0 INFO-VAX	Wed, 19 Jan 2005	Volume 2005 : Issue 38      Contents:% An Interview With Mr. OpenVMS Himself  AS 1000 boot failure (battery?)  Re: carly(tm) transcipt  Re: carly(tm) transcipt  GnuPG in batch1 Re: Marcello predicts 500% perfomance improvement 1 Re: Marcello predicts 500% perfomance improvement  Minimerge on HSG80 in V8.2?  Re: Minimerge on HSG80 in V8.2? & New Command Language (CL) for OpenVMS?* Re: New Command Language (CL) for OpenVMS?* Re: New Command Language (CL) for OpenVMS?* Re: New Command Language (CL) for OpenVMS?* Re: New Command Language (CL) for OpenVMS?* Re: New Command Language (CL) for OpenVMS? Re: Next Gen Fabs & Itanium  Re: vms versus solaris Re: vms versus solaris Re: vms versus solaris Re: vms versus solaris Re: vms versus solaris Re: vms versus solaris Re: vms versus solaris Re: vms versus solaris Re: vms versus solaris Re: vms versus solaris Re: vms versus solaris Re: vms versus solaris Re: vms versus solaris Re: vms versus solaris Re: vms versus solaris Re: vms versus solaris Re: vms versus solaris Re: vms versus solaris Re: vms versus solaris Re: vms versus solaris Re: vms versus solaris Re: vms versus solaris Re: vms versus solaris Re: vms versus solaris Re: vms versus solaris Re: vms versus solaris Re: vms versus solaris Re: vms versus solaris Re: vms versus solaris Re: Webcast and VMS   F ----------------------------------------------------------------------  # Date: Wed, 19 Jan 2005 17:47:30 GMT 6 From: "Kenneth Farmer" <kfarmer@NOSPAM.spyderbyte.com>. Subject: An Interview With Mr. OpenVMS Himself= Message-ID: <S2xHd.4851$K72.1215750@twister.southeast.rr.com>    ShannonKnowsHPC.com:A http://www.shannonknowshpc.com/stories.php?story=05/01/18/0435366        Ken    OpenVMS.org % _____________________________________  Kenneth R. Farmer <>< % SpyderByte: http://www.SpyderByte.com    ------------------------------  # Date: Wed, 19 Jan 2005 17:08:01 GMT , From: "Ransom Fitch" <rlf_vms@earthlink.net>( Subject: AS 1000 boot failure (battery?)@ Message-ID: <RtwHd.131$r27.110@newsread1.news.pas.earthlink.net>  6 When I attempt a cold boot on a AS 1000 it fails with:  , EISA Data in non-volatile storage is corrupt EISA Configuration Error.....     H After running ECU I can boot normally (requires setting date and time).  Reboots (warm boots) work fine.   M I suspect that this is due to a dead battery, and if so may have been caused  I by a KVM switch. The only thing that I find that looks like a battery is  J ~0.5(w) x 0.375(h) x 2.0(l) inches that's marked "lithium battery...." on  the main board(?).   Any suggestions on: ! 1) determining if battery is dead $ 2) finding a replacement (if needed)M 3) installing a replacement (if needed).  I believe that this is (multi-)pin  3 socket but may require some tool(s) for extraction.   , If the problem is not the battery then what?     thanks,    Ransom Fitch rlf_vms at earthlink dot net     ------------------------------    Date: 19 Jan 2005 06:30:46 -0800$ From: "AEF" <spamsink2001@yahoo.com>  Subject: Re: carly(tm) transciptC Message-ID: <1106145046.812812.194540@z14g2000cwz.googlegroups.com>    JF Mezei wrote:  > Dave Froble wrote: > > > Are you for real?  > > >  > > >  > > G > > Don't go there.  It's a bit like, "Be careful what you ask for, you 
 just might > > get it." > D > Oh come on guys. There have been a constant flow of comments aboutF > Carly's presentation, hairdo, private jets etc etc.  Why make such a big 2 > fuss about yet another comment about her image ?    F Hmmm. I thought we were above that in c.o.v. Do you make equally inane> comments in the other groups, thereby attracting undesirables?   ------------------------------  # Date: Wed, 19 Jan 2005 17:34:25 GMT * From: "FredK" <fred.nospam@nospam.dec.com>  Subject: Re: carly(tm) transcipt2 Message-ID: <BSwHd.6128$yW5.1845@news.cpqcorp.net>   >  > JF Mezei wrote:  > > Dave Froble wrote: > > > > Are you for real?  > > > >  > > > >  > > > I > > > Don't go there.  It's a bit like, "Be careful what you ask for, you  > just might > > > get it." > > F > > Oh come on guys. There have been a constant flow of comments aboutH > > Carly's presentation, hairdo, private jets etc etc.  Why make such a > big 4 > > fuss about yet another comment about her image ? >   = You really don't get it.  I imagine that you thought you were = being funny or witty.  What you come off as is a sexist fool.   9 If anyone *was* inclined to take you seriously, it simply 1 lessens that possibility (or reduces the number).    ------------------------------  % Date: Wed, 19 Jan 2005 14:44:31 +0000 % From: David Gray <police@spamcop.net>  Subject: GnuPG in batch 8 Message-ID: <ccssu09lrehgkq3a6rik3nvdluhblah07p@4ax.com>   Hi all,    OpenVMS	 7.3-2   GnuPG  		v1.2.3   D Has anyone got an example of how to use the PASSPHRASE-FD command ofC GnuPG when running in batch?  I've searched the net but cannot find # any examples of it's use under VMS.   B I'm currently writing the GPG command into a temporay command file: with the passphrase on the second line but wondered if the1 PASSPHRASE-FD would provide a 'neater' solution.       Thanks in advance 	 	Dave.       ------------------------------  # Date: Wed, 19 Jan 2005 17:53:48 GMT * From: "FredK" <fred.nospam@nospam.dec.com>: Subject: Re: Marcello predicts 500% perfomance improvement2 Message-ID: <M8xHd.6130$226.1652@news.cpqcorp.net>  4 "Dave Froble" <davef@tsoft-inc.com> wrote in message$ news:41EDD481.60201@tsoft-inc.com... > John Smith wrote:   K > Hey, now is now.  The itanic is all we got.  If it's working well enough,  fineL > by me.  Yes, I think EPIC is rather a stupid thing to do.  So what.  If it runs@ > my programs, I'm a whole bunch happier than having no options. > G > My big concern is whether the itanic will survive.  If Alpha couldn't 
 survive in0 > the low volume world, why would any other CPU?  K The volume is better, even if it isn't the size of the x86 business.  HP-UX H is a bigger business than Tru64 was.  The UNIX world is shrinking to theJ point of only having 2-3 serious vendors HP, IBM and SUN.  Of those three,G SUN is the one who has not been able to at least squeek out break even. B POWER and Itanium will be the only 2 surviving RISC architectures.  G So the only question remains is if Windows and Linux squeeze out higher > end UNIX on RISC platforms from the bottom (IA32) up (x86-64).  K I haven't seen "mission critical" Linux or Windows yet, but it doesn't mean ; that it will never happen.  In any case, it isn't here yet.   B We've (VMS) invested 3-1/2 years of time and money in Itanium, andE for the forseable future, I don't see large scale AMD 64 systems with  mission H critical Windows and Linux.  The latest public move with Itanium removesD any stigma of having HP (a system competetor) from controlling IA64,J or having an unfair inside advantage.  If we or Intel thought that ItaniumC was going away, it's likely that any move would have been the other 
 direction.  B Just because there are dissapointed people out there mad about theA demise of Alpha, does not mean that Itanium cannot succeed as one F of the 2 premier high-end, mission critical, UNIX (and VMS) platforms.H Will x86 be able to be grown "up" into this role?  Time will tell.  With enough thrust, a pig will fly.   ------------------------------  % Date: Wed, 19 Jan 2005 13:12:53 -0500 # From: "John Smith" <a@nonymous.com> : Subject: Re: Marcello predicts 500% perfomance improvement, Message-ID: <BMednc1uOcG6PnPcRVn-oA@igs.net>   FredK wrote:6 > "Dave Froble" <davef@tsoft-inc.com> wrote in message& > news:41EDD481.60201@tsoft-inc.com... >> John Smith wrote: > D >> Hey, now is now.  The itanic is all we got.  If it's working wellE >> enough, fine by me.  Yes, I think EPIC is rather a stupid thing to C >> do.  So what.  If it runs my programs, I'm a whole bunch happier  >> than having no options. >>? >> My big concern is whether the itanic will survive.  If Alpha E >> couldn't survive in the low volume world, why would any other CPU?  > F > The volume is better, even if it isn't the size of the x86 business.? > HP-UX is a bigger business than Tru64 was.  The UNIX world is G > shrinking to the point of only having 2-3 serious vendors HP, IBM and B > SUN.  Of those three, SUN is the one who has not been able to atC > least squeek out break even. POWER and Itanium will be the only 2  > surviving RISC architectures.  > B > So the only question remains is if Windows and Linux squeeze outG > higher end UNIX on RISC platforms from the bottom (IA32) up (x86-64).  > @ > I haven't seen "mission critical" Linux or Windows yet, but itE > doesn't mean that it will never happen.  In any case, it isn't here  > yet.  K That does not stop people/corporations from believeing that it is there, or I nearly so. And that comes back to the issue of perception vs. reality, or K perhaps more properly put in this context - the Linux/Windows perception of H reality vs. the VMS perception of reality. If you want VMS to grow againE people have to be persuaded that the VMS perception of reality is the  'truth'.    D > We've (VMS) invested 3-1/2 years of time and money in Itanium, andG > for the forseable future, I don't see large scale AMD 64 systems with 	 > mission B > critical Windows and Linux.  The latest public move with Itanium< > removes any stigma of having HP (a system competetor) from > controlling IA64, D > or having an unfair inside advantage.  If we or Intel thought thatG > Itanium was going away, it's likely that any move would have been the  > other direction. > D > Just because there are dissapointed people out there mad about theC > demise of Alpha, does not mean that Itanium cannot succeed as one H > of the 2 premier high-end, mission critical, UNIX (and VMS) platforms.D > Will x86 be able to be grown "up" into this role?  Time will tell.  H Hopefully enough companies will buy into the Intel/HP bet on Itanic. AndB hopefully HP will see the light enough to advertise and market VMSJ effectively. At this stage of the game I'd like to devote my spare time toK recreational pursuits vs. learning how to change my earn-my-living paradigm  to Windows.     % > With enough thrust, a pig will fly. " F-4 Phantom II - the flying brick.   ------------------------------  # Date: Wed, 19 Jan 2005 14:08:02 GMT / From: "Jeff Goodwin" <jgoodwin@maine.rrr-r.com> $ Subject: Minimerge on HSG80 in V8.2?8 Message-ID: <6RtHd.151340$Uf.42792@twister.nyroc.rr.com>  B Whilst reading the V8.2 white paper, I noticed this in Chapter 10:   Shadowing Minimerge for Fibre  Channel (HSG80) - Improving the performance of shadow-set merge 2 operations, this shadowing feature is supported by/ OpenVMS V7.3-1 +TIMA kit, along with an ACS 8.7 6 W/R release from Storage (this is the firmware for the: HSG80 controller). This feature allows the merging of only4 the physical disk changes, rather than a full merge.2 The changes are kept in the HSG80 memory cache via4 write history logging (WHL) done via the ACS 8.7 W/R8 code. The W/R variants are identical to the F/S variants0 that customers use today for switch and snapshot9 capabilities, except they add the WHL capability. Similar 6 Minimerge functionality exists with the HSJ controller today for CI storage.   # It then went on to talk about HBMM.   = Wasn't the HSG80 Minimerge feature scrapped in favor of HBMM?      The white paper also said this:   - OpenVMS Alpha V8.2 and OpenVMS VAX V7.3-2 are 1 also a supported mixed-architecture cluster pair.   ' Where do I get this new VAX version? :)    -Jeff    ------------------------------    Date: 19 Jan 2005 10:46:19 -0500/ From: brooks@cuebid.zko.dec.nospam (Rob Brooks) ( Subject: Re: Minimerge on HSG80 in V8.2?- Message-ID: <AfKZRG0vjiFx@cuebid.zko.dec.com>   1 "Jeff Goodwin" <jgoodwin@maine.rrr-r.com> writes: D > Whilst reading the V8.2 white paper, I noticed this in Chapter 10: > / > Shadowing Minimerge for Fibre Channel (HSG80)   / > Improving the performance of shadow-set merge 4 > operations, this shadowing feature is supported by1 > OpenVMS V7.3-1 +TIMA kit, along with an ACS 8.7 8 > W/R release from Storage (this is the firmware for the > HSG80 controller). [...]   % > It then went on to talk about HBMM.  > ? > Wasn't the HSG80 Minimerge feature scrapped in favor of HBMM?  	 ) Yes; the white paper needs to be updated.    --    M Rob Brooks    VMS Engineering -- I/O Exec Group     brooks!cuebid.zko.dec.com    ------------------------------  % Date: Wed, 19 Jan 2005 13:55:37 +0100 ( From: "Rudolf Wingert" <win@fom.fgan.de>/ Subject: New Command Language (CL) for OpenVMS? 3 Message-ID: <000d01c4fe26$2c494c30$994614ac@wat153>    Hello,  H I did hear about a new CL for OpenVMS (HPCL). Is that true? If yes, when? will we see this one (8.2?)? What's happen with all our command ) procedure? Is there a compatibility mode?    TIA and best regards R. Wingert    ------------------------------  % Date: Wed, 19 Jan 2005 13:17:06 +0000 - From: Roy Omond <Roy.Omond@BlueBubble.UK.Com> 3 Subject: Re: New Command Language (CL) for OpenVMS? , Message-ID: <35752gF4gldg1U1@individual.net>   Rudolf Wingert wrote:  > Hello, > J > I did hear about a new CL for OpenVMS (HPCL). Is that true? If yes, whenA > will we see this one (8.2?)? What's happen with all our command + > procedure? Is there a compatibility mode?    Es war ein Scherz !    ------------------------------   Date: 19 Jan 2005 14:02:31 GMT( From: bill@cs.uofs.edu (Bill Gunshannon)3 Subject: Re: New Command Language (CL) for OpenVMS? , Message-ID: <3577jnF4h517hU3@individual.net>  3 In article <0IntsTrIDtHq@eisner.encompasserve.org>, > 	koehler@eisner.nospam.encompasserve.org (Bob Koehler) writes:` > In article <000d01c4fe26$2c494c30$994614ac@wat153>, "Rudolf Wingert" <win@fom.fgan.de> writes:	 >> Hello,  >>  K >> I did hear about a new CL for OpenVMS (HPCL). Is that true? If yes, when B >> will we see this one (8.2?)? What's happen with all our command, >> procedure? Is there a compatibility mode? > H >    Aaaaaarrrrrgh.  Please file this next to the difference between VMS >    and OpenVMS.    > Cut him some slack.  After all, it was a joke and at least one= english speaker here fell for it too, so having a non-english I speaker fooled by the subtlety shouldn't be that much of a surprise.  :-)    bill   --  J Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesD bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. University of Scranton   |A Scranton, Pennsylvania   |         #include <std.disclaimer.h>       ------------------------------    Date: 19 Jan 2005 07:55:54 -0600; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) 3 Subject: Re: New Command Language (CL) for OpenVMS? 3 Message-ID: <0IntsTrIDtHq@eisner.encompasserve.org>   ^ In article <000d01c4fe26$2c494c30$994614ac@wat153>, "Rudolf Wingert" <win@fom.fgan.de> writes: > Hello, > J > I did hear about a new CL for OpenVMS (HPCL). Is that true? If yes, whenA > will we see this one (8.2?)? What's happen with all our command + > procedure? Is there a compatibility mode?   F    Aaaaaarrrrrgh.  Please file this next to the difference between VMS    and OpenVMS.    ------------------------------    Date: 19 Jan 2005 10:31:02 -0600- From: Kilgallen@SpamCop.net (Larry Kilgallen) 3 Subject: Re: New Command Language (CL) for OpenVMS? 3 Message-ID: <zq73y755gTit@eisner.encompasserve.org>   ^ In article <000d01c4fe26$2c494c30$994614ac@wat153>, "Rudolf Wingert" <win@fom.fgan.de> writes: > Hello, > J > I did hear about a new CL for OpenVMS (HPCL). Is that true? If yes, whenA > will we see this one (8.2?)? What's happen with all our command + > procedure? Is there a compatibility mode?   D http://groups-beta.google.com/group/comp.os.vms/msg/5407514334cf160c   ------------------------------  # Date: Wed, 19 Jan 2005 16:42:28 GMT 6 From: "Kenneth Farmer" <kfarmer@NOSPAM.spyderbyte.com>3 Subject: Re: New Command Language (CL) for OpenVMS? = Message-ID: <U5wHd.4837$K72.1209357@twister.southeast.rr.com>   ; "Larry Kilgallen" <Kilgallen@SpamCop.net> wrote in message  - news:zq73y755gTit@eisner.encompasserve.org... G > In article <000d01c4fe26$2c494c30$994614ac@wat153>, "Rudolf Wingert"   > <win@fom.fgan.de> writes: 	 >> Hello,  >>K >> I did hear about a new CL for OpenVMS (HPCL). Is that true? If yes, when B >> will we see this one (8.2?)? What's happen with all our command, >> procedure? Is there a compatibility mode? > F > http://groups-beta.google.com/group/comp.os.vms/msg/5407514334cf160c    ( You just had to do that, didn't you.  :)     Ken    OpenVMS.org % _____________________________________  Kenneth R. Farmer <>< & SpyderByte: http://www.SpyderByte.com    ------------------------------  % Date: Wed, 19 Jan 2005 06:17:39 -0800 # From: "Tom Linden" <tom@kednos.com> $ Subject: Re: Next Gen Fabs & Itanium( Message-ID: <opskusnpiizgicya@hyrrokkin>  4 On Wed, 19 Jan 2005 04:10:35 GMT, Robert Deininger  % <rdeininger@mindspringdot.com> wrote:   7 > In article <41ED4783.F705C654@teksavvy.com>, JF Mezei ' > <jfmezei.spamnot@teksavvy.com> wrote:  >  >> Robert Deininger wrote:E >>> Alpha systems absolutely cost more to make than similar Integrity F >>> systems.  You'd lose your bet if the numbers were available to theA >>> public.  Since they're not, you're safe with your wild public  >>> speculations.  >>7 >> Why should an Alpha box cost more than an IA64 box ?  >> Is the the power supply ? >  > Almost always. >  >> Is it the case ?  > 
 > Usually. >  >> Is it the disk drives ? >  > Probably not.  >  >> Is it the connectors ?  >  > Yes. >  >  >> Is it the type of memory ?  >  > Often. >  > > >> Are alpha servers built with gold solder  instead of lead ? >  > No.  >  >>K >> If it is just the chip which is different, then I doubt that Alpha chips K >> actualy cost more than IA64, especially if HP ordered one large batch of $ >> them made for a final production. > H > Alpha CPUs do in fact cost a lot more than IA64.  I don't know why.  I> > don't think either benefits greatly from economies of scale. > G > But if you spend 5 minutes inside a recent alpha server and a similar D > Integrity server, you will see that it's NOT just the CPU which is > different. > ' >> (Higher volum the smaller quaktities K >> of IA64s being made in light of the fact that the chip is still evolving C >> and you don't want to stockpile years worth of Merced supplies).  >>K >> If IA64 and Alpha servers are of same quality, I do not buy the argument I >> that the Alpha has to be more expensive. Remember that many were  done G >> under Compaq which also had good "low cost manufacturing" expertise.  >  > K > Alpha systems might not HAVE TO cost more, but they do.  Low-cost designs I > were simply not a strength of the Digital server design teams, and were J > not a particular interest of the management food chain.  Could they haveH > done better?  Yes, but they rarely did in practice.  Alpha servers areF > (mostly) excellent designs, but low-cost was not high on the list of: > goals.  That might be one reason Digital didn't survive. > E > They merger with Compaq did not affect the way Alpha systems were   
 > designedH > at all.  The Digital teams pretty much continued as always, until theyI > were dispersed.  The supply chain overlap between Alpha systems and the H > Compaq PC business was pretty minor -- disk drives, CD-ROMs, sometimes$ > memory.  Not a heck of a lot else. > B > The Digital and Compaq manufacturing facilities weren't combinedG > particularly well either.  Compaq may have had low cost manufacturing K > expertise for PCs, but it never transfered to the server business.  There 0 > were some notable failures at trying to do so. > J > The HP teams that design and build IA64 systems pay particular attentionH > to cost.  There is increasing overlap in the supply chains for PCs andJ > servers.  There are fewer minor variations in parts, meaning fewer typesI > of spares to buy and stock.  HP is just better at low-cost servers than H > Digital was, and the prices for customers reflect this.  Folks might   > enjoy I > bashing HP, but they deserve credit for doing a pretty good job in this  > area.   K Alphas suffered from lack of standardization, which is a failing of central D management. With how many Alphas, for example, can you interchange  
 memory?  QED.      --  C Using Opera's revolutionary e-mail client: http://www.opera.com/m2/    ------------------------------  % Date: Wed, 19 Jan 2005 01:51:24 -0500 - From: JF Mezei <jfmezei.spamnot@teksavvy.com>  Subject: Re: vms versus solaris , Message-ID: <41EE035F.308ABC31@teksavvy.com>   re: RMS bundled into VMS.   F Perhaps the right approach would be to follow Quicktime. It used to beB included with all MACs. Then Apple realised that to make quicktimeE relevant, it had to be available for free to other platforms as well.   @ Consider that the vast number of applications that make use of aH database engine could run very easily on simple indexed files, but theytN go through the more complex DB engine because that is all they have available.  A If RMS were available for free on Linux, it would give RMS a much A greater profile and allow simple apps to start making use of RMS.    ------------------------------   Date: 19 Jan 2005 07:03:26 GMT2 From: "Dave Weatherall" <djw-nothere@nospam.nohow> Subject: Re: vms versus solaris ? Message-ID: <DTiotGxQ0bj6-pn2-YdbjuWuK9weF@dave2_os2.home.ours>   D On Wed, 19 Jan 2005 03:05:11 UTC, Dave Froble <davef@tsoft-inc.com>  wrote:   > Bill Gunshannon wrote: >  > J > > I am sure (and I certainly hope) that other readers hear realize thereL > > is no animosity between me and Dave or Bob or anyone else here.  I don'tJ > > know about them, but I enjoy the lively debates and I even learn stuffM > > from them. (I'm still waiting for someone to explain to me the difference J > > between Symbols and Logicals as I really thought they were pretty much5 > > the same except for the method of defining them.)  >  > 0 > I wish someone else would do this.  Oh well... > P > Symbols exist within a process.  They can be global to the process, or can be Q > local to a DCL procedure.  Symbols have a value, and can be evaluated, set and  % > re-set by DCL and library routines.  > R > Logicals are entries in various tables, process, group, system, cluster, and at R > various levels in some of the catagories.  A "logical name" can also be used to R > represent another logical name.  It can represent multiple values, as when used R > to represent multiple disk/directory locations.  Logical names are evaluated by  > the filespec parser. > N > And now you probably know less than before.  It's a bit tough for me, I use Q > logical names, but have never gave much thought to a formal definition of what  M > they are.  I'm sure someone else could give a better 'design view' of what   > logical names are.  F Picking up on Bob's scope comments, one further advantage of a logicalF over a symbol is that it allows information to be stored in a logical D name table that can be seen by chosen users or groups of user's. It A does require some manipulation of the LNM table scanning control  D (itself a logical name), but it does provide a method of setting up B project-wide definitions without taking up excessive process/user F memory, as happens when the definitions are made in the process table 
 during login.   A Similarly, access to logical name tables can be controlled by ACL     --   Cheers - Dave W.   ------------------------------   Date: 19 Jan 2005 07:03:25 GMT2 From: "Dave Weatherall" <djw-nothere@nospam.nohow> Subject: Re: vms versus solaris ? Message-ID: <DTiotGxQ0bj6-pn2-Nk38g9oIUdJI@dave2_os2.home.ours>   E On Wed, 19 Jan 2005 06:17:52 UTC, Bill Todd <billtodd@metrocast.net>   wrote:  
 <Big Snip>  H > Given its splendid isolation from the rest of the industry, just like J > many other useful aspects of VMS RMS simply doesn't constitute anything I > like a decisive plus, save perhaps (if you consider this an advantage)  H > to the degree that it locks in VMS customers who can't easily migrate G > applications which depend on it elsewhere.  Just one more example of  C > where Digital Had It Then but failed to turn it into the lasting  @ > leverage of a true (sanctioned or de facto) industry standard.   BillC        am I parsing this right when I take 'decisive plus'  in the  ? context of 'bringing new customers to the VMS platform'.  As a  C record-oriented user/programmer, it _is_ a decisive plus but then,  ; perhaps, I fall into the 'locked in' category (mentally or   philosophically).    --   Cheers - Dave W.   ------------------------------   Date: 19 Jan 2005 07:03:23 GMT2 From: "Dave Weatherall" <djw-nothere@nospam.nohow> Subject: Re: vms versus solaris ? Message-ID: <DTiotGxQ0bj6-pn2-deEeZEqSBvKc@dave2_os2.home.ours>   + On Wed, 19 Jan 2005 04:02:56 UTC, JF Mezei  % <jfmezei.spamnot@teksavvy.com> wrote:    > Dave Froble wrote:P > > The default on VMS it to grant TMPMBX and NETMBX to a user.  I'm a bit fuzzyO > > about DECnet at this time, so I'm not sure, but I think this gives non-priv 9 > > users capability to send and receive DECnet messages.  > H > You need privs (I believe SYSNAM) to create a decnet object anyone canJ > connect to without any authentication. (or have read/write access to the+ > NCP database to define the object there).   > Correct,. I was going to pick up on the discussion about what B priveleges one needs to set up a TCP/IP socket service but then I F remembered that my only actual experience is with my application that @ receives DecNet comms from a VMS client but uses TCP/IP to pass A simulation data to/from a host simulation running under Solaris.  E Because I need SYSNAM for the DecNet service I don't know, for sure,  C whether the TCP/IP runs without it. My colleague who developed the  ? TCP/IP interface also has SYSNAM, for the DecNet circuit, so I  0 couldn't extrapolate from his experience either.   --   Cheers - Dave W.   ------------------------------  % Date: Wed, 19 Jan 2005 10:09:57 +0000 - From: John Laird <nospam@laird-towers.org.uk>  Subject: Re: vms versus solaris 8 Message-ID: <04csu0tvu0gelnuddt47v1msnter98jqk8@4ax.com>  K On Tue, 18 Jan 2005 23:16:19 -0500, JF Mezei <jfmezei.spamnot@teksavvy.com>  wrote:   >Dave Froble wrote: P >> Symbols exist within a process.  They can be global to the process, or can beQ >> local to a DCL procedure.  Symbols have a value, and can be evaluated, set and & >> re-set by DCL and library routines. >  > D >Would it be more precise to state that symbols exist at the CLI/DCLI >level, whereas logicals exist in various tables, including a process and 
 >job tables ?  > A >Is it correct to state that RUN/DETACHED image.exe will not give : >image.exe access to any symbols for reading and writing ?  F Yes and yes.  I think it used to be the case that images couldn't evenK access symbols wihout dipping into process structures - I'm sure I remember J LIB$GET[SET]_SYMBOL arriving.  By and large, symbols are for DCL.  LogicalL names are for whatever you wish to use them for - storing state information,I controlling images, reporting results, and of course they have always had H special meaning in file-parsing, their original purpose (fair to say ?).   --  : Try to look unimportant; the bad guys may be low on ammo.    Mail john rather than nospam...    ------------------------------  % Date: Wed, 19 Jan 2005 05:52:57 -0500 ( From: Bill Todd <billtodd@metrocast.net> Subject: Re: vms versus solaris = Message-ID: <0LudneELkKGboXPcRVn-iw@metrocastcablevision.com>    Dave Weatherall wrote:   ...       As a E > record-oriented user/programmer, it _is_ a decisive plus but then,  = > perhaps, I fall into the 'locked in' category (mentally or   > philosophically).   H The point is that as a record-oriented user/programmer, you'd also have G access to the kinds of packages I mentioned on Unix (or - blech - even  ? to some extent on Windows) - in fact, to a considerably larger  I collection of options in this area than is available to you on VMS, many  G of them free or very low cost but still very functional and stable and  G quite a few of them also somewhat more technically current than RMS is  G (e.g., in their integral use of journaling to improve both performance   and reliability).   D The fact that RMS is bundled and supported as part of the OS may be I convenient and reassuring, but is hardly of *major* significance to most  F users (or applications, though not having a third-party dependency in & this area is part of the convenience).   - bill   ------------------------------   Date: 19 Jan 2005 13:23:22 GMT( From: bill@cs.uofs.edu (Bill Gunshannon) Subject: Re: vms versus solaris , Message-ID: <3575aaF4g3o73U1@individual.net>  * In article <41EDCE67.50107@tsoft-inc.com>,* 	Dave Froble <davef@tsoft-inc.com> writes: > Bill Gunshannon wrote: >  > I >> I am sure (and I certainly hope) that other readers hear realize there K >> is no animosity between me and Dave or Bob or anyone else here.  I don't I >> know about them, but I enjoy the lively debates and I even learn stuff L >> from them. (I'm still waiting for someone to explain to me the differenceI >> between Symbols and Logicals as I really thought they were pretty much 4 >> the same except for the method of defining them.) >  > 0 > I wish someone else would do this.  Oh well... > P > Symbols exist within a process.  They can be global to the process, or can be Q > local to a DCL procedure.  Symbols have a value, and can be evaluated, set and  % > re-set by DCL and library routines.  > R > Logicals are entries in various tables, process, group, system, cluster, and at R > various levels in some of the catagories.  A "logical name" can also be used to R > represent another logical name.  It can represent multiple values, as when used R > to represent multiple disk/directory locations.  Logical names are evaluated by  > the filespec parser. > N > And now you probably know less than before.  It's a bit tough for me, I use Q > logical names, but have never gave much thought to a formal definition of what  M > they are.  I'm sure someone else could give a better 'design view' of what   > logical names are. >   B No, actually that was a good explanation.  I can now say that UnixC has the equivalent in it's system variables but that VMS appears to B have a bit more flexibility.  All unix scoping is downward.  I can@ not change the value of a variable defined in a process above meB except for how it will appear to processes below me.  Any variableC I create is only valid for processes below me and goes away when my B process ends.  Assuming a Logical can be changed system wide (thatB is also assuming adequate priveledge) that would be a good featureA that unix lacks.  Interestingly enough, I don't think it would be A that hard to implement.  Hmmmm.....  Maybe it is but I just never B needed to do it so I never learned how.  Worth more investigation.  C Thanks for the explanation.  I told you I actually learn stuff from : these debates.  And they tend to kick the brain into gear.   bill   --  J Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesD bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. University of Scranton   |A Scranton, Pennsylvania   |         #include <std.disclaimer.h>       ------------------------------   Date: 19 Jan 2005 13:38:31 GMT( From: bill@cs.uofs.edu (Bill Gunshannon) Subject: Re: vms versus solaris , Message-ID: <35766mF4h517hU1@individual.net>  D In article <nospam.News.Bob-F38BA1.22511618012005@news.verizon.net>,< 	Bob Harris <nospam.News.Bob@remove.Smith-Harris.us> writes:, > In article <41EDCE67.50107@tsoft-inc.com>,+ >  Dave Froble <davef@tsoft-inc.com> wrote:  >  >> Bill Gunshannon wrote:  >>   >>  K >> > I am sure (and I certainly hope) that other readers hear realize there M >> > is no animosity between me and Dave or Bob or anyone else here.  I don't K >> > know about them, but I enjoy the lively debates and I even learn stuff N >> > from them. (I'm still waiting for someone to explain to me the differenceK >> > between Symbols and Logicals as I really thought they were pretty much 6 >> > the same except for the method of defining them.) >>   >>  1 >> I wish someone else would do this.  Oh well...  >>  Q >> Symbols exist within a process.  They can be global to the process, or can be  N >> local to a DCL procedure.  Symbols have a value, and can be evaluated, set  >> and u& >> re-set by DCL and library routines. >> pP >> Logicals are entries in various tables, process, group, system, cluster, and  >> at P >> various levels in some of the catagories.  A "logical name" can also be used  >> to N >> represent another logical name.  It can represent multiple values, as when  >> used P >> to represent multiple disk/directory locations.  Logical names are evaluated  >> by  >> the filespec parser.  >> aO >> And now you probably know less than before.  It's a bit tough for me, I use  M >> logical names, but have never gave much thought to a formal definition of t >> what N >> they are.  I'm sure someone else could give a better 'design view' of what  >> logical names are.r >> e >> Daves > E > More on symbols.  Local symbols have what I like to call "Umbrella sD > Scope"  that is to say, a sub procedure can look up and see local K > variables created in a parent procedure, unless the subprocedure creates r$ > a local symbol with the same name. > K > UNIX script variables do not have this characteristic.  And while script rH > variables can local or global (depending on which shell is being run) K > they are not the same kind of global as a VMS Global Symbol.  That is to  F > say they are only visible to scripts running in the current process ? > which and not to any subprocess or program that might be run.-  ; Not true.  Depending on shell variables are either exported1> automatically or can be using the "export" command.  For every> modern shell I am aware of, it is automatic.  If this were not? true, there would be no way for the system to pass to all users  things like PATH, TERMCAP, etc.w   > D > A shell symbol that needs to be visible to a child process and/or F > program invoked from a script can be made an environment variable.  I > Environment variables are passed down (inherited) by child processes.     B I should have read on.  But the above still stands.  There really  is only one type of variable.0  H > BUT, unlike a Global Symbol, changes to the environment variable by a K > child process are _NOT_ visible to the parent.  Changes are only visible IG > to the child process that changes the environment variable or to any  J > child processes that it creates.  This is because environment variables J > are copied from the parent onto the stack of the child process, and the H > parent does not have access to that address space, and when the child K > terminates, the child's stack is deleted.  So any return information has hF > to use a different means (and we have all kinds of tricks, the most G > common of which is to have the parent read what we write to standard e > out).e  D This is true, but the trick you mention is only usable if the parentB is expecting the child to change something.  This still limits the& scope to something much less than VMS.   > G > And another point about logicals is that there does not need to be a -H > parent child relationship for some classes of logicals to be seen (or K > there can if you use the right class).  And logicals can be used to pass 5D > information from the child back to the parent, if that is desired.   That's a plus.     > J > And Global Symbols can also be used to return information to the parent E > when running a program that does not spawn a subprocess, because a  I > program is run in the same process space as DCL, just a different mode.l   Don't understand this one.   > K > Passing information from child to parent or to non-related processes can rJ > not be down via variables or environment variables.  Child to parent is H > frequently done by the parent establishing a pipe between the child's H > standard out and reading that to capture any information to be passed J > back.   Non-parent/child information passing could use the file system, J > as in either the name of a file, the contents of a symbolic link (which J > does not need to be a file spec, but rather can be almost anything), or K > the content of a file, or a named pipe, a message queue for interprocess  $ > message passing, or shared memory. > F > I would say the closest thing to a logical name would be a symbolic 	 > link.     @ A symbolic link is merely a different logical name for a file or? directory.  Other than that, it passes no information and being ? as it can point at a non-existant filename sometimes it doesn't  pass any information.o  H >        I'm not saying it does everything a logical does, just that it J > can be used to pass information around, as well as be used to hold file I > spec information.  And a symbolic link is accessible anywhere the file mI > system can be seen, such as Tru64 TruCluster or even NFS server/client cI > mounted disk. Although not too many applications actually use symbolic   > links to pass information.    F I just tried it for grins.  I have never seen a symbolic link used forF this, but I can see how it wold work.  I also don't think the original* designers eevr had this idea in mind.  :-)  I >                            File and shared memory are much more common n > for unrelated processes.   True.E   > . > And just to put my credentials on the table: > D > o  12 years OpenVMS as System Admin, Applications programmer, and H > PATHWORKS for Mac file server developer (with a smidge of kernel code  > for intrusion detection).  > G > o  12 years UNIX as System Admin, Applications programmer, and Tru64  = > UNIX AdvFS file system developer working inside the kernel.D > J > I have _ALWAYS_ had SYSTEM or ROOT access (and the scars to prove it :-) > I > And for the most part I tend to agree with most of the stuff that Bill EK > Gunshannon has been try to say.  OpenVMS has its strengths, and UNIX has mJ > it strengths.  If you know one and not the other, you are going to tend  > to favor the one you know. > J > And since this is comp.os.vms, I say, buy more OpenVMS systems and keep   > the ZKO3-4 fully populated :-) > 4 >                                         Bob Harris   bill   --  J Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesD bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. University of Scranton   |A Scranton, Pennsylvania   |         #include <std.disclaimer.h>   i   ------------------------------    Date: 19 Jan 2005 07:44:03 -0600; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler)  Subject: Re: vms versus solaris 3 Message-ID: <BFZDUFkHfeiw@eisner.encompasserve.org>t  f In article <ksgHd.41307$W33.855142@news20.bellglobal.com>, "Neil Rieck" <n.rieck@sympatico.ca> writes:  5 > It would be nice if someone in this newsgroup with  D > experience on both UNIX and OpenVMS would set the record straight.  $    The records, they are a changin'.   ------------------------------   Date: 19 Jan 2005 13:46:32 GMT( From: bill@cs.uofs.edu (Bill Gunshannon) Subject: Re: vms versus solarise, Message-ID: <3576loF4h517hU2@individual.net>  3 In article <0rF4L0yqRgL6@eisner.encompasserve.org>, 0 	Kilgallen@SpamCop.net (Larry Kilgallen) writes:P > In article <opsktoxpk4zgicya@hyrrokkin>, "Tom Linden" <tom@kednos.com> writes: > M >>> As some people have already mentioned to me by email, modern flavors of  S >>> UNIXM >>> do implement quota-based resource management. (most of my UNIX experiencenM >>> came from BSD 4.4 on PDP-11/44 back in the mid 1980's and lots of stuff    >>> has1K >>> changed since then). It would be nice if someone in this newsgroup withrF >>> experience on both UNIX and OpenVMS would set the record straight. >> e >>   >>  H >> ulimit(1)                                                             >> ulimit(1) >> n >> NAME  >> h/ >>    ulimit - Sets or reports a resource limiti > G > I did not see anything in that description specifying the name of theE > user affected.  ? A user can only change his own limimits so there is no need forl the ability to specify a user.   > D > Does this mean that all users on the system have the same quotas ?  E No, the initial values are set at login based on a class.  The valuesC are defined in /etc/login.conf.    bill   -- tJ Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesD bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. University of Scranton   |A Scranton, Pennsylvania   |         #include <std.disclaimer.h>       ------------------------------    Date: 19 Jan 2005 07:41:48 -0600; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler)m Subject: Re: vms versus solarise3 Message-ID: <UyHBNaiFQkmg@eisner.encompasserve.org>b  W In article <355cdiF4ik1fqU1@individual.net>, bill@cs.uofs.edu (Bill Gunshannon) writes:uI >>                                                  I've actually had one J >>    (1), count them : ONE programmer who set up cd, pwd, ... for his VMS >>    account. >> sI >>    All true VMS users know that sd is the symbol for some command file - >>    that does nice things with set default.i > D > What happened to the idea of making the machine work for the users4 > rather than making the users work for the machine?  F    I don't believe foisting my ideas on users is a way to make it work    for anyone but me.a  H >>    My admin friend down the hall recognizes "admin account".  Some OS( >>    are even more prevalent than UNIX. >  > Name one!L  9    Guess quickly:  he does a lot of Microsoft type stuff.g > H > I am sure (and I certainly hope) that other readers hear realize thereJ > is no animosity between me and Dave or Bob or anyone else here.  I don'tH > know about them, but I enjoy the lively debates and I even learn stuffK > from them. (I'm still waiting for someone to explain to me the difference H > between Symbols and Logicals as I really thought they were pretty much3 > the same except for the method of defining them.)S  G    I have a fellow who refers to them as "logical symbols".  I tell newl&    users these are the most important:  H    1 they are both memory resident strings which stand for other stringsH    2 symbols are per-process, but usually inherited by spawned processesF    3 symbols are global to all procs in a process or local to one procD    4 logicals are per-process, job-wide, group-wide, system-wide, or       cluster-wide%    5 logicals are global to all procs =    6 DCL will use symbols like UNIX aliases, but not logicals0E    7 the file system will automatically use logicals, but not symbolsj    ME    I don't bother new users with inner vs. outer mode, concealed, andn0    all those other attributes logicals can have.   ------------------------------    Date: 19 Jan 2005 07:45:49 -0600; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler)  Subject: Re: vms versus solarisu3 Message-ID: <84S2hUkSEi9G@eisner.encompasserve.org>i  q In article <zDGF$Snaq4hE@eisner.encompasserve.org>, koehler@eisner.nospam.encompasserve.org (Bob Koehler) writes:rY > In article <3555ujF4j0dtaU3@individual.net>, bill@cs.uofs.edu (Bill Gunshannon) writes:  >> SI >> No, it's not.  ISAM and DBMS are both available on unix. It's just not J >> the default.  And before you you bring up the additional cost of buyingH >> an additional commercial package.  What is the difference between VMSI >> and pretty much any unix, pricewise?  Basicly, VMS provides it, wetheroJ >> you need it or not and makes you pay for it, wether you need it or not. > B >    In the real world, that's like selling soup without the bowl.  ,    Oops, I meant milkshakes without the cup.   ------------------------------  % Date: Wed, 19 Jan 2005 06:25:24 -0800h# From: "Tom Linden" <tom@kednos.com>  Subject: Re: vms versus solaris-( Message-ID: <opskus0mlzzgicya@hyrrokkin>  . On Tue, 18 Jan 2005 23:16:19 -0500, JF Mezei  % <jfmezei.spamnot@teksavvy.com> wrote:n   > Dave Froble wrote:K >> Symbols exist within a process.  They can be global to the process, or  r	 >> can behK >> local to a DCL procedure.  Symbols have a value, and can be evaluated,  w
 >> set and& >> re-set by DCL and library routines. >  > E > Would it be more precise to state that symbols exist at the CLI/DCLiJ > level, whereas logicals exist in various tables, including a process and > job tables ? >(B > Is it correct to state that RUN/DETACHED image.exe will not give; > image.exe access to any symbols for reading and writing ?m   What about lib$[s,g]et_symbol?   --  C Using Opera's revolutionary e-mail client: http://www.opera.com/m2/    ------------------------------  % Date: Wed, 19 Jan 2005 06:28:11 -0800-# From: "Tom Linden" <tom@kednos.com>, Subject: Re: vms versus solarism( Message-ID: <opskus49vjzgicya@hyrrokkin>  H On 18 Jan 2005 22:46:49 -0600, Larry Kilgallen <Kilgallen@SpamCop.net>   wrote:  J > In article <opsktoxpk4zgicya@hyrrokkin>, "Tom Linden" <tom@kednos.com>  	 > writes:f >sK >>> As some people have already mentioned to me by email, modern flavors ofc >>> UNIXD >>> do implement quota-based resource management. (most of my UNIX   >>> experienceK >>> came from BSD 4.4 on PDP-11/44 back in the mid 1980's and lots of stuffO >>> hasBK >>> changed since then). It would be nice if someone in this newsgroup with-F >>> experience on both UNIX and OpenVMS would set the record straight. >> >> >> >> ulimit(1) >> ulimit(1) >> >> NAME  >>/ >>    ulimit - Sets or reports a resource limit. >?G > I did not see anything in that description specifying the name of thet > user affected. >nD > Does this mean that all users on the system have the same quotas ?I It has been a while, but I believe it is per user, as determine at login.d5 Of course that was Tru64, don't know what Linux does.h       -- aC Using Opera's revolutionary e-mail client: http://www.opera.com/m2/l   ------------------------------  % Date: Wed, 19 Jan 2005 06:31:42 -0800d# From: "Tom Linden" <tom@kednos.com>R Subject: Re: vms versus solarisw( Message-ID: <opskuta4cgzgicya@hyrrokkin>  H On Wed, 19 Jan 2005 01:17:52 -0500, Bill Todd <billtodd@metrocast.net>   wrote:   > Larry Kilgallen wrote:H >> In article <355hpdF4g6skfU2@individual.net>, bill@cs.uofs.edu (Bill   >> Gunshannon) writes: >t > ...t > H >>> If you offer a commercial package that depends on another commercialJ >>> package you don't expect to find on the customers machine, you license1 >>> that package and provide it along with yours. E >>   That works if the product is the reason people are running the  i >> machine, F >> but not for utility programs.  Consider all the VMS Freeware that  
 >> depends
 >> on RMS. > L > And if RMS weren't there, either those utilities wouldn't depend upon it  6 > or they would bundle in whatever they did depend on. >-L > Look, guys:  I'm about as record-biased as one can get.  While I believe  K > that RMS could (and should) be made much more approachable and possibly  CH > more streamlined internally to avoid imposing overheads on those who  D > don't need them (the implicit system cost of bundling it in is a  J > separate, though similar, issue, but I firmly believe that if RMS-type  L > facilities were *standard* in the industry the industry would appreciate  H > them - and I tried to get DEC to take the lead in standardizing them  G > across Unix and Windows almost 20 years ago), when it comes to data  S: > management I think byte streams represent the stone age.  J The ISAM packaged developed to support PL/I under Ultrix was spun off as aG separately licensable package, but I don't think it ever sold very much5 outside of PL/I usage.   >gG > But the incontrovertible fact is that most of the world doesn't see  "B > things that way.  They're quite happy with Unix- and (shudder)  J > Windows-level byte-stream file systems, plus add-on packages when they  F > need them such as dbm (free, I think), Postgres and IBPhoenix (nee  J > Interbase), both free and the latter with a relatively small footprint  K > and excellent reputation, Berkeley db and C-ISAM-like products (free or  2F > at least low-cost), and Oracle (not free at all, but still popular). >:I > Given its splendid isolation from the rest of the industry, just like  5K > many other useful aspects of VMS RMS simply doesn't constitute anything  tJ > like a decisive plus, save perhaps (if you consider this an advantage)  I > to the degree that it locks in VMS customers who can't easily migrate  aH > applications which depend on it elsewhere.  Just one more example of  D > where Digital Had It Then but failed to turn it into the lasting  @ > leverage of a true (sanctioned or de facto) industry standard. >i > - bill       -- /C Using Opera's revolutionary e-mail client: http://www.opera.com/m2/o   ------------------------------  % Date: Wed, 19 Jan 2005 06:34:06 -0800_# From: "Tom Linden" <tom@kednos.com>  Subject: Re: vms versus solaris/( Message-ID: <opskute4r2zgicya@hyrrokkin>  H On Wed, 19 Jan 2005 05:52:57 -0500, Bill Todd <billtodd@metrocast.net>   wrote:   > Dave Weatherall wrote: >  > ...I >i	 >    As a-G >> record-oriented user/programmer, it _is_ a decisive plus but then,  e? >> perhaps, I fall into the 'locked in' category (mentally or  t >> philosophically). >eK > The point is that as a record-oriented user/programmer, you'd also have  rJ > access to the kinds of packages I mentioned on Unix (or - blech - even  B > to some extent on Windows) - in fact, to a considerably larger  L > collection of options in this area than is available to you on VMS, many  J > of them free or very low cost but still very functional and stable and  J > quite a few of them also somewhat more technically current than RMS is  J > (e.g., in their integral use of journaling to improve both performance   > and reliability).   D An interesting aside, IBM has added record I/O support to C for Z-OS > G > The fact that RMS is bundled and supported as part of the OS may be  oL > convenient and reassuring, but is hardly of *major* significance to most  I > users (or applications, though not having a third-party dependency in   ( > this area is part of the convenience). >f > - bill       -- .C Using Opera's revolutionary e-mail client: http://www.opera.com/m2/w   ------------------------------  + Date: Wed, 19 Jan 2005 15:20:21 +0000 (UTC)u( From: m.kraemer@gsi.de (Michael Kraemer) Subject: Re: vms versus solarise5 Message-ID: <csltrk$u73$1@lnx107.hrz.tu-darmstadt.de>h  N In article <opskuta4cgzgicya@hyrrokkin>, "Tom Linden" <tom@kednos.com> writes: > L > The ISAM packaged developed to support PL/I under Ultrix was spun off as a;                                          ^^^^^^^^^^^^^^^^^^ ! Such a beast is still available ?q   ------------------------------  % Date: Wed, 19 Jan 2005 10:56:59 -0500i# From: "John Smith" <a@nonymous.com>> Subject: Re: vms versus solaris , Message-ID: <mcidndJdK9DRHnPcRVn-2w@igs.net>   Bill Todd wrote: > Larry Kilgallen wrote:F >> In article <355hpdF4g6skfU2@individual.net>, bill@cs.uofs.edu (Bill >> Gunshannon) writes: >d > ...  >oH >>> If you offer a commercial package that depends on another commercialB >>> package you don't expect to find on the customers machine, you9 >>> license that package and provide it along with yours.  >> >>A >> That works if the product is the reason people are running the  >> machine,rD >> but not for utility programs.  Consider all the VMS Freeware that
 >> depends
 >> on RMS. >tG > And if RMS weren't there, either those utilities wouldn't depend upons9 > it or they would bundle in whatever they did depend on.q > B > Look, guys:  I'm about as record-biased as one can get.  While ID > believe that RMS could (and should) be made much more approachableF > and possibly more streamlined internally to avoid imposing overheadsG > on those who don't need them (the implicit system cost of bundling itnG > in is a separate, though similar, issue, but I firmly believe that ifrB > RMS-type facilities were *standard* in the industry the industryD > would appreciate them - and I tried to get DEC to take the lead inG > standardizing them across Unix and Windows almost 20 years ago), whenoF > it comes to data management I think byte streams represent the stone > age. >aE > But the incontrovertible fact is that most of the world doesn't seee@ > things that way.  They're quite happy with Unix- and (shudder)H > Windows-level byte-stream file systems, plus add-on packages when theyD > need them such as dbm (free, I think), Postgres and IBPhoenix (neeH > Interbase), both free and the latter with a relatively small footprintF > and excellent reputation, Berkeley db and C-ISAM-like products (free? > or at least low-cost), and Oracle (not free at all, but stilla > popular).2 >0G > Given its splendid isolation from the rest of the industry, just liker@ > many other useful aspects of VMS RMS simply doesn't constituteF > anything like a decisive plus, save perhaps (if you consider this anC > advantage) to the degree that it locks in VMS customers who can't E > easily migrate applications which depend on it elsewhere.  Just onehF > more example of where Digital Had It Then but failed to turn it intoB > the lasting leverage of a true (sanctioned or de facto) industry > standard.,    ' Interesting that you mention Interbase.g  J Firebird, the renamed build of Interbase 6, still has the VMS conditionals! in it - including use of the DLM.   C Jim Starkey (Datatrieve, Rdb, and Interbase) is still involved with G Firebird. In speaking with Ann Harrison (IBPhoenix) today, her sense ofDH resurrecting a VMS version of Interbase/Firebird is that it would not be very difficult.D  B We last used Interbase/VMS at v4 (as an embedded db for one of ourI applications - a stock exchange 'ticker plant' for lack of a better term)hK and found it to be fast, extremely stable, easy to administer, and consumed J very little system resource. We went to Interbase primarily as a result ofB the Rdb runtime being removed from the base VMS license years ago.  G I don't know how Firebird compares to current versions of MySQL , which @ would be what I'd otherwise consider for an 'open source' rdbms.   ------------------------------  % Date: Wed, 19 Jan 2005 07:55:10 -0800h# From: "Tom Linden" <tom@kednos.com>i Subject: Re: vms versus solarish( Message-ID: <opskuw58q7zgicya@hyrrokkin>  ; On Wed, 19 Jan 2005 15:24:46 +0000 (UTC), Michael Kraemer  o <m.kraemer@gsi.de> wrote:h  J > In article <opskute4r2zgicya@hyrrokkin>, "Tom Linden" <tom@kednos.com>  	 > writes:  >>G >> An interesting aside, IBM has added record I/O support to C for Z-OSd > : > that was available even way back in MVS of 1993 vintage, > from my sw attic:t > B > fopen( cFile, "rb,type=record,recfm=F,lrecl=4096,blksize=4096" ) >  Shows you what I know.     -- lC Using Opera's revolutionary e-mail client: http://www.opera.com/m2/i   ------------------------------    Date: 19 Jan 2005 10:19:42 -0600- From: Kilgallen@SpamCop.net (Larry Kilgallen)b Subject: Re: vms versus solariso3 Message-ID: <kUttlL2eZNv$@eisner.encompasserve.org>   q In article <UyHBNaiFQkmg@eisner.encompasserve.org>, koehler@eisner.nospam.encompasserve.org (Bob Koehler) writes:   F >    4 logicals are per-process, job-wide, group-wide, system-wide, or >       cluster-wide  @ 	or in a custom table available to particular users (independent 	of UIC groups).   ------------------------------    Date: 19 Jan 2005 10:17:06 -0600- From: Kilgallen@SpamCop.net (Larry Kilgallen)  Subject: Re: vms versus solarisn3 Message-ID: <IJvm$arp78vj@eisner.encompasserve.org>o  t In article <DTiotGxQ0bj6-pn2-Nk38g9oIUdJI@dave2_os2.home.ours>, "Dave Weatherall" <djw-nothere@nospam.nohow> writes:  E >        am I parsing this right when I take 'decisive plus'  in the -A > context of 'bringing new customers to the VMS platform'.  As a  E > record-oriented user/programmer, it _is_ a decisive plus but then,  = > perhaps, I fall into the 'locked in' category (mentally or s > philosophically).2  E The languages Cobol, Pascal and Ada all have the notion of record-IO,iA and probably PL/I does also.  So I would say you are locked in by @ programming paradigms that have been found to actually be needed in real applications.    ------------------------------    Date: 19 Jan 2005 10:25:02 -0600- From: Kilgallen@SpamCop.net (Larry Kilgallen)e Subject: Re: vms versus solarisp3 Message-ID: <6LOwnuAnaCF1@eisner.encompasserve.org>r  W In article <35766mF4h517hU1@individual.net>, bill@cs.uofs.edu (Bill Gunshannon) writes:eF > In article <nospam.News.Bob-F38BA1.22511618012005@news.verizon.net>,> > 	Bob Harris <nospam.News.Bob@remove.Smith-Harris.us> writes:  K >> And Global Symbols can also be used to return information to the parent nF >> when running a program that does not spawn a subprocess, because a J >> program is run in the same process space as DCL, just a different mode. >  > Don't understand this one.  G Within a process, one command procedure or program can change the value 8 of a global DCL symbol to be read by an invoking entity.  B For logical names those need not be in the same process - data canD be passed back and forth between processes re-using the same logical names (of appropriate scope).l   ------------------------------    Date: 19 Jan 2005 10:26:31 -0600- From: Kilgallen@SpamCop.net (Larry Kilgallen)t Subject: Re: vms versus solaris 3 Message-ID: <XKswSWxjDAFb@eisner.encompasserve.org>d  W In article <3576loF4h517hU2@individual.net>, bill@cs.uofs.edu (Bill Gunshannon) writes:c5 > In article <0rF4L0yqRgL6@eisner.encompasserve.org>,(2 > 	Kilgallen@SpamCop.net (Larry Kilgallen) writes:Q >> In article <opsktoxpk4zgicya@hyrrokkin>, "Tom Linden" <tom@kednos.com> writes:o  I >>> ulimit(1)                                                             
 >>> ulimit(1)  >>>  >>> NAME >>> 0 >>>    ulimit - Sets or reports a resource limit >>  H >> I did not see anything in that description specifying the name of the >> user affected.  > A > A user can only change his own limimits so there is no need fork  > the ability to specify a user. >  >> iE >> Does this mean that all users on the system have the same quotas ?Y > G > No, the initial values are set at login based on a class.  The valuese! > are defined in /etc/login.conf.   C Does that mean a user can raise values with the ulimit command even0> though the system manager has set them to a particular limit ?   ------------------------------    Date: 19 Jan 2005 10:27:40 -0600- From: Kilgallen@SpamCop.net (Larry Kilgallen)  Subject: Re: vms versus solarisy3 Message-ID: <GzD+OAhCA5qq@eisner.encompasserve.org>i  N In article <opskus0mlzzgicya@hyrrokkin>, "Tom Linden" <tom@kednos.com> writes:0 > On Tue, 18 Jan 2005 23:16:19 -0500, JF Mezei  ' > <jfmezei.spamnot@teksavvy.com> wrote:l >  >> Dave Froble wrote:WL >>> Symbols exist within a process.  They can be global to the process, or  
 >>> can beL >>> local to a DCL procedure.  Symbols have a value, and can be evaluated,   >>> set and ' >>> re-set by DCL and library routines.t >> >>F >> Would it be more precise to state that symbols exist at the CLI/DCLK >> level, whereas logicals exist in various tables, including a process andw >> job tables ?o >>C >> Is it correct to state that RUN/DETACHED image.exe will not give < >> image.exe access to any symbols for reading and writing ? >   > What about lib$[s,g]et_symbol?  G Those depend on a CLI, so unless image.exe was LOGINOUT.EXE those callse do not work.   ------------------------------  % Date: Wed, 19 Jan 2005 08:39:26 -0800s# From: "Tom Linden" <tom@kednos.com>n Subject: Re: vms versus solariso( Message-ID: <opskuy70bkzgicya@hyrrokkin>  5 On Wed, 19 Jan 2005 17:26:13 +0100, Keith Cayemberg  0! <keith.cayemberg@arcor.de> wrote:l   > Bill Todd wrote: >> Dave Weatherall wrote:a >>  ....
 >>    As a >>H >>> record-oriented user/programmer, it _is_ a decisive plus but then,  @ >>> perhaps, I fall into the 'locked in' category (mentally or   >>> philosophically). I >>   The point is that as a record-oriented user/programmer, you'd also  aK >> have access to the kinds of packages I mentioned on Unix (or - blech -  hH >> even to some extent on Windows) - in fact, to a considerably larger  H >> collection of options in this area than is available to you on VMS,  L >> many of them free or very low cost but still very functional and stable  L >> and quite a few of them also somewhat more technically current than RMS  B >> is (e.g., in their integral use of journaling to improve both    >> performance and reliability).I >>  The fact that RMS is bundled and supported as part of the OS may be  MH >> convenient and reassuring, but is hardly of *major* significance to  L >> most users (or applications, though not having a third-party dependency  , >> in this area is part of the convenience).
 >>  - bill >sJ > It appears to me that everyone here fails to understand the philosophy  J > behind having an universal record handling service integrated into the   > operating system.o >iJ > It's primary purpose within the context of professional programming is  E > not to make record handling "easier" or more "convenient" for the  G
 > programmer.: > L > It's primary purpose is to improve the quality and correctness of record   > storage and transfer.t >nI > This principle is similar to that which behind the use of Descriptors   K > and a defined Call Standard in OpenVMS, which are also excellent unique  w2 > quality features of the OpenVMS OS Architecture. >bJ > By making this service "default" or for some purposes even "mandatory"  H > for the operating system, programmers are encouraged to forego their  L > artistic-license to make many common record-handling errors, and provide  F > a clear definition of the record structure of a file. When another  E > program tries to use the record service to read the records in an  nL > erroneous format, this is quickly identified as an error, before passing  B > the incorrectly interpreted information to the next application. >i@ > I have been told several times by programmers, that they found >a= > 	"RMS is a frustrating obstacle to my programming freedom",r	 > or thatl; > 	"RMS is very finicky about file and record definitions".b >  > I then tell them...  > = > 	"That is why RMS exists, to be finicky and frustrate you".  > B > RMS helps programmers to be disciplined, and to get their file   > definitions right.  H Actually, you could make all the same arguments as a case for using PL/I
 instead of C.i > C > Please note, that RMS also manages the access of files with the  wF > stream-lf format. This protects, for instance, a program correctly  ? > programmed for accessing a stream-lf file from accessing an  oH > Indexed-Sequential file, or even an executable file erroneously. RMS  L > thus reduces the likelihood not only of programmer error, but user error  
 > as well. >vI > I have over the years experienced countless cases where RMS has saved   J > our customers a lot of trouble. I am especially referring to a complex  E > CIM and Production Planning environment in which a diagram of the  uD > interfaces between the processes controlling the various factory  E > processes looks like a 3D spider web. Information produced by one  WK > process is often transferred to many other processes in that processing  yL > web using some form of file transfer. Having an undiscovered error in an  H > interface would quickly propagate deeply into many other systems. It  A > would be a potentially create an unrecoverable mess! This CIM  oJ > environment also sends and receives information from external systems,  I > often programmed by external companies or consultants using their own   I > methods. By having a RMS described file as a transfer medium, we have   C > often caught these external agencies that have quietly (perhaps  dJ > unintentially through their own programming errors) changed the format  I > of these files, before accepting their information in our complex CIM  tK > environment. For us, this seems to happen regularly when information is  oI > transferred from SAP systems which have had some sort of upgrade. The  tL > SAP folks appear to be so isolated from the file details that they don't  H > always know the exact format of a file that will be produced after a  B > change. They could certainly use a service that implements the   > principles found in RMS! >mK > Although, I agree that there are customers and applications with design  rG > goals that would make the Unix minimalism desirable, I respectfully   J > disagree that this philosophy should be a goal within an Enterprise or  J > Mission-Critical oriented OS. The presence of such universal (at least  J > within the chosen OS) quality-oriented services is clearly a desirable  A > quality enhancement within the professional Enterprise OS and  u! > Mission-Critical problem space.e >n	 > Cheers!b >, > Keith Cayembergs       -- rC Using Opera's revolutionary e-mail client: http://www.opera.com/m2/c   ------------------------------  % Date: Wed, 19 Jan 2005 17:26:13 +0100c0 From: Keith Cayemberg <keith.cayemberg@arcor.de> Subject: Re: vms versus solariseB Message-ID: <41ee8a25$0$17600$9b4e6d93@newsread4.arcor-online.net>   Bill Todd wrote: > Dave Weatherall wrote: >  > .... >  >   As a > F >> record-oriented user/programmer, it _is_ a decisive plus but then, > >> perhaps, I fall into the 'locked in' category (mentally or  >> philosophically). >  > J > The point is that as a record-oriented user/programmer, you'd also have I > access to the kinds of packages I mentioned on Unix (or - blech - even pA > to some extent on Windows) - in fact, to a considerably larger aK > collection of options in this area than is available to you on VMS, many  I > of them free or very low cost but still very functional and stable and hI > quite a few of them also somewhat more technically current than RMS is pI > (e.g., in their integral use of journaling to improve both performance l > and reliability).o > F > The fact that RMS is bundled and supported as part of the OS may be K > convenient and reassuring, but is hardly of *major* significance to most -H > users (or applications, though not having a third-party dependency in ( > this area is part of the convenience). >  > - bill  G It appears to me that everyone here fails to understand the philosophy  G behind having an universal record handling service integrated into the E operating system.   G It's primary purpose within the context of professional programming is gB not to make record handling "easier" or more "convenient" for the  programmer.a  I It's primary purpose is to improve the quality and correctness of record e storage and transfer.t  F This principle is similar to that which behind the use of Descriptors H and a defined Call Standard in OpenVMS, which are also excellent unique 0 quality features of the OpenVMS OS Architecture.  G By making this service "default" or for some purposes even "mandatory" eE for the operating system, programmers are encouraged to forego their oI artistic-license to make many common record-handling errors, and provide pC a clear definition of the record structure of a file. When another ,B program tries to use the record service to read the records in an I erroneous format, this is quickly identified as an error, before passing  @ the incorrectly interpreted information to the next application.  > I have been told several times by programmers, that they found  ; 	"RMS is a frustrating obstacle to my programming freedom",g or thatd9 	"RMS is very finicky about file and record definitions".t   I then tell them....  ; 	"That is why RMS exists, to be finicky and frustrate you".o  ? RMS helps programmers to be disciplined, and to get their file o definitions right.  @ Please note, that RMS also manages the access of files with the C stream-lf format. This protects, for instance, a program correctly t< programmed for accessing a stream-lf file from accessing an E Indexed-Sequential file, or even an executable file erroneously. RMS pI thus reduces the likelihood not only of programmer error, but user error h as well.  F I have over the years experienced countless cases where RMS has saved G our customers a lot of trouble. I am especially referring to a complex sB CIM and Production Planning environment in which a diagram of the A interfaces between the processes controlling the various factory pB processes looks like a 3D spider web. Information produced by one H process is often transferred to many other processes in that processing I web using some form of file transfer. Having an undiscovered error in an eE interface would quickly propagate deeply into many other systems. It r> would be a potentially create an unrecoverable mess! This CIM G environment also sends and receives information from external systems, hF often programmed by external companies or consultants using their own F methods. By having a RMS described file as a transfer medium, we have @ often caught these external agencies that have quietly (perhaps G unintentially through their own programming errors) changed the format tF of these files, before accepting their information in our complex CIM H environment. For us, this seems to happen regularly when information is F transferred from SAP systems which have had some sort of upgrade. The I SAP folks appear to be so isolated from the file details that they don't uE always know the exact format of a file that will be produced after a h? change. They could certainly use a service that implements the o principles found in RMS!  H Although, I agree that there are customers and applications with design D goals that would make the Unix minimalism desirable, I respectfully G disagree that this philosophy should be a goal within an Enterprise or lG Mission-Critical oriented OS. The presence of such universal (at least  G within the chosen OS) quality-oriented services is clearly a desirable  > quality enhancement within the professional Enterprise OS and  Mission-Critical problem space.e   Cheers!    Keith Cayembergc   ------------------------------  + Date: Wed, 19 Jan 2005 17:13:35 +0000 (UTC)c( From: m.kraemer@gsi.de (Michael Kraemer) Subject: Re: vms versus solarisi4 Message-ID: <csm4fv$t7$1@lnx107.hrz.tu-darmstadt.de>  u In article <41ee8a25$0$17600$9b4e6d93@newsread4.arcor-online.net>, Keith Cayemberg <keith.cayemberg@arcor.de> writes:l > @ > I have been told several times by programmers, that they found > = > 	"RMS is a frustrating obstacle to my programming freedom", 	 > or thato; > 	"RMS is very finicky about file and record definitions".  >  > I then tell them...e > = > 	"That is why RMS exists, to be finicky and frustrate you".  > A > RMS helps programmers to be disciplined, and to get their file l > definitions right. >    Funny.> In former times some of the VMS bigots in my field joked aboutC the - in their view - strange concept of record structured files on M IBM's mainframes. "No need for that, just open a file and be happy bytewise".-D Strange that VMS bigots nowadays bash Unix for exactly the opposite.   ------------------------------  # Date: Wed, 19 Jan 2005 18:01:59 GMT * From: "FredK" <fred.nospam@nospam.dec.com> Subject: Re: vms versus solaris<1 Message-ID: <rgxHd.6131$c36.864@news.cpqcorp.net>l  . "Tom Linden" <tom@kednos.com> wrote in message" news:opskuy70bkzgicya@hyrrokkin...5 > On Wed, 19 Jan 2005 17:26:13 +0100, Keith Cayemberg  > J > Actually, you could make all the same arguments as a case for using PL/I > instead of C.-  K I can make arguments for buying a caddy rather than a bimmer, but the caddyu still remains an old fart car.  K Kind-of like PL/I.  Or VMS itself.  Not a lot of appeal to anyone who wants. to appear young or cool.a   ------------------------------   Date: 19 Jan 2005 18:09:15 GMT( From: bill@cs.uofs.edu (Bill Gunshannon) Subject: Re: vms versus solarise, Message-ID: <357m2aF4h22dhU3@individual.net>  3 In article <XKswSWxjDAFb@eisner.encompasserve.org>,s0 	Kilgallen@SpamCop.net (Larry Kilgallen) writes:Y > In article <3576loF4h517hU2@individual.net>, bill@cs.uofs.edu (Bill Gunshannon) writes:16 >> In article <0rF4L0yqRgL6@eisner.encompasserve.org>,3 >> 	Kilgallen@SpamCop.net (Larry Kilgallen) writes:nR >>> In article <opsktoxpk4zgicya@hyrrokkin>, "Tom Linden" <tom@kednos.com> writes: > J >>>> ulimit(1)                                                             >>>> ulimit(1) >>>> t	 >>>> NAMEs >>>> u1 >>>>    ulimit - Sets or reports a resource limits >>> I >>> I did not see anything in that description specifying the name of theh >>> user affected. >> iB >> A user can only change his own limimits so there is no need for! >> the ability to specify a user.e >> s >>> F >>> Does this mean that all users on the system have the same quotas ? >> eH >> No, the initial values are set at login based on a class.  The values" >> are defined in /etc/login.conf. > E > Does that mean a user can raise values with the ulimit command evens@ > though the system manager has set them to a particular limit ?   No.    bill   -- dJ Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesD bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. University of Scranton   |A Scranton, Pennsylvania   |         #include <std.disclaimer.h>       ------------------------------  % Date: Wed, 19 Jan 2005 10:13:37 -0500n# From: "John Smith" <a@nonymous.com>  Subject: Re: Webcast and VMS, Message-ID: <ea2dnS0KD5G_5HPcRVn-ug@igs.net>   Dave wrote:wF > You asked for suggestions for improving the perception of VMS in the7 > marketplace.  At least I think that's what you asked.l > D > I think you've already said what's required. A quote from the Q&A: >eD > "OpenVMS is one of the most powerful operating environments in the > industry." > D > All you need to do is continue to say this publically.  Often.  InA > advertisements.  Every salesman should make this statement to aoE > customer, at least once, regardless of what the customer is lookingr > at.  >nE > If a customer had to choose between a perceived "industry standard"fD > and "the best", how many people do you know that want second best.@ > All you have to do is make sure that they know that "the best"# > exists, and is available from HP.n >aE > It's my perception that HP makes more on a VMS sale than most othermG >   HP products. Most VMS customers buy lots of support.  Good support,  > forget about India.i      B Well put.   try sending this to Ann (and Marcello) using the usualH fname.lname@CUTTHIShp.com, perhaps with 'OpenVMS' as the subject line to7 escape spam filters. Also request a Read Reply receipt.   J And/Or you can use both these links to send your message with the preambleI that Ann explicitly requested feedback like this during the chat phase ofo the webcast yesterday...   via carly(tm):: http://www.hp.com/hpinfo/execteam/email/fiorina/index.html   via the Board of Directors:s6 http://www.hp.com/hpinfo/execteam/email/bod/index.html   ------------------------------   End of INFO-VAX 2005.038 ************************