1 INFO-VAX	Sun, 11 Mar 2001	Volume 2001 : Issue 139       Contents:' Re: - OpenVMS ever to be on Intel chip? ' Re: - OpenVMS ever to be on Intel chip? ' Re: - OpenVMS ever to be on Intel chip?  Re: Advocacy Website Woes  Re: Advocacy Website Woes # Re: blade vs. DS10 on "mybenchmark"   CNET OVMS Review: Closing in 200  Disk Read Rate / Disk Write Rate' 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 J Dumbing Down VMS with UNIX Elements? (was Re: OpenVMS Educational Program)/ Re: In-memory layout of PACKED ARRAY OF BOOLEAN  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 N Re: Mimer (database engine)  for OpenVMS (and others) is available fordownload& Re: New memory in VAX 4000/90 not seen& Re: New memory in VAX 4000/90 not seen' Re: Newbie question - VMS Documentation 2 Re: open vms newbie is looking far an DCL emulator Re: OpenVMS Educational Program  Re: OpenVMS Educational Program  Re: OpenVMS Educational Program  Re: OpenVMS Educational Program  Re: OpenVMS Educational Program  Re: OpenVMS Educational Program  Re: OpenVMS Educational Program  Re: OpenVMS Educational Program  Re: OpenVMS Educational Program P Re: Poor Administration? (was) Re: Russian Mafia Increases TCO Of M$Products <VBP Re: Poor Administration? (was) Re: Russian Mafia Increases TCO Of M$Products <VB VMS release dates  Re: VMS release dates  Re: VMS release dates   Re: [SURVEY] Info-VAX suscribers  F ----------------------------------------------------------------------  % Date: Sat, 10 Mar 2001 19:52:38 -0600 7 From: "David J. Dachtera" <djesys.nospam@earthlink.net> 0 Subject: Re: - OpenVMS ever to be on Intel chip?- Message-ID: <3AAADA66.EC54EF43@earthlink.net>    JF Mezei wrote:  >  > "David J. Dachtera" wrote:F > > On the radio on the way to work today, I heard that some couple ofK > > million credit card numbers have been purloined by hackers exploiting a 5 > > well known and long-standing security hole in NT.  > J > Now that NT is standard and that corporations have invested so much timeO > setting up real applications on NT and thus dependant and unlikely to change, U > Microsoft will do what it takes to convince them to upgrade to the latest releases.  > O > I would not be surprised if that latest hacker news about stolen credit cards M > might have been unofficially started by Microsoft. Think about it. The fact I > that the media focus on "older version of Microsoft's operating system" N > provides strong incentive to update to the current version and hence give MS
 > more money.   E But, what reason is there to believe that W2K has fixed this, or ever  will?    --   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.    ------------------------------  # Date: Sun, 11 Mar 2001 05:56:53 GMT + From: "Darren Peacock" <daz005@hotmail.com> 0 Subject: Re: - OpenVMS ever to be on Intel chip?> Message-ID: <FsEq6.10155$0N3.67186@news-server.bigpond.net.au>  K Cant remember the detail but back in 1988-89 there was a project in Digital  to do just that ..6 My memory is fuzzy but it was called Emerald or Gem ..  It was descibed as VMS on Intel.  J But the initial downsizing then , the project was halted. Funny to see theJ following  24 months some of the members end up in a small word processing company called Microsoft.       B "David J. Dachtera" <djesys.nospam@earthlink.net> wrote in message' news:3AA9AC35.5F986D50@earthlink.net...  > David Mathog wrote:  > > H > > In article <905F6E4CFmedleybev1net@158.234.153.155>, medleyb@ev1.net (Bert Medley) writes: J > > >Or is there a possible O/S toolkit to build an emulator on top of Dos like > > >Cygwin for UNIX?  > > J > > There's a VAX emulator around which you can run on windows and run VMS > > within it. > > H > > Or you can wait for Microsoft to finish cannibalizing VMS technology/ > > which should be done by Windows 2010 or so.  > ; > Hhmmm... Maybe by then it will have some decent security!  > D > On the radio on the way to work today, I heard that some couple ofI > million credit card numbers have been purloined by hackers exploiting a 3 > well known and long-standing security hole in NT.  >  > -- > David J. Dachtera  > dba DJE Systems  > http://www.djesys.com/ > < > Unofficial Affordable OpenVMS Home Page and Message Board:! > http://www.djesys.com/vms/soho/  > H > This *IS* an OpenVMS-related newsgroup. So, a certain bias in postings > is to be expected. > B > Feel free to exercise your rights of free speech and expression. > H > However, attacks against individual posters, or groups of posters, are > strongly discouraged.    ------------------------------  # Date: Sun, 11 Mar 2001 06:53:41 GMT 4 From: "Terry C. Shannon" <terryshannon@mediaone.net>0 Subject: Re: - OpenVMS ever to be on Intel chip?; Message-ID: <VhFq6.9855$5f.2979544@typhoon.ne.mediaone.net>   6 "Darren Peacock" <daz005@hotmail.com> wrote in message8 news:FsEq6.10155$0N3.67186@news-server.bigpond.net.au...E > Cant remember the detail but back in 1988-89 there was a project in  Digital  > to do just that ..8 > My memory is fuzzy but it was called Emerald or Gem .." > It was descibed as VMS on Intel. > L > But the initial downsizing then , the project was halted. Funny to see theL > following  24 months some of the members end up in a small word processing > company called Microsoft.  >   J Well, I dragged Charlie Matco away from his usual pursuits of wine, women,C and rumourmongering and persuaded him to come clean on this matter.   J There was in fact a DEC project called EMERALD back in the late 1980s. ItsH goal: to boldly send VMS where it had not gone before... into IA32-land.J EMERALD was scuttled right around the same time the PRISM RISC project wasI killed (late March 1988). Prototype PRISM processors existed at the time, J but apparently there were Big Delays with the complementary MICA operatingH system. MICA was, simply stated, a reimplementation of VMS for the PRISMI RISC architecture. Dave Cutler flew the coop right after Ken Olsen pulled I the plug on PRISM/MICA. Word has it that a lot of the MICA code rose from " the dead when Windows NT was born.  L Separately, there was a midnight project to port VMS to the Mach kernel. TheK project was done (half-baked, actually) at Carnegie Mellon University IIRC. H Some of the incomplete code--which may well have a few facets of Emerald8 embedded in it--is said to be floating around somewhere.  G And that's all I got from Charlie. Heck, he started mumbling some stuff K about OZIX, MERLIN, QUARTZ, CHEYENNE, and other cryptic codewords from days J gone by. Wish he'd be a tad more forward-looking and spill the beans about4 MARVEL... and the system a generation beyond MARVEL.   cheers,    Matco's Handler    ------------------------------  # Date: Sat, 10 Mar 2001 22:39:31 GMT + From: Jeff Campbell <jcampbell@ins-msi.com> " Subject: Re: Advocacy Website Woes+ Message-ID: <3AAAA792.9A730C46@ins-msi.com>    "Karl S. Erbland" wrote: > > > In article <tq7q6.1722$G76.2745936@typhoon.ne.mediaone.net>,# > terryshannon@mediaone.net says... L > > Nope, but the design and management of the site is under the aegis of anP > > outside contractor. Complaints or suggestions re: the site are best directed( > > to webmaster@compaqworkinggroup.org. > J > Terry, why don't you just pass our comments along to the folks who hired > the outside contractor.  > D > Suggestion: Hire another contractor with knowledge and experience.   Another suggestion:   < Shouldn't the "requirements" portion of the contract between compaqworkinggroup.comK (or whoever) and the contractor state that all page designs provided by the 
 contractorK function correctly will *all* browsers available for Compaq platforms? This  would - cover Lynx, Netscape x.x for VMS, Tru64, etc.   
 Jeff Campbell  n8wxs@arrl.net   ------------------------------    Date: 10 Mar 2001 17:30:58 -05009 From: Kilgallen@eisner.decus.org.nospam (Larry Kilgallen) " Subject: Re: Advocacy Website Woes3 Message-ID: <cStA$EFVG1JW@eisner.encompasserve.org>   Y In article <3AAAA792.9A730C46@ins-msi.com>, Jeff Campbell <jcampbell@ins-msi.com> writes:    > Another suggestion:  > > > Shouldn't the "requirements" portion of the contract between > compaqworkinggroup.comM > (or whoever) and the contractor state that all page designs provided by the  > contractorM > function correctly will *all* browsers available for Compaq platforms? This  > would / > cover Lynx, Netscape x.x for VMS, Tru64, etc.    But it would also cover:  1 	Larry's Extremely Poor Browser Only For Wildfire   % and mandate extra effort to no avail.   ; I think requirements can only be based on an explicit list.    ------------------------------  # Date: Sun, 11 Mar 2001 05:26:53 GMT + From: "Darren Peacock" <daz005@hotmail.com> , Subject: Re: blade vs. DS10 on "mybenchmark"> Message-ID: <x0Eq6.10100$0N3.66743@news-server.bigpond.net.au>   Well, L          Good news , I just tuned my Alphastation 200/233 with a single scsi diskI I changed the block size on the and the block read size from the disk and K block write size, I also changed the  My benchmark program to utlize an rms K call rather than the standrad C , that way the UNix write and the VMS write 
 are the same. 4 And the figures exceeded the little blade exceeded .  J Have seen this benchmark posted before, and I mentioned that it was not anJ apples to apples, the easiest thing to do is set the VMS system to reflect the UNix Op setup.          ? "David Mathog" <mathog@seqaxp.bio.caltech.edu> wrote in message & news:988qcu$h8t@gap.cco.caltech.edu...; > Disk IO benchmarks are now in for the Sun Blade 100 using L > "mybenchmark" (http://seqaxp.bio.caltech.edu/pub/software/mybenchmark.zip)? > and have been added to the table.  My thanks to Brian Stretch J > (bstretch@mindspring.com) and Paul Joslin (paul.joslin@suncor3.sdrc.com)I > for running these.  Results for a VAX 4000 were reported "Barry Treahy, 2 > Jr." <treahy@mmaz.com> and have also been added. > = > Note that contrary to some of my former statements the Unix D > machines are not (apparently) all using file caching to obtain theK > speeds obtained on /tmp - but caching on the drive itself is enabled. The J > test file created by maketest is 8326000 bytes long - do the math to get the  > actual write speeds observed.  > G > The STREAM benchmark for a Blade 100 is about 4 times lower than on a  DS10, I > but the $1000 "vanilla" blade still moves data to disk (or at least the K > memory cache on the disk) many times faster than heavily optimized IO can J > on a DS10 running VMS.  And it does it using an ATA disk. A DS10 runningK > Linux or Tru64 - and using uw or u2w scsi, is faster than a blade though.  > (No big surprise there.) > L > I don't expect shame (or anything else) from Compaq management, but I hope? > that the folks in VMS engineering are suitably red faced that G > this el cheapo Sun system just wiped the floor with the VMS DS10 with # > respect to disk IO.  (And price.)  > 
 > Regards, >  > David Mathog > mathog@seqaxp.bio.caltech.edu @ > Manager, sequence analysis facility, biology division, CaltechL > ************************************************************************** > 9 > This is a table showing MYBENCHMARK speeds on a variety ; > of platforms.  Times shown are in seconds.  The tests are : > from the MYBENCHMARK suite which measure various aspects- > of "disk" I/O and file caching performance.  >  >  > Tests to local disks: E >      maketestf    mystart       myread        mysplit      Disk  OS  HardwareD > Notes      maketest      myopenin      myopen        mysplitbinaryL > -------------------------------------------------------------------------- ------J >  2   -     0.33   0.01   0.01   0.17   0.17   0.41   1.14  /tmp  Linux64 Blazer Itanium (666Mhz) G >  3   -     0.42   0.00   0.00   0.28   0.24   0.64   0.62  /tmp  RH 7  Proliant 550 (450Mhz) L >  5   0.90  0.40   0.00   0.00   0.20   0.20   0.70   1.30  /tmp  SunOS 5.7 Ultra60 (360 Mhz) L > 18   1.50  0.30   0.00   0.00   0.10   0.20   1.50   1.30  /tmp  SunOS 5.8 Blade 100 (500 Mhz) H >  7   2.90  0.20   0.00   0.00   0.10   0.10   0.50   0.30  /tmp  Tru64 DS10L (466 Mhz) H >  9   3.40  0.20   0.00   0.00   0.10   0.10   0.60   0.30  /tmp  Tru64 DS20 (500 Mhz)I > 12   1.46  0.29   0.00   0.00   0.14   0.11   0.40   0.37  /tmp  RH 6.2  DS10 (466 Mhz)L > 16   -     0.43   0.01   0.01   0.30   0.64   1.10   0.97  RAM   VMS 7.2-1 DS10 (466 Mhz) (FASTRMS)L > 11  19.85  1.36   0.02   0.01   0.20   4.46   7.76   7.94  local OVMS7.3FT DS10L (466 Mhz)(FASTRMS)L > 15   -     1.77   0.02   0.04   0.47   9.05  16.62  14.38  local VMS 7.2-1 DS10 (466 Mhz) (FASTRMS)L > 14   -    26.46   0.11   0.01   0.54   7.75  46.51  39.10  local VMS 7.2-1 DS10 (466 Mhz) (Vanilla)J > 17  12.18  9.51   0.13   0.18   6.13   9.21  19.61  16.27  local VMS 7.2 VAX 4000/100   (FASTRMS,RAID)  >  > Tests to networked disks: E >      maketestf    mystart       myread        mysplit      Disk  OS  HardwareD > Notes      maketest      myopenin      myopen        mysplitbinaryL > -------------------------------------------------------------------------- ------J >  1   -     2.29   0.01   0.01   1.14   1.40   5.49   4.67  NFS   Linux64 Blazer Itanium (666Mhz) L >  4   1.20  0.70   0.00   0.00   3.60   0.50   1.70   3.80  NFS   SunOS 5.7 Ultra60 (360 Mhz) H >  6   3.50  1.50   0.00   0.00   0.10   2.10   6.00   4.70  NFS   Tru64 DS10L (466 Mhz) H >  8   3.50  1.40   0.00   0.00   0.10   2.20   6.20   4.70  NFS   Tru64 DS20 (500 Mhz)L > 10  35.57 34.33   0.03   0.02   0.20   5.55  43.27  40.24  MSCP  OVMS7.3FT DS10L (466 Mhz) (FASTRMS)  > H > 1. Linux64 on a Blazer Itanium (666Mhz) - the Compaq testdrive system.5 > Using /home (NFS mounted from spe253) . 08-FEB-2001  > H > 2. Linux64 on a Blazer Itanium (666Mhz) - the Compaq testdrive system. > Using /tmp . 08-FEB-2001 > G > 3. RedHat7 on a Proliant 5500 (450Mhz) - the Compaq testdrive system.  > Using /tmp . 08-FEB-2001 > H > 4. SunOS 5.7 on an Ultra60 (360 Mhz) - the Caltech ITS "clyde" system.( > Using /home (NFS mounted). 02-MAR-2001 > G > 5. SunOS 5.7 on an Ultra60 (360Mhz) - the Caltech ITS "clyde" system.  > Using /tmp. 02-MAR-2001  > E > 6. Tru64Unix 5.1 on a DS10L (466Mhz) - the Compaq testdrive system. E > Using /usr/bin/time on the login directory which is on /home, which 6 > in NFS mounted and served from "spe253". 08-FEB-2001 > E > 7. Tru64Unix 5.1 on a DS10L (466Mhz) - the Compaq testdrive system. 8 > Using /usr/bin/time on the /tmp directory. 08-FEB-2001( > MAKETESTF: Creating test file, fortran > D > 8. Tru64Unix 5.1 on a DS20 (500Mhz) - the Compaq testdrive system.E > Using /usr/bin/time on the login directory which is on /home, which 6 > in NFS mounted and served from "spe253". 08-FEB-2001 > D > 9. Tru64Unix 5.1 on a DS20 (500Mhz) - the Compaq testdrive system.8 > Using /usr/bin/time on the /tmp directory. 08-FEB-2001( > MAKETESTF: Creating test file, fortran > G > 10. OpenVMS 7.3 field test on a DS10L (466Mhz) - the Compaq testdrive  system. C > Using USER1 login directory which is served from VMSSERV which is = > an AlphaServer 1200 5/400 and the device is a DEC RZ1CB-CA. 1 > FASTRMS variant of these programs.  08-FEB-2001  > D > 11. OpenVMS 7.3 on a DS10L (466Mhz) - the Compaq testdrive system.C > Using $2$DKA100, which is a SCSI disk on "spe202".  Using FASTRMS  > variants.  09-FEB-2001 > B > 12.Redhat 6.2 on a DS10 (466Mhz U2W disks) - my "barrel" system,8 > Using /usr/bin/time on the /tmp directory. 08-FEB-2001 > C > 14. OpenVMS 7.2-1 on a DS10 (466 Mhz U2W disks). "seqaxp" system. , > RMS set to defaults.  No FASTRMS settings. >nC > 15. OpenVMS 7.2-1 on a DS10 (466 Mhz U2W disks). "seqaxp" system. ) > RMS set to defaults.  FASTRMS settings.v >tC > 16. OpenVMS 7.2-1 on a DS10 (466 Mhz U2W disks). "seqaxp" system.d6 > RMS set to defaults.  FASTRMS settings. On a RAMDISK > E > 17. OpenVMS 7.2, VAX 4000/100, no VIOC or third party disk caching, E > and all storage connected via standard 8-bit internal SCSI-1 bus torI > a Infrotrends RAID array OpenVMS 7.2 on VAX 4000/100.  FASTRMS variant. F > RMS block=0/buffer=0/extend=64.  Barry Treahy, Jr. <treahy@mmaz.com> >eA > 18. SunOS 5.8, Sun Blade 100, 128Mb RAM, UltraSparcIIe 500 Mhz, > > by Brian Stretch (bstretch@mindspring.com).  Similar resultsF > reported by Paul Joslin (paul.joslin@suncor3.sdrc.com). 08-MAR-2001.H > Note that mysplit results reproted by both testers showed user:0.3 and8 > sys:0.1, leaving the other .9 seconds unaccounted for.   ------------------------------  # Date: Sat, 10 Mar 2001 20:04:35 GMTb4 From: "Terry C. Shannon" <terryshannon@mediaone.net>) Subject: CNET OVMS Review: Closing in 200e; Message-ID: <nNvq6.9711$5f.2725235@typhoon.ne.mediaone.net>s  8 So I went and took another look at the CNET OS Survey at  L http://enterprise.cnet.com/enterprise/0-9566-601-1751615.html?tag=st.it.9566 -701-1751615.dirruo.uo    J there are now 199 responses for OpenVMS, which is far and away the highestK feedback level attained by any OS covered in the CNET user opinion surveys.o   -- Terry C. Shannon Consultant and Publisher Shannon Knows Compaq  email: terryshannon@mediaone.net$ Web (info on SKC):  www.acersoft.com   ------------------------------  $ Date: Fri, 9 Mar 2001 07:43:02 +0100! From: "Paul" <schizi@hotmail.com>R) Subject: Disk Read Rate / Disk Write Rate-% Message-ID: <3aa87bcc$1@news.cvut.cz>-   Hello,   We have DS20 with OpenVMS 7.2-12 and Oracle 7.3 DB.  * I need to find out Read/Write proportion -! Disk Read Rate / Disk Write Rate.b   I know only?   $ monitor fcp/int=1.   command.  C (And I cannot buy some Enterprise Capacity and Performance Analyzer  or Data Collector...)   D Unfortunately read statistics (Disk Read Rate) doesn't appear to me. Here is the output example:c  -                     FILE PRIMITIVE STATISTICSi.                                  on node XXXXX3                              9-MAR-2001 07:24:54.84i  K                                        CUR        AVE        MIN        MAXn  K     FCP Call Rate                     1.00      13.24       0.00     269.00sK     Allocation Rate                   0.00       0.00       0.00       1.00 K     Create Rate                       0.00       0.01       0.00       2.00i  K     Disk Read Rate                    0.00       0.00       0.00       1.00 K     Disk Write Rate                   1.00       0.26       0.00       9.00iK     Volume Lock Wait Rate             0.00       0.00       0.00       0.00t  K     CPU Tick Rate                     0.00       0.12       0.00       4.00yK     File Sys Page Fault Rate          0.00       0.00       0.00       0.00 K     Window Turn Rate                  0.00       0.01       0.00       2.00t  K     File Lookup Rate                  0.00       5.94       0.00     130.00oK     File Open Rate                    0.00       3.04       0.00      61.00nK     Erase Rate                        0.00       0.00       0.00       0.00-      + Thank you very much in advance for help me.0   Paul   ------------------------------  % Date: Sat, 10 Mar 2001 20:08:45 +0000V) From: Christof Brass <brass@infopuls.com>F0 Subject: Re: Drop C/C++ better sooner than later, Message-ID: <3AAA89CD.64A90716@infopuls.com>   Bill Gunshannon wrote: > . > In article <3AA6EADB.455E15F7@infopuls.com>,. >  Christof Brass <brass@infopuls.com> writes: > |> > |>H > |> Having the compiler written in the same language as it is compilingB > |> is an advantage if the language is better suited for compiler > |> constructiono > C > Your assuming here that Ada is somehow superior at this task.  Ani > opinion I don't share. > B > |>              than the alternative in this case C which is notE > |> only unsuitable for comiler construction but also for almost allt > |> other tasks.t > + > Also your opinion and one I do not share.b >  > |>D > |> If you once have a running Ada compiler the chicken-egg-problemF > |> is solved. This stage has obviously long reached and the decisionK > |> to (re-)write the Ada front-end in Ada can only be highly appreciated.e > J > Except that in the past it was possible to port GNAT to any machine thatL > had even a rudimentary K&R C compiler.  That covered a rather large subsetO > of the computing industry.  Requiring something as obscure as an Ada compiler   f Better a high quality Ada compiler as any compiler for an obscure and obsolete language like C or C++.  K > in order to build the Ada compiler has made it virtually impossible to doAI > private ports to new systems.  I believe this was the intent of ACT and/= > would seem to be very contrary to the ideal behind the GPL.t >  > 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   |@ > Scranton, Pennsylvania   |         #include <std.disclaimer.h>   ------------------------------  % Date: Sat, 10 Mar 2001 20:08:59 +0000e) From: Christof Brass <brass@infopuls.com>t0 Subject: Re: Drop C/C++ better sooner than later, Message-ID: <3AAA89DB.32E4F623@infopuls.com>   Paul Repacholi wrote:m > * > "Rick Cadruvi" <rick@rdperf.com> writes: > F > > I think that most any language can be used to write a compiler.  ID > > actually believe the best language for writing a compiler is the > > language itself. > I > So you have no problems with the idea of writing a compiler in say RPG?g > Or Datatrieve ;)  The statement holds only if at all for calcuation complete ("berechnungsvollstndige") languages. I don't think that using the same language is the best decision (yes, I know, it is a contradiction to what I said before), but it is the best decision if it is a good language for compiler construction. Counterexample: BASIC. I would advice not to use BASIC to write a BASIC compiler although it's definitely possible..   > A > > The reason is that it makes the compiler writer a user of theRF > > software.  I think ALL programmers should have to use the softwareC > > they write as an end user for an extended period of time.  It'shB > > amazing how much software improves when programmers become end
 > > users. > E > In practice, this is a very bad idea. Very, very bad. You have usestF > who 'know' where the docs and specs are wrong, hazy or missing. They= > carry all their preducices and pre-conceptions into use and.G > testing. There is a large body of documented practise that this in ite* > self is set the project up for disaster.   In practice this is the best idea. You need of course other users also. If the programmers are also users you get the best of both worlds.  C > It also make it very hard to isolate the use environment from thec > development environment.  A If you have both types of users you have to separate this anyway.o    D > > Cross compiling solves the porting problem more than adequately.G > > Obviously, a simple version of the compiler must be written in some H > > other language that exists with a new language that doesn't yet haveB > > any compiler, but that should only be to bootstrap a compiler. > G > Far better to flush the idea. It has nothing going for it at all. UsesE > a language that will ot cause you grief. IE, existing, well tested,69 > and with the hooks you need into the system it runs on..  Crosscompiling is the best if the language is appropriate for compiler construction at all (and this was the premise otherwise the compiler shouldn't have written in its own language in the first place). And the best compilers I know of are written and ported that way. One possibility is to have a small bootstrap compiler written in another language. Another possibility is to have a small bootstrap compiler and a translater into another language which is available on the target system.   If you want to have your own code generator problems with the intermediate language on the target system is not an issue at all.   > --> > Paul Repacholi                               1 Crescent Rd.,9 > +61 (08) 9257-1001                           Kalamunda.@B >                                              West Australia 60760 > Raw, Cooked or Well-done, it's all half baked.   ------------------------------  % Date: Sat, 10 Mar 2001 20:16:12 +0000$) From: Christof Brass <brass@infopuls.com>t0 Subject: Re: Drop C/C++ better sooner than later, Message-ID: <3AAA8B8C.C488248E@infopuls.com>  " Shane.F.Smith@Healthnet.com wrote: > F > That model only works for some languages. Anybody who writes a BASICL > compiler in BASIC, or a Java JIT compiler in Java needs to be removed from > the gene pool..... >  > Shanel  . Yes, BASIC compiler in BASIC, not a good idea.E But having the front-end Java compiler in Java isn't that a bad idea.  Actually I don't see any reason why the back-end of a Java JIT compiler shouldn't be written in Java - it has to prove that it really speeds up the code ;-)   ------------------------------  $ Date: Fri, 2 Mar 2001 07:48:45 -07005 From: "cstranslations" <cstranslations@email.msn.com>h0 Subject: Re: Drop C/C++ better sooner than later) Message-ID: <e2pkBfyoAHA.359@cpmsnbbsa07>3  : "JF Mezei" <jfmezei.spamnot@videotron.ca> wrote in message& news:3A9ECF1B.50C19A15@videotron.ca...$ > Shane.F.Smith@Healthnet.com wrote:H > > Genuine curiosity here: could one of you give me an example of where, > > pointer arithmetic is necessary, please?  H I suppose designing some piece of code that involves global sections andH (particularly variable length data structures) would be a tad bit easier1 (more straight forward) in C as opposed to BASIC.e  L Then there was last week when I put together an example (in BASIC) of how toF use ASTs for a little presentation at work ('cause we're a BASIC shop,L although after someone left recently we're a bit more open minded). I had toL pass a record from the caller into the AST. Once in the AST interpreting theG parameter as an address, using lib$movc3 to move a chunk of memory to acE locally defined record, accessing the data, modifying the data, usingeL lib$movc3 to move a chunk of memory on the way out  is a *bit* less straightJ forward than having a deference mechanism. But then most of the code we'reJ running is BASIC that was written 20 years ago (so you can imagine what itF looks like) so hell - might as well just stick it in a MAP and make it	 global...    ------------------------------  # Date: Sun, 11 Mar 2001 00:38:17 GMT 4 From: "Terry C. Shannon" <terryshannon@mediaone.net>S Subject: Dumbing Down VMS with UNIX Elements? (was Re: OpenVMS Educational Program)d; Message-ID: <ZNzq6.9788$5f.2841894@typhoon.ne.mediaone.net>b   Someone wrote...   > > > I > > > The best of both worlds is VMS. Any single element of UNIX put intos VMS reduces its quality.  K Based on totally irrelevant things like apps availability and market share, K asserting that VMS is the best of both worlds implies that these two worldsd+ collectively represent a very small planet.   I Adding Solaris-like APIs to VMS (DII COE initiative) may help address the E apps availability issue. And it darn sure will address the "Compaq iseJ dropping VMS" codswallop and calumny. Of course, that's just my opinion, I could be wrong.    ------------------------------  % Date: Sat, 10 Mar 2001 23:35:15 +0000 ) From: Christof Brass <brass@infopuls.com>d8 Subject: Re: In-memory layout of PACKED ARRAY OF BOOLEAN, Message-ID: <3AAABA33.835CA330@infopuls.com>   John Reagan wrote: >  > Christof Brass wrote:h > >  > >l > > Pascal is one of the best languages and one of the best designed languages and the modularised version of DEC/Compaq Pascal was even better than the standard and better integrated into VMS than most other Pascal implementations on their OSs.  >  > Thank you. >  > >  > > Two questions: > >is > > 1.What is the implementation language of the Compaq Pascal? Are there more than one for front-end and back-end?- > E > BLISS in around 45 modules for the front-end.  The front-end is 90% J > common between VAX and Alpha.  On the VAX side there is a little MACRO32J > code.  Also on the VAX side, we have our own code generator, object fileH > generator, listing package, etc. (another 50 modules or so) whereas on? > the Alpha side we use the same GEM code generator/compilation8E > environment that the rest of the languages use (GEM is a mixture of8 > BLISS and C++) >  > >7 > > 2.What about quality of generated code as compared to C for elementary things like arithmetical operations, pointer de-referencing, array indexing, procedure calling (I assume the standard VMS calling mechanism is used that will be equal with C). > I > On Alpha, we use the exact same GEM code generator and optimizer as theRF > C compiler.  The ability of the code generator to get efficient codeD > depends on how efficiently the front-end can describe the languageE > semantics to the back-end.  C's sometimes weakly typed pointers canVI > cause lots of problems, but they try to compensate.  However, C doesn't C > have Pascal's concept of uplevel referencing, and Pascal has some J > run-time bounds checking enabled by default, etc.  In general, on Alpha,7 > I'd rate the code quality the same for most programs.  > C > On VAX, we both have our own optimizers and code generators.  The C > comparison isn't as easy.  Both have their strong points and weakl	 > points.  >  > --
 > John Reagan- > Compaq Pascal Project Leader   Thank *you*.   A specific question about code generation on Alpha depending on the front-end. As you pointed out the language has some influence on the possibility to supply information to the code generation.With C it is possible to declare a procedure having a variable number of parameters. On some architectures I know of the consequence was to let the caller remove the pushed parameters. This was called the C calling convention whereas with Pascal the number of parameters of a procedure was always fixed, hence known to the compiler and the code for popping the parameters off the stack was located in the callee (the called procedure) which resulted in smaller programs on average.hn Is is possible to tell GEM how to treat the parameter pushing/popping wrt the location where it should happen?O Simpler question: where is the parameter popping done in Pascal and where in C?0   ------------------------------  % Date: Sat, 10 Mar 2001 19:33:15 +00000) From: Christof Brass <brass@infopuls.com>0" Subject: Re: Low cost workstations, Message-ID: <3AAA817B.22285D5C@infopuls.com>   Bill Gunshannon wrote: > . > In article <98b0tf$m95@gap.cco.caltech.edu>,7 >  mathog@seqaxp.bio.caltech.edu (David Mathog) writes:cn > |> In article <3AA88F56.A853955D@gtech.com>, Arne =?iso-8859-1?Q?Vajh=F8j?= <arne.vajhoej@gtech.com> writes: > |>> > |> >There are still some educational systems left using VMS. > |>K > |> True, they are not ALL shut down.  And likely there will still be someSO > |> running here and there in 20 years.  But just like the PDP11 grinding awayrG > |> about 30 meters from here VMS has become more of an anomoly than a3 > |> presence. > A > I can vouch for that.  The only thing they laugh at me for more.D > than my VAXen are my PDP-11's.  I even have them running at home!! >  > See, I'm really not anti-DEC.   
 Anti-Compaq??i   But that's not the point at all! The facts are important. Even if you would like to say something in favour of VMS and against UNIX: if it were wrong I wouldn't like it at all and I would throw in my comment.  There is still another question: do you really fell yourself a VMS style person? I don't think so. And I even think so that people like you who don't really understand what they talking about should *not* try to help VMS because they make things even *worse*.b  x Probably it's exactly people like you at universities who helped in abandoning VMS by using the wrong way to support it. And of course people like you contributing to this NG force other people like me to contradict the BS they say about UNIX beeing usable/helpful/appropriate etc. and by that doing harm to VMS.    > 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   |@ > Scranton, Pennsylvania   |         #include <std.disclaimer.h>   ------------------------------  % Date: Sat, 10 Mar 2001 19:39:45 +0000a) From: Christof Brass <brass@infopuls.com>r" Subject: Re: Low cost workstations, Message-ID: <3AAA8301.8A066061@infopuls.com>   Robert Deininger wrote:y > J > In article <9896mc$2i5g$1@info.cs.uofs.edu>, bill@cs.scranton.edu wrote: > - > > In article <3AA7AE9C.EF04A181@bbc.co.uk>, 4 > >  Tim Llewellyn <tim.llewellyn@bbc.co.uk> writes: > > |>B > > |> agreed, in academia cost of ownership is not such an issue, > >-J > > Excuse me??  That may fly at MIT, but that's not where the majority ofI > > students are going.  Smaller schools account for a much larger numberCH > > of grads than the big named ones anymore.  And I can assure you they8 > > pinch every penny twice before it goes out the door. > >  > > |> there are alwaysc3 > > |> willing grad students, postdocs or whatever.  > >1J > > To do what??  I tried having a student do simple Windows installs.  ItG > > was more work and more time consuming fixing his mistakes than justEJ > > doing the job myself.  Add to that the dreaded security issue.  PeopleF > > here have said they wouldn't trust software installed off the net.N > > Would you trust a machine for which a student ever had the root password?? > I > Bill, your salary, spread over the many different systems you take careiE > of, is part of TCO.  But a lot of schools don't have a guy like youhG > anymore.  The student of the day does what you are doing.  That makesaI > "TCO" a pretty invisible concept for their bean-counters.  They haven't I > noticed the drop in reliability and uptime yet, so TCO (in their narrowi) > view) is just the cost of the machines.c > H > > Another convert!!  Now let's top bashing Unix and work out some realJ > > plans to get a lab full of Alphas running VMS as an option to that labG > > full of PC's!!!  One thing I can safely say about spending here, if F > > someone were to cough up a lab's worth of machines with software IG > > would be able to get the administration to cough up the room to set  > > up the lab.  > / > What software would be important at your lab?  > H > Are there still open-topic courses for advanced undergrads?  How aboutI > some courses where groups of students port some non-trivial open-sourceeL > software to VMS?  With an instructor to point in the right direction, they0 > wouldn't need to know a ton of VMS in advance.   Great idea!o   Choose a good language! Not necessarily the one the open-source SW has been written in. Don't use a crap language like C or C++! This is especially important for educational reasons.[ Give them a sound introduction into the VMS API that the outcome will be real VMS software.0  Porting software to a different OS normally reveals a lot of bugs because the other system behaves differently and hidden bugs will show up most probable. Porting software to a different compiler on the same OS and keeping the language might be good for quality also by revealing bugs or unnecessary system dependencies. Porting to a different language may even reveal architecture/design bugs/flaws.   Gread idea!s  K > I think courses like this would be very good for the students, regardlessgI > of the platform.  It seems to me that many student programming projectsaL > these days are small, pre-cooked, and almost trivial.  They don't get much > practice working > 1) in groupsL > 2) on projects large enough that designing and organizing are non-trivial. > I > It might not be the case at your school, but I've heard of departementsoI > where seniors only do 4 or 5 "machine problems" per course, and none of % > them are more than 2 pages of code.h >  > -- > Robert Deininger > rdeininger@mindspring.comn   ------------------------------  % Date: Sat, 10 Mar 2001 19:43:22 +0000s) From: Christof Brass <brass@infopuls.com>t" Subject: Re: Low cost workstations, Message-ID: <3AAA83DA.B322486F@infopuls.com>  ! steven.reece@quintiles.com wrote:s > E > I have no information so can't claim that you're right or not abouteF > Free-BSD.  It's kindof an orthogonal argument though.  The fact thatJ > Free-BSD does well without advertising neither proves nor disproves that$ > advertising can overcome crapness. > " > bill@triangle.cs.uofs.edu wrote:I > >>>While I willingly admit that advertising can offset poor quality andu > highJ > TCO in the minds of pointy-hairs, how then do you explain the success ofH > FreeBSD??  People here would claim that it is unreliable, hard to use,H > hard to admin and has a high TCO.  And yet, unlike Linux, they have noL > marketing department and don't advertise at all.  And I would bet that theL > number of FreeBSD system being used for real work is many times the number: > of VMS systems.  And yes, some of them are rather large. > <<<g  < Thanks in helping establishing a culture of logical arguing.   ------------------------------    Date: 10 Mar 2001 12:23:31 -07001 From: nothome@spammers.are.scum (Malcolm Dunnett)5" Subject: Re: Low cost workstations, Message-ID: <SgHgDuRpgzB9@malvm1.mala.bc.ca>  M In article <rdeininger-1003011121080001@user-2ive70j.dialup.mindspring.com>, o7    rdeininger@mindspring.com (Robert Deininger) writes:: > L > Well, a budget of $0 isn't reasonable.  If you could shake loose enough $$H > for a half-dozen used alphas, you could run a small class.  (One alpha > would do in a pinch.)u  E     Getting Alphas isn't the hard part. Our CSci department was givenD> 3 Alphaserver 1000 4/266s over a year ago. The instructor evenB asked (without my prompting) if it would be possible to put VMS onM them for him to use in an Operating Systems course. Getting licenses for themoI that are actually good for anything ( such as setting up a 3 node clustera= that all the students can work on ) is expensive. As has beenvD explained numerous times in this NG, the current "education license"I program isn't much use for this. So I'd have to buy the base licenses forsJ them ( several thousand dollars per box ) and add them to the CSLG - whichH would push us up to the next tier and cost me a further several thousand
 dollars/year.   /   In the meantime, they run Linux on the boxes.e   ------------------------------   Date: 10 Mar 2001 20:35:45 GMT2 From: mathog@seqaxp.bio.caltech.edu (David Mathog)" Subject: Re: Low cost workstations, Message-ID: <98e371$6oj@gap.cco.caltech.edu>  X In article <3AAA6C24.AAD65BD7@infopuls.com>, Christof Brass <brass@infopuls.com> writes: >Bill Gunshannon wrote:  >> c; >> In article <0gbfat02o4iqje8a1k7ftl3bg6asvp4ccl@4ax.com>,n+ >>  Alan Greig <a.greig@virgin.net> writes:m >> |>eL >> |> MVS! What could be easier than $ MCR AUTHORIZE MODI USER/PASS=password >> a: >> How about: adduser userid "" "" "John Q. User" password > G >Sorry you obviously don't get it. The shell user interface is severely J >broken. There are countless examples of that. In this case: having a listH >of parameters of similar or same type denoting different things without >beeing forced to specify  <SNIP>  J But adduser is a DCL procedure, just like any equivalent csh/sh/bash/tcsh G etc procedure.  In any of these languages/shells the command line just iI passes the arguments in as P1->P8, $1->$N, or whatever, all of which are tH (initially) strings.  If you want it to identify the parameters you can @ always rewrite the script (in any of these) so that it requires:  Y $ adduser "userid=value" "uic=value" "defpro=value" "tname=John Q. User" "password=value"a  " and the order becomes arbitrary.    E However, the only scripting language of any type that I know of that tI actually requires this usage, and results in strongly typed command line hA usage is my own miniproc.  (You'll get much more use out of perl,l python, or php though!)  See:r  1   http://seqaxp.bio.caltech.edu/www/miniproc.html   L Which reminds me that I never put version 3.05 up on that page one of these  days.t   Regards,   David Mathog mathog@seqaxp.bio.caltech.edup? Manager, sequence analysis facility, biology division, Caltech 9   ------------------------------  % Date: Sat, 10 Mar 2001 20:10:36 -0500 2 From: rdeininger@mindspring.com (Robert Deininger)" Subject: Re: Low cost workstationsL Message-ID: <rdeininger-1003012010370001@user-2iveaph.dialup.mindspring.com>  F In article <SgHgDuRpgzB9@malvm1.mala.bc.ca>, nothome@spammers.are.scum (Malcolm Dunnett) wrote:  DG >     Getting Alphas isn't the hard part. Our CSci department was given @ > 3 Alphaserver 1000 4/266s over a year ago. The instructor evenD > asked (without my prompting) if it would be possible to put VMS onO > them for him to use in an Operating Systems course. Getting licenses for themSK > that are actually good for anything ( such as setting up a 3 node clustera? > that all the students can work on ) is expensive. As has beenoF > explained numerous times in this NG, the current "education license"# > program isn't much use for this.    9 For classroom use, I think the edu program is actally ok.:  5 1. College registers in program and gets access code.n  ; 2. Each student in the class uses the code to get licences.g  > 3. Instructor load's all the students' licences on the system.  2 Or have I misunderstood that crazy legalese again?   -- o Robert Deininger rdeininger@mindspring.comt   ------------------------------  % Date: Sat, 10 Mar 2001 20:02:24 -0600e7 From: "David J. Dachtera" <djesys.nospam@earthlink.net>s" Subject: Re: Low cost workstations- Message-ID: <3AAADCB0.C0F8109B@earthlink.net>i   Robert Deininger wrote:) > H > In article <SgHgDuRpgzB9@malvm1.mala.bc.ca>, nothome@spammers.are.scum > (Malcolm Dunnett) wrote: > I > >     Getting Alphas isn't the hard part. Our CSci department was givensB > > 3 Alphaserver 1000 4/266s over a year ago. The instructor evenF > > asked (without my prompting) if it would be possible to put VMS onQ > > them for him to use in an Operating Systems course. Getting licenses for themsM > > that are actually good for anything ( such as setting up a 3 node clustereA > > that all the students can work on ) is expensive. As has beennH > > explained numerous times in this NG, the current "education license"$ > > program isn't much use for this. > ; > For classroom use, I think the edu program is actally ok.( > 7 > 1. College registers in program and gets access code.l > = > 2. Each student in the class uses the code to get licences.- > @ > 3. Instructor load's all the students' licences on the system. > 4 > Or have I misunderstood that crazy legalese again?  3 Each machine is dedicated to a single student only?l  G I don't recall all the details, but I seem to recall that being a major = stumbling block. As always, I could be (probably am) wrong...S   -- i David J. DachteraX dba DJE Systemse 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.i   ------------------------------    Date: 10 Mar 2001 20:11:11 -07001 From: nothome@spammers.are.scum (Malcolm Dunnett) " Subject: Re: Low cost workstations, Message-ID: <kc62K$hAHdtc@malvm1.mala.bc.ca>  M In article <rdeininger-1003012010370001@user-2iveaph.dialup.mindspring.com>, 37    rdeininger@mindspring.com (Robert Deininger) writes:i  H > In article <SgHgDuRpgzB9@malvm1.mala.bc.ca>, nothome@spammers.are.scum > (Malcolm Dunnett) wrote: >  sH >>     Getting Alphas isn't the hard part. Our CSci department was givenA >> 3 Alphaserver 1000 4/266s over a year ago. The instructor evenlE >> asked (without my prompting) if it would be possible to put VMS onnP >> them for him to use in an Operating Systems course. Getting licenses for themL >> that are actually good for anything ( such as setting up a 3 node cluster@ >> that all the students can work on ) is expensive. As has beenG >> explained numerous times in this NG, the current "education license"e$ >> program isn't much use for this.  > ; > For classroom use, I think the edu program is actally ok.m > 7 > 1. College registers in program and gets access code.M > = > 2. Each student in the class uses the code to get licences.n > @ > 3. Instructor load's all the students' licences on the system. > 4 > Or have I misunderstood that crazy legalese again? >   C     Who knows, it seems to contradict itself in several places. But @ let's assume for a moment you are correct. First off, requiring ? each student to apply for his own license in this situtation ise: unnecessarily bureaucratic and is frankly more hassle than= the instructor or students want to tolerate. They hire people > like me to set the systems up so they can get on with the task? of teaching/learning. It's clearly impossible for me to requestp< a bunch of licenses for a class under this program. Further,C it says it's not possible to combine licenses - so only one studentf< could use each system at a given time. Strictly speaking Joe= would have to unload his licenses and Fred would have to loadd9 his whenever they changed users on the box - more uselesst@ bureaucracy. The FAQ clearly states that for multi-user licensesA one should use the CSLG, not this program. Adding 3 or 4 machines > to CSLG would push us up to the next pricing tier and cost the* college something like another $5000/year.  B    There is a confusing clause in the edu program which appears to> let an instructor allow multiple students on multiple machinesA ( does that imply only one student per machine? ) "in a classroomp= situation under your control".  Does that mean the instructoreA must be physically present whenever the computers are being used?u+ If not, exactly what constitutes "control".g  D    I have asked Compaq folks for answers to these questions, they'veA chosen to not respond. I can only take this to mean that either Ii@ am correct and they see no reason to address my concerns or they- have no idea what the program permits either.m  A    The truly frustrating thing is they announced this program was ; coming, then went away for about 8 months and designed thisp= offering, apparently with no consultation with the education  > community as to what we actually needed. Now that the concerns> have been expressed it appears they've gone into hiding again.9 I expect in  another 6 months or so they'll announce some8; tinkering with the program which makes no material changes.n   ------------------------------  % Date: Sun, 11 Mar 2001 05:00:52 +0100u  From: Paul Sture <paul@sture.ch>" Subject: Re: Low cost workstations+ Message-ID: <VA.000002fa.2231e63e@sture.ch>e  B In article <98aukb$b84$2@info.cs.uofs.edu>, Bill Gunshannon wrote:   [snip]  F > And, the last sentence just sounds like your selling yourself short.E > I have no doubt that most of the people here could master Unix in atE > very short time if the took the chip off their shoulder and decidedbF > to treat it like learning any other computer related task.  It's notF > as difficult as some would have you believe, but if you go into with> > the attitude exhibited by many here, what would you expect?? >lC Believe me I've tried to master Unix, admittedly the Linux flavour !C thereof. It is nowhere as easy as you say - it appears to me as an fH outsider to live around the philosophy of "you're not in the club of CS ; graduates or teenage geeks, so we'll keep you in the dark".e   I could go on...  ? But please folks, Bill is telling us to get off our complacent n@ backsides and keep up with the competition, otherwise we become 
 dinosaurs. ___d
 Paul Sture Switzerlandc   ------------------------------  % Date: Sat, 10 Mar 2001 23:40:13 +0000o) From: Christof Brass <brass@infopuls.com>VW Subject: Re: Mimer (database engine)  for OpenVMS (and others) is available fordownloadh, Message-ID: <3AAABB5D.C41043CA@infopuls.com>   Paul Repacholi wrote:I > - > Christof Brass <brass@infopuls.com> writes:t > H > > This is one of the best web sites. It is lightning fast and readableB > > with every browser. No fancy graphics, instead sound and solidA > > information. Well organised and easy to find information. ThetA > > product seems to be very much developed and to be in a matureeF > > state. Documentation is mostly around in HTML/PDF and some only inF > > PDF - all can directly be downloaded from the Web site - includingD > > the database itself for all major platforms (VMS, True64, Linux,* > > AIX, HP-UX, Solaris) and even Windoze. > / > Such a pleasant change. And it was READABLE!!t > D > > From what I read it is a clear Oracle competitor but without theG > > Oracle8i object and Java extensions. But with full fledged triggerssB > > and stored procedures according to SQL standards - even with a! > > precompiler for embedded SQL.t > ( > There is a Java kit available as well. > < > > I asked some questions and got fast and precise answers. > $ > > The development platform is VMS!A > > The implementation language is *not* C or a similar language..3 > > They offer evaluation and development licences:  > G > > Quote from the licence section (under products is a link "Mimer SQLu > > Licensing"): > C > > The free Mimer SQL Development version can only be used for themF > > purposes of evaluation, application development and testing; it isB > > not licensed to be used in a production environment. To put anG > > application based on Mimer SQL into production a Mimer SQL run-timed > > license is required. > C > The version I read was that the 'up-to' 10 user licence was free,oH > more, pay. Pay for support as well if you want it. Need to check that. > G > BTW, installation is so easy it can bite you. Run the exe and *pow* a E > complete subtree with everything in place. Run the startup file andeG > you are running. DON'T do this thinking it is a compessed backup savet > set or the like. > # > Looks very nice and clean so far.  >  > --> > Paul Repacholi                               1 Crescent Rd.,9 > +61 (08) 9257-1001                           Kalamunda.iB >                                              West Australia 60760 > Raw, Cooked or Well-done, it's all half baked.  : Thanks for this clearification and additional information.= I'll install the thing if I have my new Alpha up and running.t Then I'll report back.   ------------------------------    Date: 10 Mar 2001 16:42:15 -05003 From: "Gareth V. Williams" <graff@cfa0.harvard.edu>e/ Subject: Re: New memory in VAX 4000/90 not seenr, Message-ID: <3aaa9fb7.0@cfanews.harvard.edu>   Hoff Hoffman <hoffman> wrote:td : In article <3aa99f52.0@cfanews.harvard.edu>, "Gareth V. Williams" <graff@cfa0.harvard.edu> writes:  : :Hoff Hoffman <hoffman> wrote:   : :KA49-A V1.3-0BC-V4.4  71 MHZn  H :   V1.4 is current, though that is likely not relevent to this problem.   : :08-00-2B-30-92-EA : :16MB   J :   That is the VAX console program for the KA49 CPU that sees only 16MB, G :   which would imply something is wrong with the CPU or the memory (ore= :   the installation) -- a low-level failure, in other words.   . :   The MEM_BOT/MEM_TOP looks to be 16MB, too.  G : ::   And based on your comments, may I assume you used eight MS44-CA l : ::   series (16MB) SIMMs?  : :eG : :  Ah.  The boards we were shipped, which were ordered and CONFIRMED lF : :as being for a 4000/90, are MS44-DC, according to the Compaq box inH : :which they were shipped.  But they are eight 16 MB SIMMs.  What's theH : :difference?  Are we going to have to sort this out with the supplier?  B :   MS44-DC is a set of MS44-CA SIMMs.  Make sure all SIMMs match.  P :  ---------------------------- #include <rtfaq.h> -----------------------------P :       For additional, please see the OpenVMS FAQ -- www.openvms.compaq.com    P :  --------------------------- pure personal opinion ---------------------------N :    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com     -- dH ------------------------------------------------------------------------H Gareth V. Williams, MS 18, 60 Garden Street, Cambridge, MA 02138, U.S.A.+ Associate Director, IAU Minor Planet Center H gwilliams@cfa.harvard.edu        http://cfa-www.harvard.edu/iau/mpc.html7 OpenVMS & RISC OS: refined choices in operating systemst   ------------------------------    Date: 10 Mar 2001 17:39:00 -05003 From: "Gareth V. Williams" <graff@cfa0.harvard.edu>i/ Subject: Re: New memory in VAX 4000/90 not seens, Message-ID: <3aaaad04.0@cfanews.harvard.edu>   Hoff Hoffman <hoffman> wrote:IF >  And have you tried populating bank zero only; shuffling some SIMMs?  G   I have just tried this.  The system fails to come up.  The diagnostic D lights are displaying (in the terminology of the list in the 4000/90F Owner's and System Installation Guide) XXXX 00X0.  This combination isE not included as one of the common problems in the list in the manual.eG So I've no idea what it means.  But it occurs with both sets of 4 SIMMsi in both banks 0 and 1.  L   Populating all 8 sockets, the system comes up with 16 MB.  At the console,C SHOW CONFIG indicates 4 x 4 MB in bank zero and mismatched SIMMs in-I bank one.  This probably explains the 16 MB total even though 8 SIMMs had H been put in...  It also probably explains the diagnostic light sequence.  G   Anyway, I tried a second set of 8 x 16 MB SIMMs (we intend to upgrade2I our three VAX 4000/90 to 128 MB).  This time the system comes up with theiJ console banner announcing 32 MB!  At the console, typing SHOW CONFIG showsE that the memory in bank zero is 4 x 4 MB and that in bank one is alsomI 4 x 4 MB.  16 MB + 16 MB = 32 MB.  So it seeing 8 SIMMs, but only 4 MB oft each.r  H >  Specifically which set of SIMMs was used here (MS44-CA, 54-19103-CA)?  F   Looking at the original set of supposed 16 MB SIMMs, I see that each3 is marked 5419103.  As are the second set of SIMMs.t     -- mH ------------------------------------------------------------------------H Gareth V. Williams, MS 18, 60 Garden Street, Cambridge, MA 02138, U.S.A.+ Associate Director, IAU Minor Planet CenteroH gwilliams@cfa.harvard.edu        http://cfa-www.harvard.edu/iau/mpc.html7 OpenVMS & RISC OS: refined choices in operating systemsh   ------------------------------  % Date: Sat, 10 Mar 2001 22:03:57 +0100o= From: Arne =?iso-8859-1?Q?Vajh=F8j?= <arne.vajhoej@gtech.com>h0 Subject: Re: Newbie question - VMS Documentation) Message-ID: <3AAA96BD.DB0F543B@gtech.com>-   Amy Hoskins wrote:N > Wow!  Where to begin!  What is a good starting point to learn VMS?  It seems4 > to be very difficult to find books / documentation  & Try and follow some of the links from:  ) http://www.levitte.org/~ava/vms_doc.htmlxs) http://www.levitte.org/~ava/vms_faq.htmlxa   Arne   ------------------------------  % Date: Sat, 10 Mar 2001 22:02:32 +0100>= From: Arne =?iso-8859-1?Q?Vajh=F8j?= <arne.vajhoej@gtech.com>W; Subject: Re: open vms newbie is looking far an DCL emulatoro) Message-ID: <3AAA9668.4477A8F3@gtech.com>"   "g.reinders" wrote:dH > I'm looking for an DCL emulator to try DCL command under Windows (jek)N > who can tell me if such a program exists And if it does where I can find it.  
 Try f.ex.:  ) ftp://ftp.wku.edu/vms/fileserv/pc_dcl.zip   9 It is *NOT* a complete DCL, but is it impressibe how muchu5 functionality has been put into such a small package./   Arne   ------------------------------    Date: 11 Mar 2001 02:50:59 +0800, From: Paul Repacholi <prep@prep.synonet.com>( Subject: Re: OpenVMS Educational Program- Message-ID: <87r905e9v0.fsf@prep.synonet.com>e  3 bill@triangle.cs.uofs.edu (Bill Gunshannon) writes:   D > I don't suppose most here want a lesson in Unix esoterica, but for > the sake of those who do:U  D > The primary reason for the very curt Unix command names was more a= > practical than a philosophical matter.  The original common ; > interface was an ASR33 teletype.  Anybody here ever use ao< > Portacom-110??  It seems like a speed demon by comparison.  D Missed a chance here Bill. The original was a Teletype 37. It had an? even bigger apitite for fingernails than a 33, was a 6 hour gym F workout for wrists and fingers in 20 min, and intodued the abomination of LF as the end of line char.  B > Most commands consist of some portion of the spelled out command; > with letters removed that still allowed the command to be F > recognizable to someone who knew what they were trying to do anyway.  I Re BSD and the net. You are aware that there was near 15 years of the net1: before BSD? Have a look at RFC 681 for a view at the time.   -- l< Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda. @                                              West Australia 6076. Raw, Cooked or Well-done, it's all half baked.   ------------------------------  % Date: Sat, 10 Mar 2001 21:47:58 +0000a) From: Christof Brass <brass@infopuls.com>s( Subject: Re: OpenVMS Educational Program, Message-ID: <3AAAA10E.F9F74273@infopuls.com>   Alan Greig wrote:e > = > On 8 Mar 2001 14:59:32 GMT, bill@triangle.cs.uofs.edu (Billt > Gunshannon) wrote: > I > >Sadly, while this stuff gets talked about here every couple of months,oI > >and everybody agrees that something needs to be done, nothing ever is.r > H > I would agree that this has been the case. Development wise though VMSG > is moving forward again. The DII COE work if done well should give uss+ > the best of both worlds - VMS *and* Unix.l > + > What we still need is marketing and apps.g >  > >r > >bill  >  > -- > Alan  \ The best of both worlds is VMS. Any single element of UNIX put into VMS reduces its quality.   ------------------------------  % Date: Sat, 10 Mar 2001 22:28:29 +0000B) From: Christof Brass <brass@infopuls.com>g( Subject: Re: OpenVMS Educational Program, Message-ID: <3AAAAA8D.A60900B3@infopuls.com>   Bill Gunshannon wrote: > + > In article <3AA89666.D161CD47@gtech.com>,tB >  Arne =?iso-8859-1?Q?Vajh=F8j?= <arne.vajhoej@gtech.com> writes: > |> Bill Gunshannon wrote:v4 > |> > In article <8766hkrwza.fsf@prep.synonet.com>,6 > |> >  Paul Repacholi <prep@prep.synonet.com> writes:	 > |> > |>m > |> > |> PROD INS * > |> >H > |> > This is even more curt and cryptic than Arne's.  And getting real( > |> > close to Unix command syntax. :-) > |>% > |> But there are a big difference !x > |>? > |> Abbreviation is optional on VMS. It is mandatory on Unix !  > ; > Ummm.  Nothing is mandatory in Unix.  If a user finds thek? > short names (which really do have a reason and a logic behindr< > them.  Well, most of them anyway.) offensive he is free to> > change any of them.  Aliases are simple to set up.  And even@ > complete shells are easy to make.  I have seen Shells provided= > by various Unix vendors that imitated the MSDOS COMMAND.COMo= > interface and Tandy did one for their version of Xenix that > > mimiced TRSDOS-II.   I personally wrote a Shell that mimiced= > The UCSD Pascal menu interface.  One of Unix's strengths isX > infinite adaptability.  kThe important difference is that with VMS you can rely on the standard. With UNIX you aren't sure about the shell you will have to use and if *your* aliases can be set up automatically (what if you have to type them in every time you login because you don't have the right to change anything permament of youer account on a certain machine at a customer's site?).g Look at the Oracle installation on UNIX. They provide at least two different sets of scripts for two shells. Major step forward!   Congratulations BTW for beeing acquainted with the UCSD Pascal system which was by far one of the best development systems at that time.   > |>L > |> > |> of a scratch disk, and tweek the file names into dependancy order. > |> >3 > |> > And your going to donate this scratch disk??r > |>D > |> Your beloved FreeBSD install from the net also requires someone > |> to donate a disk. > C > I don't understand your meaning here.  The suggestion to copy alltE > of the programs from the CD to "a scratch disk" requires me to havetD > an additional disk with enough space to hold several CD's worth ofB > stuff in order to do an unattended install.  The FreeBSD install@ > requires only the system disk.  The LP software is eitehr on aA > single CD or loaded from the net.  No additional disk required. @ > So while the scratch disk is  not "required" on VMS either, itB > was the recommended method to use if I wanted the convenience of > the FreeBSD install.   As I understand the poster the meaning is that the disk is required in a server which will then be accessible through the internet.o   >  > |>I > |> The only difference is that someone has. Compaq has not. For obviousyE > |> reasons (installing software directly from the net is a no-no in  > |> most VMS sites).e > ? > We've heard all this before.  But as I said, blind trust in as> > product because the vendor gave it to you on a CD is just as> > risky.  There is past (documented) evidence of this.  One ofB > parts of the opensource paradigm is that if I am truly paranoid,< > I can install all of my software from sources that I build> > myself only after examining these sources to ensure they are> > what they purport to be.  Believe it or not, in the packages= > that are officially supported under FreeBSD (at FreeBSD 4.1l< > that is aprox 3600 packages) even building from sources is: > trivial and fully automated.  Except for the reading the > sources part. :-)n >  > |>E > |> I think other posts to this thread has clearly revealed that yourI > |> do not know PCSI INSTALL and are comparing your 2001 FreeBSD installr? > |> with the VMSINSTAL that were used 10-20 years ago on VMS !n > ? > I openly admit that I am a novice VMS Admin.  Been at it less > > than a year.  But the point is most of the people here don't= > know Unix any better than I know VMS.  And that makes theirt= > observations about as meaningful (or meaningless, dependingu > on your point of view) mine.  B I'm not sure if this is true. I do UNIX administration since 1996.  8 > The Unix haters here are so blinded by their religion 9 > that they are missing the whole point and have led thish6 > to deteriorate into little more than ahouting match.  m The flaws of UNIX are no invention of VMS people/advocates. They are *inherent*, they are 'architectured' in.a  a7 > I am not trying to convert anyone.  I could care lessg7 > if anyone here ever uses a Unix system again in theirw: > life.  The point is that Unix is out there in the world.8 > It is easily and freely available.  It is neither hard< > to install or difficult to administer.  It has a seemingly4 > limitless selection of applications, both free and- > commercial.  It is trivial to learn to use.l > = > What does this mean to VMS??  It means if VMS is to survivel9 > (and I think we all agree that the survival of VMS is a1; > good thing) people in a position to do something about it < > need to stop believing the myths they have heard somewhere> > about Unix and start treating it like the serious competitor > it is.  Marketing and non-technical reasons for having VMS a small market share is an entirely different discussion. UNIX is by far worse than its reputation. I'm not believing myths. I regret that I started using and trying to understand UNIX seriously because the UNIX people aren't serious. If I had read "The UNIX-Haters Handbook" earlier I would have safed a lot of time because the content of the book is true.   Alas the majority of people has the same sloppy attitude which explains why VMS has a small change to become a mass OS even if it goes open source.   < > Part of this points at the education program. How did Unix< > get it's start??  In schools and Universities where it was7 > cheaper than the alternatives.  It made more than one.< > generation of students familiar with it and they took this< > familiarity with them into the business world.  That gives< > them a big lead.  You can stand on your soapbox and scream; > at the top of your lungs that "Unix is garbage".  But the1< > world isn't listening because they know better.  They know; > that the INTERNET was built on Unix and runs primarily oni> > Unix (what do you think is inside most of those router boxes) > that make the Internet function, VMS??)B > 9 > The big question is, what are we going to do about it??b) > Attempt CPR or just continue the wake??d >  > 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   |@ > Scranton, Pennsylvania   |         #include <std.disclaimer.h>   ------------------------------    Date: 10 Mar 2001 17:27:00 -05009 From: Kilgallen@eisner.decus.org.nospam (Larry Kilgallen)o( Subject: Re: OpenVMS Educational Program3 Message-ID: <svWGduhI+3mK@eisner.encompasserve.org>n  X In article <3AAAA10E.F9F74273@infopuls.com>, Christof Brass <brass@infopuls.com> writes: > Alan Greig wrote:i >> c> >> On 8 Mar 2001 14:59:32 GMT, bill@triangle.cs.uofs.edu (Bill >> Gunshannon) wrote:d >> oJ >> >Sadly, while this stuff gets talked about here every couple of months,J >> >and everybody agrees that something needs to be done, nothing ever is. >> nI >> I would agree that this has been the case. Development wise though VMSmH >> is moving forward again. The DII COE work if done well should give us, >> the best of both worlds - VMS *and* Unix. >> m, >> What we still need is marketing and apps. >> e >> > >> >bill >> " >> --  >> Alanr > ^ > The best of both worlds is VMS. Any single element of UNIX put into VMS reduces its quality.  C I am enough of a VMS bigot that I cannot think of a counterexample,t@ but I am enough of a realist to say that I cannot endorse such a
 statement.  A If someone were to move to VMS only because of Unix compatibility B and thereafter discover some of the good features of VMS, presence8 of that Unix compatibility would have been a good thing.   ------------------------------  % Date: Sat, 10 Mar 2001 22:55:33 +0000 ) From: Christof Brass <brass@infopuls.com>l( Subject: Re: OpenVMS Educational Program, Message-ID: <3AAAB0E5.3662255E@infopuls.com>   Bill Gunshannon wrote: > 2 > In article <009F8B94.EC262094@SendSpamHere.ORG>,B >  system@SendSpamHere.ORG (Brian Schenkenberger, VAXman-) writes: > |>I > |> Again, Bill, I ask, "What medium do you get these unix products dis-  > |> tributed upon?" > ' > Some on CD and some over the network.W > I > |>                  There are a great many products in a quarterly dis- 2 > |> tribution.  Too many to fit onto a single CD. > % > I don't deny or begrudge them that.b >  > |>I > |> FYI, when I was on site at a major candy mfg. about two years ago, arJ > |> NT guys was running about with the latest shipment of PeeCee productsJ > |> from Micro$haft.  It dwarfed the quarterly VMS distribution by an or-J > |> der of magnitude and then some.  So, what I see is you are pissing onJ > |> VMS for having too many products?  Or Sony and Philips for not devis-5 > |> ing a CD standard with greater storage capacity.  > D > I don't deny this either.  But it is not germane to the discussionC > at hand.  My argument is with people that say that "Installing LPAD > software is easier on VMS than it is on any Unix.  Period.  End ofE > discussion" because it just isn't true.  And when I give a concretef? > repeatable example the subject changes and all kinds of other6C > conditions that were not part of the discussion are added.  I waseC > merely trying to say that for an experienced Unix Admin tasks arerB > just as easy, if not easier on Unix then they are for an equallyD > experienced VMS Admin on a VMS system.  Somehow, people here, mostH > of whom have little if any experience with Unix systems administration# > think this is somehow immposible.i >  > |>1 > |> BTW, what's your budget for other platforms?I > 7 > They spend many thousands of $$ on PeeCees evey year.- > F > |>                                              Just because the uniJ > |> won't cough up money for your VMS platforms, you want it for nothing? > |> That is absurd thinking.o > F > I think your missing this point as well.  It doesn't affect me or myH > job one bit if they make me throw all my VMS stuff in the dumpster andG > never set up another one.  I am not the only one here saying that VMS1G > needs to fight it's way back into academia.  And it's not going to be K > an easy fight.  But if it doesn't happen, who is going to be the looser??.H > Not Unix and not Microsoft.  They are entrenched.  They can sell theirF > products without even trying.  Right now, there is no reason for theI > University to think about VMS, much less spend even a dime on it.  ThatrJ > puts the ball firmly in Compaq's court.  If they are satisfied with thisK > then so be it.  But don't shoot the messenger.  I'm pointing out reality.lH > I didn't put Unix in all those universities so that it became a commonF > term in the computing industry.  Believe it or not, DEC is very muchF > responsible for that.  Now, if they want it to survive, they need toH > use the same strategy with VMS.  Myself and others think this would beH > a good thing.  Others appear not to agree as they think that by makingH > up ridiculous claims about Unix inferiority that they are going change= > this tide.  Which do you think is the more likely scenario.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   |@ > Scranton, Pennsylvania   |         #include <std.disclaimer.h>  The point is not what other people think about UNIX or VMS. The point is what is true: VMS is *by far* the superior OS. This can't be changed by looking at the market share or whatever. This NG isn't there to spread the good news of VMS. This NG is for discussion of VMS related topics and for helping each other with using VMS. We know that the majority likes crap, fast food and UNIX. We won't change that.   ------------------------------  % Date: Sat, 10 Mar 2001 23:01:00 +00004) From: Christof Brass <brass@infopuls.com> ( Subject: Re: OpenVMS Educational Program, Message-ID: <3AAAB22C.B404A6A5@infopuls.com>   Bill Gunshannon wrote: > . > In article <9889gn$597@gap.cco.caltech.edu>,7 >  mathog@seqaxp.bio.caltech.edu (David Mathog) writes:. > |> > |>J > |> OS, no.  Culture, yes.  Linux and netBSD are as often, or more often,' > |> loaded over the net than from CDs.- > D > Actually, with the ready availability of CD burners today the normD > is to grab the ISO image of the base package and burn your own CD.C > It's faster and easier to install the base system from a bootable D > CD.  The packages are best installed over the net although they to > can be done from a CD. > L > |> This isn't to suggest that all is sweetness and light in the Unix/LinuxL > |> world.  I upgraded a file server a couple of weeks ago from RH 5.1->6.2P > |> and it was very difficult to keep it from wiping all the other disks in the > |> system. > J > Read my previous comments regarding Linux.  If your only experience withI > a Unix like OS is with Linux, you have no experience with Unix.  It wasDM > another case of NIH syndrome leading to not doing things in a proven mannerDC > and adopting a harder, less reliable system just to be different.  >  > 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   |@ > Scranton, Pennsylvania   |         #include <std.disclaimer.h>  Linux is in fact not a good example. But even FreeBSD has the desinged in flaws like missing regularity and minimalism of service as attitude. The well organised FreeBSD community did a good job but couldn't change the well know problems of UNIX like the shell diversity, the broken user API wrt the shell 'syntax' and the lack of abstractions (e.g. only structureless byte stream files).   ------------------------------  % Date: Sat, 10 Mar 2001 17:43:36 -0600n7 From: "David J. Dachtera" <djesys.nospam@earthlink.net>e( Subject: Re: OpenVMS Educational Program- Message-ID: <3AAABC28.20590D2D@earthlink.net>c   Larry Kilgallen wrote: > Z > In article <3AAAA10E.F9F74273@infopuls.com>, Christof Brass <brass@infopuls.com> writes: > > Alan Greig wrote:a > >>@ > >> On 8 Mar 2001 14:59:32 GMT, bill@triangle.cs.uofs.edu (Bill > >> Gunshannon) wrote:a > >>L > >> >Sadly, while this stuff gets talked about here every couple of months,L > >> >and everybody agrees that something needs to be done, nothing ever is. > >>K > >> I would agree that this has been the case. Development wise though VMS J > >> is moving forward again. The DII COE work if done well should give us. > >> the best of both worlds - VMS *and* Unix. > >>. > >> What we still need is marketing and apps. > >> > >> >
 > >> >bill > >> > >> -- 	 > >> Alan  > > ` > > The best of both worlds is VMS. Any single element of UNIX put into VMS reduces its quality. > E > I am enough of a VMS bigot that I cannot think of a counterexample,tB > but I am enough of a realist to say that I cannot endorse such a > statement. > C > If someone were to move to VMS only because of Unix compatibilityrD > and thereafter discover some of the good features of VMS, presence: > of that Unix compatibility would have been a good thing.   I have a question, though.  G When we ran RSTS/E back around, oh V7.2 or so, using RMS was an option. F For example, BASIC-Plus had no RMS hooks (I think), while BASIC-Plus/2+ used RMS for all file access (fairly sure).   E So, if RMS is one of the many elements of VMS that UN*X systems lack,lF shouldn't the ability to easily side-step it be an enhancement to VMS?  F You can sort of do that now, I suppose, with the resultant files beingE seen by RMS as Stream_LF. However, I wouldn't know how to resolve the-A difference between UN*X text files and binary files other than tor, distinguish between Stream_LF and Fixed-512.  F I dunno. I just had to ask that. Other than restrictions pertaining toF ODS-2, most of which will be eliminated by the successor to ODS-5, RMSE seems (to me) to be one of the next biggest "hitch"-es in the portingu
 effort(s).   --   David J. Dachtera6 dba DJE Systemss http://www.djesys.com/  : Unofficial Affordable OpenVMS Home Page and Message Board: http://www.djesys.com/vms/soho/T  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: Sun, 11 Mar 2001 01:04:06 +0000n) From: Christof Brass <brass@infopuls.com>f( Subject: Re: OpenVMS Educational Program, Message-ID: <3AAACF05.49854345@infopuls.com>   Larry Kilgallen wrote: > Z > In article <3AAAA10E.F9F74273@infopuls.com>, Christof Brass <brass@infopuls.com> writes: > > Alan Greig wrote:  > >>@ > >> On 8 Mar 2001 14:59:32 GMT, bill@triangle.cs.uofs.edu (Bill > >> Gunshannon) wrote:l > >>L > >> >Sadly, while this stuff gets talked about here every couple of months,L > >> >and everybody agrees that something needs to be done, nothing ever is. > >>K > >> I would agree that this has been the case. Development wise though VMShJ > >> is moving forward again. The DII COE work if done well should give us. > >> the best of both worlds - VMS *and* Unix. > >>. > >> What we still need is marketing and apps. > >> > >> >
 > >> >bill > >> > >> --a	 > >> Alan  > >)` > > The best of both worlds is VMS. Any single element of UNIX put into VMS reduces its quality. > E > I am enough of a VMS bigot that I cannot think of a counterexample,nB > but I am enough of a realist to say that I cannot endorse such a > statement. > C > If someone were to move to VMS only because of Unix compatibility D > and thereafter discover some of the good features of VMS, presence: > of that Unix compatibility would have been a good thing.  dPlease, please, please: don't mix marketing and technique. I was referring to "The DII COE work if done well should give us the best of both worlds - VMS *and* Unix.". I'm already suspecting that this DII COE will reduce VMS' quality - at least by offering additional ways of doing something the UNIX way and by that making the OS complexer as is could be.   ------------------------------  % Date: Sun, 11 Mar 2001 06:54:33 +0100r  From: Paul Sture <paul@sture.ch>( Subject: Re: OpenVMS Educational Program+ Message-ID: <VA.000002fb.2299fb96@sture.ch>e  F In article <3AAABC28.20590D2D@earthlink.net>, David J. Dachtera wrote:9 > From: "David J. Dachtera" <djesys.nospam@earthlink.net>d > Newsgroups: comp.os.vms * > Subject: Re: OpenVMS Educational Program' > Date: Sat, 10 Mar 2001 17:43:36 -0600r >  > Larry Kilgallen wrote: > > \ > > In article <3AAAA10E.F9F74273@infopuls.com>, Christof Brass <brass@infopuls.com> writes: > > > Alan Greig wrote:  > > >>B > > >> On 8 Mar 2001 14:59:32 GMT, bill@triangle.cs.uofs.edu (Bill > > >> Gunshannon) wrote:t > > >>N > > >> >Sadly, while this stuff gets talked about here every couple of months,N > > >> >and everybody agrees that something needs to be done, nothing ever is. > > >>M > > >> I would agree that this has been the case. Development wise though VMS L > > >> is moving forward again. The DII COE work if done well should give us0 > > >> the best of both worlds - VMS *and* Unix. > > >>0 > > >> What we still need is marketing and apps. > > >> > > >> > > > >> >bill > > >>	 > > >> --  > > >> Alanx > > >rb > > > The best of both worlds is VMS. Any single element of UNIX put into VMS reduces its quality. > > G > > I am enough of a VMS bigot that I cannot think of a counterexample, D > > but I am enough of a realist to say that I cannot endorse such a > > statement. > > E > > If someone were to move to VMS only because of Unix compatibilityyF > > and thereafter discover some of the good features of VMS, presence< > > of that Unix compatibility would have been a good thing. >  > I have a question, though. > I > When we ran RSTS/E back around, oh V7.2 or so, using RMS was an option. H > For example, BASIC-Plus had no RMS hooks (I think), while BASIC-Plus/2- > used RMS for all file access (fairly sure).h >hN After my experience with CTS-300, which sat on top of RT-11, I was amazed thatJ RSTS didn't support some form of indexed file system. Indeed an OEM I cameR across was doing a great trade with a product called ACCESS which provided exactly that.l  G > So, if RMS is one of the many elements of VMS that UN*X systems lack,iH > shouldn't the ability to easily side-step it be an enhancement to VMS? > H > You can sort of do that now, I suppose, with the resultant files beingG > seen by RMS as Stream_LF. However, I wouldn't know how to resolve theaC > difference between UN*X text files and binary files other than toh. > distinguish between Stream_LF and Fixed-512. > I 20 years or so ago, those same ex-RSTS programmers used Fixed-512 and didp the indexing themselves.  M A total swine if you wanted a slightly different report than already provided I from one of those files, as you had to go cap in hand to the programmers,a; rather than do it yourself. Yes, RMS was a big improvement.e  > I am reminded of the truly appalling performance of the DEFRAGG database. It's been a couple of years since I looked, but IIRC it wouldnD quite happily do 3000 I/Os on an empty database with a SHOW command.  G Why do folks have to reinvent the wheel, especially when they get it soa drastically wrong?  H > I dunno. I just had to ask that. Other than restrictions pertaining toH > ODS-2, most of which will be eliminated by the successor to ODS-5, RMSG > seems (to me) to be one of the next biggest "hitch"-es in the portinge > effort(s). > D It's a good question. Coming from a mainframe background I've alwaysJ found it hard to understand why folks put up with an OS which doesn't comeK with indexed file support. Of course if my name were Larry Ellison I'd haveo a different answer to that...s ___h
 Paul Sture Switzerlandt   ------------------------------  % Date: Sat, 10 Mar 2001 22:13:55 +0100r= From: Arne =?iso-8859-1?Q?Vajh=F8j?= <arne.vajhoej@gtech.com>tY Subject: Re: Poor Administration? (was) Re: Russian Mafia Increases TCO Of M$Products <VBd) Message-ID: <3AAA9913.EF31ADAF@gtech.com>b   "Bradford J. Hamilton" wrote:nP > Can it be argued (after reading the article) that it was not the use of WinNT,J > but rather the *mis*use (not applying patches to fix several-years-knownB > vunerabilities) of the OS that led to this current tragi-comedy?  = But the bugs were made in the first place and MS is not quitee9 as pushy with patch-info as with info about new products.    N > Were these systems administered by MCSEs?  Is the quality of MCSE courses soI > poor that security becomes an afterthought, rather than a basic?  Is itlO > possible to mis-administer a VMS system in such a way that a similar incidentn > could occur on that system?t  G A sufficient bad VMS system manager can ofcourse screw up security on aw VMS: system.G  G I think the difference is that VMS is OK default and the system manageruD has actively to do something to screw up and that on NT it is enough  not to do something to screw up.  C I do not recall any similar holes in VMS since VMS 5.0 tigthened uph some things many years ago.d   Arne   ------------------------------  % Date: Sat, 10 Mar 2001 21:11:41 -0700a% From: Dean Woodward <deanw@rdrop.com> Y Subject: Re: Poor Administration? (was) Re: Russian Mafia Increases TCO Of M$Products <VBs) Message-ID: <3AAAFAFD.1F6F1008@rdrop.com>    "Bradford J. Hamilton" wrote:  > P > Can it be argued (after reading the article) that it was not the use of WinNT,J > but rather the *mis*use (not applying patches to fix several-years-knownB > vunerabilities) of the OS that led to this current tragi-comedy?  ? I also administer a handful of NT systems.  Out of the box, thepA first thing I have to do is remove "FULL CONTROL" from "EVERYONE"SA for most of the filesystem.  What more sign of lack of commitment.( to security from the vendor do you need?  N > Were these systems administered by MCSEs?  Is the quality of MCSE courses soI > poor that security becomes an afterthought, rather than a basic?  Is itpO > possible to mis-administer a VMS system in such a way that a similar incidente > could occur on that system?e  C Administered by MCSEs?  Who cares?  For years I wasn't an MCSE, and @ only became "certified" to impress customers.  I know of no MCSEC criteria (for NT) about how to set up a fully secure system, and it E would be flawed info in the first place, because of starting with the>G assumption that the system's not secure in the first place (see above).r  B Of course you can mis-manage a VMS system to be as insecure as NT-D but a VMS administrator has to work at it; on NT, it comes that way.   ------------------------------  % Date: Sun, 11 Mar 2001 11:44:29 +1100h- From: "Nick Ogbourne" <okeburnit@bigpond.com>r Subject: VMS release dates8 Message-ID: <0Gzq6.12186$v5.53177@newsfeeds.bigpond.com>  L Can anyone point me to a table that lists the release dates for the versions of VMS?a    
 Nick Ogbourney   ------------------------------  % Date: Sun, 11 Mar 2001 04:47:44 +0100n2 From: martin@radiogaga.harz.de (Martin Vorlaender) Subject: Re: VMS release dates; Message-ID: <3aaaf560.524144494f47414741@radiogaga.harz.de>i  , Nick Ogbourne (okeburnit@bigpond.com) wrote:N > Can anyone point me to a table that lists the release dates for the versions	 > of VMS?d  E http://www.openvms.compaq.com/openvms/os/openvms-release-history.htmlt   cu,o   Martin --J One OS to rule them all       | Martin Vorlaender  |  VMS & WNT programmer7 One OS to find them           | work: mv@pdv-systeme.deaJ One OS to bring them all      |   http://www.pdv-systeme.de/users/martinv/> And in the Darkness bind them.| home: martin@radiogaga.harz.de   ------------------------------   Date: 11 Mar 2001 04:05:08 GMT) From: leslie@clio.rice.edu (Jerry Leslie)$ Subject: Re: VMS release dates' Message-ID: <98ethk$1b4$1@joe.rice.edu>   , Nick Ogbourne (okeburnit@bigpond.com) wrote:> : Can anyone point me to a table that lists the release dates  : for the versions of VMS? :  : Nick Ogbourneg   From the VMS FAQ, available at:f  7   http://www.openvms.compaq.com/wizard/openvms_faq.htmlS   the section titled:   0   VMS1.   What is OpenVMS?  What is its history?   mentions the release history:   E http://www.openvms.compaq.com/openvms/os/openvms-release-history.htmls  4 --Jerry Leslie     (my opinions are strictly my own)   ------------------------------  % Date: Sat, 10 Mar 2001 22:09:17 +0100 = From: Arne =?iso-8859-1?Q?Vajh=F8j?= <arne.vajhoej@gtech.com>s) Subject: Re: [SURVEY] Info-VAX suscribersb) Message-ID: <3AAA97FD.4972ACAB@gtech.com>    Didier Morandi wrote:iQ > How come there are only 79 people currenty subscribed to the info-VAX list? Are'M > we only 79 persons dealing with VMS daily operations in the world (we couldtP > create a club) or most of the readers/participants use a news browser instead?  ? INFO-VAX/comp.os.vms consist of 3 forums that are all gatewayedf to each other:   - the news-group comp.os.vms5   - the "internet" mail-list INFO-VAX at mvb.saic.com B   - a collection of "BITNET" mail-lists INFO-VAX are several sitesE ["BITNET" is ofcourse not real BITNET anymore - they have switched to B internet a long time ago, but they use the real LISTSERV software]  A My estimate of readers would be 10000, 2000 and 500 respectively.m  ; I assume that the 79 subscribers are to one of the "BITNET"e mail-lists.    Arne   ------------------------------   End of INFO-VAX 2001.139 ************************