SpecInt2000  E Time for some remedial math, Rob.  The best current SPECint score fort+ McKinley is 810.  Add 40% and you get 1134.    > 1800 Specfp2000   L On the other hand, 1356 * 1.4 = 1898 - nice for HPTC, but that doesn't bringL in the revenue that commercial computing does, and those folks can often useK larger numbers of more cost-effective processors (Itanic showing no sign of-H being cost-effective any time soon, unless Intel is willing to lose evenG more money on each one than it is currently - and that still won't helpsE performance/Watt, which is also an issue in some HPTC installations).t  # > 115000 tpmC for a 4 processor boxl  I 78454 * 1.4 = 109835.6:  0 for 3 now, Rob.  The other obvious question ispL what kind of performance Itanic can attain running a more serious enterpriseI system/database combination than Windows/SQL Server (which was presumablymK chosen for the test - as it apparently is for virtually all platforms whererJ running Windows is an option - because it yielded better tpmC numbers thanF the Unix/Oracle and Unix/Sybase combinations used in the Alpha tests).   >i? > Maybe it doesn't show up until mid-2003, but I'm sure we willo > hear a lot more about it.   J Actually, the closer it gets to shipping the more we should hear about howK great Montecito is going to be:  selling the future processor each time thee= current one comes up short has become a tradition for Itanic.v  $   It should be at or near the top of > performance when it ships.  L I don't think so, Rob.  Hammer is already ahead of where you project MadisonK to be in SPECint - and using only 32-bit mode, with 64-bit mode expected toCK improve that number by about 20% (to over 1400).  And that's when it debutsuE at the end of Q1 next year:  later speed-bumps are expected, plus the H appearance of a 90 nm. version of Hammer before Montecito ships.  As forH TPC-C, you've been contending recently that on-chip cache-size increasesK would do wonders for future Itanic family members, so what do you expect to L happen when Alpha gets a *14-fold* increase in on-chip cache size in EV7 (toC complement radically reduced memory latency that should bring EV7's2I main-memory accesses up to about half the speed of EV6's current off-chipm cache)?w  G Yes, Madison will likely lead in SPECfp, but Sandia seems to prefer theyL cost-effectiveness and power-effectiveness of Hammer, so whether Itanic willJ pick up much more in the way of supercomputing installations is debatable.F Madison should make a good high-end engineering workstation processor, though.k   ...t   > By the way, when you say:s >hJ > "If Itanium's performance doesn't improve drastically the move will make VMS  > unviable." >bD > Just what metric do you have in mind?  Of those metrics, is Power4G > or UltraSparc 3 any better at any of Itanium2's ?  If so, which ones?cB > I overlooked them if they are out there and directly comparable.  L My guess would be that he's looking ahead to 2004, when at least most of VMSH will have been ported and it can be sold on Itanic with something like aH straight face.  The problem is that Itanic seems destined to lose groundH between now and that date:  while Madison and Montecito are getting onlyK process shrinks and more on-chip cache, other architectures will be getting I the same process shrinks and cache boosts plus other *major* enhancementsiG (Alpha's and Hammer's on-chip memory and MP glue being perhaps the mostdI dramatic, but POWER is scheduled to get SMT in this period and both SPARCoH and PA-RISC get dual cores on their chips).  There's some indication nowG that Itanic may get dual cores in 2005 (when I had been assuming it waseK likely to get EV7-style on-chip glue, which there now may be some reason toBG doubt), but real core enhancements still seem no earlier than 2006 (anda maybe even later).  L Incidentally (since you ask above), perhaps the most important metric ItanicK fails dismally on compared with SPARC and POWER is performance/Watt.  PoweroI and cooling requirements are starting to assume first-class status in theeG overall scheme of things - at exactly the wrong time for a power-hungryu$ architecture like EPIC to take hold.  F So I'd change Shane's statement to:  "If Itanium's performance doesn'tJ improve drastically, Itanium itself will be unviable (at least as anythingH other than a proprietary HP niche architecture)."  Unfortunately, a side4 effect of that will be to make VMS unviable as well.   - bill   ------------------------------  $ Date: Fri, 1 Nov 2002 23:11:12 -0500* From: "Bill Todd" <billtodd@metrocast.net>@ Subject: Re: Intel Conceding 64-Bit Desktops to IBM, AMD - eWeek2 Message-ID: <jMadnSLmn539zV6gXTWcpg@metrocast.net>  8 "Rob Young" <young_r@encompasserve.org> wrote in message- news:QnMC59o01$oy@eisner.encompasserve.org...1> > In article <01C281A6.6053E3E0@sulfer.icius.com>, Shane Smith <ssmith@icius.com> writes:K > > So far we've seen two versions of Itanium not so much in the wild as in  > > wildlife parks.X > + > That isn't accurate, if accuracy matters:! >h* > http://www.theinquirer.net/?article=6025 >kI > AN EARLIER REPORT here that HP was suffering a lack of availability fork itsgH > Itanium 2 machines - the ZX 6000, and the ZX2000 - was scotched by the firm > today.L > A glitch on the Web site meant that the ZX6000 wasn't easily visible while theiE > ZX 2000, listed as on back order, is actually available, said Larryv Mahoney,G > Itanium Product Manager at HP's workstation division in the US today.i > B > In fact, said Mahoney, although the HP site showed the ZX2000 on
 backorder, itrL > and the ZX8000 were readily available and could ship in four to five days. >a( > "Both machines are shipping," he said.  L Well, since they supposedly launched almost 4 months ago, one would hope so.H The question then becomes *exactly how long* have they been shipping?  AL week?  A month?  Since people have tried and failed to obtain them (not justL looked at the order pages) quite recently, probably not much more than that.   >qE > Mahoney claimed that the HP machines, which use its own proprietaryn chipset,D > were the fastest available Itanium 2 based machines on the market.  J Can't quibble with that:  they're the *only* Itanic2-based machines on theK market, save possibly for the 32-processor NEC beasts (and one can't reallyh? expect a 32-processor box to compete equally with a 4-processoru3 contemporary:  just ask the GS and ES Alpha folks).r   >N; > > Both apparently lag behind contemporary chips in enoughr3 > > areas that they're pretty much laughing stocks.- >- > Name an area, any area.o  D Well, for Merced it's far easier to name the single area in which itK *wasn't* a complete laughingstock:  raw FP performance.  Have you forgottenF( the 'smoking brick of death' so quickly?  H So my comments below apply only to McKinley, which whatever else one may< think of it is sure as hell a major improvement over Merced.  '   There may be a few but they aren't oft > the greatest importance.  I I guess that depends on one's priorities.  For example, you don't seem tolJ consider price/performance very important, since IA32 trounces McKinley inD that area (as will Hammer) and so does the much-maligned SPARC.  AndG McKinley hasn't improved upon Merced's bottom-of-the-barrel position in1H performance/Watt, which, among other things, dictates viability in bladeI configurations (and, of course, also feeds back into price/performance in . terms of operational and cooling power usage).  J McKinley certainly uses chip area far less efficiently than IA32 or HammerG or Alpha or SPARC (and probably POWER4 and PA-RISC as well, but I don'ttF happen to know their core sizes), and there's an overall Rube GoldbergI quality to the design, but those are more issues of technical respect (orp= lack thereof) than something of primary importance to a user.n  %   Again, any of these areas less thana% > Power4 and UltraSparc3 performance?"  H They're both significantly superior to McKinley in performance/Watt, and+ SPARC is superior in performance/$ as well.s   >fI > > things it's good at but that's not enough. It has to have enough of apL > > performance /advantage/ over its competitors in enough areas that peopleA > > sit up and take notice. It's got to knock people's socks off.u >c	 > It doesl  J No, it doesn't.  Published benchmarks have been selected with extreme careG to avoid head-to-head comparisons, save for SPECfp (where Itanic indeediJ excels, with the aid of feedback-directed optimization in the base tests -J which might bring the newest Alpha within striking distance if it had usedJ them as well) and SPECint (where McKinley doesn't knock anyone's socks offH but does more or less equal the best of the current 64-bit competition -K though, again, only by virtue of using feedback-directed optimizations thate the competition doesn't use).o  J But as for other benchmarks:  SPECweb99_SSL is nowhere nearly as popular aI benchmark as SPECweb99, hence does not expose Itanic to competition from,vK e.g., other chips with large on-chip caches (the performance of PA-RISC andeI Xeon in SPECweb99_SSL suggests that they may be a very important factor -tI which should become much more obvious once that benchmark is run on EV7).'I Unisys' recent use of the 300 GB TPC-H test series to attempt to positioneJ its 32-processor box as a serious enterprise machine for 'large-scale' useL was laughable:  all the *really* serious enterprise competitors had used theL larger (3 TB or at least 1 TB) databases for their tests, so its choice of aD small version with no directly-comparable competition whatsoever was telling.  8  and it will , especially in floating point and database
 > metrics.  K FP, yes, but I still have major doubts about Itanic having any advantage inIG DB use, especially over EV7 and Hammer (both of which I expect to leave7G Itanic in the dust if anyone has the inclination, time, and money to doi serious tests with them).s   - bill   ------------------------------  % Date: Fri, 01 Nov 2002 22:38:01 -0600o1 From: "David J. Dachtera" <djesys.nospam@fsi.net>h> Subject: Re: Itanic ordered to halt by court. Iceberge free...' Message-ID: <3DC356A8.B736518D@fsi.net>h   Paddy O'Brien wrote: >  > John Smith wrote:e9 > > "Bob Ceculski" <bob@instantwhip.com> wrote in message : > > news:d7791aa1.0210311538.6a2fe8c@posting.google.com... > >c< > >>for 400 million they could have had ev8 which would have8 > >>blown away everything else around ... just as bad as; > >>everyone thinking linux is saving them money when a vms48 > >>solution would be saving them even more money in the* > >>long run ... it's called stupidity ... > >k > >tK > > And that's why the executives get paid the big bucks. You can't be that , > > stupid unless you get paid $117 million. > >f > "unless" ?? = "if"  @ I think he had it right: as has been posted here numerous times,F knowledge and salary tend to be inversely proportional, though I wouldE qualify that as not knowledge possessed but knowledge exercised. ManynG executives are very highly educated people who simply never learned how G to use their knowledge effectively, or learned more about how to misuse>H it than how to use it to benefit customers, stock-holders *AND* company,/ rather than one or more at the other's expense.f   -- e David J. Dachterao dba DJE Systemsm http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/a   ------------------------------  $ Date: Fri, 1 Nov 2002 20:12:37 +01004 From: Didier.Morandi.nospam@Free.fr (Didier Morandi)/ Subject: Re: Looking for contract opportunitiesmC Message-ID: <1fkzjwe.158r9vw1osel3gN%Didier.Morandi.nospam@Free.fr>t  3 My understanding of the situation is the following:t  H I have more than 20 years of consulting experience, not only on VMS, butH also in organization, security, and other companies management analysis.  B If I lean the technique of one or two among the most requested SAPD modules, i.e. Fi and Co, I think that I may help any customer having some needs on SAP Fi or Co.r  C I may be wrong, and in that case, I would love to hear from this ng  before I register.   D.  5 Tim Llewellyn <tim.llewellyn@blueyonder.co.uk> wrote:C   > Didier Morandi wrote:  > > 	 > > Yeah!  > >  > > :-)< > >  > > D. > > K > > (SAP Basis training: 2nd of January 2003 for 54 days... and 6800 euros,r > > ie 6800 US$) > >  > O > Is training adequate to get work in the current climate? Seems to me you need M > two years commercial experience of exactly the technologies required at theu	 > moment.  >  R >      ------------------------------  # Date: Fri, 01 Nov 2002 20:57:21 GMTp- From: "Kristian Strickland" <nospam@dev.null>g  Subject: Netware and OpenVMS 7.36 Message-ID: <RUBw9.52$wy.6792@ursa-nb00s0.nbnet.nb.ca>  D Go ahead, laugh all you want:  we're still running the Leverage HostH Services v4.2r3 product, from (at the time) Emulex, to allow old netwareL desktops to access our OpenVMS 7.1 box for automated batch file transfering.  J Well, we're finally in the middle of testing an upgrade to OpenVMS 7.3 butK have found that the IPX driver (QXDRIVER 4.0r7) fails to start on the newers VMS.  J Natually, since Emulex hasn't supported this product in years, I'll get noI help from them.  I know that I'm grasping at straws in hoping someone out'G there reading this has gotten LHS to run on OpenVMS 7.3, but of course,b< noone here wants to make any code changes, so I have to ask.  F If you've had any luck, and can help, I'd appreciate it.  After a "setJ verify", here's the fragment from IPX_LOAD.COM (called by IC_STARTUP) that7 actually starts the driver, and the error it generates:e  4 $ xcp start /qxdev=_qxa0:-      ! QXDRIVER to 0start?         /bufsiz=1500-           ! Maximum IPX message data size 5         /esckmax=90- ! Maximum # of ephemeral socketsbA         /wsckmax=30-            ! Maximum # of well-known sockets =         /nibuffers=10-  ! Number of Ethernet Received Buffersc4         /spxconmax=64-  ! Maximum of SPX Connections3         /protyp=0000-   ! Ethernet protocol number"c5         /internalnetwork=0- ! Internal Network Number '         /nidev=_EWA:- ! Ethernet driverhG         /network=0000A002 ! Modified by LHSconsole  7-OCT-2000 11:27:276  9 %QXCP-E-BADPARAM, bad parameter value, parameter ID #2772i' %SYSTEM-F-BADPARAM, bad parameter valuee( %QXCP-E-NOTSTARTED, QXDRIVER not started  G I've tried the Microsoft approach (reinstall), in this case because the < installation links the programs, but I get the same results.   TIA,
 --Kristian   ------------------------------    Date: 02 Nov 2002 04:35:44 +0800, From: Paul Repacholi <prep@prep.synonet.com>@ Subject: Re: Not sure if anyone else I seen this from LinuxWorld- Message-ID: <878z0d3u5b.fsf@prep.synonet.com>   % "John Smith" <a@nonymous.com> writes:c  F > So to steal a 'rugged' page from Hummer...maybe we'll put this ad inA > 'Field & Stream', after all don't some CEO's hunt and fish too?t  : But they stand up in canoes and can't do up their shirt :) Nice fish though...a   -- w< Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda.n@                                              West Australia 6076. Raw, Cooked or Well-done, it's all half baked.F EPIC, The Architecture of the future, always has been, always will be.   ------------------------------  # Date: Fri, 01 Nov 2002 17:42:56 GMTo4 From: Tim Llewellyn <tim.llewellyn@blueyonder.co.uk>9 Subject: Re: OpenVMS printing to printer via windows 2000g0 Message-ID: <3DC2BD02.27F7D61B@blueyonder.co.uk>   Mark Bowman wrote: > F > Does anyone know how to print to local attached printer from OpenVMS  > which is running Windows 2000?  ? I assume the printer is attached to the w2k box and you want tob8 print to it from VMS though this is not  entirely clear.   > B > I have configured the printing port on the Windows PC as ABC600. >  > From TCPIP$PRINTCAP.DATr >  > OCE_TEST|oce_test:\n- >         :lf=/TCPIP$LPD/SPOOL/OCE_TEST.LOG:\e >         :lp=OCE_TEST:\ >         :rm=x.x.x.x:\d( >         :sd=/TCPIP$LPD/SPOOL/OCE_TEST:  F You need to specify a remote printer :rp=ABC600 if my assumption above is correct.n   > & > Maybe the above file is not correct? > @ > Server queue OCE_TEST, idle, on FALCON::, mounted form DEFAULT; >   /BASE_PRIORITY=4 /DEFAULT=(FEED,FORM=DEFAULT) Lowercase8 > /OWNER=[X,SYSTEM]tH >   /PROCESSOR=TCPIP$LPD_SMB /PROTECTION=(S:M,O:D,G:R,W:S) /RETAIN=ERROR > " > All jobs go into a black hole... > 	 > Regardse  I Is lpd server running on the remote machine? Is lpd security/proxy setup c$ to allow lpd access from VMS to w2k?   regards,   > Mark > http://www.bonefire.org/guy/   -- e tim.llewellyn@blueyonder.co.uk r  H * PLEASE NOTE tim.llewellyn@cableinet.co.uk address is NO LONGER VALID *   ------------------------------  % Date: Fri, 01 Nov 2002 15:15:19 -05008- From: JF Mezei <jfmezei.spamnot@videotron.ca>t- Subject: Re: Was OpenVMS left out on purpose?n, Message-ID: <3DC2E0D5.7B0CD909@videotron.ca>  ! VAXman-, @SendSpamHere.ORG wrote: H > >>    This survey was sponsored and written by the NonStop DevelopmentL > >> department within HP so it's scope was limited to the NonStop platform.  I > BULLSHIT!  Since when is Windows a non-stop platform?  Just because itsbH > crappy GUI has only a [Start] button doesn't make it *NON-STOP* by any > stretch of the imagination!   H It isn't the inclusion of Windows in the survey that is a problem. SinceL Tandem has a very basic user interface, one could potentially see many usingG Widnows as front ends to Tandems for development, system management, orb client-server applications.v  D HOWEVER, if they went through the trouble of including various Unix,M especially including Tru64, then they have no excuse for having omitted VMS.  N And since Tandem boxes often co-exist with IBM mainframes, shouldn't they have- included IBM mainframes in the list as well ?   J When I see such surveys, I am given the impression that the goal is not toN research the installed base to help you make a decision, but rather to collectM information that will help you justify a decision that has already been made.i   ------------------------------  # Date: Fri, 01 Nov 2002 21:08:51 GMTo! From: "Scratch" <scratch@aol.com>s- Subject: Re: Was OpenVMS left out on purpose?E> Message-ID: <D3Cw9.61438$dn3.2014117@twister.southeast.rr.com>  5 "Dave Gudewicz" <dgudewicz@core.com> wrote in messagep3 news:3dc2b758$0$17648$1dc6e903@news.corecomm.net...sC > Explains what?  That LINUX and Windows are no considered non-stope
 platforms, > said the spider to the fly.m  L I was just passing the message along.  I guess by "explains" I meant that itL was the Nonstop group and not anyone related to OpenVMS.  Heck, I don't know why they left it out.h     Kent   -- Kenneth Farmer http://www.OpenVMS.org        5 "Dave Gudewicz" <dgudewicz@core.com> wrote in messagea3 news:3dc2b758$0$17648$1dc6e903@news.corecomm.net... C > Explains what?  That LINUX and Windows are no considered non-stopt
 platforms, > said the spider to the fly.s >  > --	 > Dave...e >fI > It is noble to teach oneself, but still nobler to teach others-and less 
 > trouble. > -----Mark Twainl >t. > "Scratch" <scratch@aol.com> wrote in message: > news:k8xw9.61154$dn3.1979508@twister.southeast.rr.com...# > > I guess this kinda explains it.  > >s   > >d > >w > >o  > > ----- Original Message -----3 > > From: "Moser, Jessica " <Jessica_Moser@sba.com>e0 > > To: "'Kenneth Farmer'" <kfarmer@openvms.org>, > > Sent: Friday, November 01, 2002 10:28 AM# > > Subject: RE: Instapoll questionp > >a > >m > > > Kennth - > > >dI > > >    This survey was sponsored and written by the NonStop DevelopmentMC > > > department within HP so it's scope was limited to the NonStop 	 platform.e > WeH > > > apologize if this wasn't communicated properly. Thank you for your > > interestD > > > in the advocacy program. Please continue to check our web siteG > > > (www.hpuseradvocacy.org) for future advocacy related initiatives.a > > >(  > > > -----Original Message-----7 > > > From: Kenneth Farmer [mailto:kfarmer@openvms.org]m/ > > > Sent: Thursday, October 31, 2002 12:12 PMR" > > > To: admin@hpuseradvocacy.org! > > > Subject: Instapoll questions > > >  > > >m& > > > Was OpenVMS left out on purpose? > > > A > > > http://www.hpuseradvocacy.com/Surveys/2002October/index.cfme > > >P > > >aG > > > For which platform(s) are you developing enterprise applications?e > > > NonStop Kernel > > > HP-UXi > > > Tru64 UNIX
 > > > Windowsa > > > Linuxg > > > Other (please elaborate) > >r > >c > >. > >f > >a >y >    ------------------------------   Date: 1 Nov 2002 16:11:33 -0800 ( From: bob@instantwhip.com (Bob Ceculski)- Subject: Re: Was OpenVMS left out on purpose?d= Message-ID: <d7791aa1.0211011611.1b5d515a@posting.google.com>   h koehler@encompasserve.org (Bob Koehler) wrote in message news:<plp1T4EmzszR@eisner.encompasserve.org>...d > In article <k8xw9.61154$dn3.1979508@twister.southeast.rr.com>, "Scratch" <scratch@aol.com> writes: > > 
 > >> Kennth -i > >>H > >>    This survey was sponsored and written by the NonStop DevelopmentO > >> department within HP so it's scope was limited to the NonStop platform. WeiG > >> apologize if this wasn't communicated properly. Thank you for yourr >  interestkC > >> in the advocacy program. Please continue to check our web sitetF > >> (www.hpuseradvocacy.org) for future advocacy related initiatives. > D >    Why would the NonStop Development department want to know aboutC >    NonStop Kernel, HP-UX, Tru64 UNIX, Windows, and Linux, but notk& >    about other HP operating systems?   because they are morons ...    ------------------------------  $ Date: Sat, 2 Nov 2002 01:01:33 -0500  From: John Santos <JOHN@egh.com>- Subject: Re: Was OpenVMS left out on purpose?X4 Message-ID: <1021102010045.400D-100000@Ives.egh.com>   On 1 Nov 2002, Rob Young wrote:e  W > In article <00A16521.4E63FD58@SendSpamHere.ORG>,   VAXman-  @SendSpamHere.ORG writes:  >  > > K > > BULLSHIT!  Since when is Windows a non-stop platform?  Just because itskJ > > crappy GUI has only a [Start] button doesn't make it *NON-STOP* by any > > stretch of the imagination!c > >  >  > 	ROTFL >  > 			Rob  ; Windows *needs* a START button, but it stops all by itself?    --   John Santos  Evans Griffiths & Hart, Inc. 781-861-0670 ext 539   ------------------------------  % Date: Sat, 02 Nov 2002 01:49:48 -0500I- From: JF Mezei <jfmezei.spamnot@videotron.ca>i- Subject: Re: Was OpenVMS left out on purpose?n, Message-ID: <3DC37577.10A8515F@videotron.ca>   John Santos wrote:= > Windows *needs* a START button, but it stops all by itself?   L No, the START button  is the first step towards the shutdown of the machine.   ------------------------------  $ Date: Fri, 1 Nov 2002 15:29:13 -0600/ From: "Stuart, Ed" <Ed.Stuart@austinenergy.com>n. Subject: Why Is ALTPRI Needed to Set Affinity?T Message-ID: <92EFB80E551BD511B39500D0B7B0CDCC0642CAFF@ohms.electric.ci.austin.tx.us>  K Why should someone have to have elevated privileges to set their process tomJ have affinity with a particular CPU?  If the program is single threaded itH should perform better if the overhead of moving it between processors is eliminated, right?     EdE **Please apply a generous amount of all the usual disclaimers here.**m   ------------------------------   Date: 1 Nov 2002 10:26:35 GMT)7 From: sy18889@rabmbit.famrp.cosm (Bradford J. Hamilton)'J Subject: [OT] - Sub-standard network guys (was: Re: Remote Console access)! Message-ID: <DmumPwz7vbKU@rabbit>e  G Sadly, "sub-standard network guys" seem to be the rule, rather than thepK exception.  I know of a major corporation in which the "network guys" thinkl that  	 "DECKnet"h  B is the correct spelling for a non-industry-standard (and therefore non-supported) network protcol.    In article <rdeininger-3110022322110001@1cust49.tnt1.nashua.nh.da.uu.net>, rdeininger@mindspring.com (Robert Deininger) writes:r <snip>E >>Currently, I've got an old copy of PCM (Polycenter Console Manager)oI >>running. But our network guys think that LAT is a four-letter-word. AndU >>DECserver are evil incarnate.r > G > Sounds like you have sub-standard network guys.  They can't cope withsJ > multiple network protocols?  Can they explain what they don't like about
 > DECservers?  > I > If LAT is a problem, there are TCPIP-based terminal servers.  The laternJ > ones from DEC have this ability (in addition to LAT), and so do the ones& > available now from Digital Networks. > I > I think a multi-port terminal server is the right hardware solution forhD > console connectivity.  A software layer that gives you logging andF > scripting is also very useful.  A web-based or windows-based consoleK > manager is a great waste of time unless it is always up and running.  ThemL > ones I have seen are less reliable and less functional than plain terminal
 > servers. -- q Bradford J. Hamilton& braMdhamAilPtoSn@aMtAtPbi.cSom		(home)& sMy1A88P89S@rabMbit.fAmPr.coSm		(work)  ; "All opinions that I express are my own, not my employer's"  "Lose the MAPS"h   ------------------------------   End of INFO-VAX 2002.605 ************************