1 INFO-VAX	Sat, 29 Dec 2001	Volume 2001 : Issue 723       Contents:( Bochs x86 Emulator - VMS build procedure2 Re: Compaq still tries to spin Alphacide both ways1 Re: historical evidence of what went wrong at DEC 1 Re: historical evidence of what went wrong at DEC 1 Re: historical evidence of what went wrong at DEC 1 Re: historical evidence of what went wrong at DEC 1 Re: historical evidence of what went wrong at DEC 1 Re: historical evidence of what went wrong at DEC 1 Re: historical evidence of what went wrong at DEC 0 Re: HP admits it will kill VMS if merger suceeds0 Re: HP admits it will kill VMS if merger suceeds0 Re: HP admits it will kill VMS if merger suceeds0 Re: HP admits it will kill VMS if merger suceeds; Re: PDP-10 architectural flaws? (was: VMS missing features)  Personal WS 600 power supply testL VMS as a career move ( was Re: HP admits it will kill VMS if merger suceeds)P Re: VMS as a career move ( was Re: HP admits it will kill VMS if merger suceeds)P Re: VMS as a career move ( was Re: HP admits it will kill VMS if merger suceeds)P Re: VMS as a career move ( was Re: HP admits it will kill VMS if merger suceeds)P Re: VMS as a career move ( was Re: HP admits it will kill VMS if merger suceeds)! Re: VMS Mail sends HTML messages! ! Re: VMS Mail sends HTML messages! ) Re: Where oh where has The Register gone?   F ----------------------------------------------------------------------  % Date: Sat, 29 Dec 2001 14:22:35 +0000 . From: Graham Burley <100625.30@compuserve.com>1 Subject: Bochs x86 Emulator - VMS build procedure . Message-ID: <3C2DD1AB.5950E270@compuserve.com>  * Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit   ) I've been tinkering with BOCHS on VMS ...      WHAT IS BOCHS?  @   Bochs is a highly portable open source IA-32 (x86) PC emulatorC   written in C++, that runs on most popular platforms.  It includes B   emulation of the Intel x86 CPU, common I/O devices, and a customB   BIOS.  Currently, bochs can be compiled to emulate a 386, 486 orB   Pentium CPU.  Bochs is capable of running most Operating Systems<   inside the emulation including Linux, Windows 95, DOS, andC   Windows NT 4.  Bochs was written by Kevin Lawton and is currently D   maintained by the Bochs project at "http://bochs.sourceforge.net".  F   Bochs can be compiled and used in a variety of modes, some which areA   still in development.  The 'typical' use of bochs is to provide B   complete x86 PC emulation, including the x86 processor, hardwareG   devices, and memory.  This allows you to run OS's and software within @   the emulator on your workstation, much like you have a machine;   inside of a machine.  Bochs will allow you to run Win '95 :   applications on a Solaris machine with X11, for example.  *   Bochs is distributed under the GNU LGPL.    H I've had success building versions 1.2.1 and 1.3 on VMS, these work wellF enough to boot, install, and run Linux and DOS. I've put together some) bits to build 1.3 that can be found here:   8   http://www.encompasserve.org/~burley/bochs-1_3_vms.zip  F The zip file includes a DCL build procedure and some VMS specific codeA for limited CD-ROM support, the 1.3 source kit can be found here:   ;   http://prdownloads.sourceforge.net/bochs/bochs-1.3.tar.gz   H To build you need C++ and C compilers, to run you need DECwindows, these versions work for me:      OpenVMS Alpha 7.2-2    Compaq C++ V6.2-016    Compaq C V6.2-00   DECwindows V1.2-6   / I'd be grateful for any feedback, preferably to  burley(at)encompasserve.org     
 Graham Burley    ------------------------------  + Date: Sat, 29 Dec 2001 13:36:08 +0000 (UTC) 0 From: "Rupert Pigott" <Darkb00ng@btinternet.com>; Subject: Re: Compaq still tries to spin Alphacide both ways / Message-ID: <a0kgs6$lua$1@helle.btinternet.com>   3 Nick Maclaren <nmm1@cus.cam.ac.uk> wrote in message * news:9vtop7$lsd$1@pegasus.csx.cam.ac.uk...D > The whole point of the EPIC design was that compilers could use itE > to produce fantastically fast code - and, as some of us said in the F > very beginning would happen, this has not proved possible except forC > a subset of HPC codes.  I cannot see that an EV8 engine executing   I I'd be interested to see the IA-64 compiler writer hate & wish lists in a  years time. :)   Cheers,  Rupert   ------------------------------  # Date: Sat, 29 Dec 2001 15:38:04 GMT 2 From: Arthur Krewat <krewat@bartek.dontspamme.net>: Subject: Re: historical evidence of what went wrong at DEC5 Message-ID: <3C2DE213.7A92DF0C@bartek.dontspamme.net>    "David J. Dachtera" wrote: >  > Michael Zarlenga wrote:  > > I > > In comp.os.vms Bill Pechter <pechter@i4got.pechter.dyndns.org> wrote: * > > : As far as the  Unix vs. VMS issue...M > > : Considering I make my living on Unix boxes now... it took years of work < > > : to get me as productive as I was in days with VAX/VMS., > > : That's pretty high praise for OpenVMS. > > B > > This 15+yr VMS user who also has 2+yrs of Unix finds Unix muchC > > easier to use for most of the tasks I do.  I really hate to say A > > that, but its scripting languages are light years beyond DCL.  >  > In what way? > J > How do I get a shell to check/verify command line options and values (if% > any) before it invokes the program?   H Write the script to verify command line options, then do it's main work. Step 1, step 2...   . > How do I enable command/option abbreviation?  H This is UNIX, you don't! The best you can do is make commands/options asI short as possible (like -f <filename>). Can't get much shorter than that!   9 > How do I get filespec expansion to recurse directories?   H "find". In Bourne shell, a "for a in `find . -type f`" will loop through all files from "." and below.   J > How do I get the shell to perform simple numeric calculations? ...stringE > manipulation (concatenation, reduction ala DCL)? ...conversion from 6 > binary datatypes to string datatypes (ala F$CVxI())?  D In shell, there is no distinction between numbers and strings. DoingA "A=`expr 2 + 2`" will give you 4 ... bc is another way. Strings?  F A=${B}${C} - this concatenates B and C and assigns it to A - can't get3 much easier than that. Reduction? use sed/awk, etc.   I > How do I get the shell to "export" variables to "outer" levels (ala DCL  > global symbols)?  H From ksh/sh, do ". script", this runs the script and anything defined inI it with "A=thisandthat; export A" will wind up being in your main shell's  environment.  I > How do I get the shell to open/process an indexed file (i.e., do C-ISAM D > the way DCL does RMS)? ...return information about the system (alaI > F$GETSYI())? ...return info. re: the crontab or "atd" (ala F$GETQUI())?   H There are basic "read" primitives in the shell, but for more complicatedI tasks, awk works well and there are others. Information about the system? % A=`uname -a` sets A to (on my system) I "SunOS kilowatt 5.8 Generic_108529-10 i86pc i386 i86pc". uname -r returns  "5.8",etc. etc.   H > I'm sure there are features of the shells, perl, python, etc. that youI > find useful. To me, it's still all stone-age computing (Linux, Solaris,  > *BSD).  F When I first went to UNIX from VMS, I was considerably lost - within aD few months I was using Bourne shell like it was my first love. It is( considerable easier to program than DCL.  . > Of course, and as always, YMMV considerably.  
 same here!   aak    ------------------------------  # Date: Sat, 29 Dec 2001 16:05:00 GMT < From: "Stuart R. Fuller" <stu@cc49395-b.wodhvn1.mi.home.com>: Subject: Re: historical evidence of what went wrong at DEC0 Message-ID: <2lok0a.sh9.ln@dadsys1.fuller.local>  @ In comp.sys.dec David J. Dachtera <djesys.nospam@fsi.net> wrote: : Michael Zarlenga wrote:  :>  H :> In comp.os.vms Bill Pechter <pechter@i4got.pechter.dyndns.org> wrote:) :> : As far as the  Unix vs. VMS issue... L :> : Considering I make my living on Unix boxes now... it took years of work; :> : to get me as productive as I was in days with VAX/VMS. + :> : That's pretty high praise for OpenVMS.  :>  A :> This 15+yr VMS user who also has 2+yrs of Unix finds Unix much B :> easier to use for most of the tasks I do.  I really hate to say@ :> that, but its scripting languages are light years beyond DCL.   : In what way?  J : How do I get a shell to check/verify command line options and values (if& : any) before it invokes the program?   M Why is that an issue?  If, in DCL, you type an invalid command line option or H value, you get an error message saying you can't do that.  In Unix, manyI programs will not only give you an error message, but will also provide a  short help screen.  . : How do I enable command/option abbreviation?  N In a script (remember, scripting languages are the subject here), command line" abbreviations are NOT recommended.  : : How do I get filespec expansion to recurse directories?   L That's not a scripting language issue - that's RMS.  Besides, how do you getK DCL to give you a list of all the files that have been accessed (not simply M created) in the past 2 days, with a link to one or more other filenames, that G have a specific set of protections, and when done, display that list of 5 filenames in variety of formats to suit the occasion.   J : How do I get the shell to perform simple numeric calculations? ...stringE : manipulation (concatenation, reduction ala DCL)? ...conversion from 6 : binary datatypes to string datatypes (ala F$CVxI())?  % Simple numeric calculations?  In ksh:            let x=4*8          echo $x           String manipulation?  Use awk.  $ Datatype conversion?  Not necessary.  I : How do I get the shell to "export" variables to "outer" levels (ala DCL  : global symbols)?  % Good point.  How do you do it in DCL?    $ show sym test = %DCL-W-UNDSYM, undefined symbol - check validity and spelling  $ test:==this is a test  $ show sym test    TEST == "THIS IS A TEST" $ spawn ( %DCL-S-SPAWNED, process FULLER_1 spawned: %DCL-S-ATTACHED, terminal now attached to process FULLER_1 $ show sym test    TEST == "THIS IS A TEST" $ test:==this is a new test  $ show sym test    TEST == "THIS IS A NEW TEST" $ log 8   Process FULLER_1 logged out at 29-DEC-2001 10:42:07.795 %DCL-S-RETURNED, control returned to process FULLER_0  $ show sym test    TEST == "THIS IS A TEST"
             I : How do I get the shell to open/process an indexed file (i.e., do C-ISAM D : the way DCL does RMS)? ...return information about the system (alaI : F$GETSYI())? ...return info. re: the crontab or "atd" (ala F$GETQUI())?   H When was the last time you really needed to know the cluster incarnation
 date/time?  M In DCL, how does one do looping?  How does one perform a series of actions on N a group of filenames, or just a group of names (not necessarily files) without a lot of GOTO here and there.            Stu M         (not anti-VMS or anti-DCL - just tired of the "my OS or CLI is better F         than yours" crap.  It's just a different way of doing things.)   ------------------------------  # Date: Sat, 29 Dec 2001 17:53:35 GMT 1 From: "David J. Dachtera" <djesys.nospam@fsi.net> : Subject: Re: historical evidence of what went wrong at DEC' Message-ID: <3C2E03B6.C0D6F6A0@fsi.net>    Arthur Krewat wrote: >  > "David J. Dachtera" wrote: > > L > > How do I get a shell to check/verify command line options and values (if' > > any) before it invokes the program?  > J > Write the script to verify command line options, then do it's main work. > Step 1, step 2...   + ...but then *I* am doing it, not the shell.     > [snip]; > > How do I get filespec expansion to recurse directories?  > J > "find". In Bourne shell, a "for a in `find . -type f`" will loop through > all files from "." and below.   - "find" is an external program, not the shell.    L > > How do I get the shell to perform simple numeric calculations? ...stringG > > manipulation (concatenation, reduction ala DCL)? ...conversion from 8 > > binary datatypes to string datatypes (ala F$CVxI())? > F > In shell, there is no distinction between numbers and strings. DoingB > "A=`expr 2 + 2`" will give you 4 ... bc is another way. Strings?H > A=${B}${C} - this concatenates B and C and assigns it to A - can't get5 > much easier than that. Reduction? use sed/awk, etc.   8 "sed", "awk", etc. are external programs, not the shell.   K > > How do I get the shell to "export" variables to "outer" levels (ala DCL  > > global symbols)? > J > From ksh/sh, do ". script", this runs the script and anything defined inK > it with "A=thisandthat; export A" will wind up being in your main shell's  > environment.  G O.k. So, how do I get the same effect from a script at, say, a depth of D 8 or so to make a variable (or a change in its value) visible to theG script at depth 2 or 1? (Answer: you can't without jumping thru a lotta  hoops to make a work-around.)    K > > How do I get the shell to open/process an indexed file (i.e., do C-ISAM F > > the way DCL does RMS)? ...return information about the system (alaK > > F$GETSYI())? ...return info. re: the crontab or "atd" (ala F$GETQUI())?  > J > There are basic "read" primitives in the shell, but for more complicated- > tasks, awk works well and there are others.   ; ...but here again, that's "awk" or whatever, not the shell.    > Information about the system? ' > A=`uname -a` sets A to (on my system) K > "SunOS kilowatt 5.8 Generic_108529-10 i86pc i386 i86pc". uname -r returns  > "5.8",etc. etc.   A How do I retrieve (again, using *ONLY* the shell, not an external C program) memory size? ...system parameters? ...nodename? ..., etc.?   J > > I'm sure there are features of the shells, perl, python, etc. that youK > > find useful. To me, it's still all stone-age computing (Linux, Solaris, 
 > > *BSD). > H > When I first went to UNIX from VMS, I was considerably lost - within aF > few months I was using Bourne shell like it was my first love. It is* > considerable easier to program than DCL.  @ In what way? (as I asked in my first post) To quote my 5th grade8 teacher, "be specific, cite examples, make comparisons".  H I got my first System 3.7 UNIX system on a no-name M68K box when I stillH a relative VMS newbie, but with about five years of BASIC Plus-2 and twoD years of VAX BASIC and DCL behind me - had to learn DIBOL and MS-DOS> (ala IBM) at the same time. At the time, thought it (UN*X) wasE ultra-primitive and feature-poor. Hasn't changed much over the years, H IMO. Wish I still had my .exrc script for making "vi" somewhat EDT-like.  D Dunno. Still sounds to me like DCL has the shells beat hands down inF many ways, but the shells have advantages in others. Tough call. Kinda depends what you need.   --   David J. Dachtera  dba DJE Systems  http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/    ------------------------------    Date: 29 Dec 2001 12:13:03 -0600- From: Kilgallen@SpamCop.net (Larry Kilgallen) : Subject: Re: historical evidence of what went wrong at DEC3 Message-ID: <bki4smQvjDY0@eisner.encompasserve.org>   o In article <2lok0a.sh9.ln@dadsys1.fuller.local>, "Stuart R. Fuller" <stu@cc49395-b.wodhvn1.mi.home.com> writes: B > In comp.sys.dec David J. Dachtera <djesys.nospam@fsi.net> wrote:  K > : How do I get the shell to "export" variables to "outer" levels (ala DCL  > : global symbols)? > ' > Good point.  How do you do it in DCL?  >  > $ show sym test ? > %DCL-W-UNDSYM, undefined symbol - check validity and spelling  > $ test:==this is a test  > $ show sym test  >   TEST == "THIS IS A TEST"	 > $ spawn * > %DCL-S-SPAWNED, process FULLER_1 spawned< > %DCL-S-ATTACHED, terminal now attached to process FULLER_1 > $ show sym test  >   TEST == "THIS IS A TEST" > $ test:==this is a new test  > $ show sym test   >   TEST == "THIS IS A NEW TEST" > $ log : >   Process FULLER_1 logged out at 29-DEC-2001 10:42:07.797 > %DCL-S-RETURNED, control returned to process FULLER_0  > $ show sym test  >   TEST == "THIS IS A TEST"   >         Stu O >         (not anti-VMS or anti-DCL - just tired of the "my OS or CLI is better H >         than yours" crap.  It's just a different way of doing things.)  D Fine, but the counter-example you gave did not use outer DCL levels,D it used subprocesses.  The example David had in mind was nested DCL,; using "@" or similar techniques, not subprocess invocation.e   ------------------------------  # Date: Sat, 29 Dec 2001 18:15:16 GMTu1 From: "David J. Dachtera" <djesys.nospam@fsi.net>a: Subject: Re: historical evidence of what went wrong at DEC' Message-ID: <3C2E08C7.D874209D@fsi.net>    "Stuart R. Fuller" wrote:  > B > In comp.sys.dec David J. Dachtera <djesys.nospam@fsi.net> wrote: > : Michael Zarlenga wrote:c > :>J > :> In comp.os.vms Bill Pechter <pechter@i4got.pechter.dyndns.org> wrote:+ > :> : As far as the  Unix vs. VMS issue... N > :> : Considering I make my living on Unix boxes now... it took years of work= > :> : to get me as productive as I was in days with VAX/VMS.g- > :> : That's pretty high praise for OpenVMS.e > :>C > :> This 15+yr VMS user who also has 2+yrs of Unix finds Unix much D > :> easier to use for most of the tasks I do.  I really hate to sayB > :> that, but its scripting languages are light years beyond DCL. >  > : In what way?   No response, eh?  eL > : How do I get a shell to check/verify command line options and values (if' > : any) before it invokes the program?( > O > Why is that an issue?  If, in DCL, you type an invalid command line option orsJ > value, you get an error message saying you can't do that.  In Unix, manyK > programs will not only give you an error message, but will also provide ag > short help screen.  G ...but at that point, you've already expended the resources and machineaH cycles to invoke the program instead of catching the error prior to that# point. *THAT* is why it's an issue!M  u0 > : How do I enable command/option abbreviation? > P > In a script (remember, scripting languages are the subject here), command line$ > abbreviations are NOT recommended.  H ...or even possible! Digital also recommended against it, but of course,B everyone does it anyway ("SUBM/NOT" instead of "SUBMIT/NOTIFY" for	 example).m   ; > : How do I get filespec expansion to recurse directories?v > 6 > That's not a scripting language issue - that's RMS.   G Wrong-O! It's an ODS issue! RMS only comes into play when the system istG reading a directory file (or maybe not - not sure if that's done by theO% primitives or the higher level code).t   > Besides, how do you getrM > DCL to give you a list of all the files that have been accessed (not simply O > created) in the past 2 days, with a link to one or more other filenames, thatnI > have a specific set of protections, and when done, display that list of 7 > filenames in variety of formats to suit the occasion.e  E Again, this is an ODS issue. Remember: /bin/ls is *EXTERNAL* program!mC ...*NOT* the shell! If any of that were important, one could easily F write a (system of) DCL "script(s)" to perform those functions, within  the limits of ODS functionality.   L > : How do I get the shell to perform simple numeric calculations? ...stringG > : manipulation (concatenation, reduction ala DCL)? ...conversion fromd8 > : binary datatypes to string datatypes (ala F$CVxI())? > ' > Simple numeric calculations?  In ksh:e >  >         let x=4*8b >         echo $x  >   > String manipulation?  Use awk.  + Here again, that's "awk" - *NOT* the shell!D  M& > Datatype conversion?  Not necessary.   Don't bet on it!  dK > : How do I get the shell to "export" variables to "outer" levels (ala DCL  > : global symbols)? > ' > Good point.  How do you do it in DCL?e  ; $ A = 1		! Local symbol - visible at current depth or lowerC? $ A == 1	! Global symbol - visible at all depths (including 0).k  r > $ show sym testy? > %DCL-W-UNDSYM, undefined symbol - check validity and spelling- > $ test:==this is a test  > $ show sym test0 >   TEST == "THIS IS A TEST"	 > $ spawn0* > %DCL-S-SPAWNED, process FULLER_1 spawned< > %DCL-S-ATTACHED, terminal now attached to process FULLER_1 > $ show sym testc >   TEST == "THIS IS A TEST" > $ test:==this is a new test  > $ show sym teste  >   TEST == "THIS IS A NEW TEST" > $ logf: >   Process FULLER_1 logged out at 29-DEC-2001 10:42:07.797 > %DCL-S-RETURNED, control returned to process FULLER_0h > $ show sym testg >   TEST == "THIS IS A TEST"  # This is a documented behavior. Per:    $ CREATE D1.COMt $ A = 1 
 $ SH SYM A $ @D2t
 $ SH SYM A $ EXIT ^Z $o   $ CREATE D2.COMi
 $ SH SYM A $ A == 2 $ EXIT ^Z $e   Try THAT and see what you get.  K > : How do I get the shell to open/process an indexed file (i.e., do C-ISAM F > : the way DCL does RMS)? ...return information about the system (alaK > : F$GETSYI())? ...return info. re: the crontab or "atd" (ala F$GETQUI())?p > J > When was the last time you really needed to know the cluster incarnation > date/time?  H Recently wrote a SHCLU.COM procedure. Can't cut-and-paste a sample rightE now - I'd have to log off the internet and dial into work instead (noxH VPN tunnel). It shows not only when the cluster was founded, but by whom@ (which node), as well similar info. on each node of the cluster.8 Couldn't get that from SHOW CLUSTER the way I wanted it.  # > In DCL, how does one do looping?     $label:     .    .    . $ GOTO label  - > How does one perform a series of actions on P > a group of filenames, or just a group of names (not necessarily files) without > a lot of GOTO here and there.a  H DCL is a non-structured language. One of its many limitations. This does> not, however and IMO, reduce the value of its many advantages.   
 >         StueO >         (not anti-VMS or anti-DCL - just tired of the "my OS or CLI is betteroH >         than yours" crap.  It's just a different way of doing things.)  G Agreed. Depends what you need. The "problems" arise when you need to dorB something the environment doesn't support, such as long strings orE looping constructs in DCL, or indexed files or string reductions in ag
 shell script.   F I guess if an "ideal" CLI were possible, it would support the featuresA of the shells and the versatility and functionality of DCL. Maybe : DII-COE will bring something interesting - y'never know...   -- o David J. Dachteran dba DJE Systems  http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/a   ------------------------------    Date: 29 Dec 2001 12:39:49 -0600- From: Kilgallen@SpamCop.net (Larry Kilgallen)l: Subject: Re: historical evidence of what went wrong at DEC3 Message-ID: <EJRLJ4t3KlHN@eisner.encompasserve.org>h  [ In article <3C2E08C7.D874209D@fsi.net>, "David J. Dachtera" <djesys.nospam@fsi.net> writes:m > "Stuart R. Fuller" wrote:  >>  C >> In comp.sys.dec David J. Dachtera <djesys.nospam@fsi.net> wrote:   1 >> : How do I enable command/option abbreviation?t >> sQ >> In a script (remember, scripting languages are the subject here), command line % >> abbreviations are NOT recommended.n > J > ...or even possible! Digital also recommended against it, but of course,D > everyone does it anyway ("SUBM/NOT" instead of "SUBMIT/NOTIFY" for > example).l   Not everybody.   ------------------------------  % Date: Sat, 29 Dec 2001 14:25:04 +0000u, From: hmv@port.ac.uk (Mike Meredith at home): Subject: Re: historical evidence of what went wrong at DEC( Message-ID: <0ojk0a.76k1.ln@lucifer-ec0>  ' In article <3C2D48CE.FA4FF25A@fsi.net>, 4 	"David J. Dachtera" <djesys.nospam@fsi.net> writes: > Michael Zarlenga wrote:  >> rA >> This 15+yr VMS user who also has 2+yrs of Unix finds Unix muchuB >> easier to use for most of the tasks I do.  I really hate to say@ >> that, but its scripting languages are light years beyond DCL.  > What scripting languages ? Unix scripting is pretty attrocious@ once you get to something serious (anything more than 100 lines, and I switch to Perl). 1  J > How do I get a shell to check/verify command line options and values (if& > any) before it invokes the program?   ; Under Unix, it's the responsibility of the program to checkg@ command line options and values. This may be a mistake, but it's> too late to change now. And there is a utility to help process2 command line options in a shell script --- getopt.  . > How do I enable command/option abbreviation?    Two different things under Unix.  > Command abbrieviations under Unix are done with shell aliases.  3 Option abbrieviations are dealt with by the script.,  : > How do I get filespec expansion to recurse directories?   ? Filespec expansion is done by the shell. If you want to recurse 8 through directories, use a shell that supports it (zsh).  @ > How do I get the shell to perform simple numeric calculations?   See the expr utility   >  ...string3 > manipulation (concatenation, reduction ala DCL)?    ? No idea as I don't know DCL, but there's plenty of utilities tov manipulate strings.r   > ...conversion from6 > binary datatypes to string datatypes (ala F$CVxI())?  = This is an area where traditional scripting is weak ... you'da) probably have to use something like Perl.t  I > How do I get the shell to "export" variables to "outer" levels (ala DCLg > global symbols)?   Can't. Why do you need to ?   I > How do I get the shell to open/process an indexed file (i.e., do C-ISAMa  ? Can't unless your C-ISAM/whatever library supports utilities to, do the job.a  D > the way DCL does RMS)? ...return information about the system (alaI > F$GETSYI())? ...return info. re: the crontab or "atd" (ala F$GETQUI())?e  > No idea what F$GETSYSI() does, but there's plenty of utilities, that'll return information about the system.  ? And 'crontab -l' or 'atq' should show the information you need.?  H > I'm sure there are features of the shells, perl, python, etc. that youI > find useful. To me, it's still all stone-age computing (Linux, Solaris,o > *BSD).  > It may be stone-age, but it's a little harsh to criticise UnixA for not being VMS. I've never written a DCL script (fiddling with ? LOGIN.COM doesn't count), but if I ever do, I hope I won't makeh4 the mistake of trying to get VMS to behave like Unix   ------------------------------  % Date: Sat, 29 Dec 2001 03:16:33 -0500c- From: JF Mezei <jfmezei.spamnot@videotron.ca> 9 Subject: Re: HP admits it will kill VMS if merger suceeds=, Message-ID: <3C2D7BD4.29EC7509@videotron.ca>   the augur wrote:H > retire.  And of course, "awhile" means different things if you're like( > me, a 23 year old programmer for VMS,   N I regret getting involved in VMS. It was a very bad business move. Don't focusN on a losing operating system where people debate whether to create a dedicated) newsgroup to discuss its impending death.e  G > The main point is that VMS makes money, it makes a good enough amountt, > of money for any small-mid-large cap firm.  L VMS doesn't make money. It is a tool used by companies to get work done. AndN if you can get the same work done at a lesser cost, then your accountants willL get you to move to the cheaper solution, even if it means payiung someone toC babysit it and pressing alt-ctrl-del every 2 hours on 100 machines.t  M And when corporations are seing Compaq purposefully ignore VMS, do you reallysF thing that they see a bright future for VMS ? Do you really think thatJ Compaq's treatmeht of VMS (or lack thereof) is attracting the applicationsL that that your competitord alrteady have on their Unix servers but you can't get because you're on VMS ?2   > No one is going to justr > dump it overnight.  M Correct. But a VMS machine that is put in maintenance mode at a company whilemI the new applications are being developped on a Unix or NT machine doesn't K generate much programming/consultant business, especially because VMS takes.B carte of itself and requires nobody to babysit it like NT or Unix.  1 > Nobody is going to throw it away along with all D > its source code and customers into some black hole/bottomless pit.  K Humm, Look at Amiga, OS2, and now MPE. When vendors stop developping an OS,iD customers stop adding applications to it, hence no more business forK developping/selling applications on that platform. And if you know that younL will migrate to another platform within 5 years, do you really want to spendL much money developing something on VMS instead of starting to build your new' infrastructure right away, bit by bit ?2  G > VMS is not going to the farthest depths of Hell, never to return.  ItcF > will be here, sitting still, running its tasks 24/7, 30 years out ofD > 30, you better bet your bottom dollar on that one, uncle sam does.  L Yes, just like there are still a few Data General machines running AOS-VS in/ some corner. But nobody is developping for it.    G > high-end OS-es have been coming around for ages.  It would be hard tonH > catch up to those systems in terms of user favortism, even if VMS were > open-sourced n  F I disagree. It would not be that hard. But it would take an effort andJ marketing money and serious commitment from Compaq to make VMS part of itsL core product offerings. I still maintain that Compaq could have killed LinuxL if it had pitched VMS as an alternative to Microsoft. But Compaq isn't aboutK to do a 360 to its friend Microsoft. I would like to remind you that CompaqiS was the only large company that openly supported Microsoft in its anti-trust trial.a    C > If the economy were to turn around say in the new year, effectiveuE > immediately, right before the merger voting, I'm sure as hell there H > would be a New HP OpenVMS or whatever the hell they decide to call it.  I How do you reconcile the above with Carly's multiple statements that theykM would focus on NT and open/industry standard systems and reduce the number of. OSs they sell ?I    H > Remember, the New HP will be a merger of bloodsuckers, bigger, badder, > and hairier than ever.  K Nop. The new HP will be a bloated combination or 2 "old" companies that are,L too fat to be nimble and compete. It will make one bigger piece of lard thatK will be too busy cleaning the lint from its own navel to be able to compete F against the others. And it is likely that the merger will generate badJ financials and that Carly will have to play musical chairs ever quarter toO compensate for the bad news and not make Wall Street Casino Analysts too angry."  N The "new" HP will be made up of the same people such as Winkler who put sticksN into VMS's wheels. Things will not change for the better. At best, we can hope; for status quo, at worse, the same treatement that MPE got.L  B > Anyway, all you vms customers and sales people should relax--VMSD > wusn't raised by no idiot mother, she was raised correctly and can! > stand on her own two damn feet!s  K You forget that VMS is on life support because of over a decade of negative J marketing. I guess you were too young to see the ads by Digital urging VMS customers to migrate to NT.n  H > quite intact.  The only question will be whether or not it will be NewB > HP OpenVMS, Compaq OpenVMS, Redhat OpenVMS, IBM OpenVMS, US Army > OpenVMS, Microsoft OpenVMS,     M No. You should consider  what happened to Alpha. The same fate is waiting for0C VMS. They won't sell VM'S as a viable product, they will donate itsjJ intellectual property to someone such as Microsoft. They know that VMS hasN potential if someone wants to make it succeed, and they are not about to allowM VMS to be unleashed against the Microsoft products Compaq depends so much on..   ------------------------------  % Date: Sat, 29 Dec 2001 10:27:22 +0100t1 From: John McLean <mcleanj@swissonline.delete.ch>(9 Subject: Re: HP admits it will kill VMS if merger suceedso5 Message-ID: <3C2D8C7A.DE94780C@swissonline.delete.ch>Y   Just a few brief points ...i   the augur wrote: >  ....G > The main point is that VMS makes money, it makes a good enough amountp, > of money for any small-mid-large cap firm.  C Why doesn't Compaq do more to promote the use of VMS ?  What's yourh
 thoughts ?     ...TB > Let's face it, there are sexier os-es out there that claim to doC > the same, and Compaq/Digital has already damaged the image of VMS1H > enough.  Linux, UNIX, and even the IBM high-end OS-es have been comingD > around for ages.  It would be hard to catch up to those systems in< > terms of user favortism, even if VMS were open-sourced ...  E Maybe these other OSes get some publicity.  Maybe potential customers'B get told about their features and the benefits that these features2 offer.  Does the "S" in VMS really mean "secret" ?  C > If the economy were to turn around say in the new year, effective@E > immediately, right before the merger voting, I'm sure as hell there-H > would be a New HP OpenVMS or whatever the hell they decide to call it.F >  There would be some dump of money into VMS, but not much.  It would< > appear to me that HP/Compaq would want a more streamlined,D > industry-standard type of package.  That means UNIX, what every ITB > manager has heard of, Linux, the new catch phrase, and win crap.  H When Compaq say that they want to be "industry standard" in hardware and@ their aims in Global Services are mainly for "industry standard"G services, it does not bode well.  Sure there is the occasional customereG who wants IS software on VMS, but unfortunately this seems rare despiteh the idea making a lot of sense.   J > That is... unless you customers make enough noise and enough promises ofC > $$$, that those blood suckers in the top office decide to listen.,  C That a big ask when a lot of customers have seen that the seller isaH somewhat unenthusiastic about VMS.  Many - far too many - customers seemC to be working towards other platforms because Compaq give so littleoE visble reason to be confident about the future.  Maybe there are manymG reasons to be confident and Compaq is just incapable of saying them....d  H > Whether or not VMS gets sold or not, it seems to me that its future isH > quite intact.  The only question will be whether or not it will be NewB > HP OpenVMS, Compaq OpenVMS, Redhat OpenVMS, IBM OpenVMS, US ArmyF > OpenVMS, Microsoft OpenVMS, Fujitsu OpenVMS, VMS, EDS OpenVMS, AppleF > OpenVMS (chuckle, fat chance), Novell OpenVMS, Smurf OpenVMS, Debian6 > OpenVMS, Suse OpenVMS, or maybe just plain OpenVMS.  > G > I say OpenVMS finds some vc money and flies solo from here on out, weoC > could be a spinoff from HP called OpenVMS Technologies, much like , > Agilent Technologies (symbol on NYSE: A ).  H Rebuilding from the years of neglect would be a rather tough call on theG new owner, not impossible just tough.   They may have to actually spendt some money on promotion.  C A year of two back CPQ spent $12 million on advertising VMS when itiC (with storage and other bits and pieces) was a revenue stream of $4vE billion.  That's 0.3% of revenue being spent to promote the sales.  IgG wonder how that percentage compares with recommended business practicesr ...)     John McLeany   ------------------------------  % Date: Sat, 29 Dec 2001 20:55:46 +1100 , From: israel r t <israelrt@optushome.com.au>9 Subject: Re: HP admits it will kill VMS if merger suceedsi8 Message-ID: <lm4r2uk7tlb17au9ibc1jij41i63r3efmg@4ax.com>  / On Sat, 29 Dec 2001 10:27:22 +0100, John McLean & <mcleanj@swissonline.delete.ch> wrote:  D >Why doesn't Compaq do more to promote the use of VMS ?  What's your >thoughts ?   D It is seen as too high risk a strategy to stake money on a niche os.A Compaq sees NT and possibly Unix as being the horses to hitch itsc fortunes to.   ------------------------------    Date: 29 Dec 2001 06:46:28 -0800) From: P.Young@unsw.EDU.AU (Patrick Young)?9 Subject: Re: HP admits it will kill VMS if merger suceeds = Message-ID: <55f85d77.0112290646.4ddafe10@posting.google.com>-  j theaugur@hotmail.com (the augur) wrote in message news:<d39e6c16.0112282257.ed5b46a@posting.google.com>...  G > me, a 23 year old programmer for VMS, in which I then can't speculate?H > for, but I'm targeting the major audience, probably those 30+ people.   ) 35 (just) - who is counting - i'm not :-(1  G > The main point is that VMS makes money, it makes a good enough amounttE > of money for any small-mid-large cap firm.  No one is going to just E > dump it overnight.  Nobody is going to throw it away along with allT  B Thank you! Just the answer I was waiting for. OpenVMS is around as long as people buy it.  H Simple as that folks. I guess this is the answer to all those pesky 120+ threads floated around.n  H (Compaq still need to know that people ***WANT*** to buy it, but may notH be able to since they are forced into crapware by budget constraints and crap employees)e  . Well, we have a 23 year old and a 35 year old.  C Anyone else want to join the party of folks wishing to use OpenVMS?A  < Need to corner those 18+ year olds - Should not be too hard.  F > OpenVMS, Microsoft OpenVMS, Fujitsu OpenVMS, VMS, EDS OpenVMS, AppleF > OpenVMS (chuckle, fat chance), Novell OpenVMS, Smurf OpenVMS, DebianF > OpenVMS, Suse OpenVMS, or maybe just plain OpenVMS.  Take your pick, > baby!a  F The folks doing the current development have my vote - I'd not want to3 see it go to Novell, Fujitsu, Apple or anyone else.e  D M$ - They would kill it as soon as look at it - let us not go there.  % Many thanks - a most refreshing post!u   ------------------------------  % Date: Sat, 29 Dec 2001 12:29:58 +0100 ( From: Paul Sture <paul.sture@bluewin.ch>D Subject: Re: PDP-10 architectural flaws? (was: VMS missing features)4 Message-ID: <VA.00000505.23dc370e@bluewin.delete.ch>  G In article <a0cleu$31l$1@citadel.in.taronga.com>, Peter da Silva wrote:s/ > In article <VA.00000502.14358048@bluewin.ch>, , > Paul Sture  <paul.sture@bluewin.ch> wrote:R > >Splutter, cough. Was recommended to do that by IBM more than 20 years ago when E > >we had a mainframe printer problem. It was a hardware problem, butuG > >I still had to go through the hoop of reinstalling software to proveO > >it wasn't software's fault. > > > You had to reinstall the whole OS? Or just that one package? >u  I Just stuff relevant to the printer, although there was probably more thanuA one component. It was a political thing, as we knew that IBM wereeJ throwing their engineers at a major problem at the biggest IBM customer inF town, and we were somewhat upset at the initial fobbing off. I did theM reinstallation as a preemptive strike, so that on my next call I could simplyLN say "Done that, please elevate the problem". IBM's elevation procedures workedJ beautifully :-) Given that it was indeed a h/w problem, the reinstallation< was probably not needed, but it got us the required results.  K > I can see reinstalling a program, that's easy and fast and doesn't impactwO > other use of the machine (except on Windows, where Microsoft uses application2M > vendors as OS update shippers, so you have to reboot). But reinstalling theo > OS?v >nF No, definitely not the whole OS. That would have been time to grab the
 backup tapes.n ___c
 Paul Sture Switzerland.   ------------------------------    Date: 29 Dec 2001 08:03:48 -0800) From: P.Young@unsw.EDU.AU (Patrick Young)c% Subject: Personal WS 600 power supplye= Message-ID: <55f85d77.0112290803.3e5f209c@posting.google.com>t  < The power supply in a Personal Workstation 600au I work with3 seems to have died badly (exploded) over the break.r  = Has anyone here experienced this failure of the power supply?   = It is an API-6029 unit (noname "MADE IN CHINA MAINLAND") that   uses a IRFPC50 switching MOSFET.  : I've used an IRF450 MOSFET from a broken TA90 power supply= to replace the failed part. Good or bad move? Seems to exceedk the replaced part.  = DEC appeared to be big on International Rectifier. I'm not so : sure of their quality (although the demand placed on these6 power supply parts is high) so I am most likely wrong.   What have others used?  4 In Australia getting any of these parts is not easy.   Many thanks.   ------------------------------  % Date: Sat, 29 Dec 2001 14:57:48 +0100 / From: Paul Sture <paul.sture@bluewin.delete.ch>n
 Subject: testt4 Message-ID: <VA.00000507.24638e9c@bluewin.delete.ch>   ___n
 Paul Sture Switzerlandn   ------------------------------  % Date: Sat, 29 Dec 2001 20:56:37 +1100h, From: israel r t <israelrt@optushome.com.au>U Subject: VMS as a career move ( was Re: HP admits it will kill VMS if merger suceeds) 8 Message-ID: <3q4r2uc6ad3r4kbclnejf61fn08q5po4ie@4ax.com>  , On Sat, 29 Dec 2001 03:16:33 -0500, JF Mezei% <jfmezei.spamnot@videotron.ca> wrote:e  O >I regret getting involved in VMS. It was a very bad business move. Don't focustO >on a losing operating system where people debate whether to create a dedicatedh* >newsgroup to discuss its impending death.  E I have a client who looks after the VMS systems at one of the leading  teaching hospitals in Sydney.   F He sees more and more Unix and NT boxes being brought in. He says thatD he is often asked "when do you plan to retire" . He is 53 years old.@ They cannot sack him even when the VMS machines eventually go inA another 3 to 4 years .  Australian industrial law makes it almost.5 impossible to sack someone except for criminal acts .L  D He found it funny in the beginning . After all, he has  little or noC work to do, he has guaranteed  lifetime job security and receives a F six figure salary . However , he now has started to feel humiliated byE the frequent "when do you plan to retire" and is retraining in Oracle  on Unix boxes.   ------------------------------    Date: 29 Dec 2001 07:33:28 -0800) From: P.Young@unsw.EDU.AU (Patrick Young) Y Subject: Re: VMS as a career move ( was Re: HP admits it will kill VMS if merger suceeds)s= Message-ID: <55f85d77.0112290733.417441af@posting.google.com>b  8 israel r t <israelrt@optushome.com.au> wrote in message   sG > I have a client who looks after the VMS systems at one of the leadingF  > teaching hospitals in Sydney.  > H > He sees more and more Unix and NT boxes being brought in. He says that  H Funny that - I did at one point. Everyone wanted their own web site, andF their own "Server". The resultant mess was something indeed to behold.  F My only competitor these days is Novell. NT sort of stalled (yes I didA have a hand in that) on the "server" (I *$#$%HATE$#& that word toh7 describe a computer running a non M$ OS) end of things.   @ Tru64 will be around for a while for the UNIX only requirements.  ' All the web stuff - OpenVMS/Apache now.t  1 Now, would you please go troll someplace else????-   ------------------------------  % Date: Sat, 29 Dec 2001 09:09:02 -0700<$ From: "rcarter" <rcarter@zianet.com>Y Subject: Re: VMS as a career move ( was Re: HP admits it will kill VMS if merger suceeds)r* Message-ID: <3c2debee.0@oracle.zianet.com>  J I wonder how long True64 will be around. Compaq has already announced theyI will stop making Alpha chips - it's a lot cheaper to buy an intel box and K run Linux. At work we are in the process of replacing our aging Alphas withwH Dell supplied RH7.1 boxes - much faster, fairly inexpensive, and minimal porting problems so far.    4 Patrick Young <P.Young@unsw.EDU.AU> wrote in message7 news:55f85d77.0112290733.417441af@posting.google.com...m9 > israel r t <israelrt@optushome.com.au> wrote in message  >yI > > I have a client who looks after the VMS systems at one of the leading ! > > teaching hospitals in Sydney.t > >eJ > > He sees more and more Unix and NT boxes being brought in. He says that >2J > Funny that - I did at one point. Everyone wanted their own web site, andH > their own "Server". The resultant mess was something indeed to behold. > H > My only competitor these days is Novell. NT sort of stalled (yes I didC > have a hand in that) on the "server" (I *$#$%HATE$#& that word too9 > describe a computer running a non M$ OS) end of things.M >2B > Tru64 will be around for a while for the UNIX only requirements. >V) > All the web stuff - OpenVMS/Apache now.a > 3 > Now, would you please go troll someplace else????p   ------------------------------   Date: 29 Dec 2001 18:29:23 GMT) From: leslie@clio.rice.edu (Jerry Leslie)eY Subject: Re: VMS as a career move ( was Re: HP admits it will kill VMS if merger suceeds) ' Message-ID: <a0l223$6m7$1@joe.rice.edu>o  # rcarter (rcarter@zianet.com) wrote: + : I wonder how long True64 will be around. O :   9    http://news.zdnet.co.uk/story/0,,t269-s2100182,00.html 8    ZDNet |UK| - News - Story - HP brings Unixes together      HP brings Unixes together  "    15:29 Friday 30th November 2001    Martin Veitch, IT Weekn  H   "The HP-UX system will form the bulk of the post-merger company's Unix    server offering  H    Hewlett-Packard's HP-UX is to continue to form the bulk of the firm'sH    Unix offering after the acquisition of Compaq is completed next year.A    This will mean upheaval for firms using Compaq's Tru64 Unix as I    elements of Tru64 are added to HP-UX and, as previously announced, them.    AlphaServer family is slowly phased out..."  *    http://www.theinquirer.net/28120102.htm     Compaq readying Tru64 layoffs       Compaq readying Tru64 layoffs    HP or no HP  )    By Mike Magee, 28/12/2001 09:32:14 BSTt  H   "SOURCES AT COMPAQ Germany tell the INQUIRER that the firm is readyingF    layoffs of staff working on its version of Unix, Tru64, whether theB    takeover deal between it and Hewlett Packard happens or not..."    4 --Jerry Leslie     (my opinions are strictly my own)   ------------------------------  % Date: Sat, 29 Dec 2001 13:13:35 -0500o' From: Howard S Shubs <howard@shubs.net> Y Subject: Re: VMS as a career move ( was Re: HP admits it will kill VMS if merger suceeds)c< Message-ID: <howard-B99BCE.13133529122001@enews.newsguy.com>  J In article <3c2debee.0@oracle.zianet.com>, "rcarter" <rcarter@zianet.com>  wrote:  L > I wonder how long True64 will be around. Compaq has already announced they  ; You might as well spell it right while it's around: "Tru64"  -- u Howard S ShubsD "Run in circles, scream and shout!"  "I hope you have good backups!"- Xerox is the anti-Microsoft.  And visa-versa.    ------------------------------    Date: 29 Dec 2001 06:59:05 -0800( From: bob@instantwhip.com (Bob Ceculski)* Subject: Re: VMS Mail sends HTML messages!= Message-ID: <d7791aa1.0112290659.39647ebc@posting.google.com>a  \ "John Vottero" <John@mvpsi.com> wrote in message news:<u2podqsrkjin87@news.supernews.com>...M > MIME headers are being added to VMSmail messages so the MIME headers hiddent > in the subject are not used. >    is this also true in vms 7.2?l   ------------------------------  % Date: Sat, 29 Dec 2001 12:33:12 -0500n From: gce <ge@gce.com>* Subject: Re: VMS Mail sends HTML messages!' Message-ID: <3C2DFE58.1FB2698A@gce.com>u  2 Might using "mail/old" instead of "mail" fix this?D Mail/old is the older Bliss implementation. Mail gets the C version.G I have been under the impression the Bliss version is being left alone.u     John Vottero wrote:s > M > MIME headers are being added to VMSmail messages so the MIME headers hiddenc > in the subject are not used. > @ > "John McLean" <mcleanj@swissonline.delete.ch> wrote in message1 > news:3C2CB4E0.DD3C7B86@swissonline.delete.ch...o	 > > John,t > >sK > > What breaks and in what way ?  Is it with the html header stuff or doeso > > something else fail ?r > >o > >w > > John McLeanO > >S > >p > > John Vottero wrote:S > > >p1 > > > Don't get too excited, this breaks in V7.3.  > > > ; > > > "Bob Ceculski" <bob@instantwhip.com> wrote in messaget= > > > news:d7791aa1.0112271655.42c7dfea@posting.google.com...iK > > > > someone posted this awhile ago on this site ... I needed to send an  > htmlL > > > > confirmation message to customers on our site and found a simple way > to doM	 > > > it!e/ > > > > this was what was originally posted ...d > > > >  > > > > $ cr[0,8] = 13 > > > >    $ lf[0,8] = 10,4 > > > >    $ subject = "This is a test." + cr + lf -/ > > > >    _$ + "Mime-Version: 1.0" + cr + lf -s- > > > >    _$ + "Content-Type: text/enriched"D% > > > >    $ write sys$output subject. > > > >    This is a test. > > > >    Mime-Version: 1.0& > > > >    Content-Type: text/enrichedL > > > >    $ mail/subject="''subject'" tt: "SMTP%""anybody@xyzcompany.com"""M > > > >    Enter your message below. Press CTRL/Z when complete, or CTRL/C tou > > > >    quit:! > > > >    <bold>Sometimes</bold>s1 > > > >    <x-color><param>red</param>I</x-color> . > > > >    even amaze <italic>myself</italic>!1 > > > >    ctrl-Z <-- i.e., here I typed a ctrl-zh > > > >    $ > > > >  > > > > 7 > > > > just do this, changing "enriched" to "html" ...yE > > > > if you create in any language a variable and define it as the  > following 	 > > > ...g > > > >uF > > > > SUB = 'Web Order<CR><LF>Mime-Version: 1.0<CR><LF>Content-Type: > text/html' > > > > N > > > > and use this variable as the subject content in the mail command, then > youo	 > > > candI > > > > send an html based file as a the message and it works great!  Thet > abovesN > > > > <CR><LF> were created w/EDT GOLD KEY, then type decimal 13 or 10, then > EDTl, > > > > GOLD 3 (specins) ... it's that easy!   ------------------------------  % Date: Sat, 29 Dec 2001 12:29:58 +0100m( From: Paul Sture <paul.sture@bluewin.ch>2 Subject: Re: Where oh where has The Register gone?4 Message-ID: <VA.00000506.23dc3722@bluewin.delete.ch>  D In article <3C2B3EC8.AD0E2B0B@eisner.decus.org>, Duncan Brown wrote: > Tim Llewellyn wrote: > >  > > Duncan Brown wrote:r > > >t > > > Tim Llewellyn wrote: > > > >oJ > > > > FWIW, I have just connected to the register site fine from the UK. > > > K > > > Can you please tell us what the IP address is?  I found an article in  > >  > > OK, sorry about the DOSe > > 1 > > C:\WINDOWS\Desktop>ping www.theregister.co.ukl > > H > > Pinging www.theregister.co.uk [213.40.196.64] with 32 bytes of data: >  > Bingo - that does it.  > F > http://213.40.196.64/       will get you to their main page.  ThoughB > there is only one article new from today, and just a couple from2 > yesterday.  Something odd is obviously going on! > I The trick I used was to feed www.theregister.co.uk into the "What's that pA site running?" facility at www.netcraft.com . The history recordsi revealed the IP address. ___s
 Paul Sture Switzerlandl   ------------------------------   End of INFO-VAX 2001.723 ************************