1 INFO-VAX	Fri, 02 Mar 2001	Volume 2001 : Issue 121       Contents:1 Advance Server not starting after replacing drive 6 AS 200 4/233 power supply replaced, system won't start: Re: AS 200 4/233 power supply replaced, system won't start6 Re: Compaq to transfer clustering technology to Oracle6 Re: Compaq to transfer clustering technology to Oracle6 Re: Compaq to transfer clustering technology to Oracle6 Re: Compaq to transfer clustering technology to Oracle Re: DCL content (long) Re: DCL content (long) DCPS/Printer Suggestion  RE: DCPS/Printer Suggestion ' Re: Drop C/C++ better sooner than later ' Re: Drop C/C++ better sooner than later ' Re: Drop C/C++ better sooner than later ' Re: Drop C/C++ better sooner than later ' Re: Drop C/C++ better sooner than later ' Re: Drop C/C++ better sooner than later ' Re: Drop C/C++ better sooner than later ' Re: Drop C/C++ better sooner than later ' Re: Drop C/C++ better sooner than later ' Re: Drop C/C++ better sooner than later ' Re: Drop C/C++ better sooner than later ' Re: Drop C/C++ better sooner than later ' Re: Drop C/C++ better sooner than later ' Re: Drop C/C++ better sooner than later ' Re: Drop C/C++ better sooner than later ' Re: Drop C/C++ better sooner than later ' Re: Drop C/C++ better sooner than later ' Re: Drop C/C++ better sooner than later ' Re: Drop C/C++ better sooner than later ' Re: Drop C/C++ better sooner than later ' Re: Drop C/C++ better sooner than later ' Re: Drop C/C++ better sooner than later ' Re: Drop C/C++ better sooner than later ' Re: Drop C/C++ better sooner than later ' Re: Drop C/C++ better sooner than later ' Re: Drop C/C++ better sooner than later ' Re: Drop C/C++ better sooner than later & Re: Future outlook for OpenVMS admins?& Re: Future outlook for OpenVMS admins?& Re: Future outlook for OpenVMS admins?& Re: Future outlook for OpenVMS admins?& Re: Future outlook for OpenVMS admins? Re: Gospel of the Tux 5 How to convert an indexed file to a sequential file?? 9 Re: How to convert an indexed file to a sequential file?? C Re: How to get Polycenter Performance Advisor Graphs to a web page? < Re: How to replace a multi-volme disk set with larger disks?< Re: How to replace a multi-volme disk set with larger disks?< Re: How to replace a multi-volme disk set with larger disks?< Re: How to replace a multi-volme disk set with larger disks? Re: Janitor fixes 90L  Re: Janitor fixes 90L  Re: Janitor fixes 90L  Re: Janitor fixes 90L  RE: Janitor fixes 90L  Re: Janitor fixes 90L  Re: Janitor fixes 90L  Re: Janitor fixes 90L  Re: Janitor fixes 90L # Re: Looking for VI emulator for TPU # Re: Looking for VI emulator for TPU # Re: Looking for VI emulator for TPU # Re: Looking for VI emulator for TPU # Re: Looking for VI emulator for TPU  Re: Low cost workstations  Re: Low cost workstations  Re: Low cost workstations  Re: Low cost workstations  Re: Low cost workstations  Re: Low cost workstations  Re: Low cost workstations  Re: Low cost workstations  Re: Low cost workstations / Re: My name is Dean, and I'm a [Un*x|NT] hater. / Re: My name is Dean, and I'm a [Un*x|NT] hater.  Re: new freeware: SIPU.COM0 Pls help with $PRODUCT INSTALL on this PCSI file  Re: Possible security hole in... PROD Utility Failure' Re: QUESTION: BZIP2 For OpenVMS Anyone? ' Re: QUESTION: BZIP2 For OpenVMS Anyone? % Re: Recommend a VMS directory browser ! Recommend a VMS directory browser 0 Re: So what Are we meant to use on the 'desktop' Re: UCX/TCPIP 5.1 for Hobbyist Re: UCX/TCPIP 5.1 for Hobbyist) Re: VMS 7.2-1 Bugcheck In DECW$REINIT.EXE ) Re: VMS 7.2-1 Bugcheck In DECW$REINIT.EXE  vms programming question Re: vms programming question Re: vms programming question Re: vms programming question3 VMS V7.2+ bug - directory modification date changes  Re: [info] PCSI restriction ! Re: [Request] f$element delimiter   F ----------------------------------------------------------------------  $ Date: Thu, 1 Mar 2001 13:52:30 -0500% From: "Tom Steuver" <steuver@nku.edu> : Subject: Advance Server not starting after replacing drive& Message-ID: <3a9e9ab2_2@news2.one.net>  I AlphaServer 2100, OpenVMS 7.2-1, Advance Server 7.2a, Mutltinet 4.3 rev A   J After a disk crash which contained the Advance Server data files, the diskG was replaced and restored.  Now Advance Server will not start properly. : When I run the pwrk$config procedure, I get the following:  = Checking to see if OpenVMS Registry Services are available... 5 The OpenVMS Registry server has successfully started. 1 %REG-E-DBNOTYETLOADED, database is not yet loaded 3 %REGUTL-W-NOTINREG, known but not found in registry 9 %PWRK-E-NOTMIGRATED, entire lanman.ini file not processed    What can I do to correct this?   Thomas Steuver Servers and Operations Manager Northern Kentucky University   ------------------------------   Date: 1 Mar 2001 20:18:34 -0500 + From: kuhrt@encompasserve.org (Marty Kuhrt) ? Subject: AS 200 4/233 power supply replaced, system won't start 3 Message-ID: <PIBCAt+iOtOi@eisner.encompasserve.org>   B Had a power supply go blammo on one of my AlphaStation 200 4/233s.A Got a replacement through Datamarc, a reseller I use quite often. D Put in the new power supply, and carefully hooked up all the variousE cables.  Powered up, and got the 1-2-1 beep code and the D6 LED code.   " So the question is, what do the...  F Beep code 1-2-1, (1) The DROM code detected a realtime clock (BQ3287) ; failure, or (2) the realtime clock interrupt did not occur.    and   # LED D9, realtime clock test failed.   8 errors mean, with regards to what needs to be fixed now?   ------------------------------  " Date: Fri, 2 Mar 2001 04:38:24 GMT( From: Terry Kennedy <terry@gate.tmk.com>C Subject: Re: AS 200 4/233 power supply replaced, system won't start & Message-ID: <G9JzK0.yq@spcuna.spc.edu>  - Marty Kuhrt <kuhrt@encompasserve.org> writes: $ > So the question is, what do the...: > errors mean, with regards to what needs to be fixed now?  N   You probably need to replace the clock battery (about the size of a quarter,6 on the mainboard). Should be available at Radio Shack.  4         Terry Kennedy             http://www.tmk.com5         terry@tmk.com             Jersey City, NJ USA    ------------------------------   Date: 1 Mar 2001 14:16:38 -0500 + From: young_r@encompasserve.org (Rob Young) ? Subject: Re: Compaq to transfer clustering technology to Oracle 3 Message-ID: <yAsLkOawC6$D@eisner.encompasserve.org>   f In article <iUwn6.365$Up.17662@sea-read.news.verio.net>, "Nikita V. Belenki" <public@kits.net> writes:: > "Rob Young" <young_r@encompasserve.org> wrote in message/ > news:Gt4YPj0rHXjp@eisner.encompasserve.org...  > 2 >> After all, Compaq is very pro NT/W2K was that aF >> big slip-up leaving out NT/W2K?  When pigs fly!!  Think about it...H >> just how easy is it to undo all the 64-bit stuff in the Tru64 DLM and* >> make sure it runs right under W2K/IA32? > 4 > Does it really matter? VMS buyers will finance it. >   > 	Look.... I wouldn't ever state:  "Will never happen".  I wishC 	to remain as slippery as an Enterprise Architect on this one.  But A 	let's suppose it does get funded and a go ahead.  I predict with < 	a great deal of certainty it will have timelines as long or= 	longer than a FRS IA64 CPU!  Hence, my conclusion Winkler is ; 	selling futures.  Point us to something to support another B 	conclusion.  I won't beat you up... maybe we all learn something.1 	But these little jabs of nothingness get tiring.    				Rob    ------------------------------   Date: 1 Mar 2001 15:21:36 -0500 + From: young_r@encompasserve.org (Rob Young) ? Subject: Re: Compaq to transfer clustering technology to Oracle 3 Message-ID: <ZHR$BG81MvSW@eisner.encompasserve.org>   a In article <Gt4YPj0rHXjp@eisner.encompasserve.org>, young_r@encompasserve.org (Rob Young) writes:  >  >>  B >> When I reported Winkler's speech at the time I pointed out thisE >> clustering  technology transfer and several people told me that it G >> wasn't true to the best of their knowledge. Then a month or so later H >> comes the official announcement. I would tend to assume that the restH >> of his speech reflected Compaq policy and the only thing he did wrong( >> was to reveal too much in public. No? >>   > F > 	I suspect Mr. Winkler is selling a good batch of futures.  Betcha!  >   A 	Follow-up.. sure.. W2K and WNT maybe are roadmappable.  In fact, B 	that may be part of the sign-off on getting the Tru64 technology.@ 	The other part of the sign-off that shows that Mr. Capellas andB 	crew are much better negotiators, is the fact that Tru64 precedesE 	other tier-1 platforms by 2 or 3 quarters.  For Larry to tweek Bill, = 	I wonder if the "very difficult"  *backporting* of the Tru64 G 	technology into IA32 W2K/WNT is left until last.  One strong motivator D 	there eh?  Besides the DLM, what about very large memory support?  D 	I understand WNT/W2K IA32 can be hacked up to handle address space @ 	beyond 4 GByte and RDBMS to work that magic too... imagine the A 	backporting of Tru64 to play that funky pointer address stuff.   C 	Yeah.. no wonder *it appears* they are leaving IA32 WNT/W2K until   	last (at all?) ...   5 	But wait until McKinley!  Twice as fast as Merced!!!   < 	But hey, I've checked and can't find a linkage WNT/W2K and ; 	Oracle/Tru64 announcement.  But have found other OS links:   U http://yahoofin.cnet.com/news/0-1003-200-4954031.html?tag=pt.yahoofin.financefeed..ne   P "Oracle will release a new version of its clustering software for Compaq's Tru64J Unix operating system by the end of the year. Then the database giant willI release new versions that support other versions of Unix, including Sun's H Solaris and HP-UX, by the second or third quarter of 2002, Jones said. "  > 	So feel free to hop in a point out where IA32 WNT/W2K fits in
 	all this!     				Rob    ------------------------------  % Date: Thu, 01 Mar 2001 21:32:19 +0000 % From: Alan Greig <a.greig@virgin.net> ? Subject: Re: Compaq to transfer clustering technology to Oracle * Message-ID: <3A9EBFE2.840E305F@virgin.net>   Rob Young wrote:P "Oracle will release a new version of its clustering software for Compaq's Tru64  L > Unix operating system by the end of the year. Then the database giant willK > release new versions that support other versions of Unix, including Sun's J > Solaris and HP-UX, by the second or third quarter of 2002, Jones said. " > G >         So feel free to hop in a point out where IA32 WNT/W2K fits in  >         all this!  >  >   b It fits in because Winkler said (I paraphrase) "Won't this upset Microsoft? No because Compaq willd assign all the engineers it takes to make sure that the Oracle/Compaq clustering code is implemented promptly on Windows."   a I still don't disagree with most of your logic Rob. We'll just have to wait and see whether logic 	 wins out.    --
 Alan Greig   ------------------------------   Date: 1 Mar 2001 23:19:13 -0500 + From: young_r@encompasserve.org (Rob Young) ? Subject: Re: Compaq to transfer clustering technology to Oracle 3 Message-ID: <9qAR$zS0E4$E@eisner.encompasserve.org>   R In article <3A9EBFE2.840E305F@virgin.net>, Alan Greig <a.greig@virgin.net> writes: >  >  > Rob Young wrote:R > "Oracle will release a new version of its clustering software for Compaq's Tru64 > M >> Unix operating system by the end of the year. Then the database giant will L >> release new versions that support other versions of Unix, including Sun'sK >> Solaris and HP-UX, by the second or third quarter of 2002, Jones said. "  >>H >>         So feel free to hop in a point out where IA32 WNT/W2K fits in >>         all this! >> >> >   N It fits in because Winkler said (I paraphrase) "Won't this upset Microsoft? NoK because Compaq will assign all the engineers it takes to make sure that the B Oracle/Compaq clustering code is implemented promptly on Windows."   O I still don't disagree with most of your logic Rob. We'll just have to wait and  see whether logic wins out.     > 	This sounds good doesn't it?  But as we look at "whisper down= 	the lane" in action here we see where some very difficult OS @ 	features were incorporated in Tru64 (DLM and shared filesystem)= 	and not overnight.  Now that special sauce makes it way into F 	Oracle Cluster, such that goofy things like failover and partitioningA 	and function shipping are for "broken" Clusters *until* they too E 	get Oracle Cluster.  Sun is probably the closest as they have gotten + 	a much better Cluster out the door in 3.0. D 	So we see Tru64, then Sun , then number three Unix partner HP/UX...  > 	Now how easy does that fit in the NT world?  Squeeeeeze that > 	Unix paradigm into NT ... ayup.  So in a dramatic move to let? 	the world know that Oracle and NT and Compaq are teaming up we @ 	hear about it a few weeks ago.  Fast forward to this week.  NowC 	what we heard a few weeks ago makes sense looking back.  It really ? 	is about Unix and Oracle "and oh by the way, Oracle and NT get @ 	this too" is what should have been the statement.  So NT Oracle> 	gets to steal the thunder.  But why?  Calm the masses.  We'llB 	see how resources are dedicated and just how quckly the code baseA 	makes it to the NT folks.  Larry might be a little slow at that. E 	We see he has a few Unixes lined up beforehand.  After all, no love   	lost between Larry and Bill.   B 	If it really was a big thing for Oracle Cluster and NT, you wouldE 	have seen big press linkage there wouldn't you?  I don't think Larry = 	is up to promoting NT anymore at all.  Seems that way to me. : 	Why should he?  SQL 7.0 and follow-ons are out to destroyD 	Oracle.  Maybe soon Bill bundles a couple hundred SQL licenses with@ 	NT and "chokes off Larry's air."  The Justice Department would E 	probably look the other way this time.  This whole monopoly business  	is a myth.   @ 	Oh, one other thing.  Winkler mentions promptly.  Make a mental@ 	note of that.  We'll see just how promptly Oracle Cluster shows
 	up on NT.   				Rob    ------------------------------  $ Date: Thu, 1 Mar 2001 19:11:06 -0500  From: John Santos <JOHN@egh.com> Subject: Re: DCL content (long) 6 Message-ID: <1010301190824.35173D-100000@Ives.egh.com>  3 My cluster has 2 Alphas (V7.2-1) and one VAX (V7.1) F ARCH_NAME works on Alpha's (including showing the VAXes architecture),C but returns "" for remote nodes when run on the VAX.  Remote use of  ARCH_NAME must be new in V7.2.    " On Thu, 1 Mar 2001, Syltrem wrote:   > OVMS 7.2-1 >=20, > I don't have another cluster to try it on. > -- >=20 > Sytrem" > http://pages.infinit.net/syltrem >=20> > <steven.reece@quintiles.com> a =E9crit dans le message news:? > OF12B73D42.B8F4E1FB-ON80256A02.00579FC9@qedi.quintiles.com...  > > L > > Which version did you try it on?  It doesn't work on another cluster no= de > > in v7.1-1H2....... > > ; > > Sylterm (syltrem at videotron dot ca) wrote/responded : G > > >>>> Another "wish list" item for me would be to have the ARCH_NAME 	 > keyword I > > > work between cluster nodes. For now, HW_NAME is all you can get for   > > > other than the local node. > >  > > ???  > > & > > $ say f$gets("arch_name","kronos")	 > > Alpha & > > $ say f$gets("arch_name","pallas") > > VAX  > > <<<    --=20  John Santos  Evans Griffiths & Hart, Inc. 781-861-0670 ext 539   ------------------------------  % Date: Thu, 01 Mar 2001 21:37:26 -0600 7 From: "David J. Dachtera" <djesys.nospam@earthlink.net>  Subject: Re: DCL content (long) - Message-ID: <3A9F1576.98CE07D3@earthlink.net>    Syltrem wrote: > J > > Another "wish list" item for me would be to have the ARCH_NAME keywordG > > work between cluster nodes. For now, HW_NAME is all you can get for  > > other than the local node. >  > ???  > $ > $ say f$gets("arch_name","kronos") > Alpha $ > $ say f$gets("arch_name","pallas") > VAX    Thanx! I'll look into that.   ? Now - do you happen to know how to get ALLOCLASS between nodes? 1 (Don't ask why - you don't wanna know (do you?).)    --   David J. Dachtera  dba DJE Systems  http://www.djesys.com/  : Unofficial Affordable OpenVMS Home Page and Message Board: http://www.djesys.com/vms/soho/   F This *IS* an OpenVMS-related newsgroup. So, a certain bias in postings is to be expected.  @ Feel free to exercise your rights of free speech and expression.  F However, attacks against individual posters, or groups of posters, are strongly discouraged.n   ------------------------------  # Date: Thu, 01 Mar 2001 21:19:44 GMTp) From: "J.G. Peters" <jpeters@sctcorp.com>e  Subject: DCPS/Printer Suggestion0 Message-ID: <01c0a295$3470d310$32341895@jpeters>  
 Greetings,  $ OpenVMS Version 7.2 on an Alpha 2100 DCPS Version 1.5  4 We have an LN20 printer, approximately 2 years old -7 standard configuration plus the optional duplexer. It'sy2 never been an, uh, outstanding performer - "hangs"8 fairly often and has to be either soft or hard rebooted.0 Print quality has not been good either and toner6 cartridges quite expensive. At any rate, there's money8 in the budget for a replacement. HP 8000 series printers5 are listed as supported by DCPS 1.8. Does anyone have 6 experience with this series? I am assuming HP 8100 and9 and 8150 series are not presently supported. Any idea if/a3 when they will? Does anyone have a suggestion for ar5 printer under $4000 that will work with DCPS? Any/alle* input is greatly appreciated. Thanks. ~Joe   ------------------------------  % Date: Thu, 01 Mar 2001 22:12:47 -0500+- From: "Mitchell, David R." <mitchell@WPI.EDU>e$ Subject: RE: DCPS/Printer SuggestionH Message-ID: <FE1835D68492D311BF7900508B5BEB0D2CA30B@petra.admin.wpi.edu>  A We use the 8100's here with DCPS 1.8 which seem to work fine in = 
 "unsupported"pG mode.  You wont have much in the way of tray selection in unsupported =  modeE without doing a few tweaks to a device control library.  The last I = 	 heard the G 8100 was on a list of printers to be supported in the next release of =a DCPS.eJ As the 8150's are relatively new I don't know if they're due for support = in the next release or not.   Davidm   > -----Original Message-----0 > From: J.G. Peters [mailto:jpeters@sctcorp.com]( > Sent: Thursday, March 01, 2001 4:20 PM > To: Info-VAX@Mvb.Saic.ComV" > Subject: DCPS/Printer Suggestion >=20 >=20 > Greetings, >=20& > OpenVMS Version 7.2 on an Alpha 2100 > DCPS Version 1.5 >=206 > We have an LN20 printer, approximately 2 years old -9 > standard configuration plus the optional duplexer. It'sv4 > never been an, uh, outstanding performer - "hangs": > fairly often and has to be either soft or hard rebooted.2 > Print quality has not been good either and toner8 > cartridges quite expensive. At any rate, there's money: > in the budget for a replacement. HP 8000 series printers7 > are listed as supported by DCPS 1.8. Does anyone havee8 > experience with this series? I am assuming HP 8100 and; > and 8150 series are not presently supported. Any idea if/o5 > when they will? Does anyone have a suggestion for ao7 > printer under $4000 that will work with DCPS? Any/allo, > input is greatly appreciated. Thanks. ~Joe >=20 >=20 >=20   ------------------------------  # Date: Thu, 01 Mar 2001 18:53:01 GMT + From: "Nikita V. Belenki" <public@kits.net>i0 Subject: Re: Drop C/C++ better sooner than later8 Message-ID: <hUwn6.364$Up.17662@sea-read.news.verio.net>  : "Bob Koehler" <koehler@encompasserve.org> wrote in message- news:Y$aqbQdRtYFw@eisner.encompasserve.org...   E >    The following simple little C program has a security hole a milei >    wide in it.J >    I know I could plug the hole.  I also know a lot of C programmers who9 >    can't see the hole and have no idea that it's there.m  1 I don't think you should call them "programmers".r  G >    I know I could massage this to get through a C++ compiler, so this # >    critisism also applies to C++.o  J Sure. But why don't you use std::string? It is much easier to program with it.E  J >    I know I need a lot of technical information to take advantage of the
 >    hole.H >    Most importantantly I know of no other language in which this is so8 >    easy to do, or so hard to see.  Which is the point.  F Why? *I* can easily put a security hole into the program in almost any language ;)-   Kit.   ------------------------------  % Date: Thu, 01 Mar 2001 11:24:03 -0800r! From: Shane.F.Smith@Healthnet.com 0 Subject: Re: Drop C/C++ better sooner than laterD Message-ID: <OFFAFA898D.FD99694B-ON88256A02.006A0760@foundation.com>  F AFAICR, I have never had to resort to pointer arithmetic, in C or C++.G Maybe I've just been lucky in the kinds of projects I've had to do, butaJ it's never seemed necessary. It's not that I've been particularly avoiding it, either.l  D Genuine curiosity here: could one of you give me an example of where( pointer arithmetic is necessary, please?   Shanet          = Christof Brass <brass@infopuls.com> on 02/28/2001 06:47:55 PMR   To:   Info-VAX@Mvb.Saic.Coms cc:   1 Subject:  Re: Drop C/C++ better sooner than laters     JF Mezei wrote:  >c > Rick Cadruvi wrote:SG > > it because I don't use it.  When I do need to I always type cast or- declare-
 > > thingsI > > as (char *).  Anyone who faults C based on the way pointer arithmeticM is > > doneJ > > and the excessive type casting it causes will get NO argument from me. >PK > Whenever I am not sure of what it actually does, I end up writing a small G > "test.c" program that does the actual operation and I then either usee debug or@ > lots of printfs to verify that the compiler generate the right arithmetic.r >sB > This is the same with bit fields/arithmetic or other constructs. > F > What I don't like about DEC-C is that it sometimes forces me to make strangecI > constructs to shut the compiler up even though the "illegal" constructst wouldt > be much easier to understand.i >cK > I despise having to typecast. There are times where I will move a pointer? to aD > void and then to a char just to make sure I can do real arithmetic without then > compiler complaining.  >jD > When working with dynamically allocated memory structures, pointer
 arithmetic > and handling is a must.a   *Only* in C/C++.   ------------------------------   Date: 1 Mar 2001 15:23:41 -0500o9 From: Kilgallen@eisner.decus.org.nospam (Larry Kilgallen)o0 Subject: Re: Drop C/C++ better sooner than later3 Message-ID: <oAQASDCMbntd@eisner.encompasserve.org>e  h In article <OFFAFA898D.FD99694B-ON88256A02.006A0760@foundation.com>, Shane.F.Smith@Healthnet.com writes: > H > AFAICR, I have never had to resort to pointer arithmetic, in C or C++.I > Maybe I've just been lucky in the kinds of projects I've had to do, butcL > it's never seemed necessary. It's not that I've been particularly avoiding
 > it, either.e > F > Genuine curiosity here: could one of you give me an example of where* > pointer arithmetic is necessary, please?  C I don't know about _necessary_, but some people will take a pointersD to a string and successively increment it to point to each character of the string.  F In languages that don't support pointer arithmetic that is impossible.F According to the definition of those languages, it might not even workC if it were possible.  That is why Ada has "access variables" ratheriC than pointers.  There is no guarantee that it is represented as thenC address of anything, much less the object for which it is an accessaF variable.  It could be an index into a table of addresses, or anythingH else the compiler writer fancied.  The implementation is _not_ something  on which the programmer depends.   ------------------------------   Date: 1 Mar 2001 11:14:47 CDTg= From: wayne@tachysoft.xxx.344932.killspam.0142 (Wayne Sewell)o0 Subject: Re: Drop C/C++ better sooner than later. Message-ID: <CHtLZwhwOeSS@tachxxsoftxxconsult>  [ In article <3A9D6097.EDEDADF@videotron.ca>, JF Mezei <jfmezei.spamnot@videotron.ca> writes:r > I > Similarly, using C to call system services on VMS is somewhat of a pains* > because descriptors are not native to C.  M You got that right.  zero terminated strings basically suck, especially sincesJ they are impossible to use without overhead.  How can you possibly get the9 length of a string without scanning the whole damn thing?s  2 I'm not really that wild about descriptors either.  M I personally think that DEC Pascal has the best approach.  It's essentially arJ counted string with a 16-bit length.  It's built into the language, so youN don't have to access the length directly.  It is set automatically by a stringN assignment.  Most of the time you can just forget about the length, as long asM the *allocated* size of the string is always adequate to contain the *actual*n= size.  Which of course one has to do with any type of string.t   -- -O ===============================================================================tM Wayne Sewell, Tachyon Software Consulting  (281)812-0738  wayne@tachysoft.xxx : http://www.tachysoft.xxx/www/tachyon.html and wayne.html  K change .xxx to .com in addresses above, assuming you are not a spambot  :-) O =============================================================================== O Dean Wormer to Flounder: "Fat, drunk, and stupid is no way to go through life."n   ------------------------------  % Date: Thu, 01 Mar 2001 16:45:54 -0500n- From: JF Mezei <jfmezei.spamnot@videotron.ca> 0 Subject: Re: Drop C/C++ better sooner than later, Message-ID: <3A9EC310.917EC996@videotron.ca>   Bill Gunshannon wrote:J > Why would you not trust one compiler any more than another??  This makesJ > absolutely no sense.  Are you saying that just because the language is CH > the guys in the DEC Compiler Engineering Group are somehow sloppier or > more careless??   M No, not at all. I mistrust the compiler's interpretation of my wishes. In therM case of ansi-C, I very much dislike the inability to use the & symbol on someiM constructs because it then makes it less obvious that you really want to passs
 a pointer.  J Similarly, pointer arithmetic is something I often check when dealing withE non-char constructs to make sure that the + 1 does what I reall want.   M I prefer using a printf statement to test how the compiler has interpreted my-I wishes than trying to figure out with the manuals what the precedence and<K conversions rules are prior to compiling. Some may call it trial-and-error,o) and otherts mightcall it quality control.r   ------------------------------  # Date: Thu, 01 Mar 2001 21:39:58 GMTl= From: system@SendSpamHere.ORG (Brian Schenkenberger, VAXman-) 0 Subject: Re: Drop C/C++ better sooner than later0 Message-ID: <009F85F6.B77F5FFD@SendSpamHere.ORG>  n In article <CHtLZwhwOeSS@tachxxsoftxxconsult>, wayne@tachysoft.xxx.344932.killspam.0142 (Wayne Sewell) writes:\ >In article <3A9D6097.EDEDADF@videotron.ca>, JF Mezei <jfmezei.spamnot@videotron.ca> writes: >>  J >> Similarly, using C to call system services on VMS is somewhat of a pain+ >> because descriptors are not native to C.g >eN >You got that right.  zero terminated strings basically suck, especially sinceK >they are impossible to use without overhead.  How can you possibly get the : >length of a string without scanning the whole damn thing? >-3 >I'm not really that wild about descriptors either.e  L People tend to forget that descriptors can be used for more than just stringL data.  The descriptor is sort of an object definition.  For a string object,K the descriptor tells us the size of the data, the location of the data, andnL the type and class of the data.  For a string, its descriptor is rather sim-M ple but take a look at something more complex like an n-dimentional array de- L scriptor and you begin to appreciate its usefulness.  It's also a common wayL to pass object information about in the mixed language environment -- one of really nice things about VMS.t   --O VAXman- OpenVMS APE certification number: AAA-0001     VAXman(at)TMESIS(dot)COMi             O city, n., 1. a place where trees are cut down and streets are named after them.e   ------------------------------  % Date: Thu, 01 Mar 2001 16:58:45 -0500q- From: JF Mezei <jfmezei.spamnot@videotron.ca> 0 Subject: Re: Drop C/C++ better sooner than later, Message-ID: <3A9EC613.386FA2CB@videotron.ca>   Jan Vorbrueggen wrote:K > The PASS (primary avionics software system, or somesuch) is definitely in N > HAL/S, and is the only one that has ever flown the thing in practice. As theL > BFS (backup flight system, or somesuch) was available for the first flightK > twenty years ago, I doubt it is in Ada, and it has never flown a shuttle. B > (They were one failure away once on an early flight, I believe.)  G There was a failure a few flights ago. (It was one flight with a femaleeL commander, one or two years ago). That failure prompted the grounding of the1 shuttle fleet while all the wiring was inspected.   K The architecture worked as designed, with the shuttle surviving the loss of  part of its systems.   ------------------------------  % Date: Thu, 01 Mar 2001 17:37:20 -0500.- From: JF Mezei <jfmezei.spamnot@videotron.ca> 0 Subject: Re: Drop C/C++ better sooner than later, Message-ID: <3A9ECF1B.50C19A15@videotron.ca>  " Shane.F.Smith@Healthnet.com wrote:F > Genuine curiosity here: could one of you give me an example of where* > pointer arithmetic is necessary, please?  J Perhaps not "necessary", but for building and decoding variable structuresI such as image files with variable length headers consisting of a variableeT number of variable size index entries, pointers arithmetic makes things much easier.  L When the data that follows is determined by the data that precedes, youcan'tL really use predefined structures and must calculate your pointer position on the fly as you parse the data.    " Another example, not related to C:  K  In a previous life, I had the pleasure of writing the program that printedl the payroll cheques.M This was done in COBOL.  The cheques had to be printed in the language of theb employee (french or english).n  N When the time came to generate the textual representation of the amount, I wasJ given permission to "escape" from cobol and write the routine in assemblerN because it would have been just too ackward to do it in Cobol. (Cobol sucks atL concatenating variable length strings). (This was in IBM 370 assembler). TheL routine filled a cobol fixed width string (padded with trailing blanks) with@ the right text based on the amount and language of the employee.  H It made the Cobol program mush simpler to read. And by having a separateG module to generate the textual representation of the amount, it allowedpM testing of various combinations of numbers (french grammar is quite picky foreH numbers, sometimes requiring a "-", sometimes requiring an "s" etc etc.)  N Asked to do the same today, I would probably write the assembler routine in C.  ; For writing reports, COBOL's report writer was pretty good.f   ------------------------------  % Date: Thu, 01 Mar 2001 23:11:32 +0000s) From: Christof Brass <brass@infopuls.com>20 Subject: Re: Drop C/C++ better sooner than later, Message-ID: <3A9ED724.3D2CC671@infopuls.com>   Rick Cadruvi wrote:e > 8 > I'm not sure what the point is with size of compilers.? > Most of the size of the code for a compiler isn't necessarily B > related to the complexity of the language.  Compilers use tablesA > for parsing/lexical analysis.  The tricky part of compilers andt3 > what tends to add size is the code optimizations.  > = > The more a language does checking of syntax and constructs,h> > the more it relies on data it gets from the sources/includes > (hidden or otherwise). > A > Compiler size has very little if anything to do with complexity0D > of the language.  Run-time support may.  The best thing to look atB > would be image size for similar tasks having the runtime supportA > linked in without using shareable images.  Better yet, keep thesA > task to something done without using libraries and see how muchk5 > extra stuff the compiler throws in the final image.  > 	 > Rick...t > A > "Robert Deininger" <rdeininger@mindspring.com> wrote in messagerH > news:rdeininger-0103011208040001@user-2ive7nq.dialup.mindspring.com...L > > In article <y44rxdfro3.fsf@mailhost.neuroinformatik.ruhr-uni-bochum.de>,L > > Jan Vorbrueggen <jan@mailhost.neuroinformatik.ruhr-uni-bochum.de> wrote: > >c9 > > > bill@triangle.cs.uofs.edu (Bill Gunshannon) writes:i > > >rA > > > > I would be more likely to mistrustt he Ada compiler basedmE > > > > solely on the several orders of magnitude more complex it is.i > > >nN > > > Is somebody willing and able to post a DIR/SIZE of ADA.EXE, DECC.EXE and > > > DECXX.EXE? > >a > >s& > > ADA.EXE;1              19590/19593% > > ACS.EXE;1               4072/4077  > >-& > > CXX$COMPILER.EXE;3     16805/16857$ > > CXX$DEMANGLE.EXE;4        30/144$ > > CXX$LINK.EXE;3            87/144 > >D > > DECC$COMPILER.EXE;1e& > >                        12517/12528 > >>% > > FORT$MAIN.EXE;1         9126/9126n > >l& > > F90$MAIN.EXE;1         14025/14031 > >  > > All somewhat out of date...d > >t > > ACS> show vern; > > DEC Ada V3.4-2 with the Professional Development Optionn > >p
 > > $ cxx/verg0 > > Compaq C++ V6.2-035 for OpenVMS Alpha V7.1-2 > >o > > $ cc/ver* > > DEC C V5.7-004 on OpenVMS Alpha V7.1-2 > >g% > > DIGITAL Fortran 77 X7.1-156-337CGs > >y% > > DIGITAL Fortran 90 X7.1-574-337CG0 > >rM > > In particular, I'm pretty sure that C++ has grown significatly since thise > > one was current. > >u > >HB > > By this measure, "orders of magnitude" doesn't look realistic. > >h > > -- > > Robert Deininger > > rdeininger@mindspring.comn  The complexity of a PL can be at different places. Syntax can be complex, i.e. not very regular which leads to complexer and bigger scanning/parsing routines. The semantics of constructs can be complex this adds to the complexity and size of the high level code generation. High level optimisation is complexer if the language offers more powerful constructs. Complexity of the PL contributes substantially to the size of the compiler.n  But one effect wasn't taken into account so far: if the compiler is written in its language the quality/attributes of the language have a double effect on the compiler: the more powerful the language is the more elegant the compiler can be constructed, and by that limiting the effect on increasing the compiler size. The better the language is in terms of safety and architectural quality the better the compiler will be.  Q This double effect is one main reason why C and C++ compilers are pieces of shit.    ------------------------------  $ Date: Thu, 1 Mar 2001 18:18:57 -0500  From: John Santos <JOHN@egh.com>0 Subject: Re: Drop C/C++ better sooner than later6 Message-ID: <1010301181655.35173C-100000@Ives.egh.com>  % On 1 Mar 2001, Bill Gunshannon wrote:i  . > In article <3A9D58D3.92B3B87A@videotron.ca>,2 >  JF Mezei <jfmezei.spamnot@videotron.ca> writes: > |> Bob Koehler wrote:dH > |> > Really?  Then why is it I've got tons of Java with massive use ofO > |> > dynamically allocated memory structures working quite well in a languageD$ > |> > which does not have pointers? > |>  = > |> Is Java a compiled language, or a run-time environment ?o > J > Java is a compiled language.  The Java Virtual Machine has nothing to doJ > with wether or not this is true.  Unless you somhow believe that becauseJ > it ran in the P-machine somehow UCSD Pascal was not a compiled language.M > Yes, Java, Byte Code and the JVM are just the UCSD Pascal idea re-invented.n  ' Actually, it is BASIC-PLUS re-invented./   > D > |>                                            Doesn't Java executeL > |> in its own virtual machine with lots of support from the Java engine ?  > O > It does.  And for political reasons Sun wanted to keep it this way.  But theyrN > lost.  GNU has a native mode java compiler.  But even that is not germane toO > the argument.  The Java Virtual Machine is an architecture just like any real N > machine.  Byte Code is it's machine language.  Using your argument, when youO > run any compiler on the CHARON emulator they are not really compilers becauseiP > they don't generate object code for the host machine and they run in a virtual
 > machine. > N > |>                                                                      Does/ > |> this qualify as a true compiled language ?y >  > As much as any other.i >  > bill >  > --  L > Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesF > bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. > University of Scranton   |C > Scranton, Pennsylvania   |         #include <std.disclaimer.h>   l >  >    -- j John Santos  Evans Griffiths & Hart, Inc. 781-861-0670 ext 539   ------------------------------  % Date: Thu, 01 Mar 2001 23:28:10 +0000d) From: Christof Brass <brass@infopuls.com>+0 Subject: Re: Drop C/C++ better sooner than later, Message-ID: <3A9EDB0A.ADC432DC@infopuls.com>   Rick Cadruvi wrote:t > M > > We get used to a certain syntax in a certain context. Just as DCL makes a   > > difference between = and ==. > A > One of the nice things about BLISS (eql, neq, lss, gtr, etc...)f > ; > One of the BAD things about C is all the coding shortcutsn% > (like (?:) and pre/post decrement).  > F > It seems to me that this discussion has pointed out that much of ourC > individual philosophies about programming enters into how we feellE > about language features along with past/present frustrations with afC > language.  I also suspect that how much experience we have with ah6 > language makes a difference in how we feel about it. > K > It would be interesting to have a discussion about programming philosophy D > and see how the Ada versus C/C++ discussion correlates to people's8 > philosophy about how to write code in the first place. >  > Rick Cadruvi...e  Q I explicitly said "educated person" as who I don't regard your 9 year old nephew.t  There *are* better and worse solutions for syntactic problems. One major problem was that the limited set of characters didn't provide enough symbols to come to a good solution. This is another reason why a PL has to evolve. Nowadays we should be able to use all necessary symbols as one 8 bit or unicode character by that eliminating a bunch of problems and discussions (e.g. neq vs <> vs != vs #).i   I'm personally convinced that *chosing* (as opposed to beeing forced to use) a PL says a lot about style and attitude of the programmer. Like it is chosing an OS. My experience is that most of C/C++ isn't worth the time it had needed to be written and isn't worth to be read by other people. The common C/C++ programmer is challanged and satisfied by mastering a clumsy syntax with error prone semantical constructs and by thinking to have total control about the compiler's code generation by using so called lo   The spirit of VMS is related to the spirit of Ada or MODULA-3 (a very high quality DEC language) i.e. clearness of thoughts expressed and reflected with the appropriate tools like DCL and TPU as opposed to csh or sed.e  YAdditionally to this philosophical point of view we have the brutal statistics telling us that C/C++ is the loser in productivity and quality like UNIX. Or PC compared to Mac. But I'm sure the majority of people (managers, sysadmins, programmers) is fond of wasting their and others time and feeling superior in mastering the mission impossible.d   ------------------------------  % Date: Thu, 01 Mar 2001 23:38:13 +0000e) From: Christof Brass <brass@infopuls.com>i0 Subject: Re: Drop C/C++ better sooner than later, Message-ID: <3A9EDD65.BD1B9AD3@infopuls.com>   Rick Cadruvi wrote:e > D > True.  I am always amazed at the failure of programmers, even when > instructedE > to isolate OS depencenies, to understand what things are dependent.o > L > Besides writing code that runs on several platforms, try writing code that > willE > work in USER and KERNEL mode and all of a sudden the liberal use ofsE > C runtime libraries goes away.  This actually points out one of thet > strengths ofM > a language like C or BLISS.  They don't have a lot of builtin features.  As> > aML > result, you can do a LOT, but you have to write more code to do so than inD > other languages.  That's why C is not well suited for applications > development,$ > but quite suited for Systems work. >  > Rick Cadruvi...y > N > "Jan Vorbrueggen" <jan@mailhost.neuroinformatik.ruhr-uni-bochum.de> wrote inL > message news:y4n1b5zv5c.fsf@mailhost.neuroinformatik.ruhr-uni-bochum.de.../ > > Christof Brass <brass@infopuls.com> writes:e > > Q > > > Secondly, portability is a no issue. If you strive for portability you willnL > > > have to solve a lot and very different problems than using a/the C/C++ > > > library. > >nP > > Nonsense. Partablility in all its colours is _the_ issue. Why did IBM inventJ > > S/370? To achieve portability. Why were "higher" programming langaugesP > > invented? To achieve portability. Why were OSes invented? X windows? Tcl/Tk? > > Posix? and so on...  > >gP > > I can tell you from personal experince that if you haven't made at least oneP > > application run on about half a dozen platforms, you haven't had a chance toR > > understand what your programming language (we're talking C, C++, Fortran, Ada,R > > Pascal, ... here) really means, what is part of the standardized semantics and4 > > what is implementation dependent. Very sobering. > >  > > Janr     Been there, done that, sigh.   Sorry, both of you deliver arguments in favour of my statement which leads to the conclusion that both of you didn't understand the implications of what I wrote.u   Portability is nothing in favour or against most PLs!! That's why it is an no issue wrt our discussion. Of course, I know we need the language to be there or to have a cross development system to be able to port a program. But there are hundreds ofZ languages available of dozens of OSs and there is no portability excuse for chosing C/C++.   ------------------------------  % Date: Thu, 01 Mar 2001 23:40:28 +0000a) From: Christof Brass <brass@infopuls.com> 0 Subject: Re: Drop C/C++ better sooner than later, Message-ID: <3A9EDDEC.845CA702@infopuls.com>   "Nikita V. Belenki" wrote: > 8 > "Christof Brass" <brass@infopuls.com> wrote in message( > news:3A9DB85B.5B8D9B84@infopuls.com... > H > > > When working with dynamically allocated memory structures, pointer( > > > arithmetic and handling is a must. > > *Only* in C/C++. > M > *Everywhere* on the current hardware. If *you* are writing the library thatcM > does memory management and hides pointer arithmetic from the other parties.r >  > Kit.   *Only* in crap languages. BTW this has nothing to do with the HW. Instead with the language, namely the compiler/interpreter. Ever heard of a type system or a GC?   ------------------------------  % Date: Thu, 01 Mar 2001 23:42:07 +00009) From: Christof Brass <brass@infopuls.com>c0 Subject: Re: Drop C/C++ better sooner than later, Message-ID: <3A9EDE4F.FE275901@infopuls.com>   Jan Vorbrueggen wrote: > 1 > JF Mezei <jfmezei.spamnot@videotron.ca> writes:d > Q > > Is Java a compiled language, or a run-time environment ? Doesn't Java executehO > > in its own virtual machine with lots of support from the Java engine ? Doesh. > > this qualify as a true compiled language ? >  > Yes. > 
 >         Janb  8 While the answer might be correct the question is wrong.   ------------------------------   Date: 01 Mar 2001 23:47:58 GMT& From: "Rick Cadruvi" <rick@rdperf.com>0 Subject: Re: Drop C/C++ better sooner than later0 Message-ID: <97mn3e$kva@dispatch.concentric.net>  K I think we finally get to the point.  You must not work for others.  I havet met fewuJ programmers/situations where the choice could be made by the programmer as! to language, OS, and other tools.r  D As to worth, I suspect that the $$ value of C/C++ code significantly eclipses ANYK other language.  Take the value of MS products and UNIX products written ins itG alone and I think my case is made.  Some of us have to live in the realn world.   Rick Cadruvi...   I > I'm personally convinced that *chosing* (as opposed to beeing forced to K use) a PL says a lot about style and attitude of the programmer. Like it isaJ chosing an OS. My experience is that most of C/C++ isn't worth the time itH had needed to be written and isn't worth to be read by other people. TheI common C/C++ programmer is challanged and satisfied by mastering a clumsysK syntax with error prone semantical constructs and by thinking to have totalSI control about the compiler's code generation by using so called low level,I "optimisations". The C/C++ programmers fits to the UNIX guru - they shareF the same philosophy. >oL > The spirit of VMS is related to the spirit of Ada or MODULA-3 (a very highH quality DEC language) i.e. clearness of thoughts expressed and reflectedE with the appropriate tools like DCL and TPU as opposed to csh or sed.l >tE > Additionally to this philosophical point of view we have the brutalgI statistics telling us that C/C++ is the loser in productivity and quality E like UNIX. Or PC compared to Mac. But I'm sure the majority of people K (managers, sysadmins, programmers) is fond of wasting their and others time-9 and feeling superior in mastering the mission impossible.F   ------------------------------  % Date: Fri, 02 Mar 2001 00:13:29 +0000e) From: Christof Brass <brass@infopuls.com>50 Subject: Re: Drop C/C++ better sooner than later, Message-ID: <3A9EE5A9.8964DD5F@infopuls.com>  & "Brian Schenkenberger, VAXman-" wrote: > p > In article <CHtLZwhwOeSS@tachxxsoftxxconsult>, wayne@tachysoft.xxx.344932.killspam.0142 (Wayne Sewell) writes:^ > >In article <3A9D6097.EDEDADF@videotron.ca>, JF Mezei <jfmezei.spamnot@videotron.ca> writes: > >>L > >> Similarly, using C to call system services on VMS is somewhat of a pain- > >> because descriptors are not native to C.c > > P > >You got that right.  zero terminated strings basically suck, especially sinceM > >they are impossible to use without overhead.  How can you possibly get thev< > >length of a string without scanning the whole damn thing? > > 5 > >I'm not really that wild about descriptors either.u >    Counted strings have the advantage that they can even contain null characters although this might not be very helpful. Anyway these sorts of simple strings like in C or Standard Pascal are obsolete; a string class is the way to go. The DEC Pascal seems toj offer something like that with automatic size management. I don't remember: were there no problems with too short declared strings? E.g. if you declare a string of a certain length (if ever possible), initialise it with an appropriate value (not too long)f and then append something which will result in a longer string than in the originally declared would fit. Would this lead to a string overflow runtime error or to a silent expansion of the string? Is the space for the strings dynamically managed andr
 reclaimed?  N > People tend to forget that descriptors can be used for more than just stringN > data.  The descriptor is sort of an object definition.  For a string object,M > the descriptor tells us the size of the data, the location of the data, andeN > the type and class of the data.  For a string, its descriptor is rather sim-O > ple but take a look at something more complex like an n-dimentional array de-iN > scriptor and you begin to appreciate its usefulness.  It's also a common wayN > to pass object information about in the mixed language environment -- one of > really nice things about VMS.    Bingo!   >  > --Q > VAXman- OpenVMS APE certification number: AAA-0001     VAXman(at)TMESIS(dot)COMt > Q > city, n., 1. a place where trees are cut down and streets are named after them.c   ------------------------------  % Date: Fri, 02 Mar 2001 00:16:27 +0000o) From: Christof Brass <brass@infopuls.com>i0 Subject: Re: Drop C/C++ better sooner than later, Message-ID: <3A9EE65B.E54BE6EB@infopuls.com>  & paddy.o'brien@zzz.tg.nsw.gov.au wrote: >  > Kit wrote, > 9 > >"Christof Brass" <brass@infopuls.com> wrote in messagei) > >news:3A9DB85B.5B8D9B84@infopuls.com...t > >eI > >> > When working with dynamically allocated memory structures, pointerm
 > >arithmetict > >> > and handling is a must. > >> *Only* in C/C++.k > > N > >*Everywhere* on the current hardware. If *you* are writing the library thatN > >does memory management and hides pointer arithmetic from the other parties. >  > Really a reply to Christof.r  F Is this irony? :-) Please explain after reading my reply to the above.  I > I may have misread what you are saying, but I would hazard a guess thatuL > VAXman and a few others here could do the Full Monty in macro (even lycra)L > without twitching a G-string.  (It's that bloody Sydney Mardi Gras weekend4 > coming up -- what a thing to be world famous for). > L > All languages can be simulated at a lower level, the ultimate is assemblerL > (macro).  All languages have a "depth" purpose and C tries to get as closeC > to the bone as possible.  Many of us do not have a need for that.  > H > Regarding foot-shooting: read Andrew Koenig's (spelling?) "C traps andK > Pitfalls."  He was a Ma Bell worker with similar renown to the other guys1! > whose names are bandied around.  >  > Regards, Paddy >  > Paddy O'Brien, > Transmission Development,> > TransGrid, > PO Box A1000, Sydney South,s > NSW 2000, Australiao >  > Tel:   +61 2 9284-3063 > Fax:   +61 2 9284-3050( > Email: paddy.o'brien@zzz.tg.nsw.gov.au > G > Either "\'" or "\s" (to escape the apostrophe) seems to work for most5	 > people, = > but that little whizz-bang apostrophe gives me little spam.u   ------------------------------   Date: 02 Mar 2001 00:15:47 GMT& From: "Rick Cadruvi" <rick@rdperf.com>0 Subject: Re: Drop C/C++ better sooner than later0 Message-ID: <97monj$l0g@dispatch.concentric.net>  F I am not sure this qualifies as necessary and I suspect with some well designed: data structures, you could get around this, but here goes.  E Complex file structures such as ISAM files that have LOTs of variablec internalF data structures that are pointed to or whose location depends on other
 variables.  L I am sure I could go through such structures using languages without pointerK arithmetic, but I don't think it adds anything to a project to have to fool a C language into doing something that can easilty be done with pointere arithmetic.uE Sometimes code complexity makes software less reliable because we tryc1 to get around using something that can be abused.   G It's like the earlier discussion of GOTOs.  We eliminated them from our G vocabularly because they were badly abused.  However, used with extremetK care, they have their place.  I haven't used one in a long time (when did If lastK do VAX BASIC code?), but I know I have spent a fair amount of time thinking-F hard, every blue moon or so, about how to get around using them.  Many thingsI I do would be even harder without pointer arithmetic.  I mostly avoid it,  but3G it does help sometimes with complex data elements that I may or may NOTo have control over.     Rick Cadruvi..    . <Shane.F.Smith@Healthnet.com> wrote in message> news:OFFAFA898D.FD99694B-ON88256A02.006A0760@foundation.com... >iH > AFAICR, I have never had to resort to pointer arithmetic, in C or C++.I > Maybe I've just been lucky in the kinds of projects I've had to do, butoL > it's never seemed necessary. It's not that I've been particularly avoiding
 > it, either.  >aF > Genuine curiosity here: could one of you give me an example of where* > pointer arithmetic is necessary, please? >s > Shanei >  >t >c >e > ? > Christof Brass <brass@infopuls.com> on 02/28/2001 06:47:55 PMn >g > To:   Info-VAX@Mvb.Saic.Comt > cc:a >j3 > Subject:  Re: Drop C/C++ better sooner than latert >  >I > JF Mezei wrote:/ > >  > > Rick Cadruvi wrote: I > > > it because I don't use it.  When I do need to I always type cast orn	 > declare  > > > thingsK > > > as (char *).  Anyone who faults C based on the way pointer arithmetic  > is
 > > > doneL > > > and the excessive type casting it causes will get NO argument from me. > >mG > > Whenever I am not sure of what it actually does, I end up writing a  smallyI > > "test.c" program that does the actual operation and I then either uset
 > debug orB > > lots of printfs to verify that the compiler generate the right
 > arithmetic.  > >dD > > This is the same with bit fields/arithmetic or other constructs. > > H > > What I don't like about DEC-C is that it sometimes forces me to make	 > strange K > > constructs to shut the compiler up even though the "illegal" constructsy > would ! > > be much easier to understand.h > > E > > I despise having to typecast. There are times where I will move ac pointeru > to aF > > void and then to a char just to make sure I can do real arithmetic
 > without then > > compiler complaining.d > >dF > > When working with dynamically allocated memory structures, pointer > arithmetic > > and handling is a must.- >- > *Only* in C/C++. >  >, >  >    ------------------------------  $ Date: Thu, 1 Mar 2001 19:40:21 -0500' From: "Bill Todd" <billtodd@foo.mv.com> 0 Subject: Re: Drop C/C++ better sooner than later( Message-ID: <97mpqp$f3o$1@pyrite.mv.net>  / Rick Cadruvi <rick@rdperf.com> wrote in message * news:97monj$l0g@dispatch.concentric.net...H > I am not sure this qualifies as necessary and I suspect with some well
 > designed< > data structures, you could get around this, but here goes. >aG > Complex file structures such as ISAM files that have LOTs of variableo
 > internalH > data structures that are pointed to or whose location depends on other > variables. >uF > I am sure I could go through such structures using languages without pointeraH > arithmetic, but I don't think it adds anything to a project to have to fool > anE > language into doing something that can easilty be done with pointeri
 > arithmetic.mG > Sometimes code complexity makes software less reliable because we trys3 > to get around using something that can be abused.  > I > It's like the earlier discussion of GOTOs.  We eliminated them from our I > vocabularly because they were badly abused.  However, used with extremerK > care, they have their place.  I haven't used one in a long time (when dide If > lastD > do VAX BASIC code?), but I know I have spent a fair amount of time thinkingH > hard, every blue moon or so, about how to get around using them.  Many > thingsK > I do would be even harder without pointer arithmetic.  I mostly avoid it,d > butsI > it does help sometimes with complex data elements that I may or may NOTg > have control over.  I Even when I have control over the on-disk data structures, I'd rather useLH structures that efficiently support the task at hand than use structuresL that a language (whatever language) happens to handle conveniently.  I guessJ it's the (or at least a) difference between the view of someone who writesG system or other 'grunt' code that will be executed zillions of times by J countless applications and that of an application developer - but even forE applications it annoys me when someone suggests that one should avoidiI variable-size constructs (when they're clearly the most effective on-disk K representation) because they feel they're awkward to handle in the languageiH (again, whatever language) - a sentiment I usually don't agree with, butK then I learned on assembler which tends to support (or not to support) most  data structures fairly equally.t   - bill   >p >l > Rick Cadruvi..   ------------------------------  % Date: Fri, 02 Mar 2001 00:04:25 +0000e) From: Christof Brass <brass@infopuls.com> 0 Subject: Re: Drop C/C++ better sooner than later, Message-ID: <3A9EE389.16D94B57@infopuls.com>   Jan Vorbrueggen wrote: > 1 > JF Mezei <jfmezei.spamnot@videotron.ca> writes:- > O > > In that case, the language best suited is the one that interfaces best withs > > those routines.e > H > As they all compile down to the same machine code, it is a quality-of-G > implementation issue whether this is a problem or not. VMS has alwaysoG > made interoperablity of languages easy; I was very surprised to learnR: > that almost all other systems make this quite difficult. > M > > If those routines were written with C formats for strings etc, then usingaP > > Pascal ill just make your life harder because you have to constantly convertD > > pascal strings into null terminated strings before passing them. > E > Or define the interface in such a way that the compiler can do this  > automatically. > K > > Similarly, using C to call system services on VMS is somewhat of a painn, > > because descriptors are not native to C. > ; > Again, a little footwork up front will make this trivial.m > 
 >         Janm   Exactly! But this a the reason why a good language implementation needs time for adapting it to the OS in question. And help from the OS vendor could be an important factor.    ------------------------------   Date: 02 Mar 2001 00:04:49 GMT& From: "Rick Cadruvi" <rick@rdperf.com>0 Subject: Re: Drop C/C++ better sooner than later0 Message-ID: <97mo31$l03@dispatch.concentric.net>  C Dan Esbensen of Touch Technologies has his own programming language L specifically for applicaitons.  It's very powerful.  Internally, he uses his ownoK descriptors for strings that allow him to relocate data without the programuC needing to know and dynamically shrink/increase string size withoutn re-allocations within boundaries.  K If the implementation of descriptors can be hidden from the programmer, youhF can do a LOT with that type of concept that can make the language both pwerfultK and VERY fast.  I have NEVER seen a language that can do strings as fast orb8 well as Intouch on VMS or his new ShearPower on Windows.  L The language doesn't do pointers at all.  I doubt there is a better business typeD applications development language anywhere.  It's amazing how fast a
 programmerH can do REALLY complex things in these languages.  How fast you can debug$ code in these is equally impressive.  E I know I have been defending C/C++.  Now I acknowledge that there aregH WAY better languages for probably most applications.  My point was NEVERK that C/C++ was the best choice, only that it can be a reasonable choice for0 someH things and a LOT of VERY good and important code has been written in it.  K I find it interesting that C/C++ are disparaged in the same paragraph wheretJ VMS is lauded.  A lot of NEW VMS code is being written in C.  I didn't see4 Ada or other languages supported for device drivers.   Rick Cadruvi...w    J "Wayne Sewell" <wayne@tachysoft.xxx.344932.killspam.0142> wrote in message( news:CHtLZwhwOeSS@tachxxsoftxxconsult...6 > In article <3A9D6097.EDEDADF@videotron.ca>, JF Mezei& <jfmezei.spamnot@videotron.ca> writes: > >eK > > Similarly, using C to call system services on VMS is somewhat of a pain*, > > because descriptors are not native to C. >tI > You got that right.  zero terminated strings basically suck, especiallyh since L > they are impossible to use without overhead.  How can you possibly get the; > length of a string without scanning the whole damn thing?m > 4 > I'm not really that wild about descriptors either. >-A > I personally think that DEC Pascal has the best approach.  It's 
 essentially aeL > counted string with a 16-bit length.  It's built into the language, so youI > don't have to access the length directly.  It is set automatically by ae stringH > assignment.  Most of the time you can just forget about the length, as long asgF > the *allocated* size of the string is always adequate to contain the *actual*? > size.  Which of course one has to do with any type of string.r >0 > -- >8L ============================================================================ === : > Wayne Sewell, Tachyon Software Consulting  (281)812-0738 wayne@tachysoft.xxxy: > http://www.tachysoft.xxx/www/tachyon.html and wayne.htmlH > change .xxx to .com in addresses above, assuming you are not a spambot :-)e >sL ============================================================================ === J > Dean Wormer to Flounder: "Fat, drunk, and stupid is no way to go through life."   ------------------------------  % Date: Fri, 02 Mar 2001 00:20:56 +0000o) From: Christof Brass <brass@infopuls.com> 0 Subject: Re: Drop C/C++ better sooner than later, Message-ID: <3A9EE768.72CB335F@infopuls.com>   Dan Sugalski wrote:a > 3 > At 03:00 AM 3/1/2001 +0000, Christof Brass wrote:h > >Dan Sugalski wrote: > > >h8 > > > At 12:22 AM 2/28/2001 +0000, Christof Brass wrote:G > > > >I decided to drop PERL beginning of 2000 as I got the impressionoG > > > >that its capabilities of handling complex/nested data structuresrF > > > >and its flexibility in extending data structures are very poor.E > > > >To my understanding the knowledge about the data structures istE > > > >in the code not in the data structure itself. There is no easy G > > > >way to understand a complex data structure or to find out how it ( > > > >is built up. Do I miss something? > > >eN > > > Apparently, yes. It's dead-simple to build up complex data structures inN > > > perl, and generally simpler to extend them at runtime than in most other > > > languages. > > I > >With extending I meant statically extending them because of additionalv > >requirements. >  > That's trivial as well.c  E Technically yes. Practially tedious and error prone. Want an example?)  L > >And about understanding complex data structures: How do you find out whatJ > >type a certain hash element has if you have deeply nested hashes and/or > >arrays (lists of lists)?  > M > Generally I read the documentation. I find that's the best way to deal withtK > things. Runtime introspection is certainly possible, and straightforward,XH > but if you're not sure of things when you write the code, no amount of( > runtime snooping is going to help any. > - >                                         Dane > K > --------------------------------------"it's like this"------------------- 4 > Dan Sugalski                          even samuraiA > dan@sidhe.org                         have teddy bears and evenu= >                                       teddy bears get drunki   This is exactly where the problem of PERL comes in: it is very time consuming to read the "data structures" to find out what is in there. I'm not talking about other people's sources. Runtime introspection is no help because this doesn't get you thea semantics. Want an example?    ------------------------------  # Date: Fri, 02 Mar 2001 01:11:54 GMTt= From: system@SendSpamHere.ORG (Brian Schenkenberger, VAXman-)a0 Subject: Re: Drop C/C++ better sooner than later0 Message-ID: <009F8614.53205468@SendSpamHere.ORG>  Y In article <97mn3e$kva@dispatch.concentric.net>, "Rick Cadruvi" <rick@rdperf.com> writes:.L >I think we finally get to the point.  You must not work for others.  I have >met fewK >programmers/situations where the choice could be made by the programmer as " >to language, OS, and other tools. >lE >As to worth, I suspect that the $$ value of C/C++ code significantlyr
 >eclipses ANY L >other language.  Take the value of MS products and UNIX products written in >itlH >alone and I think my case is made.  Some of us have to live in the real >world.t >g >Rick Cadruvi...  % There is more to life than $$$, Rick.  --O VAXman- OpenVMS APE certification number: AAA-0001     VAXman(at)TMESIS(dot)COM             eO city, n., 1. a place where trees are cut down and streets are named after them.    ------------------------------  % Date: Thu, 01 Mar 2001 18:26:51 -0800 ! From: Shane.F.Smith@Healthnet.comp0 Subject: Re: Drop C/C++ better sooner than laterD Message-ID: <OF2253DC41.DB6B5919-ON88256A03.00083164@foundation.com>  J Descriptors don't eliminate overheads. Here's a little cautionary tale for y'all.  H Once there was a report in (ahem) a certain commercial package I used toF work on. The user could tailor it down to the level of which fields toK print, using what field and record separators, etc. The original programmeroH wrote it in BASIC (quiet in the back, there!), building a dynamic stringH for each line of output by adding each field and separator individually.J After printing, he'd drop it back to an empty string. There were hundreadsG of selectable fields, and some customers would have a couple of hundreduE thousand of lines. It became a very popular report, but it ran pretty K slowly and ended up being up a substantial percentage of various customers'nJ overnight batch times. At this point, I was called in. I looked at it, hadE a little think, and turned it into a static string with a little MID$rK seasoning. It ended up taking only 30% the time it used to, on average, andi( cut hours off some peoples' batch times.  J The programmer in question was actually damn good, but he made the mistakeJ of treating that string as if it were magic. When he did all those A$ = A$H + B$'s, he wasn't thinking about what the computer actually had to do toJ fill that request.  As it grew the string was being copied to and fro, andK there was memory being allocated and deallocated left right and centre, allsJ unnecessarily. In a way, you might argue the descriptor contributed to theF problem by helping hide the details of how the string was really being manipulated.  H I've always said that programmers should be forced to write in assemblerA for a year before being let loose in the real world. The more youhC understand, the better you do regardless of the tools you're using.    Shane           8 Rick Cadruvi <rick@rdperf.com> on 03/01/2001 04:04:49 PM   To:   Info-VAX@Mvb.Saic.Comu cc:i  1 Subject:  Re: Drop C/C++ better sooner than later-    C Dan Esbensen of Touch Technologies has his own programming language H specifically for applicaitons.  It's very powerful.  Internally, he uses hisl ownzK descriptors for strings that allow him to relocate data without the program C needing to know and dynamically shrink/increase string size without> re-allocations within boundaries.  K If the implementation of descriptors can be hidden from the programmer, yourF can do a LOT with that type of concept that can make the language both pwerfuluK and VERY fast.  I have NEVER seen a language that can do strings as fast orh8 well as Intouch on VMS or his new ShearPower on Windows.  C The language doesn't do pointers at all.  I doubt there is a betteri business typeD applications development language anywhere.  It's amazing how fast a
 programmerH can do REALLY complex things in these languages.  How fast you can debug$ code in these is equally impressive.  E I know I have been defending C/C++.  Now I acknowledge that there areeH WAY better languages for probably most applications.  My point was NEVERK that C/C++ was the best choice, only that it can be a reasonable choice fore someH things and a LOT of VERY good and important code has been written in it.  K I find it interesting that C/C++ are disparaged in the same paragraph whereeJ VMS is lauded.  A lot of NEW VMS code is being written in C.  I didn't see4 Ada or other languages supported for device drivers.   Rick Cadruvi...i    J "Wayne Sewell" <wayne@tachysoft.xxx.344932.killspam.0142> wrote in message( news:CHtLZwhwOeSS@tachxxsoftxxconsult...6 > In article <3A9D6097.EDEDADF@videotron.ca>, JF Mezei& <jfmezei.spamnot@videotron.ca> writes: > > K > > Similarly, using C to call system services on VMS is somewhat of a pain2, > > because descriptors are not native to C. > I > You got that right.  zero terminated strings basically suck, especiallyj sincenH > they are impossible to use without overhead.  How can you possibly get thes; > length of a string without scanning the whole damn thing?h > 4 > I'm not really that wild about descriptors either. >aA > I personally think that DEC Pascal has the best approach.  It's 
 essentially atH > counted string with a 16-bit length.  It's built into the language, so you I > don't have to access the length directly.  It is set automatically by ab stringH > assignment.  Most of the time you can just forget about the length, as long as F > the *allocated* size of the string is always adequate to contain the *actual*? > size.  Which of course one has to do with any type of string.e >y > -- >aL ============================================================================   === : > Wayne Sewell, Tachyon Software Consulting  (281)812-0738 wayne@tachysoft.xxxI: > http://www.tachysoft.xxx/www/tachyon.html and wayne.htmlH > change .xxx to .com in addresses above, assuming you are not a spambot :-)v >fL ============================================================================   ===tJ > Dean Wormer to Flounder: "Fat, drunk, and stupid is no way to go through life."   ------------------------------   Date: 02 Mar 2001 05:17:23 GMT& From: "Rick Cadruvi" <rick@rdperf.com>0 Subject: Re: Drop C/C++ better sooner than later0 Message-ID: <97nad3$kt6@dispatch.concentric.net>   Shane,  K The case I was talking about with Intouch using descriptors is a case wherecG the language optimizes a LOT of that out.  The descriptors in this caselF are NOT standard VMS string descriptors.  The language does predictiveJ adaptation of those exact types of cases and is smart enough to pre-extendI as needed and add extra stuff when a string gets extended.  It also isn'thJ too quick to do de-allocations.  The language is willing to waste a little) virtual memory to save a LOT of overhead.r  J As to your assembler comment, I agree (although less so now).  I basicallyI write code as if it was written in assembler so the compiler never has tor thinkqJ about my intent.  As a result, I NEVER experience the compiler bugs I hearK others talk about.  I think in the last 20+ years of BLISS and C (and quiteeI a few other languages), I have only seen a compiler bug once in the AlphauD DECC compiler in my own code.  The problem was scoping and I felt itH was my fault for getting a little lazy with my coding and allowing for a routinen! that was way too complex and big.n   Rick...n  . <Shane.F.Smith@Healthnet.com> wrote in message> news:OF2253DC41.DB6B5919-ON88256A03.00083164@foundation.com... > L > Descriptors don't eliminate overheads. Here's a little cautionary tale for > y'all. >oJ > Once there was a report in (ahem) a certain commercial package I used toH > work on. The user could tailor it down to the level of which fields toB > print, using what field and record separators, etc. The original
 programmerJ > wrote it in BASIC (quiet in the back, there!), building a dynamic stringJ > for each line of output by adding each field and separator individually.L > After printing, he'd drop it back to an empty string. There were hundreadsI > of selectable fields, and some customers would have a couple of hundreddG > thousand of lines. It became a very popular report, but it ran prettyoB > slowly and ended up being up a substantial percentage of various
 customers'L > overnight batch times. At this point, I was called in. I looked at it, hadG > a little think, and turned it into a static string with a little MID$oI > seasoning. It ended up taking only 30% the time it used to, on average,  andi* > cut hours off some peoples' batch times. >tL > The programmer in question was actually damn good, but he made the mistakeL > of treating that string as if it were magic. When he did all those A$ = A$J > + B$'s, he wasn't thinking about what the computer actually had to do toL > fill that request.  As it grew the string was being copied to and fro, andI > there was memory being allocated and deallocated left right and centre,s alleL > unnecessarily. In a way, you might argue the descriptor contributed to theH > problem by helping hide the details of how the string was really being > manipulated. >iJ > I've always said that programmers should be forced to write in assemblerC > for a year before being let loose in the real world. The more you:E > understand, the better you do regardless of the tools you're using.f >l > Shane  >  >b >g >  >t: > Rick Cadruvi <rick@rdperf.com> on 03/01/2001 04:04:49 PM >  > To:   Info-VAX@Mvb.Saic.Comb > cc:t >i3 > Subject:  Re: Drop C/C++ better sooner than latern >e > E > Dan Esbensen of Touch Technologies has his own programming languageoJ > specifically for applicaitons.  It's very powerful.  Internally, he uses > hise > ownpE > descriptors for strings that allow him to relocate data without thes programeE > needing to know and dynamically shrink/increase string size withouti > re-allocations > within boundaries. >sI > If the implementation of descriptors can be hidden from the programmer,t younH > can do a LOT with that type of concept that can make the language both	 > pwerfuldJ > and VERY fast.  I have NEVER seen a language that can do strings as fast or: > well as Intouch on VMS or his new ShearPower on Windows. >sE > The language doesn't do pointers at all.  I doubt there is a better 
 > business > typeF > applications development language anywhere.  It's amazing how fast a > programmerJ > can do REALLY complex things in these languages.  How fast you can debug& > code in these is equally impressive. >oG > I know I have been defending C/C++.  Now I acknowledge that there areeJ > WAY better languages for probably most applications.  My point was NEVERI > that C/C++ was the best choice, only that it can be a reasonable choicec fort > someJ > things and a LOT of VERY good and important code has been written in it. >fG > I find it interesting that C/C++ are disparaged in the same paragraphw wherecL > VMS is lauded.  A lot of NEW VMS code is being written in C.  I didn't see6 > Ada or other languages supported for device drivers. >h > Rick Cadruvi...s >i >nL > "Wayne Sewell" <wayne@tachysoft.xxx.344932.killspam.0142> wrote in message* > news:CHtLZwhwOeSS@tachxxsoftxxconsult...8 > > In article <3A9D6097.EDEDADF@videotron.ca>, JF Mezei( > <jfmezei.spamnot@videotron.ca> writes: > > > H > > > Similarly, using C to call system services on VMS is somewhat of a pain. > > > because descriptors are not native to C. > > K > > You got that right.  zero terminated strings basically suck, especially  > sincezJ > > they are impossible to use without overhead.  How can you possibly get > thea= > > length of a string without scanning the whole damn thing?w > >v6 > > I'm not really that wild about descriptors either. > >AC > > I personally think that DEC Pascal has the best approach.  It'sR > essentially atJ > > counted string with a 16-bit length.  It's built into the language, so > youTK > > don't have to access the length directly.  It is set automatically by at > stringJ > > assignment.  Most of the time you can just forget about the length, as	 > long as H > > the *allocated* size of the string is always adequate to contain the
 > *actual*A > > size.  Which of course one has to do with any type of string.4 > >k > > -- > >  >tL ============================================================================ >E > ===d< > > Wayne Sewell, Tachyon Software Consulting  (281)812-0738 > wayne@tachysoft.xxxi< > > http://www.tachysoft.xxx/www/tachyon.html and wayne.htmlJ > > change .xxx to .com in addresses above, assuming you are not a spambot > :-)t > >r >sL ============================================================================ >a > === L > > Dean Wormer to Flounder: "Fat, drunk, and stupid is no way to go through > life." >  >' >o >e >y >a >i   ------------------------------   Date: 02 Mar 2001 05:06:17 GMT& From: "Rick Cadruvi" <rick@rdperf.com>0 Subject: Re: Drop C/C++ better sooner than later0 Message-ID: <97n9o9$jha@dispatch.concentric.net>   Brian,  L I agree, but there are the ex-wives, several mortgages, kids in college, andI a few other nuts to crack.  I don't have the luxury of not worrying aboutl
 the money.  L My comment was really directed at the notion that programmers worth anythingK choose their language and OS and that there isn't any code worth looking at= written=	 in C/C++.=  H I agree in an ideal world we would write our software in some WAY better	 language. J I have long wanted to write my own language with all the things I want.  I almostJ did it before I realized that RISC and all the implications of multi-issue instructionsE and whatever, were going to make keeping my language running on every= platformL I might need to run on a MAJOR task.  Maybe I just got old along the way and gothL tired of fighting.  Now I just want to get the job done and NOT have to come back andF fix it later.  I learned to write good code and change how I do things rather thanl* depending on the language to keep me safe.     Rick...   J "Brian Schenkenberger, VAXman-" <system@SendSpamHere.ORG> wrote in message* news:009F8614.53205468@SendSpamHere.ORG...A > In article <97mn3e$kva@dispatch.concentric.net>, "Rick Cadruvi"p <rick@rdperf.com> writes:II > >I think we finally get to the point.  You must not work for others.  Ie have
 > >met fewJ > >programmers/situations where the choice could be made by the programmer as$ > >to language, OS, and other tools. > >eG > >As to worth, I suspect that the $$ value of C/C++ code significantly. > >eclipses ANY K > >other language.  Take the value of MS products and UNIX products written  in > >itlJ > >alone and I think my case is made.  Some of us have to live in the real	 > >world.  > >e > >Rick Cadruvi... > ' > There is more to life than $$$, Rick.  > --4 > VAXman- OpenVMS APE certification number: AAA-0001 VAXman(at)TMESIS(dot)COM >tK > city, n., 1. a place where trees are cut down and streets are named aftert them.a   ------------------------------  % Date: Fri, 02 Mar 2001 01:04:41 -0500a- From: JF Mezei <jfmezei.spamnot@videotron.ca>t0 Subject: Re: Drop C/C++ better sooner than later+ Message-ID: <3A9F37F8.FF164B8@videotron.ca>e  " Shane.F.Smith@Healthnet.com wrote:J > I've always said that programmers should be forced to write in assemblerC > for a year before being let loose in the real world. The more youiE > understand, the better you do regardless of the tools you're using.    I agree entirely.o  M The 2 courses where I learned the most at uni were IBM 370 assembler followed N by computer architecture (where you learn how additions/multiplication etc are made with and/or gates).   ------------------------------  % Date: Thu, 01 Mar 2001 11:34:00 -0800 ! From: Shane.F.Smith@Healthnet.come/ Subject: Re: Future outlook for OpenVMS admins?dD Message-ID: <OF75D5376D.72B7962E-ON88256A02.006ABBFF@foundation.com>  I Motif isn't part of VMS, it's an optional add-on product. I have an AlphaiH that will be running VMS without a GUI of any kind shortly, and accessed only via set host or telnet.   Shanee          = Christof Brass <brass@infopuls.com> on 02/28/2001 08:54:33 PM-   To:   Info-VAX@Mvb.Saic.Com1 cc:M  0 Subject:  Re: Future outlook for OpenVMS admins?     Larry Kilgallen wrote: >o= > In article <3A9DBDF4.A3D41E43@infopuls.com>, Christof Brass8 <brass@infopuls.com> writes: > > Jan Vorbrueggen wrote: > >>8 > >> "Dave Gudewicz" <david.gudewicz@abbott.com> writes: > >>7 > >> > I think Tru64, like VMS is Compaq's own product.  > >>K > >> Coming as it does from OSF/1 (which was its (close to) original name),a I think I > >> the OSF is still collecting a license fee for every copy sold. Motif  as well, > >> I presume.c > >> > >>         Jan > >rJ > > If Motif still costs something although it is open source now VMS will be loaded with that also.e > 0 > Motif is free only for free operating systems.  G Exactly what I thought. This invalidates the argument further up in themH thread that Q would get all the reveneus from VMS because it is its own.   ------------------------------  % Date: Thu, 01 Mar 2001 16:48:05 -0300 ) From: fabio_compaq@ep-bc.petrobras.com.br-/ Subject: Re: Future outlook for OpenVMS admins?0L Message-ID: <OFE90404C2.B2E9EB30-ON03256A02.006C16F3@ep-bc.petrobras.com.br>  @ An operating system without a GUI is a regression nowadays . . .8 Of course the OpenVMS engineering prefers to improve the; internals (bless of God) ....but  in fact I would like morea4 management tools to help in my job, even a Java GUI, Motif GUI or MMC ...  : =x=x=x=x=x=x=x=x=x=x=x=x=x=x=x=x=x=x=x=x=x=x=x=x=x=x=x=x=x  D I dont know why some people like to:   # ps -ef  and # df -k   ! ! != Probably they feel powerful in knowing something which nobody  knows  ! ! !   :-)   Regards    FC        2 Shane.F.Smith@Healthnet.com em 01/03/2001 16:34:00             Info-VAX@Mvb.Saic.Comg      / Assunto: Re: Future outlook for OpenVMS admins?c      I Motif isn't part of VMS, it's an optional add-on product. I have an AlphauH that will be running VMS without a GUI of any kind shortly, and accessed only via set host or telnet.   Shaner          = Christof Brass <brass@infopuls.com> on 02/28/2001 08:54:33 PM    To:   Info-VAX@Mvb.Saic.Como cc:a  0 Subject:  Re: Future outlook for OpenVMS admins?     Larry Kilgallen wrote: >u= > In article <3A9DBDF4.A3D41E43@infopuls.com>, Christof Brasse <brass@infopuls.com> writes: > > Jan Vorbrueggen wrote: > >>8 > >> "Dave Gudewicz" <david.gudewicz@abbott.com> writes: > >>7 > >> > I think Tru64, like VMS is Compaq's own product.a > >>K > >> Coming as it does from OSF/1 (which was its (close to) original name),r I thinkeI > >> the OSF is still collecting a license fee for every copy sold. Motifd as well, > >> I presume.s > >> > >>         Jan > >uJ > > If Motif still costs something although it is open source now VMS will be loaded with that also.a >e0 > Motif is free only for free operating systems.  G Exactly what I thought. This invalidates the argument further up in the H thread that Q would get all the reveneus from VMS because it is its own.   ------------------------------  % Date: Thu, 01 Mar 2001 15:06:42 -0500i From: William_Bochnik@acml.com/ Subject: Re: Future outlook for OpenVMS admins? > Message-ID: <OFEB47E962.29DB0031-ON85256A02.006E3E9C@acml.com>  = if a gui is not properly thought out (or the devleloper has ac: blind spot), sometimes it can be a hindrance - sometimes a> cmdline way of doing things can be your friend - use "find" on? windoze nt 4 and try to cut and paste the resulting file list -l@ can't be done w/o a 3rd party utility- nt thinks you want to cut the files, not the list.  Dumb.       g                                                                                                        eg                     fabio_compaq@ep-bc.petrob                                                          eg                     ras.com.br                               To:  Info-VAX@Mvb.Saic.Com                bg                                                              cc:                                       ag                     03/01/2001 02:48 PM              Subject:     Re: Future outlook for OpenVMS       eg                                                      admins?                                           lg                                                                                                        t        @ An operating system without a GUI is a regression nowadays . . .8 Of course the OpenVMS engineering prefers to improve the; internals (bless of God) ....but  in fact I would like more 4 management tools to help in my job, even a Java GUI, Motif GUI or MMC ...  : =x=x=x=x=x=x=x=x=x=x=x=x=x=x=x=x=x=x=x=x=x=x=x=x=x=x=x=x=x  @ I dont know why some people like to:   # ps -ef  and # df -k   ! ! ! = Probably they feel powerful in knowing something which nobodyn knows  ! ! !   :-)   Regardso   FC        2 Shane.F.Smith@Healthnet.com em 01/03/2001 16:34:00             Info-VAX@Mvb.Saic.Come      / Assunto: Re: Future outlook for OpenVMS admins?e      @ Motif isn't part of VMS, it's an optional add-on product. I have an Alpha? that will be running VMS without a GUI of any kind shortly, ande accessed only via set host or telnet.   Shanes          = Christof Brass <brass@infopuls.com> on 02/28/2001 08:54:33 PM    To:   Info-VAX@Mvb.Saic.Comr cc:s  0 Subject:  Re: Future outlook for OpenVMS admins?     Larry Kilgallen wrote: >s= > In article <3A9DBDF4.A3D41E43@infopuls.com>, Christof Brasst <brass@infopuls.com> writes: > > Jan Vorbrueggen wrote: > >>8 > >> "Dave Gudewicz" <david.gudewicz@abbott.com> writes: > >>7 > >> > I think Tru64, like VMS is Compaq's own product.s > >>; > >> Coming as it does from OSF/1 (which was its (close to)  original name),a I think = > >> the OSF is still collecting a license fee for every copy  sold. Motif  as well, > >> I presume.= > >> > >>         Jan > >=A > > If Motif still costs something although it is open source nowt VMS will be loaded with that also.x > 0 > Motif is free only for free operating systems.  @ Exactly what I thought. This invalidates the argument further up in the? thread that Q would get all the reveneus from VMS because it is= its own.                            F ______________________________________________________________________  : The information contained in this transmission may contain< privileged and confidential information and is intended only< for the use of the person(s) named above. If you are not the< intended recipient,  or an employee or agent responsible for? delivering this message to the intended recipient,  any review, @ dissemination, distribution or duplication of this communication> is strictly prohibited. If you are not the intended recipient,A please contact the sender immediately by reply e-mail and destroyn# all copies of the original message.    ------------------------------   Date: 1 Mar 2001 15:45:45 -0500o+ From: young_r@encompasserve.org (Rob Young)t/ Subject: Re: Future outlook for OpenVMS admins? 3 Message-ID: <AfXutKuCdLjC@eisner.encompasserve.org>a  x In article <OFE90404C2.B2E9EB30-ON03256A02.006C16F3@ep-bc.petrobras.com.br>, fabio_compaq@ep-bc.petrobras.com.br writes:   > B > An operating system without a GUI is a regression nowadays . . .: > Of course the OpenVMS engineering prefers to improve the= > internals (bless of God) ....but  in fact I would like morei6 > management tools to help in my job, even a Java GUI, > Motif GUI or MMC ... >   : 	Sounds like you might be interested in OpenVMS Management= 	Station.  Disclaimer:  It is okay and has some nice featuresd@ 	but like anything else is limited.  As an example, identify andF 	disuser accounts that are inactive for 90 days.  Do this the first ofC 	every month and for the GUI that can do that, we will have to finda3 	other pathological examples that GUIs don't do).  h  D 	I will admit the GUI is prettier though and can augment the command 	line.   				Rob    ------------------------------  # Date: Fri, 02 Mar 2001 01:35:11 GMT,/ From: StevenU@POBoxes.com (Steven P. Underwood)I/ Subject: Re: Future outlook for OpenVMS admins? 1 Message-ID: <3a9ef67c.95769005@news.telocity.com>e  @ I just had a business partner in Taiwan tell me that they can noC longer order an ES40 with a VT510 terminal as Compaq is telling him F "they are being phased out".  So instead he has to configure it with aE graphics monitor, keyboard, and mouse to be used only when originally C confiuring the box.  We were trying to clone a system we have here, @ which has not had the console turned on in "Uptime  42 09:15:56"C )since we had a power failure and determined that the generator hasD- recently had too many things added to it) :7(e  A Now we need to re-test everything with the system loaded with thea graphics monitor.h   Stevem  F On Thu, 01 Mar 2001 11:34:00 -0800, Shane.F.Smith@Healthnet.com wrote:   >hJ >Motif isn't part of VMS, it's an optional add-on product. I have an AlphaI >that will be running VMS without a GUI of any kind shortly, and accessede >only via set host or telnet.a >e >Shane >t >n >r >a > > >Christof Brass <brass@infopuls.com> on 02/28/2001 08:54:33 PM >e >To:   Info-VAX@Mvb.Saic.Com >cc: >,1 >Subject:  Re: Future outlook for OpenVMS admins?  >r >e >Larry Kilgallen wrote:a >>> >> In article <3A9DBDF4.A3D41E43@infopuls.com>, Christof Brass ><brass@infopuls.com> writes:u >> > Jan Vorbrueggen wrote:m >> >> 9 >> >> "Dave Gudewicz" <david.gudewicz@abbott.com> writes:r >> >> 8 >> >> > I think Tru64, like VMS is Compaq's own product. >> >>iL >> >> Coming as it does from OSF/1 (which was its (close to) original name), >I thinkJ >> >> the OSF is still collecting a license fee for every copy sold. Motif	 >as well,b >> >> I presume. >> >>  >> >>         Jans >> >K >> > If Motif still costs something although it is open source now VMS willi >be loaded with that also. >>1 >> Motif is free only for free operating systems.o >tH >Exactly what I thought. This invalidates the argument further up in theI >thread that Q would get all the reveneus from VMS because it is its own.s >  >h >b >d   Steven P. Underwood,DNRC Whitinsville,MAp StevenU@POBoxes.como   ------------------------------  % Date: Thu, 01 Mar 2001 16:00:56 -0800 ! From: Don Sykes <don@alphase.com>  Subject: Re: Gospel of the Tux+ Message-ID: <3A9EE2B7.92DC9249@alphase.com>r  , This is a multi-part message in MIME format.& --------------BBC07EB685DD7E165AA89FC3) Content-Type: text/html; charset=us-asciit Content-Transfer-Encoding: 7bitc  > <!doctype html public "-//w3c//dtd html 4.0 transitional//en"> <html>G What of the vile works of Palmer, who seized the throne of VMS and solddH off its parts to be devoured by the likes of Oracle and Compaq, who knew, no value in it and kept it from the people ? <p>"Mark E. Levy" wrote:- <blockquote TYPE=CITE>Gospel of Tux unearthedNI <p>Every generation has a mythology. Every millenium has a doomsday cult.oK <br>Every legend gets the distortion knob wound up until the speaker melts.eJ <br>Archeologists at the University of Helsinki today uncovered what couldD <br>be the earliest known writings from the Cult of Tux, a fanaticalG <br>religious sect that flourished during the early Silicon Age, aroundr thec% <br>dawn of the third millenium AD...  <p>The Gospel of Tux (v1.1)a/ <p>In the beginning Turing created the Machine.eI <p>And the Machine was crufty and bogacious, existing in theory only. AndAG <br>von Neumann looked upon the Machine, and saw that it was crufty. He I <br>divided the Machine into two Abstractions, the Data and the Code, andaG <br>yet the two were one Architecture. This is a great Mystery, and the  <br>beginning of wisdom.G <p>And von Neumann spoke unto the Architecture, and blessed it, saying,  "GoeI <br>forth and replicate, freely exchanging data and code, and bring forthgJ <br>all manner of devices unto the earth." And it was so, and it was cool.L <br>The Architecture prospered and was implemented in hardware and software.5 <br>And it brought forth many Systems unto the earth. G <p>The first Systems were mighty giants; many great works of renown did F <br>they accomplish. Among them were Colossus, the codebreaker; ENIAC, the J <br>targeter; EDSAC and MULTIVAC and all manner of froody creatures endingJ <br>in AC, the experimenters; and SAGE, the defender of the sky and fatherL <br>of all networks. These were the mighty giants of old, the first childrenH <br>of Turing, and their works are written in the Books of the Ancients., <br>This was the First Age, the age of Lore.H <p>Now the sons of Marketing looked upon the children of Turing, and sawI <br>that they were swift of mind and terse of name and had many great and I <br>baleful attributes. And they said unto themselves, "Let us go now andtF <br>make us Corporations, to bind the Systems to our own use that they may=L <br>bring us great fortune." With sweet words did they lure their customers,I <br>and with many chains did they bind the Systems, to fashion them after C <br>their own image. And the sons of Marketing fashioned themselvesTF <br>Suits to wear, the better to lure their customers, and wrote grave andeF <br>perilous Licenses, the better to bind the Systems. And the sons ofF <br>Marketing thus became known as Suits, despising and being despised by4 <br>the true Engineers, the children of von Neumann.F <p>And the Systems and their Corporations replicated and grew numerous uponF <br>the earth. In those days there were IBM and Digital, Burroughs andF <br>Honeywell, Unisys and Rand, and many others. And they each kept to their I <br>own System, hardware and software, and did not interchange, for their H <br>Licences forbade it. This was the Second Age, the age of Mainframes.H <p>Now it came to pass that the spirits of Turing and von Neumann lookedJ <br>upon the earth and were displeased. The Systems and their CorporationsG <br>had grown large and bulky, and Suits ruled over true Engineers. Anda the J <br>Customers groaned and cried loudly unto heaven, saying, "Oh that thereF <br>would be created a System mighty in power, yet small in size, able toL <br>reach into the very home!" And the Engineers groaned and cried likewise,F <br>saying, "Oh, that a deliverer would arise to grant us freedom from theseeG <br>oppressing Suits and their grave and perilous Licences, and send us. a.K <br>System of our own, that we may hack therein!" And the spirits of Turing@F <br>and von Neumann heard the cries and were moved, and said unto eachI <br>other, "Let us go down and fabricate a Breakthrough, that these cries  <br>may be stilled."F <p>And that day the spirits of Turing and von Neumann spake unto Moore ofH <br>Intel, granting him insight and wisdom to understand the future. AndI <br>Moore was with chip, and he brought forth the chip and named it 4004.uH <br>And Moore did bless the Chip, saying, "Thou art a Breakthrough; with myF <br>own Corporation have I fabricated thee. Thou thou art yet as small as aC <br>dust mote, yet shall thou grow and replicate unto the size of aoJ <br>mountain, and conquer all before thee. This blessing I give unto thee:F <br>every eighteen months shall thou double in capacity, until the end of? <br>the age." This is Moore's Law, which endures unto this day. H <br>And the birth of 4004 was the beginning of the Third Age, the age ofH <br>Microchips. And as the Mainframes and their Systems and Corporations had@I <br>flourished, so did the Microchips and their Systems and Corporations.t' <br>And their lineage was on this wise: I <p>Moore begat Intel. Intel begat Mostech, Zilog and Atari. Mostech begatyG <br>6502, and Zilog begat Z80. Intel also begat 8800, who begat Altair;  andrH <br>8086, mother of all PCs. 6502 begat Commodore, who begat PET and 64; andeG <br>Apple, who begat 2. (Apple is the great Mystery, the Fruit that waseF <br>devoured, yet bloomed again.) Atari begat 800 and 1200, masters of the0D <br>game, who were destroyed by Sega and Nintendo. Xerox begat PARC.H <br>Commodore and PARC begat Amiga, creator of fine arts; Apple and PARCI <br>begat Lisa, who begat Macintosh, who begat iMac. Atari and PARC begat I <br>ST, the music maker, who died and was no more. Z80 begat Sinclair the I <br>dwarf, TRS-80 and CP/M, who begat many machines, but soon passed from K <br>this world. Altair, Apple and Commodore together&nbsp; begat Microsoft,  theVG <br>Great Darkness which is called Abomination, Destroyer of the Earth,? the  <br>Gates of Hell.F <p>Now it came to pass in the Age of Microchips that IBM, the greatest ofG <br>the Mainframe Corporations, looked upon the young Microchip Systemsd and"K <br>was greatly vexed. And in their vexation and wrath they smote the earth G <br>and created the IBM PC. The PC was without sound and colour, cruftys and(E <br>bogacious in great measure, and its likeness was a tramp, yet theaJ <br>Customers were greatly moved and did purchase the PC in great numbers.G <br>And IBM sought about for an Operating System Provider, for in their H <br>haste they had not created one, nor had they forged a suitably grave andfF <br>perilous License, saying, "First we will build the market, then we willI <br>create a new System, one in our own image, and bound by our Licence." I <br>But they reasoned thus out of pride and not wisdom, not forseeing thea <br>wrath which was to come.F <p>And IBM came unto Microsoft, who licensed unto them QDOS, the child ofH <br>CP/M and 8086. (8086 was the daughter of Intel, the child of Moore). AndsI <br>QDOS grew, and was named MS-DOS. And MS-DOS and the PC together waxed.H <br>mighty, and conquered all markets, replicating and taking possessionH <br>thereof, in accordance with Moore's Law. And Intel grew terrible andH <br>devoured all her children, such that no chip could stand before her. AndxF <br>Microsoft grew proud and devoured IBM, and this was a great marvel inG <br>the land. All these things are written in the Books of the Deeds ofi <br>Microsoft.H <p>In the fullness of time MS-DOS begat Windows. And this is the lineage ofG <br>Windows: CP/M begat QDOS. QDOS begat DOS 1.0. DOS 1.0 begat DOS 2.0R byI <br>way of Unix. DOS 2.0 begat Windows 3.11 by way of PARC and Macintosh.oI <br>IBM and Microsoft begat OS/2, who begat Windows NT and Warp, the losta OSJ <br>of lore. Windows 3.11 begat Windows 95 after triumphing over MacintoshE <br>in a mighty Battle of Licences. Windows NT begat NT 4.0 by way of I <br>Windows 95. NT 4.0 begat NT 5.0, the OS also called Windows 2000, TheO? <br>Millenium Bug, Doomsday, Armageddon, The End Of All Things.DF <p>Now it came to pass that Microsoft had waxed great and mighty among the J <br>Microchip Corporations; mighter than any of the Mainframe CorporationsF <br>before it had it waxed. And Gates heart was hardened, and he swore unto> <br>his Customers and their Engineers the words of this curse:H <p>"Children of von Neumann, hear me. IBM and the Mainframe CorporationsH <br>bound thy forefathers with grave and perilous Licences, such that yeI <br>cried unto the spirits of Turing and von Neumann for deliverance. Nowc IoK <br>say unto ye: I am greater than any Corporation before me. Will I loosenrI <br>your Licences? Nay, I will bind thee with Licences twice as grave andtJ <br>ten times more perilous than my forefathers. I will engrave my LicenceJ <br>on thy heart and write my Serial Number upon thy frontal lobes. I willE <br>bind thee to the Windows Platform with cunning artifices and withoF <br>devious schemes. I will bind thee to the Intel Chipset with crufty codeJ <br>and with gnarly APIs. I will capture and enslave thee as no generationE <br>has been enslaved before. And wherefore will ye cry then unto the G <br>spirits of Turing, and von Neumann, and Moore? They cannot hear ye.t I amJ <br>become a greater Power than they. Ye shall cry only unto me, and shallL <br>live by my mercy and my wrath. I am the Gates of Hell; I hold the portalG <br>to MSNBC and the keys to the Blue Screen of Death. Be ye afraid; be  ye- <br>greatly afraid; serve only me, and live."pH <p>And the people were cowed in terror and gave homage to Microsoft, andI <br>endured the many grave and perilous trials which the Windows platformXF <br>and its greatly bogacious Licence forced upon them. And once again did J <br>they cry to Turing and von Neumann and Moore for a deliverer, but noneK <br>was found equal to the task until the birth of Linux, the book of whichV <br>is next but one.J <p>While all this was happening, there lived a quiet worker who complainedH <br>not of his toils, nor did he don garish aparrel forasmuch as to call the G <br>world to cast an eye upon him. He never rested from his labours and  hisuG <br>tools never grew cold. His engineers toiled upon extensions without G <br>measure; his users knew not of the blight of crash or bsod. He only H <br>wished one boot, secure in the knowledge that with it he could standF <br>forever. His masters were gratified as he required not of upgrade; he <br>knew notI <br>obsolescence. His name was VMS, and he was begat of VAX, the daughtere ofH <br>Pdp the Queen of the land of Mini. In the fullness of time VMS begatE <br>Alpha, who was as swift as the wind. VMS and his house dwelt in a E <br>cluster apart from all others, and it came to pass that the worldpG <br>noticed them not, and did not herald their good works. However, theoG <br>great Warriors did not forget the diligence of VMS, and called upond himtG <br>to prepare their weapons and gird their minds with knowledge of thei mostG <br>secret thoughts of the people, all at once. In taking their charge,b VMS.I <br>was everywhere and nowhere; a shimmering arcangel in the night, whichi" <br>the Warriors called "Echelon".& <p>These are the generations of Linux:H <p>SAGE begat ARPA, which begat TCP/IP, and Aloha, which begat Ethernet.H <br>Bell begat Multics, which begat C, which begat Unix. Unix and TCP/IPJ <br>begat Internet, which begat the World Wide Web. Unix begat RMS, fatherG <br>of the great GNU, which begat the Libraries and Emacs, chief of theeF <br>Utilities. In the days of the Web, Internet and Ethernet begat theJ <br>Intranet LAN, which rose to renown among all Corporations and preparedK <br>the way for the Penguin. And Linus and the Web begat the Kernel through F <br>Unix. The Kernel, the Libraries and the Utilities together are theJ <br>Distribution, the one Penguin in many forms, forever and ever praised.F <p>Now in those days there was in the land of Helsinki a young scholarF <br>named Linus the Torvald. Linus was a devout man, a disciple of RMS and_H <br>mighty in the spirit of Turing, von Neumann and Moore. One day as he wasrL <br>meditating on the Architecture, Linus fell into a trance and was grantedB <br>a vision. And in the vision he saw a great Penguin, serene andI <br>well-favoured, sitting upon an ice floe eating fish. And at the sighte ofF <br>the Penguin Linus was deeply afraid, and he cried unto the spirits ofE <br>Turing, von Neumann and Moore for an interpretation of the dream.iI <p>And in the dream the spirits of Turing, von Neumann and Moore answeredlF <br>and spoke unto him, saying, "Fear not, Linus, most beloved hacker. You G <br>are exceedingly cool and froody. The great Penguin which you see is  anF <br>Operating System which you shall create and deploy unto the earth. TherE <br>ice-floe is the earth and all the systems thereof, upon which the1I <br>Penguin shall rest and rejoice at the completion of its task. And the K <br>fish on which the Penguin feeds are the crufty Licensed codebases whichcK <br>swim beneath all the earth's systems. The Penguin shall hunt and devouroJ <br>all that is crufty, gnarly and bogacious; all code which wriggles likeF <br>spaghetti, or is infested with blighting creatures, or is bound by graveeE <br>and perilous Licences shall it capture. And in capturing shall itsI <br>replicate, and in replicating shall it document, and in documentationdJ <br>shall it bring freedom, serenity and most cool froodiness to the earth <br>and all who code therein."H <p>Linus rose from meditation and created a tiny Operating System Kernel asF <br>the dream had foreshewn him; in the manner of RMS, he released theI <br>Kernel unto the World Wide Web for all to take and behold. And in the J <br>fulness of Internet Time the Kernel grew and replicated, becoming mostJ <br>cool and exceedingly froody, until at last it was recognised as indeed a9F <br>great and mighty Penguin, whose name was Tux. And the followers of LinuseD <br>took refuge in the Kernel, the Libraries and the Utilities; theyF <br>installed Distribution after Distribution, and made sacrifice unto thenF <br>GNU and the Penguin, and gave thanks to the spirits of Turing, vonH <br>Neumann and Moore, for their deliverance from the hand of Microsoft. And E <br>this was the beginning of the Fourth Age, the age of Open Source."D <p>Now there is much more to be said about the exceeding strange andG <br>wonderful events of those days; how some Suits of Microsoft plottedt waryG <br>upon the Penguin, but were discovered on a Halloween Eve; how Gatesn fellG <br>among lawyers and was betrayed and crucified by his former friends,t thenG <br>apostles of Media; how the mercenary Knights of the Red Hat broughti the F <br>gospel of the Penguin into the halls of the Corporations; and even ofE <br>the dispute between the brethren of Gnome and KDE over a trollishsF <br>Licence. But all these things are recorded elsewhere, in the Books ofF <br>the Deeds of the Penguin and the Chronicles of the Fourth Age, and IyH <br>suppose if they were all narrated they would fill a stack of DVDs as, <br>deep and perilous as a Usenet Newsgroup.C <p>Now may you code in the power of the Source; may the Kernel, thedJ <br>Libraries and the Utilities be with you, throughout all Distributions,% <br>until the end of the Epoch. Amen.aJ <p>Excessively modified by Werner L. Stunkel (werner@fsi.net)</blockquote> </html>S  & --------------BBC07EB685DD7E165AA89FC3- Content-Type: text/x-vcard; charset=us-ascii;m  name="don.vcf"  Content-Transfer-Encoding: 7bit ' Content-Description: Card for Don Sykes   Content-Disposition: attachment;  filename="don.vcf">   begin:vcard  n:Sykes;Dono$ tel;cell:Available to customers only tel;fax:415-485-6895 tel;work:415-457-8532h x-mozilla-html:TRUE  org:Alpha Software Express, LLC 8 adr:;;1380 Lincoln Ave - Suite 5;San Rafael;CA;94901;USA version:2.1, email;internet:don@alphase.com! title:Principal Software Engineer-^ note;quoted-printable:Website:  http://alphase.com=0D=0AResume: http://alphase.com/DonsCV.html x-mozilla-cpt:;5904T fn:Don Sykes	 end:vcard.  ( --------------BBC07EB685DD7E165AA89FC3--   ------------------------------  $ Date: Thu, 1 Mar 2001 19:53:54 -0600( From: "Bill Ames" <billames@accunet.net>> Subject: How to convert an indexed file to a sequential file??/ Message-ID: <t9tv57rtj1fla5@corp.supernews.com>h  I Im trying to convert an indexed file to a sequential file so that records B can be merged faster.  I would like to use a DCL command for this.   ------------------------------   Date: 02 Mar 2001 05:08:13 GMT& From: "Rick Cadruvi" <rick@rdperf.com>B Subject: Re: How to convert an indexed file to a sequential file??0 Message-ID: <97n9rt$kva@dispatch.concentric.net>  H Create an FDL file that makes the file sequential instead of Indexed and use CONVERT.   Rick...e    3 "Bill Ames" <billames@accunet.net> wrote in message ) news:t9tv57rtj1fla5@corp.supernews.com...tK > Im trying to convert an indexed file to a sequential file so that records D > can be merged faster.  I would like to use a DCL command for this. >  >i   ------------------------------  $ Date: Thu, 1 Mar 2001 15:19:40 -0500- From: "Peter Weaver" <peter.weaver@stelco.ca>kL Subject: Re: How to get Polycenter Performance Advisor Graphs to a web page?4 Message-ID: <aayn6.135578$Z2.1833450@nnrp1.uunet.ca>  1 "Scott Vieth" <svieth@wi.rr.com> wrote in messageM" news:3A9B2048.8070908@wi.rr.com... > Hi:  >AB > We've got Performance Advisor running on our main VMS system and it's1 > happily generating lots of graphs in PostscriptnD > form.  I'd like to build a web page so that anyone on our internal* > network could take a look at performance/ > on that system over the last couple of weeks.e >dD > Performance Advisor doesn't have a lot of choices as far as output( > formats.  Since I can't generate JPEGsE > directly with the Advisor, is there way a way to convert Postscripts  > files to JPEGs (something like > GhostScript)?D >   F Here is a .COM I hacked together today to convert the graphs into .PDFD format using GhostScript and then copy the .PDF to a Pathworks ShareD (thanks for the idea). You can change PDFWRITE to JPEG but make sureF that each .PS file contains only one graph otherwise the JPEG gets way too big.   $ ty adviseit-pdf.comf $!& $!      ADVISEIT-PDF.COM PRW  1-MAR-01 $!< $!      Procedure to run daily and create PDF reports of the performancee, $!      of nodes we are running AdviseIt on. $! $!6 $!      Find our procedure name and clean up old logs. $!5 $ command_procedure = "''f$environment("procedure")'"e $ command_procedure = -n    "''command_procedure'" -e+ "''F$PARSE(COMMAND_PROCEDURE,,,"VERSION")'"n; $ submit/after="tomorrow+1:00"/keep/noprint/que=MAINT$BATCH  'command_procedureF $ log_file = "''f$parse("sys$login:",".log",command_procedure)'" - ";" $ purge 'log_filel $!. $!      Call the routine for each node we want $! $ Call do_report node1 $ Call do_report node2 $ Call do_report node3 $! $! $ goto dones $! $do_report: subroutine $!/ $ assign 'p1'::psdc$database: psdc$database/job  $! $ advise performance select/beg=yesterday/end=todayD graph/type=all/format=POSTSCRIPT=CHARACTERISTICS=COLOR/out=graphs.ps $!> $ gs "-dNOPAUSE" "-sDEVICE=pdfwrite" "-sOutputFile=GRAPHS.PDF"	 GRAPHS.PSh $!# $ delete/nolog/noconfirm graphs.ps.wB $ rename graphs.pdf 'p1'_'f$cvtime("yesterday","ABSOLUTE","DATE")' $! $endsoubtroutine $! $done: $!8 $ copy *_'f$cvtime("yesterday","ABSOLUTE","DATE")'.pdf -=        pathworks_node::PCSA_DISK:[GLOBCOM.VAX__20PERFORMANCE]t $!   ------------------------------  $ Date: Thu, 1 Mar 2001 20:51:14 +0100" From: "Hans Vlems" <hvlems@iae.nl>E Subject: Re: How to replace a multi-volme disk set with larger disks?u( Message-ID: <97m93s$f5t$1@news.IAEhv.nl>   Fred,e  J restoring volume sets is a pain. The main reason for the long restore time is thateG all members of the set need to be updated during the restore operation. K The fastest way is to build the smallest volumeset that will hold all data,a keeping in minddI that the cluster size might increase (depends on which VMS version you'rec on).D Next, mount the volume set and do a restore to the regularly mounted
 volumeset.H So this cannot be an image backup (target must be mounted foreign) but a regular backup andD you've got to add the qualifier /owner=original to preserve security information., Next, add the other disks to the volume set.L This will still take a long time but will be faster. You'll find though that the disks are unevenlyD used and that will take some time before all members share the files equally.   Hans  ' Fred Zwarts heeft geschreven in berichti% <97k2sk$a24$1@info.service.rug.nl>... - We have a multi volume set of 5 disks for thep home directories of our users.1 These are 2GB RZ28 disks. I would like to replaceg# them with 5 other 4.3Gb RZ29 disks.n- I tried to do it this evening, but it failed,e* because I don't seem to understand BACKUP.   This is what I tried.d. Since I have no room in the box for ten disks,* I first made a saveset on tape (TZ88) with  8 $ BACKUP /IMAGE/NOALIAS source-disk tape:SAVESET /rewind  0 This took about 3 hours, not unreasonable, since4 the disks are about 85% full, which is about 8.5 GB.2 I dismounted the disks, replaced them with the new1 disks and tried to restore the saveset onto these21 disks. Since I wanted a cluster size of 4 blocks,e. instead of the default 9 blocks, I first did a   $ INIT /system /cluster=4 DISK  , for the 5 new disks. I mounted them /FOREIGN	 and trieda  J $ BACKUP /image /noinit tape:SAVESET /rewind dsk1:,dsk2:,dsk3:,dsk4:,dsk5:  - This worked in principle, although there weree, messages that BACKUP increased INDEXF.SYS to+ accomodate the files from the source disks.a% I don't think that this is a problem.e4 However, this BACKUP RESTORE operation was extremely5 slow: after 30 minutes only 80 MB was copied. A quicka1 calculation told me that it would cost a few days81 to complete the operation. Therefore I stopped itd and tried something else.a  1 I have room for one additional disk. So I thoughtt/ that maybe a disk-to-disk copy would be faster.i, I placed back the old disks and placed a new0 disk in the extra position. I again did the INIT for the new disk and tried5 $ BACKUP /image/volume=1 source-disk new-disk /NOINIT 3 However, BACKUP did not like the combination of thee1 /VOLUME and the /NOINIT qualifiers. It complained 3 about /VOLUME being inconsistent with the operationt type.a  8 So I decided to give up on the non-default cluster size. I triede- $ BACKUP /image/volume=1 source-disk new-diskg5 This seemed to work much faster, however, in about 30 6 minutes I have seen 6 messages from BACKUP that it had9 a problem to close files because the file header is full.a8 So I don't think that it will work either. I stopped the6 operation and restored the old situation. I will retry4 to install the new disks another time, when I have a better understanding of BACKUP.h  4 So the general question I have is, how does one copy9 (in a reasonable time) the contents of the 5 source diskso5 to 5 other disks in order to get a logical equivalent 9 multi volume set, if one can have only six of those disksl online at a time?   " The specific questions I have are:  8 1) Why is the BACKUP RESTORE operation so slow? Almost a7 factor of 20 slower than the BACKUP SAVE operation? Areh7 there qualifiers for BACKUP to enhance the performance?r  ; 2) Why can't I use the /VOLUME and the /NOINIT qualifier ind1 one BACKUP COPY operation? Is there a way to do aa= disk-to-disk copy while keeping the cluster size at 4 blocks, 9 even if one cannot have all disks online at thesame time?   3 3) Why did BACKUP complain about closing errors and > full file headers when the cluster size increases from 4 to 9?  9 4) (For completeness.) Is it true that the BACKUP messagef6 during a RESTORE operation about increasing INDEXF.SYS7 is no problem, or should I INIT the disks with a largerd INDEXF.SYS?r  7 We run Open VMS 7.2-1. I used an AlphaServer 2100 4/20051 with 512 MB memory. These are the process quotas:a   Account name: SYSTEME CPU limit:                      Infinite  Direct I/O limit:       400OE Buffered I/O byte count quota:    199808  Buffered I/O limit:     400oE Timer queue entry quota:             250  Open file quota:        512aE Paging file quota:                197008  Subprocess quota:        40hE Default page fault cluster:           64  AST quota:             4094aE Enqueue quota:                      3000  Shared file limit:       40 E Max detached processes:                0  Max active jobs:          0e  D   Working Set (pagelets)  /Limit=10240  /Quota=16384  /Extent=262144J   Adjustment enabled      Authorized Quota=16384  Authorized Extent=262144  @   Working Set (8Kb pages) /Limit=640  /Quota=1024  /Extent=16384H                           Authorized Quota=1024  Authorized Extent=16384   ------------------------------  % Date: Thu, 01 Mar 2001 21:12:43 -0500p2 From: "Richard B. Gilbert" <DRAGON@compuserve.com>E Subject: Re: How to replace a multi-volme disk set with larger disks? 7 Message-ID: <200103012113_MC2-C75C-C344@compuserve.com>g  D         As Fred is learning the hard way, bound volume sets can be a management nightmare!o  J         The only really good reason I can think of for creating one is th= at9 a VERY LARGE file exceeds the size of any available disk!u  D         In addition to the problems cited, if you bind two disks andJ somebody creates a file on the *second* disk, that file will not be seen = byJ a backup of the volume set.  Yes, you may ask me how I know this.  Somebo= dyJ did exactly that on one of the systems I now manage.  When we upgraded ou= r;J disk farm I did disk to disk backups of everything; I even figured out ho= weH to do the volume set.  When we brought the system up with the new disks,J the database refused to start.  Sure enough, when we went back to the old=  J disk farm and looked, there was the file except that BACKUP could not see=  $ it when we backed up the volume set.  J         There is nothing but knowledge to keep a user from making such an=  @ error.  Fortunately we caught this problem while it was fixable.  $ Message text written by "Hans Vlems" >Fred,  J restoring volume sets is a pain. The main reason for the long restore tim= eo is thatsG all members of the set need to be updated during the restore operation. J The fastest way is to build the smallest volumeset that will hold all dat= a, keeping in mindfJ that the cluster size might increase (depends on which VMS version you're=   on).D Next, mount the volume set and do a restore to the regularly mounted
 volumeset.H So this cannot be an image backup (target must be mounted foreign) but a regular backup andF you've got to add the qualifier /owner=3Doriginal to preserve security information., Next, add the other disks to the volume set.G This will still take a long time but will be faster. You'll find thougha that the disks are unevenlyD used and that will take some time before all members share the files equally.   Hans  ' Fred Zwarts heeft geschreven in berichtV% <97k2sk$a24$1@info.service.rug.nl>...a- We have a multi volume set of 5 disks for thel home directories of our users.1 These are 2GB RZ28 disks. I would like to replace # them with 5 other 4.3Gb RZ29 disks.d- I tried to do it this evening, but it failed,h* because I don't seem to understand BACKUP.   This is what I tried.s. Since I have no room in the box for ten disks,* I first made a saveset on tape (TZ88) with  8 $ BACKUP /IMAGE/NOALIAS source-disk tape:SAVESET /rewind  0 This took about 3 hours, not unreasonable, since4 the disks are about 85% full, which is about 8.5 GB.2 I dismounted the disks, replaced them with the new1 disks and tried to restore the saveset onto these 1 disks. Since I wanted a cluster size of 4 blocks, . instead of the default 9 blocks, I first did a    $ INIT /system /cluster=3D4 DISK  , for the 5 new disks. I mounted them /FOREIGN	 and tried   J $ BACKUP /image /noinit tape:SAVESET /rewind dsk1:,dsk2:,dsk3:,dsk4:,dsk5= :i  - This worked in principle, although there werew, messages that BACKUP increased INDEXF.SYS to+ accomodate the files from the source disks.l% I don't think that this is a problem.f4 However, this BACKUP RESTORE operation was extremely5 slow: after 30 minutes only 80 MB was copied. A quick 1 calculation told me that it would cost a few daysg1 to complete the operation. Therefore I stopped ite and tried something else.>  1 I have room for one additional disk. So I thoughtA/ that maybe a disk-to-disk copy would be faster.e, I placed back the old disks and placed a new0 disk in the extra position. I again did the INIT for the new disk and tried7 $ BACKUP /image/volume=3D1 source-disk new-disk /NOINITo3 However, BACKUP did not like the combination of theo1 /VOLUME and the /NOINIT qualifiers. It complainedd3 about /VOLUME being inconsistent with the operatione type.n  8 So I decided to give up on the non-default cluster size. I triedt/ $ BACKUP /image/volume=3D1 source-disk new-disky5 This seemed to work much faster, however, in about 30a6 minutes I have seen 6 messages from BACKUP that it had9 a problem to close files because the file header is full.e8 So I don't think that it will work either. I stopped the6 operation and restored the old situation. I will retry4 to install the new disks another time, when I have a better understanding of BACKUP.h  4 So the general question I have is, how does one copy9 (in a reasonable time) the contents of the 5 source diskss5 to 5 other disks in order to get a logical equivalenta9 multi volume set, if one can have only six of those disksa online at a time?n  " The specific questions I have are:  8 1) Why is the BACKUP RESTORE operation so slow? Almost a7 factor of 20 slower than the BACKUP SAVE operation? Are 7 there qualifiers for BACKUP to enhance the performance?t  ; 2) Why can't I use the /VOLUME and the /NOINIT qualifier inb1 one BACKUP COPY operation? Is there a way to do ai= disk-to-disk copy while keeping the cluster size at 4 blocks, 9 even if one cannot have all disks online at thesame time?P  3 3) Why did BACKUP complain about closing errors ande> full file headers when the cluster size increases from 4 to 9?  9 4) (For completeness.) Is it true that the BACKUP messaged6 during a RESTORE operation about increasing INDEXF.SYS7 is no problem, or should I INIT the disks with a largero INDEXF.SYS?r  7 We run Open VMS 7.2-1. I used an AlphaServer 2100 4/200s1 with 512 MB memory. These are the process quotas:o   Account name: SYSTEME CPU limit:                      Infinite  Direct I/O limit:       400nE Buffered I/O byte count quota:    199808  Buffered I/O limit:     400tE Timer queue entry quota:             250  Open file quota:        512bE Paging file quota:                197008  Subprocess quota:        40 E Default page fault cluster:           64  AST quota:             4094rE Enqueue quota:                      3000  Shared file limit:       40oE Max detached processes:                0  Max active jobs:          0   J   Working Set (pagelets)  /Limit=3D10240  /Quota=3D16384  /Extent=3D26214= 4eJ   Adjustment enabled      Authorized Quota=3D16384  Authorized Extent=3D2= 62144   F   Working Set (8Kb pages) /Limit=3D640  /Quota=3D1024  /Extent=3D16384J                           Authorized Quota=3D1024  Authorized Extent=3D16= 384r <    ------------------------------  % Date: Thu, 01 Mar 2001 22:33:58 -0600o7 From: "David J. Dachtera" <djesys.nospam@earthlink.net>eE Subject: Re: How to replace a multi-volme disk set with larger disks?i- Message-ID: <3A9F22B6.BA26AF01@earthlink.net>B   Fred Zwarts wrote: > $ > The specific questions I have are: > [snip]: > 1) Why is the BACKUP RESTORE operation so slow? Almost a9 > factor of 20 slower than the BACKUP SAVE operation? Arem9 > there qualifiers for BACKUP to enhance the performance?e  B Well, understand what is happening in a restore. You are literallyH getting a header, allocating an extent, then populating that extent withF the contents of the saveset. One file at a time - just as BACKUP findsH them in the saveset. A strictly synchronous, serial operation, even in aF disk-to-disk /IMAGE copy or a disksaveset-to-disk restore. In a /IMAGEF restore, BACKUP provides the same FID the file had before, even if the physical location changes.  H Now, look at what is happening in a save. To prepare for sending data toG the target, BACKUP opens as many files as it can and writes data to thedF target and then opens more files and the cycle repeats. Newer versionsG of BACKUP provide for asynchronous operations. Very much more efficientoC than the restore, by nature. Do SHOW DEVICE/FILES periodically on aa@ source disk during a save operation, and you'll see this happen.  ) Hope this helps with that part, at least.p   -- > David J. Dachterat dba DJE Systemsx http://www.djesys.com/  : Unofficial Affordable OpenVMS Home Page and Message Board: http://www.djesys.com/vms/soho/i  F This *IS* an OpenVMS-related newsgroup. So, a certain bias in postings is to be expected.  @ Feel free to exercise your rights of free speech and expression.  F However, attacks against individual posters, or groups of posters, are strongly discouraged.    ------------------------------  % Date: Thu, 01 Mar 2001 22:41:05 -0600-7 From: "David J. Dachtera" <djesys.nospam@earthlink.net>0E Subject: Re: How to replace a multi-volme disk set with larger disks?e- Message-ID: <3A9F2461.E86A1410@earthlink.net>b   "Richard B. Gilbert" wrote:7 > F >         As Fred is learning the hard way, bound volume sets can be a > management nightmare!0 > M >         The only really good reason I can think of for creating one is thatM; > a VERY LARGE file exceeds the size of any available disk!s  F Well, here's another: the destination of a large number of small files@ where EOF size is less than the cluster size would be on a large? stripe-set. Less wasted disk space by using multiple disks in a'F volume-set where the cluster size is smaller because the physical disk size is smaller.   -- t David J. Dachterah dba DJE Systemsc http://www.djesys.com/  : Unofficial Affordable OpenVMS Home Page and Message Board: http://www.djesys.com/vms/soho/   F This *IS* an OpenVMS-related newsgroup. So, a certain bias in postings is to be expected.  @ Feel free to exercise your rights of free speech and expression.  F However, attacks against individual posters, or groups of posters, are strongly discouraged.a   ------------------------------  % Date: Thu, 01 Mar 2001 13:52:03 -0500r- From: "Richard D. Piccard" <piccard@ohio.edu>t Subject: Re: Janitor fixes 90L( Message-ID: <3A9E9A50.21954015@ohio.edu>  E If I recall correctly those were told as personal experiences by BilldG Hancock at DECUS war stories sessions in the mid-1980s.  If he is stilloD around, he should re-tell them.  But I do remember them, vividly forH sure, and perhaps even accurately.  The same venue also had the story ofG the beserk automated fork-lift on the brewery loading dock, but I don't"" remember whose that is to re-tell.  #                                 RDP      WILLIAM WEBB wrote:1  ; > Two stories come to mind that I'm *sure* have been put ono/ > the Web somewhere because they are legendary:' >:. > The Purina kennel feeding control system and >r6 > The Polar Bears and the Point Barrow satellite dish. >r9 > Someone else would probably retell them better than me.s >d > WWWebb   --B ==================================================================B Dick Piccard                           Academic Technology ManagerB piccard@ohio.edu                                 Computer ServicesB http://oak.cats.ohiou.edu/~piccard/                Ohio University   ------------------------------  % Date: Thu, 01 Mar 2001 14:57:53 -0500-  From: norm.raphael@jamesbury.com Subject: Re: Janitor fixes 90L4 Message-ID: <C2256A02.006CDD47.00@jklh21.valmet.com>  K All this and more in "Computer Consulting is a Very Funny Business" By Bill> Hancocke! Copyright 1993 by Network-1, Inc..  ' I don't remember if there was a sequel.    However:  F WALTHAM, Mass.-- (BUSINESS WIRE) -- June 26, 2000 - Network-1 Security Solutions, Inc.tJ (NASDAQ: NSSI), a leader in distributed intrusion prevention solutions for
 e-Business? networks, today announced changes to its Board of Directors....n  M In addition, the Company announced that Dr. William Hancock has resigned frome	 the Board H of Directors to focus on his duties as Chief Security Officer for Exodus Communications - aJ recognized leader for enterprise Web hosting and mission-critical Internet operations.   B Maybe we could get them to make it available on the web somewhere.          * piccard@ohio.edu on 03/01/2001 01:52:03 PM  " Please respond to piccard@ohio.edu   To:   Info-VAX@mvb.saic.comt cc:l Subject:  Re: Janitor fixes 90Lt        E If I recall correctly those were told as personal experiences by Bill G Hancock at DECUS war stories sessions in the mid-1980s.  If he is stillhD around, he should re-tell them.  But I do remember them, vividly forH sure, and perhaps even accurately.  The same venue also had the story ofG the beserk automated fork-lift on the brewery loading dock, but I don'te" remember whose that is to re-tell.  #                                 RDPi     WILLIAM WEBB wrote:L  ; > Two stories come to mind that I'm *sure* have been put on,/ > the Web somewhere because they are legendary:  >l. > The Purina kennel feeding control system and >e6 > The Polar Bears and the Point Barrow satellite dish. >v9 > Someone else would probably retell them better than me.  >c > WWWebb   --B ==================================================================B Dick Piccard                           Academic Technology ManagerB piccard@ohio.edu                                 Computer ServicesB http://oak.cats.ohiou.edu/~piccard/                Ohio University   ------------------------------  % Date: Thu, 01 Mar 2001 12:03:17 -0800s! From: Shane.F.Smith@Healthnet.como Subject: Re: Janitor fixes 90LD Message-ID: <OFB30B45B4.9935AC11-ON88256A02.006CE9EF@foundation.com>  G Personally, I think upper management being technically clueless was thelH underlying problem. (Isn't it always?) If they weren't, they'd have paidH the money for extending the computer room complete with climate control.   Shanew          4 jlsue <jlsuexxxz@home.com> on 03/01/2001 06:46:32 AM  $ Please respond to jlsuexxxz@home.com   To:   Info-VAX@Mvb.Saic.Comt cc:    Subject:  Re: Janitor fixes 90Lk    . On 28 Feb 2001 09:10:42 +0100, Jan Vorbrueggen8 <jan@mailhost.neuroinformatik.ruhr-uni-bochum.de> wrote:  $ >Shane.F.Smith@Healthnet.com writes: >sK >> Then there was the PDP 11/73 (I think - it was the one with the switchest onG >> the front) at a company I used to work for, that crashed reliably ats 5pm,F >> but only if unattended. The company didn't have space for it in the machine F >> room so had put it in the corridor outside. When the system manager watchedoF >> from hiding at 5pm, he saw the secretaries leaving for the day, and1 >> trailing their fingers along the switches.....t >n? >What, the switches weren't disabled? How careless can you get!K >h >    Jan  E Er... I think having the system in the corridor was the main problem.e   ------------------------------   Date: 1 Mar 2001 19:54:48 GMTa) From: leslie@clio.rice.edu (Jerry Leslie)o Subject: Re: Janitor fixes 90L' Message-ID: <97m9e8$non$1@joe.rice.edu>q  , Richard D. Piccard (piccard@ohio.edu) wrote:G : If I recall correctly those were told as personal experiences by BillEI : Hancock at DECUS war stories sessions in the mid-1980s.  If he is still F : around, he should re-tell them.  But I do remember them, vividly forJ : sure, and perhaps even accurately.  The same venue also had the story ofI : the beserk automated fork-lift on the brewery loading dock, but I don't $ : remember whose that is to re-tell.  . This site has Dr. Hancock listed as a speaker:  ,   http://www.tisc.corecom.com/symposium.html   TISC 2001 Symposium)  4 --Jerry Leslie     (my opinions are strictly my own)   ------------------------------  % Date: Thu, 01 Mar 2001 11:48:24 -0800i! From: Shane.F.Smith@Healthnet.com  Subject: RE: Janitor fixes 90LD Message-ID: <OF45D9A2B1.CB105F97-ON88256A02.006C6729@foundation.com>  H I was always fond of the one about the dead pigeons they kept finding onH the steps of the Baltic Exchange in London. It took a while, but someoneF finally realised that they were flying through beam from the microwaveG communications dish on top of (I think) the Net West tower, and gettingoD cooked mid flight. The Baltic Exchange just happened to be the right distance from the tower.   Shaneu          > WILLIAM WEBB <WWEBB1@email.usps.gov> on 03/01/2001 07:08:09 AM   To:   Info-VAX@Mvb.Saic.Coma cc:i   Subject:  RE: Janitor fixes 90Le      9 Two stories come to mind that I'm *sure* have been put ont- the Web somewhere because they are legendary:r  , The Purina kennel feeding control system and  4 The Polar Bears and the Point Barrow satellite dish.  7 Someone else would probably retell them better than me.    WWWebb   ------------------------------  % Date: Thu, 01 Mar 2001 15:04:49 -0700a+ From: Linda Luik <p14175@email.sps.mot.com>c Subject: Re: Janitor fixes 90L1 Message-ID: <3A9EC781.670252DE@email.sps.mot.com>0  , This is a multi-part message in MIME format.& --------------E3EEE9B9D6D2BEEEFD32790E* Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bitd  C The first computer room I worked in did not have a raised floor. Inx@ addition to the computers, printers, and plotters, the sump pumpF plumbing for the basement restrooms was also located in the same room. It would leak occassionally.   Lindab  " Shane.F.Smith@Healthnet.com wrote: > I > Personally, I think upper management being technically clueless was the J > underlying problem. (Isn't it always?) If they weren't, they'd have paidJ > the money for extending the computer room complete with climate control. >  > Shanee > 6 > jlsue <jlsuexxxz@home.com> on 03/01/2001 06:46:32 AM > & > Please respond to jlsuexxxz@home.com >  > To:   Info-VAX@Mvb.Saic.Comk > cc:r > ! > Subject:  Re: Janitor fixes 90L  > 0 > On 28 Feb 2001 09:10:42 +0100, Jan Vorbrueggen: > <jan@mailhost.neuroinformatik.ruhr-uni-bochum.de> wrote: > & > >Shane.F.Smith@Healthnet.com writes: > >sM > >> Then there was the PDP 11/73 (I think - it was the one with the switches  > onI > >> the front) at a company I used to work for, that crashed reliably at  > 5pm,H > >> but only if unattended. The company didn't have space for it in the	 > machinesH > >> room so had put it in the corridor outside. When the system manager	 > watchedfH > >> from hiding at 5pm, he saw the secretaries leaving for the day, and3 > >> trailing their fingers along the switches.....r > >tA > >What, the switches weren't disabled? How careless can you get!  > >e
 > >    Jan > G > Er... I think having the system in the corridor was the main problem.n& --------------E3EEE9B9D6D2BEEEFD32790E- Content-Type: text/x-vcard; charset=us-ascii;m  name="p14175.vcf" Content-Transfer-Encoding: 7bitf( Content-Description: Card for Linda Luik  Content-Disposition: attachment;  filename="p14175.vcf"   begin:vcard  n:Luik;Linda tel;pager:1.888.772.5230 tel;fax:480.655.3569 tel;work:480.655.4432  @ x-mozilla-html:FALSE 	-mozilla-cpt:;3344;;;( org:Motorola  SPS ;IT-CIM maildrop: M555 version:2.1r& email;internet:linda.luik@motorola.com6 title:Regional VMS Systems Adminstrator/Backup Analyst5 adr;quoted-printable:;;2200 W. Broadway Road  =0D=0A=  fn:Luik, Linda	 end:vcardC  ( --------------E3EEE9B9D6D2BEEEFD32790E--   ------------------------------   Date: 1 Mar 2001 17:35:11 -0500Z9 From: Kilgallen@eisner.decus.org.nospam (Larry Kilgallen)T Subject: Re: Janitor fixes 90L3 Message-ID: <nYwSi2hda3cI@eisner.encompasserve.org>   _ In article <3A9EC781.670252DE@email.sps.mot.com>, Linda Luik <p14175@email.sps.mot.com> writes:o  E > The first computer room I worked in did not have a raised floor. In B > addition to the computers, printers, and plotters, the sump pumpH > plumbing for the basement restrooms was also located in the same room. > It would leak occassionally.  " Is that why Unix developed pipes ?   ------------------------------  % Date: Thu, 01 Mar 2001 16:32:53 -0700 % From: Dean Woodward <deanw@rdrop.com>Y Subject: Re: Janitor fixes 90L) Message-ID: <3A9EDC25.4EBDD6A9@rdrop.com>   " Shane.F.Smith@Healthnet.com wrote: > I > Personally, I think upper management being technically clueless was the J > underlying problem. (Isn't it always?) If they weren't, they'd have paidJ > the money for extending the computer room complete with climate control.  C Unless, of course, they're trying not to spend any money so profitsyB look better for the next quarterly report. (After this quarter, ofD course, they have to not spend money so the next quarter looks good,
 and so on...)/  C It was recently discovered here that the UPS for the nationwide NOC C and local LAN servers was wired in parallel to street power insteadcB of in series; this means that when the grid dropped, our UPS wouldA valiantly try to supply power to the entire local grid instead of A just the systems in the building.  It has, apparently, been wirede2 that way since it was installed in the early '90s.   ------------------------------  % Date: Thu, 01 Mar 2001 16:56:05 -0800b0 From: Mark Berryman <Mark.Berryman@Mvb.Saic.Com> Subject: Re: Janitor fixes 90L, Message-ID: <3A9E7F25.4563B170@Mvb.Saic.Com>   Larry Kilgallen wrote: > a > In article <3A9EC781.670252DE@email.sps.mot.com>, Linda Luik <p14175@email.sps.mot.com> writes:t > G > > The first computer room I worked in did not have a raised floor. In5D > > addition to the computers, printers, and plotters, the sump pumpJ > > plumbing for the basement restrooms was also located in the same room.  > > It would leak occassionally. > $ > Is that why Unix developed pipes ?  @ No, that is simply the origin of the term "floating-point unit".   ------------------------------  % Date: Thu, 01 Mar 2001 14:24:21 -0500   From: jamese@beast.dtsw.army.mil, Subject: Re: Looking for VI emulator for TPU0 Message-ID: <01030114242116@beast.dtsw.army.mil>  0 John Reagan <reagan@hiyall.zko.dec.com> wrote onJ Thu, 01 Mar 2001 13:40:01 -0500 in <3A9E5131.52930CED@hiyall.zko.dec.com>:  H > I seem to remember a VI emulator that sits on top of TPU a while back. > G > I've checked out the DECUS library, the WKU archives, and the OpenVMS & > Freeware CD but didn't see anything. > - > Any hints or pointers would be appreciated.I  5 VILE.  http://www.clark.net/pub/dickey/vile/vile.htmle  : Ed James                           ed.james@telecomsys.com5 TeleCommunications Systems, Inc.   voice 410-295-1919Y; 2024 West Street, Suite 300              800-810-0827 x1919m5 Annapolis, MD 21401-3556           fax   410-280-10948   ------------------------------  % Date: Thu, 01 Mar 2001 14:43:26 -0500p  From: jamese@beast.dtsw.army.mil, Subject: Re: Looking for VI emulator for TPU0 Message-ID: <01030114432691@beast.dtsw.army.mil>  F jamese@beast.dtsw.army.mil wrote on Thu, 01 Mar 2001 14:24:21 -0500 in% <01030114242116@beast.dtsw.army.mil>:   2 > John Reagan <reagan@hiyall.zko.dec.com> wrote onL > Thu, 01 Mar 2001 13:40:01 -0500 in <3A9E5131.52930CED@hiyall.zko.dec.com>: > J > > I seem to remember a VI emulator that sits on top of TPU a while back. > > I > > I've checked out the DECUS library, the WKU archives, and the OpenVMS ( > > Freeware CD but didn't see anything. > > / > > Any hints or pointers would be appreciated.r > 7 > VILE.  http://www.clark.net/pub/dickey/vile/vile.html0  ' Whoops, wrong editor. Sorry about that.a  C VITPU is probably what you wanted. It comes as a 36 part VMS-SHARE.o. See <http://ftp.digital.com/pub/VMS/vitpu-v5/>  : Ed James                           ed.james@telecomsys.com5 TeleCommunications Systems, Inc.   voice 410-295-1919i; 2024 West Street, Suite 300              800-810-0827 x1919 5 Annapolis, MD 21401-3556           fax   410-280-1094    ------------------------------  % Date: Thu, 01 Mar 2001 21:36:56 +0100 = From: Arne =?iso-8859-1?Q?Vajh=F8j?= <arne.vajhoej@gtech.com> , Subject: Re: Looking for VI emulator for TPU) Message-ID: <3A9EB2E8.DC6498BC@gtech.com>f   John Reagan wrote:H > I seem to remember a VI emulator that sits on top of TPU a while back. > G > I've checked out the DECUS library, the WKU archives, and the OpenVMS & > Freeware CD but didn't see anything. > - > Any hints or pointers would be appreciated.l  : Gregg Wonderly wrote a VI in TPU many years ago. It worked8 great on VMS 4.x - it was ported to VMS 5.x, but I think it was never is bug-free there.    [now you know how old it is]  6 Source ? I would suggest trying some old DECUS tapes !   Arne   ------------------------------  % Date: Thu, 01 Mar 2001 17:59:53 -0500o2 From: rdeininger@mindspring.com (Robert Deininger), Subject: Re: Looking for VI emulator for TPUL Message-ID: <rdeininger-0103011759550001@user-2iveaf2.dialup.mindspring.com>  H In article <3A9EB2E8.DC6498BC@gtech.com>, Arne =?iso-8859-1?Q?Vajh=F8j?= <arne.vajhoej@gtech.com> wrote:e   > John Reagan wrote:J > > I seem to remember a VI emulator that sits on top of TPU a while back. > > I > > I've checked out the DECUS library, the WKU archives, and the OpenVMSf( > > Freeware CD but didn't see anything. > > / > > Any hints or pointers would be appreciated.a > < > Gregg Wonderly wrote a VI in TPU many years ago. It worked: > great on VMS 4.x - it was ported to VMS 5.x, but I think! > it was never is bug-free there.  >   I The one I have is Wonderly's.  I've been using it on Vax and Alpha, up toD( V7.1, which is the latest VMS I've used.  O I've fixed some bugs, but there are a couple of drawing bugs lucking somewhere.o@ I've patched in buffer-change journalling (the original only hadH keystroke), but haven't finished putting in code to restore from journal files.  F I've seen tables of contents of old libraries that convince me I don'tG have (quite) the newest version.  But I've not found the actual code ofh anything newer than mine.a  ? The newest reference I found via a web search was V2-657, datediG 4-Dec-1992.  The reference was "Chris Higgins OpenVMS Software list II,iJ versino 8.50".  I don't know anything about Chris Higgins or this software8 list.  Can anyone point me to it's current home, if any?   -- s Robert Deininger rdeininger@mindspring.com    ------------------------------  % Date: Thu, 01 Mar 2001 16:06:43 -0800s! From: Don Sykes <don@alphase.com>e, Subject: Re: Looking for VI emulator for TPU+ Message-ID: <3A9EE413.A53C30D1@alphase.com>t  , This is a multi-part message in MIME format.& --------------C4C076AEF2709B510260DE6B) Content-Type: text/html; charset=us-ascii  Content-Transfer-Encoding: 7bitr  > <!doctype html public "-//w3c//dtd html 4.0 transitional//en"> <html>F I don't mean to sound rude, but why would anyone want to use an editorG like VI when they could use TPU ? I'd rather go back to EDT *line mode*I) than use VI - but I guess that's just me.h <p>John Reagan wrote:tG <blockquote TYPE=CITE>I seem to remember a VI emulator that sits on top  of TPU a while back.H <p>I've checked out the DECUS library, the WKU archives, and the OpenVMS( <br>Freeware CD but didn't see anything.. <p>Any hints or pointers would be appreciated. <p>--  <br>John Reagano4 <br>Compaq Pascal/AMACRO Project Leader</blockquote> </html>r  & --------------C4C076AEF2709B510260DE6B- Content-Type: text/x-vcard; charset=us-ascii;   name="don.vcf"n Content-Transfer-Encoding: 7bit.' Content-Description: Card for Don Sykesl  Content-Disposition: attachment;  filename="don.vcf"=   begin:vcard  n:Sykes;Don=$ tel;cell:Available to customers only tel;fax:415-485-6895 tel;work:415-457-8532y x-mozilla-html:TRUEo org:Alpha Software Express, LLC 8 adr:;;1380 Lincoln Ave - Suite 5;San Rafael;CA;94901;USA version:2.1  email;internet:don@alphase.com! title:Principal Software Engineer ^ note;quoted-printable:Website:  http://alphase.com=0D=0AResume: http://alphase.com/DonsCV.html x-mozilla-cpt:;5904C fn:Don Sykes	 end:vcard>  ( --------------C4C076AEF2709B510260DE6B--   ------------------------------  % Date: Thu, 01 Mar 2001 11:11:57 -0800,! From: Shane.F.Smith@Healthnet.com " Subject: Re: Low cost workstationsD Message-ID: <OF2BA8EABD.72037705-ON88256A02.00692798@foundation.com>  I You might try looking at it from another angle. There's plenty of people,lK myself included, who bitch about there being no low-cost VMS packages. DaveeH over at Islandco got off his butt and tried to do something about it. ItC ain't his fault he can't get Compaq to lower the baseline, but he'ssJ forgoing his profits to keep the costs down. Personally I think a round of applause is in order.g   Shanei          7 John Vottero <John@mvpsi.com> on 02/28/2001 06:05:57 PMk   To:   Info-VAX@Mvb.Saic.Comh cc:   # Subject:  Re: Low cost workstations1    F "Larry Kilgallen" <Kilgallen@eisner.decus.org.nospam> wrote in message- news:AOlyRfc3QIh6@eisner.encompasserve.org... < > In article <t9q68rip3oqbdf@corp.supernews.com>, "Islandco" <sales@islandco.com> writes: > > Good point > >iI > > We can get the QL-MT1AE-6X 12 Unit Alphastation Base license for $900t: > > We can pass this on at Cost to help us sel the systems >t6 > I think that would be an excellent competitive move,E > (even though I am not currently in the market for another machine).w  I How can that be considered an excellent competitive move?  How can a $900 K O/S License compete with a $75 O/S License?  That's only competitive if the / OpenVMS License includes an Alpha to run it on.    ------------------------------  % Date: Thu, 01 Mar 2001 21:22:44 +0100n= From: Arne =?iso-8859-1?Q?Vajh=F8j?= <arne.vajhoej@gtech.com>o" Subject: Re: Low cost workstations) Message-ID: <3A9EAF94.7F7387E6@gtech.com>=   "John Gemignani, Jr." wrote:2 > "John Vottero" <John@mvpsi.com> wrote in messageM > > How can that be considered an excellent competitive move?  How can a $900mO > > O/S License compete with a $75 O/S License?  That's only competitive if they3 > > OpenVMS License includes an Alpha to run it on.a >  > Well, not necessarily ...n > / > Are you willing to save $825 in exchange for:9 >  >     blue screens full of hex,O' >     unwelcome visits from Dr. Watson,s  >     mysterious new .chk files,A >     regular reboots (in a futile attempt to prevent the above).l > . > But then, for $75, you get what you pay for.   Where do the $75 come from ?  A AFAIK a Windows 2000 Professional License cost $400-500 (when noth; sold with a system - and when sold with a system MS have nox support obligations !).o   Arne   ------------------------------  $ Date: Thu, 1 Mar 2001 17:38:30 -0500  From: John Santos <JOHN@egh.com>" Subject: Re: Low cost workstations6 Message-ID: <1010301172914.35173A-100000@Ives.egh.com>  ' On Thu, 1 Mar 2001, John Vottero wrote:t   > H > "Larry Kilgallen" <Kilgallen@eisner.decus.org.nospam> wrote in message/ > news:tIAXGlP9E8aO@eisner.encompasserve.org...bB > > In article <t9rbkvlf52n21a@news.supernews.com>, "John Vottero" > <John@mvpsi.com> writes:L > > > "Larry Kilgallen" <Kilgallen@eisner.decus.org.nospam> wrote in message3 > > > news:AOlyRfc3QIh6@eisner.encompasserve.org...aA > > >> In article <t9q68rip3oqbdf@corp.supernews.com>, "Islandco"t" > > > <sales@islandco.com> writes: > > >> > Good point  > > >> >N > > >> > We can get the QL-MT1AE-6X 12 Unit Alphastation Base license for $900? > > >> > We can pass this on at Cost to help us sel the systems  > > >>; > > >> I think that would be an excellent competitive move,sJ > > >> (even though I am not currently in the market for another machine). > > >eJ > > > How can that be considered an excellent competitive move?  How can a > $900M > > > O/S License compete with a $75 O/S License?  That's only competitive if  > the 5 > > > OpenVMS License includes an Alpha to run it on.y > > D > > Islandco is in competition with other vendors, and thanks to theD > > Internet, locality of purchase (at least within the US) does notD > > count so much any more.  If Islandco offers a cheaper price thanG > > their competition for a VMS license, it is a good competitive move.t > >  > K > Yes, it is a good competitive move by IslandCo.  I wish Compaq would make- > some competitive moves.- > E > > If you happen to feel that some $75 operating system is the equalo@ > > of VMS, then by all means purchase it, but some of us cannot4 > > think of a $75 operating system of that caliber. > >  > J > I don't think that Solaris is equal to VMS.  Do you think that VMS is 12 > times better than Solaris? > I > No one at Sun thinks they're losing money by selling single CPU SolariseL > licenses for $75.  No one at Sun thinks they'll lose money selling the SunN > Blade 100 for $995.  Does anyone at Compaq think the DS10 is their bread andG > butter?  Do they really think people will stop buying GS140's if theyp) > offered a $1000 low-end OpenVMS system?n  F Suppose Compaq sold a low-end Alpha for about $1000 with a VMS licenseE for $75.  Should we start a pool on how long before the 1st complaintrC that 1) You can't upgrade the machine.  2) If you buy a new, higheryE capacity machine, you can't move your $75 VMS license to it for free.e@ 3) The system only supports 1GB of memory.  What use is a 64-bitF architecture when it is limited to 1GB?  4) The system only comes withB 128MB, memory upgrades are much more expensive when purchased fromF Compaq than 3rd-party PC memory, and why does it require expensive ECCG memory instead of cheap non-parity memory?  5) How come VMS is so much 2F more expensive on the higher priced machines?  Windows costs the same + whether you run it on a 486 or a 1.5Mhz P4.d   --   John Santosi Evans Griffiths & Hart, Inc. 781-861-0670 ext 539   ------------------------------  % Date: Thu, 01 Mar 2001 14:51:26 -0800l! From: Shane.F.Smith@Healthnet.coml" Subject: Re: Low cost workstationsD Message-ID: <OFEB5661EE.4159F79E-ON88256A02.007D6DFA@foundation.com>  G I hear what you're saying, John, and you're probably right. However, itlK seems to me that having a low cost workstation available with people cryingtE out for more is a step better than having NO low cost workstation ande people crying out for one.   Shane           4 John Santos <JOHN@egh.com> on 03/01/2001 02:38:30 PM   To:   Info-VAX@Mvb.Saic.Com  cc:j  # Subject:  Re: Low cost workstations2    ' On Thu, 1 Mar 2001, John Vottero wrote:    >mH > "Larry Kilgallen" <Kilgallen@eisner.decus.org.nospam> wrote in message/ > news:tIAXGlP9E8aO@eisner.encompasserve.org...hB > > In article <t9rbkvlf52n21a@news.supernews.com>, "John Vottero" > <John@mvpsi.com> writes:D > > > "Larry Kilgallen" <Kilgallen@eisner.decus.org.nospam> wrote in message.3 > > > news:AOlyRfc3QIh6@eisner.encompasserve.org...nA > > >> In article <t9q68rip3oqbdf@corp.supernews.com>, "Islandco"m" > > > <sales@islandco.com> writes: > > >> > Good point  > > >> >I > > >> > We can get the QL-MT1AE-6X 12 Unit Alphastation Base license for. $900? > > >> > We can pass this on at Cost to help us sel the systemsu > > >>; > > >> I think that would be an excellent competitive move,eJ > > >> (even though I am not currently in the market for another machine). > > >yJ > > > How can that be considered an excellent competitive move?  How can a > $900J > > > O/S License compete with a $75 O/S License?  That's only competitive if > thel5 > > > OpenVMS License includes an Alpha to run it on.a > >.D > > Islandco is in competition with other vendors, and thanks to theD > > Internet, locality of purchase (at least within the US) does notD > > count so much any more.  If Islandco offers a cheaper price thanG > > their competition for a VMS license, it is a good competitive move.2 > >. >oK > Yes, it is a good competitive move by IslandCo.  I wish Compaq would make2 > some competitive moves.- >-E > > If you happen to feel that some $75 operating system is the equali@ > > of VMS, then by all means purchase it, but some of us cannot4 > > think of a $75 operating system of that caliber. > >  > J > I don't think that Solaris is equal to VMS.  Do you think that VMS is 12 > times better than Solaris? >sI > No one at Sun thinks they're losing money by selling single CPU Solaris H > licenses for $75.  No one at Sun thinks they'll lose money selling the SunnJ > Blade 100 for $995.  Does anyone at Compaq think the DS10 is their bread and G > butter?  Do they really think people will stop buying GS140's if theya) > offered a $1000 low-end OpenVMS system?   F Suppose Compaq sold a low-end Alpha for about $1000 with a VMS licenseE for $75.  Should we start a pool on how long before the 1st complaint:C that 1) You can't upgrade the machine.  2) If you buy a new, highersE capacity machine, you can't move your $75 VMS license to it for free.h@ 3) The system only supports 1GB of memory.  What use is a 64-bitF architecture when it is limited to 1GB?  4) The system only comes withB 128MB, memory upgrades are much more expensive when purchased fromF Compaq than 3rd-party PC memory, and why does it require expensive ECCF memory instead of cheap non-parity memory?  5) How come VMS is so muchE more expensive on the higher priced machines?  Windows costs the same + whether you run it on a 486 or a 1.5Mhz P4.e   -- John Santosa Evans Griffiths & Hart, Inc. 781-861-0670 ext 539   ------------------------------  % Date: Fri, 02 Mar 2001 00:00:53 +0000 ) From: Christof Brass <brass@infopuls.com>e" Subject: Re: Low cost workstations, Message-ID: <3A9EE2B5.E0950895@infopuls.com>  " Shane.F.Smith@Healthnet.com wrote: > K > You might try looking at it from another angle. There's plenty of people,rM > myself included, who bitch about there being no low-cost VMS packages. DavetJ > over at Islandco got off his butt and tried to do something about it. ItE > ain't his fault he can't get Compaq to lower the baseline, but he'sdL > forgoing his profits to keep the costs down. Personally I think a round of > applause is in order.u >  > Shanei > 9 > John Vottero <John@mvpsi.com> on 02/28/2001 06:05:57 PM. >  > To:   Info-VAX@Mvb.Saic.Com  > cc:s > % > Subject:  Re: Low cost workstationst > H > "Larry Kilgallen" <Kilgallen@eisner.decus.org.nospam> wrote in message/ > news:AOlyRfc3QIh6@eisner.encompasserve.org...0> > > In article <t9q68rip3oqbdf@corp.supernews.com>, "Islandco" > <sales@islandco.com> writes: > > > Good point > > >tK > > > We can get the QL-MT1AE-6X 12 Unit Alphastation Base license for $900 < > > > We can pass this on at Cost to help us sel the systems > > 8 > > I think that would be an excellent competitive move,G > > (even though I am not currently in the market for another machine).  > K > How can that be considered an excellent competitive move?  How can a $900tM > O/S License compete with a $75 O/S License?  That's only competitive if theh1 > OpenVMS License includes an Alpha to run it on.@   I have to agree.  ) Buy something valuable from IslandCo now!1   ------------------------------   Date: 02 Mar 2001 00:26:45 GMT& From: "Rick Cadruvi" <rick@rdperf.com>" Subject: Re: Low cost workstations0 Message-ID: <97mpc5$kt6@dispatch.concentric.net>  I John's point is well taken.  Also, how long before everyone wants all thefH cheap PC hardware for graphics, audio, whatever supported on these cheapI systems.  I can hear it now - "Why won't my cheap made in China non-brande sound card or modem work?"  L I also think that PART of the Microsoft NT reliability problem is the crappyK hardware it has to run on.  If VMS supported all the junk that NT supports, D I doubt it would be as reliable and would probably be more maligned.  A I suspect that MOST arguments for low cost VMS systems comes fromaJ people who would like faster newer VMS systems and aren't willing to spendI the money on them, rather than a  REAL need in the marketplace that makes  it worthwhile.  J Having said that, I think a compromise between rellay expensive and reallyG cheap makes sense and I personally am in favor of  really cheap becauseJI I want a faster, nicer, etc... VMS system at home.  It pisses me off thatSC my PC is WAY faster and cheaper to upgrade.  I want my compiles and I other VMS stuff to be LOT's faster (like instant as they are with my PC).0     Rick Cadruvi...-    - "John Santos" <JOHN@egh.com> wrote in message50 news:1010301172914.35173A-100000@Ives.egh.com...) > On Thu, 1 Mar 2001, John Vottero wrote:g >r > >mJ > > "Larry Kilgallen" <Kilgallen@eisner.decus.org.nospam> wrote in message1 > > news:tIAXGlP9E8aO@eisner.encompasserve.org...oD > > > In article <t9rbkvlf52n21a@news.supernews.com>, "John Vottero" > > <John@mvpsi.com> writes:F > > > > "Larry Kilgallen" <Kilgallen@eisner.decus.org.nospam> wrote in messagee5 > > > > news:AOlyRfc3QIh6@eisner.encompasserve.org... C > > > >> In article <t9q68rip3oqbdf@corp.supernews.com>, "Islandco"t$ > > > > <sales@islandco.com> writes: > > > >> > Good pointo
 > > > >> >K > > > >> > We can get the QL-MT1AE-6X 12 Unit Alphastation Base license for- $900A > > > >> > We can pass this on at Cost to help us sel the systemsd > > > >>= > > > >> I think that would be an excellent competitive move,mL > > > >> (even though I am not currently in the market for another machine). > > > >.L > > > > How can that be considered an excellent competitive move?  How can a > > $900L > > > > O/S License compete with a $75 O/S License?  That's only competitive if > > thea7 > > > > OpenVMS License includes an Alpha to run it on.d > > >eF > > > Islandco is in competition with other vendors, and thanks to theF > > > Internet, locality of purchase (at least within the US) does notF > > > count so much any more.  If Islandco offers a cheaper price thanI > > > their competition for a VMS license, it is a good competitive move.i > > >r > >eH > > Yes, it is a good competitive move by IslandCo.  I wish Compaq would make > > some competitive moves.t > >MG > > > If you happen to feel that some $75 operating system is the equaluB > > > of VMS, then by all means purchase it, but some of us cannot6 > > > think of a $75 operating system of that caliber. > > >i > >eL > > I don't think that Solaris is equal to VMS.  Do you think that VMS is 12 > > times better than Solaris? > >uK > > No one at Sun thinks they're losing money by selling single CPU SolariswJ > > licenses for $75.  No one at Sun thinks they'll lose money selling the SuniL > > Blade 100 for $995.  Does anyone at Compaq think the DS10 is their bread and I > > butter?  Do they really think people will stop buying GS140's if theyC+ > > offered a $1000 low-end OpenVMS system?n > H > Suppose Compaq sold a low-end Alpha for about $1000 with a VMS licenseG > for $75.  Should we start a pool on how long before the 1st complaintdE > that 1) You can't upgrade the machine.  2) If you buy a new, higher-G > capacity machine, you can't move your $75 VMS license to it for free.IB > 3) The system only supports 1GB of memory.  What use is a 64-bitH > architecture when it is limited to 1GB?  4) The system only comes withD > 128MB, memory upgrades are much more expensive when purchased fromH > Compaq than 3rd-party PC memory, and why does it require expensive ECCH > memory instead of cheap non-parity memory?  5) How come VMS is so muchG > more expensive on the higher priced machines?  Windows costs the sameE- > whether you run it on a 486 or a 1.5Mhz P4.s >  > --
 > John Santosb > Evans Griffiths & Hart, Inc. > 781-861-0670 ext 539 >e   ------------------------------   Date: 1 Mar 2001 17:30:10 -0700h1 From: nothome@spammers.are.scum (Malcolm Dunnett)>" Subject: Re: Low cost workstations, Message-ID: <lwL7ixGsXf9m@malvm1.mala.bc.ca>  1 In article <97mpc5$kt6@dispatch.concentric.net>, e+    "Rick Cadruvi" <rick@rdperf.com> writes:f  K > John's point is well taken.  Also, how long before everyone wants all the J > cheap PC hardware for graphics, audio, whatever supported on these cheapK > systems.  I can hear it now - "Why won't my cheap made in China non-brandk > sound card or modem work?" > D     In general I'd tend to agree. There's not really much market forM VMS workstations anymore anyway. If this discussion was taking place 10 yearsrH ago it might be worth arguing there needed to be a VMS workstation whichA competes with Sun (in fact people did argue that - Digital didn't0P listen and the rest is history), but not today - they'd never get the volumes toM make it worthwhile. I think it's more important there be "cheap" VMS servers,.A but I don't think we're talking $1000 there, more like $2000-5000- (depending on configuration).6  H     I'd certainly argue that this Sun announcement suggests the need forE a cheap Alphastation to run Tru64 Unix though, but that's a topic forb a different ng.   D    If this had happened a year ago we might have bought a server and= 20+ workstations from Sun rather than a DS20 and X-terminals.w   > L > Having said that, I think a compromise between rellay expensive and reallyI > cheap makes sense and I personally am in favor of  really cheap becauseoK > I want a faster, nicer, etc... VMS system at home.  It pisses me off that E > my PC is WAY faster and cheaper to upgrade.  I want my compiles andlK > other VMS stuff to be LOT's faster (like instant as they are with my PC).c  D    This is where deals like the hobbyist license and the cheap boxes- Islandco is advertising should meet the need.d   ------------------------------  $ Date: Thu, 1 Mar 2001 20:21:40 -0500- From: "Island Computers" <sales@islandco.com> " Subject: Re: Low cost workstations/ Message-ID: <t9ttapdvhjt2d8@corp.supernews.com>,  B It would be even better if we were selling a bunch of these things  F What I don't understand is that even this REALLY inexpensive system isI selling slower than Cheese in a Candy Store - bad analogy but appropriatee  J Out of curiosity - anyone here buy one of these CLone PC164LX systems from us...?  K I was just wandering how they were performing - I haven't had the chance toh really build on myself yet...o  D I know they run non-commercial Unix like a bat out a hell though ;0)       -- David Turner Island Computers US Corporation" 2700 Gregory Streett Savannah GA 31404o Tel: 912 447 6622d Fax:912 201 0096 sales@islandco.com. <Shane.F.Smith@Healthnet.com> wrote in message> news:OFEB5661EE.4159F79E-ON88256A02.007D6DFA@foundation.com... |sI | I hear what you're saying, John, and you're probably right. However, it F | seems to me that having a low cost workstation available with people cryingG | out for more is a step better than having NO low cost workstation and  | people crying out for one. |  | Shanea |m |t |r |n |h6 | John Santos <JOHN@egh.com> on 03/01/2001 02:38:30 PM |o | To:   Info-VAX@Mvb.Saic.Comm | cc:F | % | Subject:  Re: Low cost workstationst |0 |0) | On Thu, 1 Mar 2001, John Vottero wrote:  |d | > J | > "Larry Kilgallen" <Kilgallen@eisner.decus.org.nospam> wrote in message1 | > news:tIAXGlP9E8aO@eisner.encompasserve.org...0D | > > In article <t9rbkvlf52n21a@news.supernews.com>, "John Vottero" | > <John@mvpsi.com> writes:F | > > > "Larry Kilgallen" <Kilgallen@eisner.decus.org.nospam> wrote in	 | messager5 | > > > news:AOlyRfc3QIh6@eisner.encompasserve.org... C | > > >> In article <t9q68rip3oqbdf@corp.supernews.com>, "Islandco"a$ | > > > <sales@islandco.com> writes: | > > >> > Good point 
 | > > >> >K | > > >> > We can get the QL-MT1AE-6X 12 Unit Alphastation Base license fort | $900A | > > >> > We can pass this on at Cost to help us sel the systemsI | > > >>= | > > >> I think that would be an excellent competitive move,iL | > > >> (even though I am not currently in the market for another machine). | > > >6L | > > > How can that be considered an excellent competitive move?  How can a | > $900L | > > > O/S License compete with a $75 O/S License?  That's only competitive | if | > the 7 | > > > OpenVMS License includes an Alpha to run it on.a | > >mF | > > Islandco is in competition with other vendors, and thanks to theF | > > Internet, locality of purchase (at least within the US) does notF | > > count so much any more.  If Islandco offers a cheaper price thanI | > > their competition for a VMS license, it is a good competitive move.e | > >r | >dH | > Yes, it is a good competitive move by IslandCo.  I wish Compaq would make | > some competitive moves.o | >IG | > > If you happen to feel that some $75 operating system is the equaliB | > > of VMS, then by all means purchase it, but some of us cannot6 | > > think of a $75 operating system of that caliber. | > >o | >mL | > I don't think that Solaris is equal to VMS.  Do you think that VMS is 12 | > times better than Solaris? | >lK | > No one at Sun thinks they're losing money by selling single CPU SolariseJ | > licenses for $75.  No one at Sun thinks they'll lose money selling the | Sun L | > Blade 100 for $995.  Does anyone at Compaq think the DS10 is their bread | andeI | > butter?  Do they really think people will stop buying GS140's if theyt+ | > offered a $1000 low-end OpenVMS system?n | H | Suppose Compaq sold a low-end Alpha for about $1000 with a VMS licenseG | for $75.  Should we start a pool on how long before the 1st complainteE | that 1) You can't upgrade the machine.  2) If you buy a new, higher G | capacity machine, you can't move your $75 VMS license to it for free.rB | 3) The system only supports 1GB of memory.  What use is a 64-bitH | architecture when it is limited to 1GB?  4) The system only comes withD | 128MB, memory upgrades are much more expensive when purchased fromH | Compaq than 3rd-party PC memory, and why does it require expensive ECCH | memory instead of cheap non-parity memory?  5) How come VMS is so muchG | more expensive on the higher priced machines?  Windows costs the samew- | whether you run it on a 486 or a 1.5Mhz P4.d |  | --
 | John Santosr | Evans Griffiths & Hart, Inc. | 781-861-0670 ext 539 |h |p |e |a |n   ------------------------------  % Date: Thu, 01 Mar 2001 21:33:41 -0600 7 From: "David J. Dachtera" <djesys.nospam@earthlink.net>6" Subject: Re: Low cost workstations- Message-ID: <3A9F1495.F186B18B@earthlink.net>    Island Computers wrote:@ > D > It would be even better if we were selling a bunch of these things > H > What I don't understand is that even this REALLY inexpensive system isK > selling slower than Cheese in a Candy Store - bad analogy but appropriateo  E Can you say, "marketing"? Who knows about these machines besides COV?e   Can you say, "direct mail"?t  . I'll give you the same info. I gave to Compaq:  	 Metromailn 360 E. 22nd St.. Lombard, IL 601480
 (800)567-6100o
 (630)620-3300b   Sue Skonetski replied:L >Just so I have it clear.  Metromail could create a distribution list of say4 >all the CTO's of all the top Fortune 500 companies? >p >Please advise.a   ...and I responded:@   Yes! ...and then some!  G They are also the home of the National Consumer Database and the DirectJC Marketing Association (to the best of my knowledge, but it's been au* while (March of 1995); so, no guarantees).   -- c David J. Dachteraa dba DJE Systems  http://www.djesys.com/  : Unofficial Affordable OpenVMS Home Page and Message Board: http://www.djesys.com/vms/soho/n  F This *IS* an OpenVMS-related newsgroup. So, a certain bias in postings is to be expected.  @ Feel free to exercise your rights of free speech and expression.  F However, attacks against individual posters, or groups of posters, are strongly discouraged.t   ------------------------------  $ Date: Thu, 1 Mar 2001 22:02:29 -0000/ From: "Gerald Marsh" <gerald@cyfer.demon.co.uk>w8 Subject: Re: My name is Dean, and I'm a [Un*x|NT] hater.@ Message-ID: <983481859.6443.0.nnrp-07.9e984c92@news.demon.co.uk>   Crikey! I'm impressed!L (But I'm sure you don't need all those apostophe's before a lexical function ;->)  , Thanks for the great dialogue in this group.   Bye for now,   Gerald.      -- Gerald Marsh - Directord Cyfer Solutions Ltdr email: gerald@cyfer.demon.co.uko        7 "Peter LANGSTOEGER" <eplan@kapsch.net> wrote in message,$ news:3a9d2f9f$1@news.kapsch.co.at...J > In article <y4d7c2lt7h.fsf@mailhost.neuroinformatik.ruhr-uni-bochum.de>,I Jan Vorbrueggen <jan@mailhost.neuroinformatik.ruhr-uni-bochum.de> writes:hB > >system@SendSpamHere.ORG (Brian Schenkenberger, VAXman-) writes:6 > >> $ SYMBOL/SET/LOCAL/DEPTH/EXECUTIVE CTL$AG_CLIDATA1 > >> $ SYMBOL/SET/LOCAL/DEPTH/EXECUTIVE PPD$L_PRC 7 > >> $ SYMBOL/SET/LOCAL/DEPTH/EXECUTIVE PRC_L_RECALLPTRe > >> $6 > >> $ PRC = 'F$fao("%x!@XL",CTL$AG_CLIDATA+PPD$L_PRC)3 > >> $ NXT = 'F$fao("%x!@XL",PRC+PRC_L_RECALLPTR)-1 ' > >> $ CMD = NXT-'F$fao("%x!@XB",NXT)-1 - > >> $ WRITE SYS$OUTPUT "''F$fao("!AC",CMD)'"  > >e0 > >Wonderful what FAO and DCL allow you to do... >e
 > Come on,I > without VAXman's SYMBOL package, DCL would not be of great help here...s >o > --> > Peter "EPLAN" LANGSTOEGER           Tel.    +43 1 81111-2651= > Network and OpenVMS system manager  Fax.    +43 1 81111-888 > > <<< KAPSCH AG  Wagenseilgasse 1     E-mail  eplan@kapsch.netJ > A-1121 VIENNA  AUSTRIA              "I'm not a pessimist, I'm a realist"   ------------------------------  # Date: Thu, 01 Mar 2001 22:38:26 GMTt= From: system@SendSpamHere.ORG (Brian Schenkenberger, VAXman-)h8 Subject: Re: My name is Dean, and I'm a [Un*x|NT] hater.0 Message-ID: <009F85FE.E26D503C@SendSpamHere.ORG>  r In article <983481859.6443.0.nnrp-07.9e984c92@news.demon.co.uk>, "Gerald Marsh" <gerald@cyfer.demon.co.uk> writes: >Crikey! I'm impressed!aM >(But I'm sure you don't need all those apostophe's before a lexical function  >;->)o  H Use only if you want to see what each line evaluate to when running withI SET VERIFY.  I place them in so that I can see the hexadecimal address oft each item as it is computed.   --O VAXman- OpenVMS APE certification number: AAA-0001     VAXman(at)TMESIS(dot)COMh            eO city, n., 1. a place where trees are cut down and streets are named after them.    ------------------------------   Date: 1 Mar 2001 21:49:23 +0100 ) From: maulis@ludens.elte.hu (Adam Maulis)a# Subject: Re: new freeware: SIPU.COM ! Message-ID: <GCXYE7ROUll1@ludens>   / SIPU Version 0.5 released for ypur pleasure :-)t      c In article <3IMzhzjh9KbX@eisner.encompasserve.org>, koehler@encompasserve.org (Bob Koehler) writes:r > < > 2)  Unziping the sipu.zip with the latest unzip I got fromF >    www.openvms.comaq.com created stral-lf files even though the dataD >    was variable length records.  I had to set file/attr=rfm:var onH >    both aaaredme.txt and sipu.com.  Perhaps something was skipped when2 >    you zipped it (I tried unzip and unzip -"V")?   fixed.     > @ > 3)  You hard coded EIHD$ symbols.  This makes sipu.com version7 >    specific.  Get and use the SYMBOL utility instead.a     new feature:  / SIPU can load any symbols from sys$share:*.req.l     from aaareadme.txt:    	0, How does 'LOAD' work?e         0," 	$ search sys$share:*.req 'prefix'= 	interpret these literals and macros and define correspondingh< 	symbols. SIPU uses the sys$scratch: directory for temporary< 	files. Because the searching or interpreting may take a lot? 	of times SIPU crate a permanent definition file for subsequenti5 	loading. File name: sys$scratch:SIPUDEF_'prefix'.COMo  
 	For example:  	$ dir sys$scratch:sipudef*.*-" 	%DIRECT-W-NOFILES, no files found 	$ 	$ @SIPU APPLE.EXE  	SIPU> LOAD EIHD$	! some seconds 	SIPU> quito 	$ DIR sys$scratch:sipudef*.*p   	Directory USER$DISK:[MAULIS]i 	i 	SIPUDEF_EIHD$.COM;1 	l 	Total of 1 file.  	$ @SIPU APPLE.EXE$ 	SIPU> LOAD EIHD$	! few hundreds :-) 	SIPU> quite 	$     > H > ----------------------------------------------------------------------A > Bob Koehler                     | Computer Sciences Corporation ? > NASA GSFC Flight Software       | Federal Sector, Civil GroupoG >                                 | please remove ".aspm" when replyingu    ( Any other bugreports or suggestions? ;-)    H Adam Maulis              maulis@ludens elte hu        VMS system managerH ........................................................................H VMS Competence Center                              VMS Szakertoi KozpontH Eotvos Lorand University                   Eotvos Lorand TudomanyegyetemH Budapest, Hungary                                               BudapestH ========================================================================   ------------------------------  % Date: Fri, 02 Mar 2001 12:08:53 +1000w  From: h_ling <h.ling@qut.edu.au>9 Subject: Pls help with $PRODUCT INSTALL on this PCSI file)* Message-ID: <3A9F00B4.8FC13A03@qut.edu.au>   Hi OpenVMS gurus,i  F I got the error messages (see the last paragraph)  when doing $PRODUCT> INSTALL on the PCSI file, DEC-AXPVMS-VMS721_PCSI-V0100--4.PCSI   I downloaded the file from= http://ftp.service.digital.com/patches/public/vms/axp/v7.2-1/rC by clicking on it.  Then I ftped to the vms host using binary mode.h6 Then I changed the file attribute by using the command	 $set file E DEC-AXPVMS-VMS721_PCSI-V0100--4.PCSI/attr=(rfm:fix,lrl:8192,mrs:8192)l  H I got the same errors on the original file before the $set file command.   Thank in advance,"   How-Hie Ling (h.ling@qut.edu.au)o# Queensland University of Technology       4 LING $ dir/full DEC-AXPVMS-VMS721_PCSI-V0100--4.PCSI  ? DEC-AXPVMS-VMS721_PCSI-V0100--4.PCSI;1    File ID:  (57435,6,0) . Size:         2376/2376       Owner:    [LING]" Created:    2-MAR-2001 11:37:11.07& Revised:    2-MAR-2001 11:41:19.46 (2) Expires:   <None specified>h Backup:    <No backup recorded>  Effective: <None specified>b Recording: <None specified>p File organization:  Sequential Shelved state:      OnlineH File attributes:    Allocation: 2376, Extend: 0, Global buffer count: 0, No version limit2 Record format:      Fixed length 8192 byte records Record attributes:  None RMS attributes:     None Journaling enabled: None; File protection:    System:RWED, Owner:RWED, Group:, World:  Access Cntrl List:  None      5 LING $ product install vms721_pcsi/source=DSKZ:[LING]o %PCSI-E-READERR, error reading9 _REDASH$DKB0:[LING]DEC-AXPVMS-VMS721_PCSI-V0100--4.PCSI;1o# -DDIS-E-TNF, invalid element syntaxo" %PCSI-E-S_OPFAIL, operation failed? %PCSIUI-E-ABORT, fatal error encountered - operation terminatedt   ------------------------------  % Date: Thu, 01 Mar 2001 21:13:33 +0000s  From: steven.reece@quintiles.com) Subject: Re: Possible security hole in... H Message-ID: <OFBC198536.5FF8C755-ON80256A02.007444BB@qedi.quintiles.com>  H But hang on, it is (as you say) their right.  If they want to keep quiet then that's up to them.s  K Personally I'm kinda glad.  Non-priv'ed system crashers and severe security H stuff gets notified to customers.  Patches are posted to the website andC also to the ITS database which is available to contract maintenancehB customers.  I'd prefer to get the fixes this way without all of myI (potential ab)users finding out and crashing/corrupting my systems first.   A Sun may think of things differently, as is again their privilege.o     Sunny Andrew commented/quoted: >>>oJ > > |>   I have asked for additional details around this problem, and have taken 0 > > |>   the report and the discussion off-line. > > |>I > > |>   And please...  *** DO NOT POST SECURITY HOLES OR SYSTEM CRASHERSw ***. > >iH > > And this is the main reason why there is this perception that VMS is4 > > totally secure while the various Unixes are not. >rI > It may indeed be A reason.  But if you claim it is THE MAIN reason, youo > ought to offer some evidence.w >m: Easy, how about Compaqs own apparent policy of not posting; responses to CERT advisories which OpenVMS was vunerable to ; instead prefering to push patches out for the vunerabilites 3 quietly documenting the fixes in the patch reports.-  : I say apparent because this may not be a Compaq policy but0 it is of course what has happened none the less.  9 Certain people have taken this silence on Compaqs part toa5 be an  illustration of how secure OpenVMS is relativec7 to other OS's when in fact all Compaq seem to have dones( is excercised their right to silence.<<<   ------------------------------  % Date: Thu, 01 Mar 2001 14:27:11 -0800 . From: Jack Trachtman <Jack.Trachtman@vmmc.org> Subject: PROD Utility Failurew( Message-ID: <3A9ECCBF.19AE19B3@vmmc.org>   Alpha OpenVMS V7.2-1   Problem: PROD (ie PCSI) aborts  ? Resolution: Compaq support and DSNlink notes state that if PCSIeG databases cannot be recovered, then delete all *.PCSI$DATABASE files ino/ SYS$COMMON:[SYSEXE], then issue a PROD REG PRODw# VMS/SOURCE=SYS$COMMON:[000000] cmd.l  H The above cmd is supposed to cause the three primary PCSI database files. to be created (PCSI$FILE_SYSTEM.PCSI$DATABASE,; PCSI$PROCESSOR.PCSI$DATABASE, and PCSI$ROOT.PCSI$DATABASE).S  + These files are not being recreated for me.0  H I have also tried copying the above files and the PCSI$MAIN.EXE from theB V7.2-1 CDROM, but PCSI will abort after the first product install.   Any help would be appreciated.   ------------------------------  % Date: Thu, 01 Mar 2001 21:40:53 +0100e= From: Arne =?iso-8859-1?Q?Vajh=F8j?= <arne.vajhoej@gtech.com>s0 Subject: Re: QUESTION: BZIP2 For OpenVMS Anyone?) Message-ID: <3A9EB3D5.52CC9087@gtech.com>e   Maulis Adam wrote:k > In article <3A9A6E73.5079DDA0@gtech.com>, Arne =?iso-8859-1?Q?Vajh=F8j?= <arne.vajhoej@gtech.com> writes:,= > > I think I only changed the suffix from ".bz2"  to "_BZ2".y >  > please: -BZ2 > & > (gnu zip has "-GZ" suffix under vms)  * compress/VMS uses "_Z". RCS/VMS uses "_V".   As I recall the story then:)   - compress/unix uses ".Z"l   - compress/VMS uses "_Z"!   - Unisys makes LZW patent claimt   - gzip replaces compress   - gzip/Unix start with ".z"--   - gzip/VMS start with "-Z" (no lowercase !)m;   - the Unix world gets wiser (".Z" and ".z" are too close)c   - gzip/Unix switch to ".gz"t=   - gzip/VMS switch to "-GZ" (could have switched to "_GZ" !)f   Arne   ------------------------------  % Date: Thu, 01 Mar 2001 22:47:28 +0100i/ From: Martin Zinser <zinser@sysdev.exchange.de>a0 Subject: Re: QUESTION: BZIP2 For OpenVMS Anyone?2 Message-ID: <3A9EC370.40815236@sysdev.exchange.de>   Hello!   Arne Vajh=F8j wrote: >=20 > Maulis Adam wrote:F > > In article <3A9A6E73.5079DDA0@gtech.com>, Arne =3D?iso-8859-1?Q?V=. ajh=3DF8j?=3D <arne.vajhoej@gtech.com> writes:? > > > I think I only changed the suffix from ".bz2"  to "_BZ2".r > >a > > please: -BZ2 > >t( > > (gnu zip has "-GZ" suffix under vms) >=20, > compress/VMS uses "_Z". RCS/VMS uses "_V". >=20   Well the port at=20a/ http://www.decus.de:8080/www/vms/sw/bzip2.htmlx9; uses -bz2 to be consistent with gzip (I did not comtemplate + the long history Arne quoted at this time).t  B FYI, I'm currently getting a port of 1.0.1 ready (with some input= =20wF =66rom Jouk Jansen), but it will take about 4 weeks until it surfaces= =20o@ on the above website since I'm on an extended bussines travel=20B currently and therefore trapped behind a tight firewall currently.   =09=09=09=09Greetings, Martino   ------------------------------  % Date: Thu, 01 Mar 2001 21:33:50 +0100 = From: Arne =?iso-8859-1?Q?Vajh=F8j?= <arne.vajhoej@gtech.com> . Subject: Re: Recommend a VMS directory browser) Message-ID: <3A9EB22E.8A757231@gtech.com>s   Clayton Kroh wrote:aM > I am looking for a VMS directory browser such as something that can show me> > the directory tree structure.n > 6 > Can you recommend one along with where I can get it?  * The two best choices are FLIST and CSWING.  = FLIST is small and fast. CSWING is bigger and more powerfull.,  C You should be able to find FLIST at ftp://ftp.wku.edu/vms/fileserv/n/ and CWING at ftp://ftp.cerritos.edu/mailserv/ !E   Arne   ------------------------------  % Date: Thu, 01 Mar 2001 14:30:01 -0700a  From: l_ricker@lto.locktrack.com* Subject: Recommend a VMS directory browser. Message-ID: <01030114300130@lto.locktrack.com>  1 >> > $    OPEN/WRITE FILE sys$login:graphtree.lisc  >> > $    DEFINE sys$output FILE            ^^^^^^^^^^^^^^^^^^^^^^> >> I had to wait until I am 50 to discover that THIS exists...  @ Then you might be interested in (or reminded of) this relatively little-known DCL-ism:   .   $ foo*bar == "@dev:[dir]any_old_comfile.com"  .   $ foo /OUTPUT=output_file (... P1 P2 etc...)         ^^^^^^^^^^^^^^^^^^^ = and, 'tho I don't use it regularly, I believe this works too:o  ;   $ @any_old_comfile /OUTPUT=output_file (... P1 P2 etc...)   B Both of these generate a .LIS text-file, and have collected all of1 the com-file's redirected-SYS$OUTPUT, er, output.v  B Yes, this is documented in VMS HELP under the topic "@", but seemsD to me to be an under-utilized trick, and saves a line or two of DCL.  B The example extracted from TREE.COM above has at least one line ofC output intended to be terminal-displayed *between* the OPEN and the.E occurrence of lines which might be redirected to the file, but my own G preference is for "less chatty" output, so I find the /OUTPUT qualifier J on foreign command symbols and com-file @-invocations to be very useful...   FWIW... cheers,t   Lorink   ------------------------------  $ Date: Fri, 2 Mar 2001 12:23:15 +1100/ From: "Phil Howell" <phowell@snowyhydro.com.au>n9 Subject: Re: So what Are we meant to use on the 'desktop'n/ Message-ID: <GzCn6.672$o4.35314@ozemail.com.au>p  8 Bob Koehler <koehler@encompasserve.org> wrote in message- news:o5vBpdLFLtQM@eisner.encompasserve.org...h> > In article <nsin6.192$o4.7358@ozemail.com.au>, "Phil Howell"# <phowell@snowyhydro.com.au> writes:  >I	 > > nuTPU  > >h >e > Have you tried this? > 5 Yes - though I wouldn't claim to have tested it fullyT and it isn't cheap. I usually use telnet and tpu/eve on our alphas4 (with set keypad edt - so you can tell how old I am)8 and on the pc I used pfe32, however there were occasions> when I found myself ftping (is that a verb?) stuff to an alpha< in order to use tpu functions or procedures, notably "learn" and "box select/cut/paste"> I also feel more comfortable when I know journal/recover works Phil   ------------------------------  $ Date: Fri, 2 Mar 2001 10:48:29 +1000B From: "Matt Muggeridge" <Matt.Muggeridge@compaq.ssppaammffree.com>' Subject: Re: UCX/TCPIP 5.1 for Hobbyist-5 Message-ID: <a9Cn6.1$ou6.125@gazette.loc1.tandem.com>4  K > Did the code improve in quality through this porting effort, one wonders?:  J We share the same code pool, so common features, common bugs, common fixesC etc.  I believe this is a good thing and aids quality improvements.g   Matt.o   ------------------------------  $ Date: Thu, 1 Mar 2001 22:08:42 -05002 From: "John Gemignani, Jr." <john@ossc.DELETE.net>' Subject: Re: UCX/TCPIP 5.1 for Hobbyistt) Message-ID: <3a9f0f34@newsfeed.vitts.com>b  L "Jan Vorbrueggen" <jan@mailhost.neuroinformatik.ruhr-uni-bochum.de> wrote inJ message news:y4k869zumj.fsf@mailhost.neuroinformatik.ruhr-uni-bochum.de...6 > "John Gemignani, Jr." <john@ossc.DELETE.net> writes: > F > > Keep in mind that they don't run on VAX, but we made the code work there. >sK > Did the code improve in quality through this porting effort, one wonders?a  H     You apparently don't use TCPIP, otherwise you wouldn't be wondering.   >h( > > Compared to the Ultrix network code,H > > the Tru64 code is very demanding on memory (ithat is, VMS' non-paged pool). >h2 > That is data memory, not code memory, I presume.  L     Incorrect.  Why do you presume that it's one and not the other?  How can: support for IPv6 be added and have no affect on code size?    > Why is that?n  K     First of all, the code in TCPIP runs on Tru64 (Alpha), whereas the code J in UCX runs on Ultrix (VAX).  Much has changed.  A lot of effort went intoK performance simply due to the demands on the systems and the sheer sizes ofsL the configurations sold nowadays.  Think about the affect the WWW has placedL on the servers in FTP and HTTP services alone?  This resulted in larger dataF tables and hashes to speed up access. The network data on VMS consumesL somewhat less memory than it does on Tru64 because the implementation itself! didn't REQUIRE 64-bit addressing.r   > Is there so muchH > leeway in TCP/IP that implmentations can differ so much in complexity?  H     There is no "leeway" as far as the protocols and interfaces go.  The? internals of an implementation is what separates them.  If youreI implementation keeps a linked list of sockets and ours uses a hash to map4L incoming datagrams to their socket, which do you think will be faster?  WhatJ if there are 1,000 active sockets?  How about 10,000?  Is this really what" you mean by leeway and complexity?   -Johne   >  > JanM   ------------------------------  # Date: Thu, 01 Mar 2001 20:35:04 GMTT4 From: LESLIE@209-16-45-102.insync.net (Jerry Leslie)2 Subject: Re: VMS 7.2-1 Bugcheck In DECW$REINIT.EXE& Message-ID: <Ynyn6.67$%L5.4096@insync>  ) Andy Stoffel (acs@fcgnetworks.net) wrote:PE : I missed the original message but this looks (painfully) familiar..U : ? : I was installing a DS20E earlier this week at a customer siteo9 : and had finished installing VMS 7.2-1 and had installedg9 : the ELSA drivers from the 7.2-1 CD (This machine has a e9 : PBXGK-BB  in PCI 0 Slot 9 (physically slot PCI slot 3))  : 9 : I had a RUNNING system, everything installed so far was  : happy, including DECWindows. : A : I then proceeded to install a "bunch" of VMS patches, starting m; : with the Update v1 from 10-Nov. This first patch requires-@ : a reboot. I then spent the next eternity trying to figure out ? : WHY I was getting a bugcheck %-) at about the time DECWindowsi : would have been starting up. : D : By disabling DECWindows I was able to get my system up and runningC : (Yes, I'm leaving out the step where I booted from another device@B :  that had the same version of VMS (but working) on it so I could> : poke around on the misbehaving system disk..) On a hunch, I @ : reinstalled the ELSA drivers from the install CD and rebooted < : and had my working system back. And then moved on to more  : productive activities....o : B : [Note: Apparently, there is a newer patch (GRAPHICS ?) for 7.2-1@ :   that has a newer version of the drivers that I installed andA : suspect to be the culprit but I wasn't aware of it at the time.f7 : Also, I've installed DS20E's with the same video carda? : before in the past and haven't had this problem. And a fellowaC : employee at my company installed the same hardware a week earliere> : and didn't have this problem, but he did the SYSUPDATE patch* : FIRST and THEN did the ELSA drivers....] :  : -Andy- :     B The system, a DPW 500au, still crashes on a reboot after applying 
 these patchs:      vms721_graphics-v0400o  C and the prerequisite patches, which were installed in the followingv order, are:h     o vms721_update-v0100e   o vms721_pcsi-v0100b   o vms721_cpu2208-v0100  > The VMS721_GRAPHICS-V0400.README file contained the following,? which indicated that the  VMS721_GRAPHICS-V0400 patch had a fixd for this problem:i  4   8  PROBLEMS ADDRESSED IN VMS721_GRAPHICS-V0300 KIT    H       o  A reboot on  a  DPWS  5/500,  5/600  or  5/433  system  with  aH          PowerStorm  4D10T  card  will  crash.  The system, must be shut3          down, a console init done, and the booted.t            Image(s) Affected:s  &           -  [SYS$LDR]SYS$GZDRIVER.EXE  '           -  [SYS$LDR]SYS$GZVDRIVER.EXEa  3 The system is a Digital Personal Workstation 500au:a     $ product show product *>   ----------------------------------- ----------- ------------7   PRODUCT                             KIT TYPE    STATE$>   ----------------------------------- ----------- ------------;   DEC AXPVMS CXML V3.59-1             Full LP     InstalledS;   DEC AXPVMS DECNET_PHASE_IV V7.2-1   Full LP     Installed$;   DEC AXPVMS DWMOTIF V1.2-5           Full LP     Installede;   DEC AXPVMS FORRTL V7.3-1            Full LP     Installed ;   DEC AXPVMS FORTRAN V7.3-1           Full LP     Installed-;   DEC AXPVMS NS_NAV_EXPORT V3.3       Full LP     Installed ;   DEC AXPVMS OPENVMS V7.2-1           Platform    Installedt;   DEC AXPVMS TCPIP V5.0-10            Full LP     Installed ;   DEC AXPVMS VMS V7.2-1               Oper System Installed >   ----------------------------------- ----------- ------------     9 items founde     $ tcpip show version  C     DIGITAL TCP/IP Services for OpenVMS Alpha Version V5.0A - ECO 2.=     on a Digital Personal WorkStation  running OpenVMS V7.2-1   ? The output from "CLUE CRASH" and "CLUE CONFIG" follow the .sig.E  / --Jerry Leslie   leslie@209-16-45-97.insync.net ;                  leslie@209-16-45-102.insync.net is invalid=2                  (my opinions are strictly my own)N ==============================================================================   SDA> clue crashg   Crashdump Summary Information:  -----------------------------* Crash Time:         1-MAR-2001 11:54:59.43; Bugcheck Type:     INVEXCEPTN, Exception while above ASTDELg' Node:              OLYWS3  (Standalone)P/ CPU Type:          Digital Personal WorkStation2 VMS Version:       V7.2-1o Current Process:   SYSTEMpI Current Image:     OLYWS3$DKA0:[SYS0.SYSCOMMON.][SYSEXE]DECW$REINIT.EXE;1i: Failing PC:        FFFFFFFF.80007268    OTS$REM_UL_C+000B8$ Failing PS:        00000000.00001504N Module:            SYS$BASE_IMAGE    (Link Date/Time: 29-DEC-1999 03:52:51.75) Offset:            00005268a  * Boot Time:          1-MAR-2001 11:46:14.00* System Uptime:               0 00:08:45.43 Crash/Primary CPU: 00/00 System/CPU Type:   1E05t Saved Processes:   22 ' Pagesize:          8 KByte (8192 bytes)1< Physical Memory:   256 MByte (32768 PFNs, contiguous memory) Dumpfile Pagelets: 35349 blocks3: Dump Flags:        olddump,writecomp,errlogcomp,dump_style' Dump Type:         compressed,selective6) EXE$GL_FLAGS:      poolpging,init,bugdump 6 Paging Files:      1 Pagefile and 1 Swapfile installed   Stack Pointers:eK KSP = 00000000.7FFA1888   ESP = 00000000.7FFA6000   SSP = 00000000.7FFAC100t USP = 00000000.7AF059E03   General Registers:K R0  = 00000000.00000000   R1  = 00000000.00000484   R2  = 00000000.7FFA1A60eK R3  = FFFFFFFF.9042DF20   R4  = 00000000.7FFA18E0   R5  = 00000000.7FFA1A50lK R6  = 00000000.7FFA1A80   R7  = 00000000.00001504   R8  = 00000000.00000000sK R9  = 00000000.7AF05B08   R10 = 00000000.0004E658   R11 = 00000000.0004E660GK R12 = 00000000.00000000   R13 = FFFFFFFF.9E7EEAF0   R14 = 00000000.00000000-K R15 = FFFFFFFF.90386760   R16 = 00000000.000001CC   R17 = 00000000.7FFA18E0rK R18 = 00000000.7FFA1A80   R19 = 00000000.9E7EEA60   R20 = 00000000.9E7EEA60tK R21 = 00000000.00000600   R22 = 00000000.00000000   R23 = 00000000.00000000 K R24 = 00000000.00000000   AI  = 00000000.00000002   RA  = FFFFFFFF.9E7E8070 K PV  = FFFFFFFF.903AA390   R28 = FFFFFFFF.FFFFFE00   FP  = 00000000.7FFA1B40u1 PC  = FFFFFFFF.800814C8   PS  = 08000000.00001500    Crashdump Summary Information:  -----------------------------   Exception Frame:K R2  = 00000000.00000000   R3  = FFFFFFFF.810A0200   R4  = 00000000.00000015tK R5  = FFFFFFFF.810A0200   R6  = 00000000.00000015   R7  = FFFFFFFF.90386760l1 PC  = FFFFFFFF.80007268   PS  = 00000000.00001504t  = Signal Array:                            64-bit Signal Array:iK Arg Count    = 00000003                  Arg Count      =          00000003gK Condition    = 00000484                  Condition      = 0000048C.00000484eK Argument #2  = 80007268                  Argument #2    = FFFFFFFF.80007268nK Argument #3  = 00001504                  Argument #3    = 00000000.00001504    Mechanism Array:K Arguments    = 0000002C                  Establisher FP = 00000000.7FFA1B40iK Flags        = 00000000                  Exception FP   = 00000000.7FFA1A80eK Depth        = FFFFFFFD                  Signal Array   = 00000000.7FFA1A50yK Handler Data = 00000000.7FFA1A10         Signal64 Array = 00000000.7FFA1A60aK R0  = 00000000.00000000   R1  = 00000000.00000000   R16 = FFFFFFFF.FFFFFFFEdK R17 = 00000000.00000000   R18 = 00000000.00000002   R19 = 00000000.9E7EEA60uK R20 = 00000000.9E7EEA60   R21 = 00000000.00000600   R22 = 00000000.00000000-K R23 = 00000000.00000000   R24 = 00000000.00000000   R25 = 00000000.00000002sK R26 = FFFFFFFF.9E7E8070   R27 = FFFFFFFF.903AA390   R28 = FFFFFFFF.FFFFFE00o   System Registers:eK Page Table Base Register (PTBR)                           00000000.000033E3aK Processor Base Register (PRBR)                            FFFFFFFF.81012000 K Privileged Context Block Base (PCBB)                      00000000.07204080mK System Control Block Base (SCBB)                          00000000.000001E1eK Software Interrupt Summary Register (SISR)                00000000.00000000OK Address Space Number (ASN)                                00000000.00000066 K AST Summary / AST Enable (ASTSR_ASTEN)                    00000000.0000000FsK Floating-Point Enable (FEN)                               00000000.00000001:K Interrupt Priority Level (IPL)                            00000000.00000015aK Machine Check Error Summary (MCES)                        00000000.00000000PK Virtual Page Table Base Register (VPTB)                   FFFFFFFC.00000000r     Crashdump Summary Information:  ----------------------------- Failing Instruction:1 OTS$REM_UL_C+000B8:     RET             R31,(R26)   * Instruction Stream (last 20 instructions):1 OTS$REM_UL_C+00068:     MULQ            R0,R17,R0,1 OTS$REM_UL_C+0006C:     SUBQ            R16,R0,R0-1 OTS$REM_UL_C+00070:     RET             R31,(R26)R4 OTS$REM_UL_C+00074:     BEQ             R17,#X00000C1 OTS$REM_UL_C+00078:     AND             R16,R1,R0 1 OTS$REM_UL_C+0007C:     RET             R31,(R26)i2 OTS$REM_UL_C+00080:     BIS             R31,R16,R01 OTS$REM_UL_C+00084:     RET             R31,(R26)b2 OTS$REM_UL_C+00088:     CMPULT          R16,R17,R12 OTS$REM_UL_C+0008C:     SUBQ            R16,R17,R01 OTS$REM_UL_C+00090:     CMOVNE          R1,R16,R0n1 OTS$REM_UL_C+00094:     RET             R31,(R26)u3 OTS$REM_UL_C+00098:     BIS             R31,R31,R31p4 OTS$REM_UL_C+0009C:     BSR             R28,#XFFFED21 OTS$REM_UL_C+000A0:     BIS             R31,R1,R0h1 OTS$REM_UL_C+000A4:     RET             R31,(R26)e7 OTS$REM_UL_C+000A8:     LDA             R16,#XFFFE(R31)o2 OTS$REM_UL_C+000AC:     BIS             R31,R31,R02 OTS$REM_UL_C+000B0:     BIS             R31,R31,R1 OTS$REM_UL_C+000B4:     GENTRAPn1 OTS$REM_UL_C+000B8:     RET             R31,(R26)Q OTS$REM_UL_C+000BC:     HALT, OTS$EXTV_C:     SRA             R17,#X03,R274 OTS$EXTV_C+00004:       BLE             R18,#X0000153 OTS$EXTV_C+00008:       ADDQ            R16,R27,R16@ SDA>N ============================================================================== SDA> clue config   System Configuration:m ---------------------  System Information: G System Type    Digital Personal WorkStation           Primary CPU ID 00 N Cycle Time     2.0 nsec (500 MHz)                     Pagesize       8192 Byte   Memory Configuration: D Cluster    PFN Start    PFN Count         Range (MByte)        UsageF  #00             0          243         0.0 MB -     1.8 MB    ConsoleE  #01           243        32524         1.8 MB -   255.9 MB    SystemiF  #02         32767            1       255.9 MB -   256.0 MB    Console  # Per-CPU Slot Processor Information:0K CPU ID         00                        CPU State    rc,pa,pp,cv,pv,pmv,pl9N CPU Type       EV56  (21164A)            Halt Request "Warm Bootstrap Request" PAL Code       1.20                      Halt PC      00000000.20000000CPU Revision   ....                      Halt PS      00000000.00001F00N Serial Number  ..........                Halt Code    "Bootstrap or Powerfail" Console Vers   V7.1-3c         Adapter Configuration: ----------------------w TR Adapter     ADP               Hose Bus   BusArrayEntry     Node CSR            Vec/IRQ Port Slot Device Name / HW-IdL| -- ----------- ----------------- ---- ----------------------- ---- ---------------------- ---- ---- ------------------------ ---n4  1 KA1E07      FFFFFFFF.81041CC0    0 BUSLESS_SYSTEM)  2 PCI         FFFFFFFF.81041EC0    0 PCIax                                             FFFFFFFF.81042140   18 FFFFFFFF.9D2A0000    0 EWA:    3 DE500-BA 100 mbit NIi                                             FFFFFFFF.81042220   38 FFFFFFFF.9D2A2000    0         7 MFPCIEt                                             FFFFFFFF.81042300   58 FFFFFFFF.9D4AE000   10 GZA:   11 PowerStorm 4D10Th                                             FFFFFFFF.810424F8   A0 FFFFFFFF.9D4B0000    0        20 PPB3)  3 PCI         FFFFFFFF.81042780    0 PCI k                                             FFFFFFFF.81042998   38 FFFFFFFF.9D2A4000    0         7 CYPRESS n                                             FFFFFFFF.810429D0   39 FFFFFFFF.9D4A8000   38 DQA:    7 IDE drivern                                             FFFFFFFF.81042A08   3A FFFFFFFF.9D4AA000   3C DQB:    7 IDE driverh                                             FFFFFFFF.81042A40   3B FFFFFFFF.9D4AC000   28         7 USBC)  4 ISA         FFFFFFFF.81042B80    0 ISAcu                                             FFFFFFFF.81042D98    0 FFFFFFFF.9D2A6000    0         0 EISA_SYSTEM_BOARD *  5 XBUS        FFFFFFFF.81043140    0 XBUSh                                             FFFFFFFF.81043318    0 FFFFFFFF.9D2A6000    C         0 MOUSg                                             FFFFFFFF.81043350    1 FFFFFFFF.9D2A6000    1         1 KBD-h                                             FFFFFFFF.81043388    2 FFFFFFFF.9D2A6000    4         2 COM1o                                             FFFFFFFF.810433C0    3 FFFFFFFF.9D2A6000    3 TTA:    3 Serial Port.|                                             FFFFFFFF.810433F8    4 FFFFFFFF.9D2A6000    7 LRA:    4 Line Printer (parallel p ort)j                                             FFFFFFFF.81043430    5 FFFFFFFF.9D2A6000    6 DVA:    5 Floppy|                                             FFFFFFFF.81043468    6 FFFFFFFF.9D2A6000    E         6 00000000.45444945 (EIDE)|                                             FFFFFFFF.810434A0    7 FFFFFFFF.9D2A6000    5         7 00003838.38315345 (ES188 8){                                             FFFFFFFF.810434D8    8 FFFFFFFF.9D2A6000    A         8 00000000.00425355 (USB)t)  6 PCI         FFFFFFFF.810440C0    0 PCI y                                             FFFFFFFF.81044378  120 FFFFFFFF.9D4B2000    C PKA:    4 Qlogic ISP1020 SCSI-20 SDA>   ------------------------------  # Date: Fri, 02 Mar 2001 02:15:11 GMT.2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman)2 Subject: Re: VMS 7.2-1 Bugcheck In DECW$REINIT.EXE5 Message-ID: <PmDn6.3$ou6.186@gazette.loc1.tandem.com>g  ] In article <Ynyn6.67$%L5.4096@insync>, LESLIE@209-16-45-102.insync.net (Jerry Leslie) writes:x    #   Increase NPAGEDYN, and try again.a   .. :u :Crashdump Summary Information:  :------------------------------ + :Crash Time:         1-MAR-2001 11:54:59.43J< :Bugcheck Type:     INVEXCEPTN, Exception while above ASTDEL( :Node:              OLYWS3  (Standalone)0 :CPU Type:          Digital Personal WorkStation :VMS Version:       V7.2-1 :Current Process:   SYSTEMJ :Current Image:     OLYWS3$DKA0:[SYS0.SYSCOMMON.][SYSEXE]DECW$REINIT.EXE;1; :Failing PC:        FFFFFFFF.80007268    OTS$REM_UL_C+000B8 % :Failing PS:        00000000.00001504AO :Module:            SYS$BASE_IMAGE    (Link Date/Time: 29-DEC-1999 03:52:51.75)g :Offset:            00005268    N  ---------------------------- #include <rtfaq.h> -----------------------------N       For additional, please see the OpenVMS FAQ -- www.openvms.compaq.com    N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------  $ Date: Thu, 1 Mar 2001 15:35:54 -0800 From: "Raziel" <me@vrwc.org>! Subject: vms programming question 1 Message-ID: <97mm92$orc$1@nntp6.u.washington.edu>U  G I am trying to convert floating point numbers from 4 byte VAX to 4 bytefE IEEE.  The VAX Alpha running v7.2-1 with DEC C v5.5-002.  The programsK compiles and links (cc program_name and link program_name) fine but crashesdI in the function call that is supposed to do the conversion (CVT$FTOF).  Ia< have included the code below, any help would be appreciated.   jims     #include <stdio.h>  ! extern unsigned long CVT$K_VAX_F;i" extern unsigned long CVT$K_IEEE_S;$ extern unsigned long CVT$M_TRUNCATE;( extern unsigned long CVT$M_ROUND_TO_POS;   unsigned long CVT$FTOF(m-  void *input_value, unsigned long input_type,t/  void *output_value, unsigned long output_type,f  unsigned long options);  
 int main() {   int i;   float fVal = 6;  unsigned long iVal = 0;  unsigned char *ch;h    i = 0;n  printf("%x\n", fVal);    ch = (unsigned char*)&fVal;  for(i=0; i < 4; i++)t   printf("%x\n", ch[i]);  &  CVT$FTOF((void*)(&fVal), CVT$K_VAX_F,5    (void*)(&iVal), CVT$K_IEEE_S, CVT$M_ROUND_TO_POS);   
  return 0; }-   --A "Those who would give up essential liberty for temporary securityL; deserve neither liberty nor security." -- Benjamin Franklin7   Jim GrossnickleC UW Redmond Plasma Physics Lab6 425-881-7706   ------------------------------   Date: 1 Mar 2001 19:27:43 -0500o, From: maulis@encompasserve.org (Adam Maulis)% Subject: Re: vms programming questionE3 Message-ID: <VPg9oLcWnHFU@eisner.encompasserve.org>i  P In article <97mm92$orc$1@nntp6.u.washington.edu>, "Raziel" <me@vrwc.org> writes: >  >  > #include <stdio.h>   > # > extern unsigned long CVT$K_VAX_F;o$ > extern unsigned long CVT$K_IEEE_S;& > extern unsigned long CVT$M_TRUNCATE;* > extern unsigned long CVT$M_ROUND_TO_POS; >  > unsigned long CVT$FTOF(S/ >  void *input_value, unsigned long input_type,01 >  void *output_value, unsigned long output_type,n >  unsigned long options);   /*   forget it.   use:   */ #include cvt$routines  #include cvtdefC    /* and your program will work */ >  > Jim Grossnickleg > UW Redmond Plasma Physics Labs > 425-881-7706 >  >   @ Your program was crashed _before_ invoking the CVT$FTOF routine.@ Crashed while trying think out what is the value of CVT$K_VAX_F.  @ You defined CVT$K_VAX_F as a variable, but it never initialized.< (and about 10 line of explanation, sorry it's 01:25 AM here, I'm so tied)    H Adam Maulis              maulis@ludens elte hu        VMS system managerH ........................................................................H VMS Competence Center                              VMS Szakertoi KozpontH Eotvos Lorand University                   Eotvos Lorand TudomanyegyetemH Budapest, Hungary                                               BudapestH ========================================================================   ------------------------------  # Date: Fri, 02 Mar 2001 00:43:13 GMTd' From: Colin Blake <colin@theblakes.com>d% Subject: Re: vms programming question , Message-ID: <3A9EEC75.CD5F78C@theblakes.com>  
 Raziel wrote:a  I > I am trying to convert floating point numbers from 4 byte VAX to 4 bytepG > IEEE.  The VAX Alpha running v7.2-1 with DEC C v5.5-002.  The program1M > compiles and links (cc program_name and link program_name) fine but crashesGK > in the function call that is supposed to do the conversion (CVT$FTOF).  I > > have included the code below, any help would be appreciated.   If you replace this:  # > extern unsigned long CVT$K_VAX_F; $ > extern unsigned long CVT$K_IEEE_S;& > extern unsigned long CVT$M_TRUNCATE;* > extern unsigned long CVT$M_ROUND_TO_POS; >  > unsigned long CVT$FTOF(,/ >  void *input_value, unsigned long input_type, 1 >  void *output_value, unsigned long output_type,Z >  unsigned long options);   with:Y   #include <cvtdef.h>  #include <cvt$routines.h>l   then the program doesn't crash.    ------------------------------  $ Date: Thu, 1 Mar 2001 18:13:54 -0800 From: "Raziel" <me@vrwc.org>% Subject: Re: vms programming question-1 Message-ID: <97mvhb$dga$1@nntp6.u.washington.edu>-  B Thanks to both of you.  Program works great now.   My examples and= documentation said nothing about including those two headers.E   thanks again   jim     ' "Raziel" <me@vrwc.org> wrote in messageS+ news:97mm92$orc$1@nntp6.u.washington.edu...eI > I am trying to convert floating point numbers from 4 byte VAX to 4 byteVG > IEEE.  The VAX Alpha running v7.2-1 with DEC C v5.5-002.  The programVE > compiles and links (cc program_name and link program_name) fine but  crasheslK > in the function call that is supposed to do the conversion (CVT$FTOF).  Ie> > have included the code below, any help would be appreciated. >  > jimt >  >n > #include <stdio.h> >i# > extern unsigned long CVT$K_VAX_F; $ > extern unsigned long CVT$K_IEEE_S;& > extern unsigned long CVT$M_TRUNCATE;* > extern unsigned long CVT$M_ROUND_TO_POS; >  > unsigned long CVT$FTOF(l/ >  void *input_value, unsigned long input_type,41 >  void *output_value, unsigned long output_type,- >  unsigned long options); >  > int main() > {i	 >  int i;t >  float fVal = 6; >  unsigned long iVal = 0; >  unsigned char *ch;= >=	 >  i = 0;= >  printf("%x\n", fVal); >  >  ch = (unsigned char*)&fVal; >  for(i=0; i < 4; i++)- >   printf("%x\n", ch[i]); >-( >  CVT$FTOF((void*)(&fVal), CVT$K_VAX_F,7 >    (void*)(&iVal), CVT$K_IEEE_S, CVT$M_ROUND_TO_POS);O >S >  return 0; > }P >y > --C > "Those who would give up essential liberty for temporary securityr= > deserve neither liberty nor security." -- Benjamin FranklinS >M > Jim GrossnickleI > UW Redmond Plasma Physics LabF > 425-881-7706 >T >E >L   ------------------------------  $ Date: Thu, 1 Mar 2001 17:55:16 -0500  From: John Santos <JOHN@egh.com>< Subject: VMS V7.2+ bug - directory modification date changes6 Message-ID: <1010301173848.35173B-100000@Ives.egh.com>  ? I have discovered that in VMS (7.2-1 on Alpha, 7.2 on VAX), the1D modification date on a directory file changes whenever the directoryA is extended or truncated by creating or deleting files in it.  NopB explicit change to the directory, just create or delete some files in it.  8 Since the point when VMS decides to extend or truncate aC directory seems completely random to a user (who is not supposed to B be aware of the undocumented internals of directory formats), this1 behaviour is completely random and unpredictable.0  ? Of course, if the directory in question is [000000] and someone0B creates a new top-level directory that extends [000000]000000.dir,< this will trigger the notorious "How come all of a sudden myC incremental backups suddenly are backing up EVERY file on my disk?"0B "feature".  (Ironically, the command to fix incremantal backups is backup/noincremental.)  B To demonstrate, create a subdirectory, check is modified date, andE then create a bunch of files in it and check the modified date again.0   $ create/dir [.foo] $ $ dir/size=all/date=modified foo.dir< $ copy nl: [.foo]abcdefghijklmnop.aaaaaaaaaaaaaabbbbbbbbbbbb@ [repeat dir and copy until the size changes and you will see the date change at the same time]0  D It helps to use long, different filenames for each file because that+ fills up the directory blocks most quickly.F  F Check on a VAX running V7.1 and it works correctly.  The modified dateB only gets change when you explicitly change the .dir file, e.g. by changing its protection code.0   -- 4 John Santos0 Evans Griffiths & Hart, Inc. 781-861-0670 ext 539   ------------------------------  % Date: Thu, 01 Mar 2001 21:57:37 -060007 From: "David J. Dachtera" <djesys.nospam@earthlink.net>-$ Subject: Re: [info] PCSI restriction- Message-ID: <3A9F1A31.2EE5BE30@earthlink.net>    Didier Morandi wrote:= > J > (I added the keyword because I saw this in f.c.s.m (fr.comp.sys.mac, oneH > of the fastest growing forum ever) and I think it could be helpful for7 > those who use MacSoup or other powerful newsreaders).0 > J > We discussed already here on the fact that, in the PCSI philosophy, only$ > one target disk can be specified.    Well, in theory, given:   9 $ define pcsi_destination disk1:[target1],disk2:[target2]   . [target1] exists on disk1:, but not on disk2:.. [target2] exists on disk2:, but not on disk1:.  ? PRODUCT INSTALL is invoked using /DESTINATION=PCSI_DESTINATION:0  A Then, files destined for the [target1] directory should end up on H disk1:. Similarly, files destined for the [target2] directory should end
 up on disk2:.0  E ...in theory. However, depending upon how PCSI$MAIN processes logical0= names and file specifications, search lists in a /DESTINATION8. specification may or may not work as expected.  D I haven't "cracked the PCSI code" yet, so I can't test that for you.  B If you find it does work, however, "jacketing" the PRODUCT INSTALLE command in, say, an INSTALL.COM procedure (shades of DOS INSTALL.BAT,BA .BATMAN!) would allow for target disk selection prior to actuallyB@ executing the installation, and PCSI should correctly record the* destination device/directory of each file.  
 ...in theory.0   -- 0 David J. Dachterau dba DJE Systems  http://www.djesys.com/  : Unofficial Affordable OpenVMS Home Page and Message Board: http://www.djesys.com/vms/soho/s  F This *IS* an OpenVMS-related newsgroup. So, a certain bias in postings is to be expected.  @ Feel free to exercise your rights of free speech and expression.  F However, attacks against individual posters, or groups of posters, are strongly discouraged.    ------------------------------   Date: 01 Mar 2001 19:37:50 GMT From: ka2doug@cs.com (KA2DOUG)* Subject: Re: [Request] f$element delimiter> Message-ID: <20010301143750.23785.00000007@ng-cn1.news.cs.com>  @ >because all the code which handles a single "=" is ready, and IG >discovered that I also need to handle "==" and I don't wish to rewrite$C >the whole stuff because I am lazy .-9  ooops swiss keyboard... :-)  >E >KA2DOUG wrote:, >> 0- >> Why not use f$locate("@",line) to do this?, >R  L I doubt that the f$element logic will be changed because it works exactly as& intended. Here's some code that works:   $ line = "CMD :==@command.com" $ AtSign = f$locate("@",line)M( $ if AtSign+1 .lt. f$length(line) then -9  whatsit = f$extract(AtSign+1,f$length(line)-AtSign,line)  $ write sys$output whatsit    -Doug  0   ------------------------------   End of INFO-VAX 2001.121 ************************