1 INFO-VAX	Sat, 26 Jul 2003	Volume 2003 : Issue 409       Contents:9 Are there PHP database plug-ins for Rdb and MySQL on VMS? 	 Re: cache 	 Re: cache 	 Re: cache 	 Re: cache 	 Re: cache " Re: Compiling PINE 3.91-2 for AXP.N Re: D.H. Brown: EV7 AlphaServers Deliver Enhanced RAS and  PowerfulPerformanceO Re: D.H. Brown: EV7 AlphaServers Deliver Enhanced RAS and Powerful  Performance O Re: D.H. Brown: EV7 AlphaServers Deliver Enhanced RAS and Powerful  Performance N Re: D.H. Brown: EV7 AlphaServers Deliver Enhanced RAS and Powerful PerformanceN Re: D.H. Brown: EV7 AlphaServers Deliver Enhanced RAS and Powerful PerformanceN Re: D.H. Brown: EV7 AlphaServers Deliver Enhanced RAS and Powerful PerformanceN Re: D.H. Brown: EV7 AlphaServers Deliver Enhanced RAS and Powerful PerformanceN Re: D.H. Brown: EV7 AlphaServers Deliver Enhanced RAS and Powerful PerformanceN Re: D.H. Brown: EV7 AlphaServers Deliver Enhanced RAS and Powerful PerformanceP Re: D.H. Brown: EV7 AlphaServers Deliver Enhanced RAS and Powerful Performance PP Re: D.H. Brown: EV7 AlphaServers Deliver Enhanced RAS and Powerful Performance PP Re: D.H. Brown: EV7 AlphaServers Deliver Enhanced RAS and PowerfulPerformance Po DEC 3000 Console Problem Re: DEC 3000 Console ProblemG Re: Does RT-11 run on the PDP-11/70?  (was Re: PDP-11 OS Release Dates) G Re: Does RT-11 run on the PDP-11/70?  (was Re: PDP-11 OS Release Dates) G Re: Does RT-11 run on the PDP-11/70?  (was Re: PDP-11 OS Release Dates) G Re: Does RT-11 run on the PDP-11/70?  (was Re: PDP-11 OS Release Dates) F Re: Does RT-11 run on the PDP-11/70? (was Re: PDP-11 OS Release Dates)F Re: Does RT-11 run on the PDP-11/70? (was Re: PDP-11 OS Release Dates) Re: DVD compatibility with VMS# Re: Elevate privileges in a program # Re: Elevate privileges in a program # Re: Elevate privileges in a program # Re: Elevate privileges in a program # Re: Elevate privileges in a program # Re: Elevate privileges in a program  Re: EMC on VMS; Re: How to operate MSL5000 tape library from OpenVMS prompt + Re: HP World: Why Alpha's Omega Makes Sense + Re: HP World: Why Alpha's Omega Makes Sense P Re: Intel's had Dirty Bombs Anthrax and everything! (Was: Re: HP World: Why AlphP Re: Intel's had Dirty Bombs Anthrax and everything! (Was: Re: HP World: Why AlphP Re: Intel's had Dirty Bombs Anthrax and everything! (Was: Re: HP World: Why AlphP Re: Intel's had Dirty Bombs Anthrax and everything! (Was: Re: HP World: Why AlphP Re: Intel's had Dirty Bombs Anthrax and everything! (Was: Re: HP World: Why AlphP Re: Intel's had Dirty Bombs Anthrax and everything! (Was: Re: HP World: Why AlphG Re: Legal status of EVE source code ?, was: RE: Creating an EVE o	r TPU 2 Re: Migrate obsolete VAX/VMS SW to OpenVMS Itanium2 Re: Migrate obsolete VAX/VMS SW to OpenVMS Itanium2 Re: Migrate obsolete VAX/VMS SW to OpenVMS Itanium2 Re: Migrate obsolete VAX/VMS SW to OpenVMS Itanium Re: New OpenVMS-to-Itanium FAQ  Re: NEWBIE: DECWindows Bit Depth  Re: NEWBIE: DECWindows Bit Depth  Re: NEWBIE: DECWindows Bit Depth  Re: NEWBIE: DECWindows Bit Depth1 Re: OpenVMS homepage and 'sales.liveperson.net' ? 1 Re: OpenVMS homepage and 'sales.liveperson.net' ? 1 Re: OpenVMS homepage and 'sales.liveperson.net' ? 1 Re: OpenVMS homepage and 'sales.liveperson.net' ? " RE: Packed decimal arithmetic in C" Re: Packed decimal arithmetic in C" Re: Packed decimal arithmetic in C" Re: Packed decimal arithmetic in C" Re: Packed decimal arithmetic in C# Re: Sun Micro Profit, Revenues Fall # Re: Sun Micro Profit, Revenues Fall # Re: Sun Micro Profit, Revenues Fall # Re: Sun Micro Profit, Revenues Fall  Re: The PDP lives 	 UCX ports 
 Re: UCX ports * Re: VMS and Intel, yet another question...  F ----------------------------------------------------------------------  # Date: Sat, 26 Jul 2003 01:38:10 GMT L From: winston@SSRL.SLAC.STANFORD.EDU ("Alan Winston - SSRL Admin Cmptg Mgr")B Subject: Are there PHP database plug-ins for Rdb and MySQL on VMS?6 Message-ID: <00A23664.5C48793E@SSRL.SLAC.STANFORD.EDU>  	 VMSers --   E For two different projects, I have a need for PHP database plugins to E communicate with Rdb and with MySQL, with the databases hosted on the  VMS server.   N I have a budget of zero, and I know a lot more about Perl than I do about PHP.  M A quick google search turned up discussion about using the ODBC PHP plugin to O talk to Rdb databases, but that won't work for Rdb running on the same machine; L the ODBC driver for Rdb is a Windows-only front-end to SQL/Services, so in a0 VMS-only solution there's nothing speaking ODBC.  L (I'm aware that this problem could be solved with Attunity or Connx, but see' the part above about a budget of zero.)   L There is a Perl-based ODBC server which can use DBD:: plugins to talk to theJ actual databases, so I suppose it's theoretically possible to use the PHP I ODBC plug-in to talk to the Perl ODBC server which talks to the database, M assuming that the PHP ODBC plug-in exists on VMS, but this seems unlikely to  2 perform well and to be extremely painful to debug.   -- Alan    --  O =============================================================================== 0  Alan Winston --- WINSTON@SSRL.SLAC.STANFORD.EDUM  Disclaimer: I speak only for myself, not SLAC or SSRL   Phone:  650/926-3056 M  Paper mail to: SSRL -- SLAC BIN 99, 2575 Sand Hill Rd, Menlo Park CA   94025 O ===============================================================================    ------------------------------    Date: 25 Jul 2003 14:14:59 -05004 From: cornelius@encompasserve.org (George Cornelius) Subject: Re: cache3 Message-ID: <$uJRm08PDwRP@eisner.encompasserve.org>   c In article <KbOC3rQ7f0mJ@eisner.encompasserve.org>, Kilgallen@SpamCop.net (Larry Kilgallen) writes: D > Perhaps not a normal prompt for MUMPS, but when I encountered thisL > problem in the presence of a MUMPS expert he said "Oh it's...<I forget>...8 > which is what you would expect as a MUMPS programmer."  K Hey, I resemble that remark!  (Oh, I see.. it was you doing the forgetting, 0 not the Mumps programmer.  I resemble that too.)  K Actually, I don't recall the problem of getting out of Mumps the first time E being nearly as difficult as getting out of TECO.  Gee, that sequence K HKEKEX<ESC><ESC> should be right on the tip of the tongue of anyone who has / ever used a DEC/Compaq/HP system, shouldn't it?   F (Yes, I know, <CTRL/C> a few times works too, at least with VMS TECO -J but that was probably added just because of the panic users went into whenL there was no response at all to QUIT<CR>, EXIT<CR>, or any other sequence of9 ordinary ASCII characters followed by a carriage return).    --? George Cornelius                    cornelius@encompasserve.org 6                                     cornelius@mayo.edu   ------------------------------    Date: 25 Jul 2003 16:26:17 -0500; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler)  Subject: Re: cache3 Message-ID: <h$f4jd6JTDKM@eisner.encompasserve.org>   j In article <$uJRm08PDwRP@eisner.encompasserve.org>, cornelius@encompasserve.org (George Cornelius) writes: > H > (Yes, I know, <CTRL/C> a few times works too, at least with VMS TECO -L > but that was probably added just because of the panic users went into whenN > there was no response at all to QUIT<CR>, EXIT<CR>, or any other sequence of; > ordinary ASCII characters followed by a carriage return).   D    Very first editor I ever used:  TECO on a DECSYSTEM-10, ^C worked-    just fine (unless you wanted that change).   G    Never had any version if TECO fail to exit on EX$$, one of the first B    commands I was taught.  Of course if you didn't have an ESC keyB    you really did need to enable ALTMODE before you got into TECO.  @    OBTW, ^C and ESC are both in the ASCII character set, no more    extra-ordinary than <CR>.   ------------------------------  % Date: Fri, 25 Jul 2003 23:15:07 +0100 4 From: "Chris Casey" <chris.caseyNOSPAM@ntlworld.com> Subject: Re: cache@ Message-ID: <l_hUa.2871$UR2.259893@newsfep2-gui.server.ntli.net>  < "David J. Dachtera" <djesys.nospam@fsi.net> wrote in message! news:3F2079BB.3EDC855A@fsi.net...  > Chris Casey wrote:L > > Actually thinking about it, it could be a standard prompt if it referred toG > > a database named INS. In which case H<RET> should exit the process.  > C > You're thinking of "namespace". This is displayed before the ">".  >   + David, I was just trying to keep it simple.    ------------------------------    Date: 25 Jul 2003 18:16:18 -0500- From: Kilgallen@SpamCop.net (Larry Kilgallen)  Subject: Re: cache3 Message-ID: <wgtWINUr2U50@eisner.encompasserve.org>   j In article <$uJRm08PDwRP@eisner.encompasserve.org>, cornelius@encompasserve.org (George Cornelius) writes:e > In article <KbOC3rQ7f0mJ@eisner.encompasserve.org>, Kilgallen@SpamCop.net (Larry Kilgallen) writes: E >> Perhaps not a normal prompt for MUMPS, but when I encountered this M >> problem in the presence of a MUMPS expert he said "Oh it's...<I forget>... 9 >> which is what you would expect as a MUMPS programmer."  > M > Hey, I resemble that remark!  (Oh, I see.. it was you doing the forgetting, 2 > not the Mumps programmer.  I resemble that too.) > M > Actually, I don't recall the problem of getting out of Mumps the first time G > being nearly as difficult as getting out of TECO.  Gee, that sequence M > HKEKEX<ESC><ESC> should be right on the tip of the tongue of anyone who has 1 > ever used a DEC/Compaq/HP system, shouldn't it?  > H > (Yes, I know, <CTRL/C> a few times works too, at least with VMS TECO -L > but that was probably added just because of the panic users went into whenN > there was no response at all to QUIT<CR>, EXIT<CR>, or any other sequence of; > ordinary ASCII characters followed by a carriage return).   & To just exit from TECO, type Control/Z' When that does not work, type Control/Z I Finally, to convince TECO you deserve to get out, type a third Control/Z.    ------------------------------  % Date: Fri, 25 Jul 2003 22:59:45 -0500 1 From: "David J. Dachtera" <djesys.nospam@fsi.net>  Subject: Re: cache% Message-ID: <3F21FCB1.28FAF8@fsi.net>    Chris Casey wrote: > > > "David J. Dachtera" <djesys.nospam@fsi.net> wrote in message# > news:3F2079BB.3EDC855A@fsi.net...  > > Chris Casey wrote:N > > > Actually thinking about it, it could be a standard prompt if it referred > toI > > > a database named INS. In which case H<RET> should exit the process.  > > E > > You're thinking of "namespace". This is displayed before the ">".  > >  > - > David, I was just trying to keep it simple.   F I didn't want to asume that a namespace can include only one database.   --   David J. Dachtera  dba DJE Systems  http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/    ------------------------------  % Date: Fri, 25 Jul 2003 15:27:27 -0400 0 From: "Brian Tillman" <Tillman@sparkingwire.com>+ Subject: Re: Compiling PINE 3.91-2 for AXP. $ Message-ID: <3f2184a7$1@news.si.com>  E >Am I correct that I need a PAK for the DEC/Compaq/HP (or whatever) C 
 >compiler?   Absolutely correct.   @ >If so, where can I find the best version of GCC for VMS on AXP?  K It used to be available at ftp://ftp.progis.de/pub/vms/gcc/ , but that site F no longer allows anonymous FTP.  It also used to be mirrored at HunterL Goatley's site when he was at Western Kentucky University, but he apparentlyL didn't take or was asked not to take those sources with him when he moved toJ Process Software.  So, there doesn't seem to be any GCC or G++ for OpenVMS Alpha any more.   J When last I knew, PINE for VMS did only IMAP V2 and not the current V4, so% it may be less useful than you think.  --  I Brian Tillman         Internet: Brian.Tillman at smiths-aerospace dot com 5 Smiths Aerospace  Addresses modified to prevent SPAM. D 3290 Patterson Ave. SE, MS 1B3 Replace "at" with "@", "dot" with "." Grand Rapids, MI 49512-1991 8        This opinion doesn't represent that of my company   ------------------------------  % Date: Fri, 25 Jul 2003 15:33:54 -0700 & From: Greg Cagle <gregc@gregcagle.com>W Subject: Re: D.H. Brown: EV7 AlphaServers Deliver Enhanced RAS and  PowerfulPerformance / Message-ID: <vi3c2kcglq6q39@corp.supernews.com>    JF Mezei wrote:   * > Andrew Harrison SUNUK Consultancy wrote: > = >>If you think back to the battles over the CPQ-HP merger you ? >>will also remember that Walter Hewlett one of your main board < >>members was clear that systems did not make HP money hence> >>his opposition to merging with another company that produced >>computer systems.  >  > L > My recollection was that Hewlett was opposed because the merger would haveI > made HP focus too much on the unprofitable wintel market and not on the - > profitable enterprise and printer business.   > Right. Walter's opposition was quite different than Andrew has characterized.  L > And I strongly suspect that as part of the deal with Deutsche Bank/BankersI > Trust, Carly probably had to agree to put a greater focus on enterprise P > systems instead of wintel, a feeling that was clearly reflected in the may 7th# > speeches that downplayed windows.   F What deal are you talking about? You mean the *rumored* deal? I really. doubt this scenario occurred as you speculate.  P > The problem with HP's enterprise business is that it has no products and won'tN > have for a couple of years. All its enterprise products are in transition orM > in the process of being killed. So it is no surprise that it, combined with $ > recession, wouldn't be profitable.  H Where do you get this stuff? HP has plenty of "enterprise products" thatE are perfectly viable, not in transition, and not being killed. Or are 2 you limiting your scope to premerger-CPQ products?   --  
 Greg Cagle gregc at gregcagle dot com   ------------------------------  # Date: Fri, 25 Jul 2003 22:03:31 GMT & From: jlsue <jefflsxxxz@sbcglobal.net>X Subject: Re: D.H. Brown: EV7 AlphaServers Deliver Enhanced RAS and Powerful  Performance8 Message-ID: <u8a3ivc0flrv5c5idr3i5bucfl3vo2qs7q@4ax.com>  H On Thu, 24 Jul 2003 13:42:44 -0400, JF Mezei <jfmezei.spamnot@istop.com> wrote:   >Fred Kleinsorge wrote: J >> The games Sun used in boosting a Spec result are widely known, and haveL >> re-sparc'ed a lively debate on how this type of stuff might be prevented.H >> In short, a specific compiler optimization for a very limited type of& >> application juiced up your numbers. > M >Interesting you would mention this. I strongly suspect that IA64 is far more O >guilty of this type of behaviour than SPARC or other chips, simply because the N >EPIC concept requires the compiler to do very specific work to make a program
 >run fast.  K That's strange.  How do you make this comparison to other compilers?  Where J do you draw the line between compilers written for their architecture, and those pulling "dirty" tricks?   E Alpha compilers had to incorporate a lot of new ideas to get the best / performance on that architecture.  Who doesn't?    ------------------------------  % Date: Fri, 25 Jul 2003 17:27:45 -0700 3 From: Robert Klute <robert_klute_removethis@hp.com> X Subject: Re: D.H. Brown: EV7 AlphaServers Deliver Enhanced RAS and Powerful  Performance8 Message-ID: <sdi3ivkmgjlv2b1f4el7hiac4n9roi50cm@4ax.com>  B On Fri, 25 Jul 2003 22:03:31 GMT, jlsue <jefflsxxxz@sbcglobal.net> wrote:  I >On Thu, 24 Jul 2003 13:42:44 -0400, JF Mezei <jfmezei.spamnot@istop.com>  >wrote:  >  >>Fred Kleinsorge wrote:K >>> The games Sun used in boosting a Spec result are widely known, and have M >>> re-sparc'ed a lively debate on how this type of stuff might be prevented. I >>> In short, a specific compiler optimization for a very limited type of ' >>> application juiced up your numbers.  >>N >>Interesting you would mention this. I strongly suspect that IA64 is far moreP >>guilty of this type of behaviour than SPARC or other chips, simply because theO >>EPIC concept requires the compiler to do very specific work to make a program  >>run fast.  > L >That's strange.  How do you make this comparison to other compilers?  WhereK >do you draw the line between compilers written for their architecture, and  >those pulling "dirty" tricks? > F >Alpha compilers had to incorporate a lot of new ideas to get the best0 >performance on that architecture.  Who doesn't?  G Compilers that gen code based on the routine name.  Compilers that look D for specific coding sequences, only commonly used in benchmarks, andC insert hand tuned code.  Compilers that restructure the data to get D performance gains.  When compilers start tuning specifically for theE benchmark and not because a customer has a similar need, then it is a F trick.  I will admit, as benchmarks get more sophisticated, there is aC large grey area as to what is customer driven and what is benchmark  sniping.     Robert Klute Cupertino Solution Center  Hewlett-Packard Company  ----- 6 The opinions are those of the poster, not the company.   ------------------------------  # Date: Fri, 25 Jul 2003 21:59:44 GMT & From: jlsue <jefflsxxxz@sbcglobal.net>W Subject: Re: D.H. Brown: EV7 AlphaServers Deliver Enhanced RAS and Powerful Performance 8 Message-ID: <mj93ivspfl9uanv2lb1o95i093rn6f4lds@4ax.com>  H On Wed, 23 Jul 2003 18:07:34 -0400, "Bill Todd" <billtodd@metrocast.net> wrote:   > 4 >"jlsue" <jefflsxxxz@sbcglobal.net> wrote in message3 >news:ta4thvcotv7hmi6i7ire682vqr39t5mt63@4ax.com... K >> On Fri, 18 Jul 2003 18:01:15 -0400, "Bill Todd" <billtodd@metrocast.net> 	 >> wrote:  >> >> >>M >> While I agree that the fall in systems after the announcement hurt things, M >> I don't know if that was planned-for or not - and I sincerely doubt you do D >> either.  It's entirely possible that it was worse than originally
 >expected,) >> but that's water under the bridge now.  > M >The point, of course, was that a dramatic fall *should* have been expected - I >by anyone other than the utterly incompetent (or those with other than a J >strictly financial agenda who hoped they could obscure their real motives >with sufficient bluster).  J Realize that I'm not saying they are/were competent.  I'm just questioningI the verasity of the conclusions based on the relatively scant input data.    >  >>F >> But there are no real numbers that you'd be able to find that would >"prove"+ >> this would have made EV8 worth pursuing.  > % >Sure there are:  I've provided them.   K You've provide some.  How do you know that they are all the pertinent data? H What inside financial information do you have that nobody else has?  AndG based on where the BOD may have wanted to go, how do you know that they + didn't choose the direction with open eyes?    >  >  There are many, many costs J >> over-and-above what you may see in the Annual Report.  Groups like ESG,J >> Tru64, OpenVMS, NSK, Customer Support, compilers, layered-products etc. >areJ >> just some examples.  And you've got to factor in the projected revenues >and0 >> margins, possible price-points, volumes, etc. > F >No, I don't - because people like Rich Marcello have (at least if oneL >assumes minimal competence on *their* part) already done so.  Rich reportedK >that annual Alpha development costs ran $150 million (one might expect for K >both the EV7 and EV8 tracks combined, but even if you consider that figure A >to have been purely EV8-related the conclusions don't change).     I I think I quite clearly stated my contention that there are other factors I besides merely the Alpha development team.  That $150Mil would only cover I the actual engineering team.  I don't believe the compilers, for example, 3 were part of it.  Nor on-going OS development, etc.   I If there plan was to go to IA64 long-term, but they couldn't afford to do K both, then they made a reasonable decision.  It may not have all turned out ? exactly as they thought, but they had reasonable justification.   
 >Rich alsoG >reported that annual VMS system profit ran $800 million (in Y2K, on $4 M >billion in annual revenue, a figure which persisted through at least late Q1 G >of 2001).  Recently, Mark Gorham has been quoted as saying current VMS L >annual system profit is $500 million on $2.5 - $3 billion in annual revenueH >(and in December, 2001, a Compaq response to a Gartner report put VMS'sK >annual revenue at only $2 billion - so it, and likely profit as well, seem L >to have rebounded a little from a low point shortly after the Alphacide but, >to nowhere near the pre-Alphacide level).    @ But you can't prove that ALL of that decline is due to the AlphaF announcement.  There's an implied, though unstated, "all other factorsH being equal", and I don't believe that all other things are equal.  LotsA and lots of things happened - Wintel grew to much larger servers,  9/11/2001, etc.    >And examination of the CompaqI >quarterly report statements shows no significant decline in VMS revenues " >before the date of the Alphacide.  I And that does not imply any commitment that the same trend will continue. H Just like investing in Wall Street, past performance is no indication of future performance.    >  >>I >> It is just far, far too complex for anyone on the outside to determine . >> based on such little information available. > J >That's why I've relied on the statements made by the insiders in the best/ >position to know what they were talking about.  >   I But why would anyone assume that the statistics made in public statements C are all of the relevant data needed to make reasonable conclusions?    ------------------------------  # Date: Fri, 25 Jul 2003 22:12:34 GMT & From: jlsue <jefflsxxxz@sbcglobal.net>W Subject: Re: D.H. Brown: EV7 AlphaServers Deliver Enhanced RAS and Powerful Performance 8 Message-ID: <3ea3ivse59mhk5g9s8ufpav8in0i8t6equ@4ax.com>  E On Wed, 23 Jul 2003 15:59:36 +0100, Andrew Harrison SUNUK Consultancy . <Andrew_No.Harrison_No@nospamn.sun.com> wrote:  
 >jlsue wrote: H >> On Mon, 21 Jul 2003 16:05:07 +0100, Andrew Harrison SUNUK Consultancy1 >> <Andrew_No.Harrison_No@nospamn.sun.com> wrote:  >>   >>     >>> > >>>Well think about it the other way. Itanium given the amount> >>>it has cost so far is unlikely to turn a profit in the next@ >>>5 years, despite this your employer and Intel show no sign of% >>>dropping consigning it to the bin.  >>   >>  M >> This is entirely conjecture on your part.  There are many places where the L >> profit could come from, and system sales margins are just one part of our >> business affected.  >>   >   >What margins on Systems Sales ? > 6 >HP don't make a profit on systems sales currently and9 >historically they havn't either. Thats one of the things ) >that people were expecting Carly to fix.  > 7 >The only conjecture here is that you make a margin out  >of systems sales, you don't.   I Re-read what I wrote.  I am saying that the total profit the company gets H is more than the what it may get on the servers themselves.  This is NOTJ saying anything about what kind of margin we are getting.  It's a relative comparison.   G I thought I explained it fairly carefully that total profit can include C add-on sales (for example in customer support, consulting, even the  ever-profitable printers).    F >> While I can agree that PCs (desktop variety) don't seem to make anyG >> margins, this is not particularly true of the higher-end Intel-based K >> servers.  And I don't see how you can make this statement about the UNIX 4 >> servers either.  Where do you get this info from? >>   > 7 >I can make the statement because its in HP's accounts. 9 >Hadn't you noticed that the Enterprise Stystems business 9 >isn't profitable, this isn't a little current difficulty - >its one of the reasons Carly was brought in.   K Right now, since the big crash after 9/11/2001, there is very little profit G in ANYONE's enterprise systems units.  Since that downturn business has E been tough all-around... and there's some indication that it actually J started a year or so before, but the terrorist attacks just precipitated a huge landslide.   ? >>>No sorry, Intel allready has the FAB's they are used to make  >>>x86 processors. >>> B >>>500 million is a rough ballpark for a new ISA, 5 billion is 10x( >>>that and the FAB costs arn't in that. >>   >>  K >> No, sorry.  If they need to design new proceses and FABs to make the ISA E >> competitive, then it is a significant part of the overall cost.  I M >> personally don't follow closely enough enough to know for sure, but I'd be I >> surprised if their processes (in the current FABs) didn't require some  >> changes to make IA64 viable.  >>   > > >But Intel didn't need to design a new FAB to produce Itaniums? >in fact the Xeons are produced in a newer FAB than Itanium is.   H Right.  What I'm saying is that, to make/keep it competitive in the longG term, i.e., after the initial release that proves it's capable, they'll J probably need new FAB processes to implement the new architecture changes.K Alpha went through similar things in the earlier years.  21064 to 21164 was  a pretty expensive time.   >  >>  D >>>Current Itanium production doesn't warrant a dedicated FAB anyway3 >>>4000 units a quarter is a small corner of a FAB.  >>   >>  M >> And it's just getting started. Compared to Alpha, what is it, EV2?  EV2.5?  >>   > F >Alpha was never that unsucessfull, less than 2000 systems per quarterD >after 2 years, even Alpha at its worst was rather better than that. >   B Do you know how many of the EV3 systems were sold?  These were the& development systems that went to ISVs.   ------------------------------  # Date: Fri, 25 Jul 2003 22:20:25 GMT & From: jlsue <jefflsxxxz@sbcglobal.net>W Subject: Re: D.H. Brown: EV7 AlphaServers Deliver Enhanced RAS and Powerful Performance 8 Message-ID: <d4b3ivoq7ksi3li3bvc6g5cokst5lscn8a@4ax.com>  I On Wed, 23 Jul 2003 16:10:28 +0000 (UTC), david20@alpha1.mdx.ac.uk wrote:   b >In article <uc5thvs11ui8g0m0cvlb906e3rt7gua1mq@4ax.com>, jlsue <jefflsxxxz@sbcglobal.net> writes:  L >>And it's just getting started. Compared to Alpha, what is it, EV2?  EV2.5? >> > C >As I recall the EV4 shipped in the first generation Alpha Systems. ; >The previous EV3 or earlier were just internal test chips.   F Actually, IIRC, there were development platforms for ISVs.  I'll stateJ right out that I may be off on that, but in the area I was in at the time, that rings true.   > ? >Hence IA64 is at least in the same generation position as EV5. N >However I'd say there were a lot more production EV5 systems sold than there N >are production IA64 systems (even if you include ALL the IA64 generations in  >the IA64 total).   J I think I've already stated this before, but imo, the initial IA64 is moreI in the EV3-EV4 stages, while the most recent, soon-to-be-released is more  in the EV4-EV5 stages.  B In any case, it's very, very early and the business environment isG completely different than when Alpha first came out.  At Alpha release, K expensive, large servers were still more common than cheaper Wintel servers F for business apps.  I don't think you can easily discount IA64 at this early stage.   ------------------------------  # Date: Fri, 25 Jul 2003 22:16:44 GMT & From: jlsue <jefflsxxxz@sbcglobal.net>W Subject: Re: D.H. Brown: EV7 AlphaServers Deliver Enhanced RAS and Powerful Performance 8 Message-ID: <sua3ivkuikl4fklf31c4bcm5a2lc043o88@4ax.com>  E On Thu, 24 Jul 2003 15:09:09 +0100, Andrew Harrison SUNUK Consultancy . <Andrew_No.Harrison_No@nospamn.sun.com> wrote:   > ; >There is a very big difference between not making a profit < >in one quarter or even a couple of years and never making a >profit.  F Well, right now almost nobody's making a profit, so singling out HP is
 disingenuous.    > ; >And your very detailed attempts to guess what Suns systems = >business made or lost last quarter does not detract from the ; >fact that HPQ's systems businesses have never consistently 8 >made a profit. It was after all one of the reasons that) >Carly was brought in to run the company.   G Again, since the landslide in the business world after 9/11, from which D we're just starting to emerge, nobody's business has been very good.7 Let's wait and see once the rebound gets in full swing.   D I've already seen a very, very large in service requests in the lastI several weeks.  I'm so busy I haven't been on a delivery in over 2 months " since I also do service proposals.   > F >Carlys defence was that the merger would allow the systems businessesD >to become profitable, this hasn't happened HP has simply chosen to E >restate systems business profitability differently e.g in a way that / >makes them look profitable even if they arn't.  >   J And we'll just have to wait until after the downturn gets over with before we know the real truth.    ------------------------------  % Date: Fri, 25 Jul 2003 17:34:11 -0700 3 From: Robert Klute <robert_klute_removethis@hp.com> W Subject: Re: D.H. Brown: EV7 AlphaServers Deliver Enhanced RAS and Powerful Performance 8 Message-ID: <2ri3ivo3m7snol45bet092movlqnd8e9fs@4ax.com>  E On Thu, 24 Jul 2003 15:09:09 +0100, Andrew Harrison SUNUK Consultancy . <Andrew_No.Harrison_No@nospamn.sun.com> wrote:   > ; >There is a very big difference between not making a profit < >in one quarter or even a couple of years and never making a >profit. > ; >And your very detailed attempts to guess what Suns systems = >business made or lost last quarter does not detract from the ; >fact that HPQ's systems businesses have never consistently 8 >made a profit. It was after all one of the reasons that) >Carly was brought in to run the company.   H While it is probably not appropriate to go into details right now (quietG period remember), it did come out during the merger wars that the 9000s F and the Pavilion line were profitable and that the Wintel servers were where profits were a problem.   < >If you think back to the battles over the CPQ-HP merger you> >will also remember that Walter Hewlett one of your main board; >members was clear that systems did not make HP money hence = >his opposition to merging with another company that produced  >computer systems. > A >He also made it clear that HP's profitable business was printing / >and that was what delivered shareholder value.  > F >Carlys defence was that the merger would allow the systems businessesD >to become profitable, this hasn't happened HP has simply chosen to E >restate systems business profitability differently e.g in a way that / >makes them look profitable even if they arn't.        Robert Klute Cupertino Solution Center  Hewlett-Packard Company  ----- 6 The opinions are those of the poster, not the company.   ------------------------------  % Date: Fri, 25 Jul 2003 23:22:36 -0400 * From: "Bill Todd" <billtodd@metrocast.net>W Subject: Re: D.H. Brown: EV7 AlphaServers Deliver Enhanced RAS and Powerful Performance 2 Message-ID: <vFGdnaHTPPyXabyiXTWJkg@metrocast.net>  3 "jlsue" <jefflsxxxz@sbcglobal.net> wrote in message 2 news:mj93ivspfl9uanv2lb1o95i093rn6f4lds@4ax.com...J > On Wed, 23 Jul 2003 18:07:34 -0400, "Bill Todd" <billtodd@metrocast.net> > wrote: >  > > 6 > >"jlsue" <jefflsxxxz@sbcglobal.net> wrote in message5 > >news:ta4thvcotv7hmi6i7ire682vqr39t5mt63@4ax.com...    ...   H > >> But there are no real numbers that you'd be able to find that would
 > >"prove"- > >> this would have made EV8 worth pursuing.  > > ' > >Sure there are:  I've provided them.  >  > You've provide some.  F More to the point, I've provided - and examined - all the numbers thatG Compaq did, but instead of just waving them around ("Gee - $150 million J annually for Alpha chip development, and maybe as much as $300 million allL told:  of *course* we couldn't afford to continue spending that much!") I'veE put them together in context to make it clear just how ridiculous the 6 statements that Compaq purportedly based on them were.  8 >  How do you know that they are all the pertinent data?  I They were the data that Compaq used to try to make its case.  It's not my  fault if they chose poorly.   E > What inside financial information do you have that nobody else has?   J The only piece of 'inside' financial information I had was Rich Marcello'sL personal statement that VMS systems brought in $800 million in annual profitJ (since his statement that they brought in $4 billion in annual revenue wasK corroborated elsewhere in public materials).  The rest, as noted many times L already, came directly from Compaq's public statements - and, once again, ifG they failed to make public other information that would have made their F financial case more solid that's hardly *my* fault (not that I see any= reason to believe that any such 'other' information existed).t     AndhI > based on where the BOD may have wanted to go, how do you know that they - > didn't choose the direction with open eyes?l  F That, of course, is the real crux of the matter:  the fact that CompaqL attempted to fabricate financial and technical bases for its actions ('whereJ the BOD may have wanted to go') rather than coming straight out and sayingJ "We just don't want to be in this business, regardless of what promises weI made earlier and how profitable it might be."  Those fabrications are thee& basis for calling them outright liars.  K Of course, even had they been more forthright there would still have been arB major *integrity* issue.  To my mind, one doesn't just break clearH commitments on a whim (i.e., while it's still eminently possible to meetJ them without endangering the company).  The Heil/Lipcon letter effectivelyE (and *very* publicly) said "By all means bet your business on Alpha'srJ long-term future, because Compaq is betting *its* business on it" - and ofJ course was accompanied by the famous alpha_ia64.pdf comparison white paperL that explained why Alpha had nothing to fear on that front.  If the BoD thenL decided it wanted to do something different, the way to do so with integrityK would have been to try to renegotiate the previous commitments to somethingPG mutually agreeable with those people to whom those commitments had beenPJ made - rather than continue actively selling Alpha systems on the basis ofJ the Heil/Lipcon letter and alpha_ia64.pdf white paper not only while CurlyH was clearly entertaining doubts about Alpha (that dated back to the timeG those documents were written) but for over half a year after the actual I decision to phase out Alpha had been made (reportedly in November, 2000).   A Not only did Compaq lie, and not only did it break its absolutelygH unequivocal commitments to Alpha's long-term future, and not only did itL continue selling Alpha based on those commitments long after the decision toL terminate Alpha had been made:  it never even *hinted* at an apology for anyH of this, but instead just tried to bull its way through with revisionistI history and happy talk.  And the irony is that this to a very significant K degree repeated the kinds of behavior it exhibited when it abruptly axed NTrJ and Win64 development on Alpha in August, 1999 - the event that caused theJ need for the Heil/Lipcon 'commitment to Alpha' letter to be written in the first place.  B Do you really wonder that I consider them to be absolute scumbags?   - bill   ------------------------------  % Date: Fri, 25 Jul 2003 15:38:28 -0700V& From: Greg Cagle <gregc@gregcagle.com>Y Subject: Re: D.H. Brown: EV7 AlphaServers Deliver Enhanced RAS and Powerful Performance Po/ Message-ID: <vi3cb6i5hnb2b6@corp.supernews.com>b  ( Andrew Harrison SUNUK Consultancy wrote:  < > And your very detailed attempts to guess what Suns systems> > business made or lost last quarter does not detract from the< > fact that HPQ's systems businesses have never consistently > made a profit.    8 Really. How can you possibly know that? Are you an ex-HP< employee? Somehow I doubt that. Please cite your references.= Also please take into account that HP/HPQ's system businessese= have been around for a long time, longer than Sun has been inl- existence. Possibly before you were born 8^).    -- e
 Greg Cagle gregc at gregcagle dot com   ------------------------------  % Date: Fri, 25 Jul 2003 20:17:41 -0400u* From: JF Mezei <jfmezei.spamnot@istop.com>Y Subject: Re: D.H. Brown: EV7 AlphaServers Deliver Enhanced RAS and Powerful Performance P,) Message-ID: <3F21C8A4.ED19E4E5@istop.com>-   jlsue wrote:M > Right now, since the big crash after 9/11/2001, there is very little profit ( > in ANYONE's enterprise systems units.   J Recession started before that. Somewhere between Nov 4 2000 and January 17N 2001. Sept 11 was just an accelerator that made things drop one big notch, butD there is no telling how the economy would have behaved without 9-11.   ------------------------------  % Date: Fri, 25 Jul 2003 20:26:28 -0400 * From: JF Mezei <jfmezei.spamnot@istop.com>Y Subject: Re: D.H. Brown: EV7 AlphaServers Deliver Enhanced RAS and PowerfulPerformance Po ) Message-ID: <3F21CAB2.1D55BE33@istop.com>H   Greg Cagle wrote: J > Where do you get this stuff? HP has plenty of "enterprise products" thatG > are perfectly viable, not in transition, and not being killed. Or area4 > you limiting your scope to premerger-CPQ products?  A Tandem: dead end on MIPS, still a few years before it is on IA64.bE VMS:	dead end on Alpha, still at least one year before it is on IA64.e	 MPE:	deade Tru64: deads8 HP-UX: dead end on PA-Risc. Customers urged to go to theN not-quite-ready-for-prime-time IA64, and then must wonder what changes/impactsH will occur as a result of bits from Tru64 being added to HP-UX later on.    L So, where are the plenty of enterprise products that are not in transition ?   ------------------------------  % Date: Fri, 25 Jul 2003 14:38:12 -0400=) From: "Jim Jennis" <jhjennis@shentel.net>=! Subject: DEC 3000 Console ProblemS/ Message-ID: <vi30cfjaut559b@corp.supernews.com>y  G Hi All,I have an old DEC 3000 running OpenVMS 7.0. (It is intentionallyCG running this obsolete version of OpenVMS).We lose the connection to our=E console when the system finishes booting. I suspect it is because the-G console is owned by a non-existant process but I am notsure why this is@& happening or how to fix it.When I do a $show dev opa0:/fu   I see the following...   show dev opa0:/fug  D  Terminal OPA0:, device type unknown, is online, enabled as operator	 terminal,r) record-oriented device, carriage control.m    <     Error count                    0    Operations completed 96  1     Owner process                 ""    Owner UIC. [1,4]h  0     Owner process ID        00000083    Dev Prot S:RWPL,O:RWPL,G,W   ;     Reference count                1    Default buffer sizeo 80  K NOTE that opa0: is owned by process id 83, however there is NO PROCESS withlC this ID running on the system. What is going on here? How can I fix , this?Thanks in advance for your help/advice!   Thanks!<   Jimo   ------------------------------  # Date: Fri, 25 Jul 2003 19:36:40 GMT/# From: hoff@hp.nospam (Hoff Hoffman)Y% Subject: Re: DEC 3000 Console Problemt0 Message-ID: <cFfUa.913$Nv5.503@news.cpqcorp.net>  [ In article <vi30cfjaut559b@corp.supernews.com>, "Jim Jennis" <jhjennis@shentel.net> writes:>H :Hi All,I have an old DEC 3000 running OpenVMS 7.0. (It is intentionally+ :running this obsolete version of OpenVMS).s  B   Um, any background on why you are staying on OpenVMS Alpha V7.0?  H   Obviously you do wish to deal with bugs that we have already found andF   fixed, right?  (Running older versions is not without its costs, andE   these costs can tend to be underestimated -- and worse, can tend torE   lead to some "surprise" upgrade requirements.  Best to manage theserD   costs, and stay more current.  TANSTAAFL, obviously.)  I certainlyE   can't encourage running revisions such as this, and would encourageu'   moving forward to V7.2-2 or V7.3-1.     F   If you must stay on OpenVMS Alpha V7.0, you will want to acquire and   apply all mandatory ECO kits.     G :                                         We lose the connection to ouriF :console when the system finishes booting. I suspect it is because theH :console is owned by a non-existant process but I am notsure why this is= :happening or how to fix it.When I do a $show dev opa0:/fu...  : L :NOTE that opa0: is owned by process id 83, however there is NO PROCESS withD :this ID running on the system. What is going on here? How can I fix- :this?Thanks in advance for your help/advice!n  E   This appears to be a problem likely fixed in a more recent version 0F   of OpenVMS -- there were cases where a command could end up with theE   OPA0: device owned by a non-existent process.  Check the accounting.D   logs and see which process was using PID 83 -- that can give you aE   clue toward the problem, obviously.  (I remember seeing reports of tE   a rather similar problem, and I know there were changes implemented D   to resolve it -- what I don't know off-hand is the specific timing<   of the fixes, the ECO kits, and the new OpenVMS releases.)  E   I'd look for OPCOM-related commands, use of SHARE privilege to gain-E   access to OPA0: (use of SHARE to override ownership and to access alI   device is a classic cause of this particular class of ownership error),UG   and similar sequences within the site-specific system startup commandpE   procedure.  And I'd certainly also apply all of the mandatory ECOs  H   for OpenVMS and for DECwindows, and ECOs related to DECwindows, OPCOM,D   and similar commands.   And I'd definitely upgrade to more current"   versions of the products in use.  N  ---------------------------- #include <rtfaq.h> -----------------------------K     For additional, please see the OpenVMS FAQ -- www.hp.com/go/openvms/faqtN  --------------------------- pure personal opinion ---------------------------E         Hoff (Stephen) Hoffman   OpenVMS Engineering   hoff[at]hp.comm   ------------------------------    Date: 25 Jul 2003 13:31:45 -05004 From: kaplow_r@encompasserve.org.TRABoD (Bob Kaplow)P Subject: Re: Does RT-11 run on the PDP-11/70?  (was Re: PDP-11 OS Release Dates)3 Message-ID: <X+Y0$$TMCNht@eisner.encompasserve.org>   g In article <bfombe$8b6$1@shell.monmouth.com>, pechter@shell.monmouth.com (Bill/Carolyn Pechter) writes:e0 > You need to sysgen RT11 for the light pattern. > E > It has a pretty slick one. IIRC it had one light on each end movingyG > to the center bit -- they touch, then bounce out to the ends and thenE" > two lights do it and then three.  J Did they bounce off each other, or just pass through each other. After all light has no mass...  1 	Bob Kaplow	NAR # 18L	TRA # "Impeach the TRA BoD"u& 		>>> To reply, remove the TRABoD! <<<K Kaplow Klips & Baffle:	http://nira-rocketry.org/LeadingEdge/Phantom4000.pdf.L     www.encompasserve.org/~kaplow_r/    www.nira-rocketry.org    www.nar.org  O  Save Model Rocketry from the HSA!   http://www.space-rockets.com/congress.html    ------------------------------   Date: 25 Jul 2003 19:00:36 GMT/ From: Thomas Dickey <dickey@saltmine.radix.net>IP Subject: Re: Does RT-11 run on the PDP-11/70?  (was Re: PDP-11 OS Release Dates)) Message-ID: <bfruok$c5$1@news1.radix.net>"  / In comp.os.vms paramucho <ian@hammo.com> wrote: 5 > On Fri, 25 Jul 2003 11:22:02 +0200, Christian Corti 4 > <Christian.Corti@studserv.uni-stuttgart.de> wrote:  : >>Bill/Carolyn Pechter <pechter@shell.monmouth.com> wrote:2 >>> You need to sysgen RT11 for the light pattern. >>B >>That would be nice, but I can't find that sysgen option in V5.3.  & > LIGH$t in RMONFB.MAC does the trick:   > .IF NE  LIGH$T  >  .ROM   DEC     LITECT,VALUE=1 >         BNE     8$ >         ADD     #512.,LITECT > 4$:     ROL     7$ >         BNE     5$ >         COM     7$ > 5$:     BCC     6$ >         ADD     #100,4$  >         BIC     #200,4$e  > 6$:     BIT     #LIGHT$,CONFG2 >         BEQ     8$ >         MOV     (PC)+,@(PC)+ > 7$:      .WORD  0,SR > .ENDCl  G more context would be interesting - 8$ is not defined in this fragment.w   -- w= Thomas E. Dickey <dickey@radix.net> <dickey@herndon4.his.com>k http://dickey.his.coma ftp://dickey.his.com   ------------------------------    Date: 25 Jul 2003 16:19:48 -0500; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler)nP Subject: Re: Does RT-11 run on the PDP-11/70?  (was Re: PDP-11 OS Release Dates)3 Message-ID: <6A24Juu2ibdc@eisner.encompasserve.org>n  j In article <X+Y0$$TMCNht@eisner.encompasserve.org>, kaplow_r@encompasserve.org.TRABoD (Bob Kaplow) writes: > L > Did they bounce off each other, or just pass through each other. After all > light has no mass...  F    Photons have no rest mass.  But then, you'll never see one resting.   ------------------------------  # Date: Sat, 26 Jul 2003 00:05:42 GMTr From: ian@hammo.com (paramucho)rP Subject: Re: Does RT-11 run on the PDP-11/70?  (was Re: PDP-11 OS Release Dates)1 Message-ID: <3f23c590.1460229@news.supernews.com>o  F On 25 Jul 2003 19:00:36 GMT, Thomas Dickey <dickey@saltmine.radix.net> wrote:  0 >In comp.os.vms paramucho <ian@hammo.com> wrote:6 >> On Fri, 25 Jul 2003 11:22:02 +0200, Christian Corti5 >> <Christian.Corti@studserv.uni-stuttgart.de> wrote:e >>; >>>Bill/Carolyn Pechter <pechter@shell.monmouth.com> wrote: 3 >>>> You need to sysgen RT11 for the light pattern.h >>>wC >>>That would be nice, but I can't find that sysgen option in V5.3.u >o' >> LIGH$t in RMONFB.MAC does the trick:o >h >> .IF NE  LIGH$Tt! >>  .ROM   DEC     LITECT,VALUE=1i >>         BNE     8$n >>         ADD     #512.,LITECTh >> 4$:     ROL     7$  >>         BNE     5$t >>         COM     7$  >> 5$:     BCC     6$e >>         ADD     #100,4$ >>         BIC     #200,4$! >> 6$:     BIT     #LIGHT$,CONFG2n >>         BEQ     8$i >>         MOV     (PC)+,@(PC)+t >> 7$:      .WORD  0,SRa >> .ENDC >wH >more context would be interesting - 8$ is not defined in this fragment.  D 8$ is an exit label, just below 7$. You'll find the complete context/ in RMONFB.MAC (available on the web somewhere).      -- Ianw" Impressive If Haughty - Q Magazine   ------------------------------  + Date: Sat, 26 Jul 2003 03:37:29 +0000 (UTC)d From: Megan <mbg@TheWorld.com>O Subject: Re: Does RT-11 run on the PDP-11/70? (was Re: PDP-11 OS Release Dates)f( Message-ID: <bfst1p$cgs$1@pcls4.std.com>  9 pechter@shell.monmouth.com (Bill/Carolyn Pechter) writes:k  + >In article <ikhlfb.4u3.ln@news.online.de>,eD >Christian Corti  <Christian.Corti@studserv.uni-stuttgart.de> wrote:H >>And what about 11/45s? At least mine shows a nice pattern with RSX11M.J >>RT11 shows nothing interesting at all, neither does any version of UNIX. >> >>Christianc  / >You need to sysgen RT11 for the light pattern.i  D >It has a pretty slick one. IIRC it had one light on each end movingF >to the center bit -- they touch, then bounce out to the ends and then! >two lights do it and then three.o  J Actually, it starts with all lights on, then shifts one direction until a K bit falls off, then in the other direction until a bit falls off... and so lJ forth until the last bit rotates out of the lights and they all come back 	 on again.     	o o o o o o o o o o o o o o o o  	. o o o o o o o o o o o o o o o  	o o o o o o o o o o o o o o o .  	o o o o o o o o o o o o o o . .  	. o o o o o o o o o o o o o o .  	. . o o o o o o o o o o o o o o  	. . . o o o o o o o o o o o o o  	. . o o o o o o o o o o o o o .  	. o o o o o o o o o o o o o . .  	o o o o o o o o o o o o o . . .  	o o o o o o o o o o o o . . . .  	. o o o o o o o o o o o o . . .  	. . o o o o o o o o o o o o . .  	. . . o o o o o o o o o o o o .  	. . . . o o o o o o o o o o o o  	. . . . . o o o o o o o o o o o  	. . . . o o o o o o o o o o o .  	. . . o o o o o o o o o o o . .  	. . o o o o o o o o o o o . . .  	. o o o o o o o o o o o . . . .  	o o o o o o o o o o o . . . . .  	o o o o o o o o o o . . . . . .  	. o o o o o o o o o o . . . . .  	. . o o o o o o o o o o . . . .  	. . . o o o o o o o o o o . . .  	. . . . o o o o o o o o o o . .  	. . . . . o o o o o o o o o o .  	. . . . . . o o o o o o o o o o  	. . . . . . . o o o o o o o o o  = (I hope you get the idea, my fingers are getting tired... :-)   I The RSX pattern, if I remember correctly, had a set of 8 lights on, four eH in the low byte and four in the high byte, as mirror images.  The bytes H would be individually shift towards the center (high byte shifts right, J low byte shifts left) and bits which shift out of a byte shift back in at  the other end of the byte...    	o o o o . . . . . . . . o o o o  	. o o o o . . . . . . o o o o .  	. . o o o o . . . . o o o o . .  	. . . o o o o . . o o o o . . .  	. . . . o o o o o o o o . . . .  	o . . . . o o o o o o . . . . o  	o o . . . . o o o o . . . . o o  	o o o . . . . o o . . . . o o o   	(rinse and repeat)e  4                                         Megan Gentry>                                         Former RT-11 Developer  H +--------------------------------+-------------------------------------+H | Megan Gentry, EMT/B, PP-ASEL   | email: mbg at world.std.com         |H | Member of Technical Staff      |        megan at savaje.com          |H | SavaJe Technologies, Inc.      |             (s/ at /@/)             |H | 100 Apollo Drive               | URL:     http://world.std.com/~mbg/ |H | Chelmsford, MA 01460           | "pdp-11 programmer - some assembler |H | (978) 256 6521   (DEC '77-'98) |  required." - mbg            KB1FCA |H +--------------------------------+-------------------------------------+   ------------------------------   Date: 26 Jul 2003 04:25:25 GMT2 From: "Zane H. Healy" <healyzh@shell1.aracnet.com>O Subject: Re: Does RT-11 run on the PDP-11/70? (was Re: PDP-11 OS Release Dates)3+ Message-ID: <bfsvrl0pfb@enews1.newsguy.com>o  / In comp.os.rsts Megan <mbg@theworld.com> wrote: " > 	. . . . o o o o o o o o o o . ." > 	. . . . . o o o o o o o o o o ." > 	. . . . . . o o o o o o o o o o" > 	. . . . . . . o o o o o o o o o  ? > (I hope you get the idea, my fingers are getting tired... :-)s  L Unfortunatly I get the idea....  You do realize that you're making me want aJ Unibus PDP-11 with a real front panel even more than I already did, right? :^)t   		Zane   ------------------------------    Date: 25 Jul 2003 12:44:52 -0700( From: bob@instantwhip.com (Bob Ceculski)' Subject: Re: DVD compatibility with VMSd= Message-ID: <d7791aa1.0307251144.5378f615@posting.google.com>t  | "eberhard heuser-hofmann" <vaxinf@chclu.chemie.uni-konstanz.de> wrote in message news:<013101c34de7$d34e1600$0501a8c0@jo>...  = so can I dump my DAT tape drives, create a VMS backup saveseta= then ZIP it then do a backup or copy to a CD-RW drive withoutl< first having to do a 2048 to 512 byte block conversion using< Glen Everhards program, or is the 512 issue only with DVD's?< Can someone here answer this question plainly and precisely?< And if so, is this done using CDRECORD or CDWRITE or both or; straight from VMS ... I am talking about both doing this onl SCSI and IDE ...   ------------------------------    Date: 25 Jul 2003 11:14:11 -0700( From: mhemker@remember.com (Mark Hemker), Subject: Re: Elevate privileges in a program= Message-ID: <df640d0b.0307251014.5c73aad9@posting.google.com>y  h Kilgallen@SpamCop.net (Larry Kilgallen) wrote in message news:<H2QYa+x5nlwY@eisner.encompasserve.org>...e > In article <5752ivskpr05bvsve4iikpnvscjlggbaee@4ax.com>, Mark Hemker <hemker@insightbb.com> writes: I > > I'm working with some of our programmers on a new product and we have,E > > run into a stumbling block with regards to needing privileges for I > > certain system function calls.  What we have is a PASCAL program that>I > > needs to be able to make several function calls where the code in thepF > > functions requires elevated privileges.  We looked at installing aI > > shareable image with these functions, but as you know that won't workvA > > We also looked into writing a User-Written System Service andVF > > discovered that some of system service calls that we need can't be > > used in a UWSS.l > I > You would get a better quality answer by specifying the system servicestH > involved.  There are _many_ ways to introduce vulnerabilities to a VMSK > system trying to do this sort of thing, and a User-Written System Service 1 > is a method of avoiding most of those problems.i >   : I need to use $GETUAI to validate a username and password.  IH > > For this standalone image, we are looking writing it in C so that weI > > can use the SYSTEM() function or one of the EXEC() functions.  We are,I > > planning to use LIB$SPAWN from the PASCAL code to run this new image.aB > > Is there a PASCAL equivalent to SYSTEM() or EXEC() or is there > > something better?w > J > SYSTEM() and  EXEC() are creations of the C Runtime Library and are thus3 > not going to be so stable as VMS system services.o > F > 95% of the time when people try something like this they introduce a > security hole.  C I am trying to avoid creating any security holes and I just haven'tiC had much luck finding a way of accomplishing my needed goal withoutt
 creating one.p  < Thanks for the help.  I am also looking into using protectedF subsystems, but we still run into an issue with the user being able to< run the program from DCL and gaining access to the protected
 directory.  8 Let me explain what the protected directory is used for:D   -Copy a file from a public directory into the protected directory.E   -Run a program that manipulates the file in the protected directoryo and creates a new file.t-   -Delete the input file we just manipulated.vE   -Queue the resulting file to a print queue with /DELETE so that then file is automatically deleted.  B We have a requirement that the files not be accessible to the userF except through our program and that is why we have all of the hoops to
 jump through.x   Thanks,  Mark   ------------------------------  # Date: Fri, 25 Jul 2003 19:03:48 GMT # From: hoff@hp.nospam (Hoff Hoffman)e, Subject: Re: Elevate privileges in a program0 Message-ID: <oafUa.912$5q5.485@news.cpqcorp.net>  h In article <df640d0b.0307251014.5c73aad9@posting.google.com>, mhemker@remember.com (Mark Hemker) writes:i :Kilgallen@SpamCop.net (Larry Kilgallen) wrote in message news:<H2QYa+x5nlwY@eisner.encompasserve.org>...M  J :> You would get a better quality answer by specifying the system servicesI :> involved.  There are _many_ ways to introduce vulnerabilities to a VMSsL :> system trying to do this sort of thing, and a User-Written System Service2 :> is a method of avoiding most of those problems. :> :s; :I need to use $GETUAI to validate a username and password.3  	   ... ...0  D :I am trying to avoid creating any security holes and I just haven'tD :had much luck finding a way of accomplishing my needed goal without :creating one.  E   You are correct; I expect you have implemented a security hole with E   your $getuai code.  Why are you checking the username and password? E   (If you want to be part of the Trusted Computing Base (TCB) as this E   password verification implies -- of the operating system kernel andoF   the privileged applications -- then you are a target, and performing<   a password check certainly makes you a target.  Big-time.)  ' :I am also looking into using protectedsG :subsystems, but we still run into an issue with the user being able to = :run the program from DCL and gaining access to the protectedj :directory.s  G   What (specific) issue(s)?  If the image has the subsystem identifier,tH   it has the access -- or the denial, since identifiers can also be used1   to negate access -- provided by the identifier.T  9 :Let me explain what the protected directory is used for:gE :  -Copy a file from a public directory into the protected directory.l  <   I assume this is a case where you want the privileged code   to pull the code over.  F :  -Run a program that manipulates the file in the protected directory :and creates a new file.  C   Okfine.  DECnet task-to-task can provide this capability easily. fG   Subsystem identifiers can work, too.  Realize that if you are runningkF   code provided by the user from a privileged context, that's, well, a   large security hole.  . :  -Delete the input file we just manipulated.F :  -Queue the resulting file to a print queue with /DELETE so that the :file is automatically deleted.t  6   Usually via a $sndjbc[w] call, if this is a program.  H   Speaking of queues, that's another solution here -- a custom symbiont.F   Though you still need to be exceedingly careful not to run untrustedI   code from a privileged context, symbiont or otherwise.  With the customaE   symbiont, the user would submit the file to a server queue, and thej!   symbiont would then process it.   C :We have a requirement that the files not be accessible to the user>G :except through our program and that is why we have all of the hoops to  :jump through.  F   Callable convert from an installed image is another option; callable5   convert is effectively a callable COPY command API.S   	--c  H   I would think you will want to describe the problem and its backgroundK   in rather more detail, rather than proposing a solution first and workingrI   backwards from that information.  (What I've read so far, I don't fullyaH   understand.  For instance, what does the privileged pre-processing do?J   What is the user's invocation context of the resulting executable image?L   Privileged?  Unprivileged?  This whole processing sequence looks unusual.)    N  ---------------------------- #include <rtfaq.h> -----------------------------K     For additional, please see the OpenVMS FAQ -- www.hp.com/go/openvms/faqeN  --------------------------- pure personal opinion ---------------------------E         Hoff (Stephen) Hoffman   OpenVMS Engineering   hoff[at]hp.como   ------------------------------  % Date: Fri, 25 Jul 2003 15:36:12 -0400r* From: JF Mezei <jfmezei.spamnot@istop.com>, Subject: Re: Elevate privileges in a program) Message-ID: <3F2186A1.80675FFE@istop.com>r   Mark Hemker wrote:G >   -Queue the resulting file to a print queue with /DELETE so that the   > file is automatically deleted.  N Do print symbionts have the equivalent of SYSPRV to access the fille for print and subsequent delete ?.  N Be careful. If User A submits a print job, you have to make sure that the fileD being submitted isn't owned by user A (so that he can't go do a SHOWL ENTRY/FULL and then copy the file), on the other hand, you also need to makeM sure that if the file is created under owner "B" but submitted by A, that theaT print symbiont will accept to print that file if A doesn't have access to that file.   ------------------------------  % Date: Fri, 25 Jul 2003 15:50:59 -0400t* From: JF Mezei <jfmezei.spamnot@istop.com>, Subject: Re: Elevate privileges in a program) Message-ID: <3F218A16.E537056F@istop.com>e   Hoff Hoffman didn't write:= > :I need to use $GETUAI to validate a username and password.p  K Gee Hoff, I would have though that by now, you'd know how to do that ?  :-)i :-) :-) :-) :-) :-)t   From the manual:   Required Access or Privileges   N Use the following list to determine the privileges required to use the $GETUAI	 service: t  M BYPASS or SYSPRV---Allows access to any record in the user authorization filed (UAF).  L GRPPRV---Allows access to any record in the UAF whose UIC group matches that of the requester. L No privilege---Allows access to any UAF record whose UIC matches that of the requester. mJ You need read access to the UAF to look up any information other than your own.      L So, if you have an application which runs under usernam,e John Doe and wantsP to verify John Doe's password, then the application doesn't need ant privileges.  N If you have an application which wants to allow John Doe to check Jane Smith'sI password, then you also need to look into the routines that will generate % audit alarms and intrusion detection..  H Aren't there newer ACME* routines that allow that ? (although they would1 probably be unusable if available only on Alpha).c   ------------------------------    Date: 25 Jul 2003 18:05:58 -0500- From: Kilgallen@SpamCop.net (Larry Kilgallen)o, Subject: Re: Elevate privileges in a program3 Message-ID: <le97qh2c+dbd@eisner.encompasserve.org>i  h In article <df640d0b.0307251014.5c73aad9@posting.google.com>, mhemker@remember.com (Mark Hemker) writes:j > Kilgallen@SpamCop.net (Larry Kilgallen) wrote in message news:<H2QYa+x5nlwY@eisner.encompasserve.org>...f >> In article <5752ivskpr05bvsve4iikpnvscjlggbaee@4ax.com>, Mark Hemker <hemker@insightbb.com> writes:J >> > I'm working with some of our programmers on a new product and we haveF >> > run into a stumbling block with regards to needing privileges forJ >> > certain system function calls.  What we have is a PASCAL program thatJ >> > needs to be able to make several function calls where the code in theG >> > functions requires elevated privileges.  We looked at installing alJ >> > shareable image with these functions, but as you know that won't workB >> > We also looked into writing a User-Written System Service andG >> > discovered that some of system service calls that we need can't beu >> > used in a UWSS. >> aJ >> You would get a better quality answer by specifying the system servicesI >> involved.  There are _many_ ways to introduce vulnerabilities to a VMS L >> system trying to do this sort of thing, and a User-Written System Service2 >> is a method of avoiding most of those problems. >> > < > I need to use $GETUAI to validate a username and password.  H If you are doing this on Alpha, use the $ACM system service under V7.3-1C or greater to do this safely.  I believe it requires no privileges.c  G >> 95% of the time when people try something like this they introduce aJ >> security hole.1 > E > I am trying to avoid creating any security holes and I just haven'tcE > had much luck finding a way of accomplishing my needed goal withoutp > creating one.u  : $ACM was provided because of the difficulty of doing this.  > > Thanks for the help.  I am also looking into using protectedH > subsystems, but we still run into an issue with the user being able to> > run the program from DCL and gaining access to the protected > directory. > : > Let me explain what the protected directory is used for:F >   -Copy a file from a public directory into the protected directory.G >   -Run a program that manipulates the file in the protected directoryt > and creates a new file.i/ >   -Delete the input file we just manipulated.vG >   -Queue the resulting file to a print queue with /DELETE so that thet  > file is automatically deleted. > D > We have a requirement that the files not be accessible to the userH > except through our program and that is why we have all of the hoops to > jump through.o  I "Protected Subsystems" is the specific nature of a particular VMS featuresH on both VAX and Alpha.  It will allow you to do what you described thereJ just fine, but it would seem you need to read and understand documentation  related to protected subsystems.  F But what you describe there has nothing at all to do with validating aB username and password.  Do you really have two problems to solve ?   ------------------------------  % Date: Fri, 25 Jul 2003 22:36:43 -0500 1 From: "David J. Dachtera" <djesys.nospam@fsi.net>t, Subject: Re: Elevate privileges in a program' Message-ID: <3F21F74B.C502264A@fsi.net>    Mark Hemker wrote: > [snip]< > I need to use $GETUAI to validate a username and password.   Hasn't VMS already done that?d   > [snip]: > Let me explain what the protected directory is used for:F >   -Copy a file from a public directory into the protected directory.G >   -Run a program that manipulates the file in the protected directory0 > and creates a new file.l/ >   -Delete the input file we just manipulated.u  F The COPY, then, is redundant. Can't the program open the original file1 in place ACCESS READ, ALLOW MODIFY (equiv. to DCLe OPEN/READ/SHARE=WRITE)?a  G >   -Queue the resulting file to a print queue with /DELETE so that theV  > file is automatically deleted.  E There's another way (one of many, I'm sure) to do that: set the printhE device of the target queue /SPOOLED and open the device directly. ThedG system will "spool" the output to a file with no directory entry (whichfE I believe may be owned by the system, but not sure just now and don't F want to take the time at this hour to set up a test). Upon the programD closing the file, the file will be entered into the queue associatedH with the spooled device. A file with no directly entry introduces a goodB number of "hoops to jump through" to find it, much less attempt to access it by FID.y  D > We have a requirement that the files not be accessible to the userH > except through our program and that is why we have all of the hoops to > jump through.d  G INSTALLing the program with privilege is not acceptable? ...even from a  CAPTIVE account?  F Not sure if that would fit your security model, but it would sure make* the whole task a lot easier to accomplish.   -- e David J. Dachteram dba DJE Systemse http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/f   ------------------------------    Date: 25 Jul 2003 13:18:34 -0500+ From: young_r@encompasserve.org (Rob Young)  Subject: Re: EMC on VMSe3 Message-ID: <vuuW7lO1ZD+5@eisner.encompasserve.org>t  [ In article <3F2164E6.91ECB7DB@fsi.net>, "David J. Dachtera" <djesys.nospam@fsi.net> writes:t > Rob Young wrote: >>  ^ >> In article <3F213E76.F2D8C738@fsi.net>, "David J. Dachtera" <djesys.nospam@fsi.net> writes: >>  H >> >>         So to cry and whine about READL/WRITEL non-support is okayI >> >>         I suppose.  But if it never comes into play, who cares?  Iti& >> >>         is a Zen thing isn't it? >> >M >> > No, it's a data integrity thing - which is what HBVS is (supposed to be)  >> > all about, no?r >> > >> iG >>         No.   Data integrity is the responsibility of the underlyingf5 >>         storage.  HBVS is all about availability. i > G > Try again. The two are not mutually exclusive, and HBVS includes dataiH > integrity features - like dropping a member when a bad block cannot be > flagged as such. >   : 	Right.  For storage that returns bad blocks or can't read= 	a block, a great necessity!  But HBVS is about availability. ? 	You pay a pretty penny for the license.  Not everyone needs tomB 	go beyond hardware RAID and spend "foolishly" on an HBVS license,7 	unless of course you are concerned about availability.h     >> If you write bad blocksJ >>         to an HBVS volume, you have bad blocks.  If you read bad blocksG >>         from an HBVS volume (whether you wrote them, or they somehowe3 >>         became bad), that is not a happy moment.b > D > ...which is why HBVS drops members when it can't setthe approriate6 > forced error flags. Starting to get the picture now? >   > 	Yes.  For storage systems that can't.  If blocks are readable) 	and have the data you wrote, why bother?e  F >>         Numerous examples of HBVS RAS.  I have mine.  The 911 folksE >>         have theirs (Commerze Bank).  Ed Wilts also comes to mind:n >> dh >> http://groups.google.com/groups?selm=UHrEbs%24k15p0%40eisner.encompasserve.org&oe=UTF-8&output=gplain >> d
 >> Ed writes:t >> tI >> Absolutely.  You simply can not trust mirrored controllers to maintaine; >> the data as much as you can host-based volume shadowing.o >  > Seems to contradict you. >    	No.  A better read is this:  G "Absolutely.  You simply can not trust mirrored controllers to maintaina?  the [storage] as much as you can host-based volume shadowing."b  @ 	Check the context of the entire secion of that discussion.  ItsC 	about members going away, not corrupt storage.  Members going awaym@ 	because controllers hang.  A reboot of a controller returns the? 	storage.  Data isn't corrupted/bad, etc., etc.  HBVS increases  	your availability.-  E 	Where we disagree is where READL/WRITEL comes into play.  Obviously CG 	need that for an MSA1000.  You don't need it for EMC Symmetrix as the i9 	Symmetrix guards against bad data in several directions.v  E >> I have personally seen a case where a single bad drive caused BOTHrJ >> controllers in a dual-redundant configuration to crash.  The drive hungF >> the SCSI bus.  When the first controller detected this, it tried toH >> reset the bus.  When it wouldn't reset, it assumed the controller wasJ >> culprit and crashed itself, causing the drive to fail over to the otherB >> controller, which promptly crashed, until I finally intervened. > H > That's a firmware problem, not an HBVS issue. HBVS could have been the > savior there.i >   > 	Assuming members were in separate cabinets, yes.  He may haveB 	had members in the same cabinet, or later made sure they weren't.  J >> I've also seen a case where a controller pair disagreed on the state ofH >> a 5 member RAID-5 set.  The entire set had to be restored from backup >> tapes (36GB worth of data). >> 2K >> I am currently using RAID-5 with redundant controllers within a cabinet,aK >> and host-based shadow that to a RAID-5 set with redundant controllers insI >> another cabinet.  In about 2 weeks, that other cabinet will be located + >> in another building.  I value my data...a > D > ...as do I, which is why I won't implement a solution that OpenVMS% > Engineering and HBVS don't support.. >   B 	HBVS doesn't support it?  Of course it does, and it works well.    < 	Doesn't meet OpenVMS engineering best practices?  You would= 	have a big problem if READL/WRITEL isn't supported on run ofo0 	the mill storage.  That FE is highly necessary.  F 	Guess what, there is a corporation that is probably in the top 30 in > 	the world in revenues (an international multi-billion dollar D 	corporation) that uses HBVS and Symmetrix, billions riding on it.  F 	I'm sure their confidence level is just as high as mine.  After all, ( 	shadowing Symmetrix volumes works well.   	"No READL/WRITEL!"y  1 	Who cares!  No bad blocks, no unreadable blocks!    > End of discussion.   				Robr   ------------------------------   Date: 25 Jul 2003 21:46:16 GMT3 From: gartmann@immunbio.mpg.de (Christoph Gartmann)uD Subject: Re: How to operate MSL5000 tape library from OpenVMS prompt0 Message-ID: <bfs8f8$klu$1@n.ruf.uni-freiburg.de>  s In article <bfr0h2$c55$1@SunSITE.icm.edu.pl>, "Micha von Grotthuss" <xxxgrotthussnospam@polantispamlot.pl> writes:s5 >I have two DS25 machines set up together in cluster.sG >Machines are connected to storagework SAN swicth 8-EL with fiber optic3 >cable.nL >Storagework SAN switch is connected to hp network storage router m2404 with >fiber optic cable.iJ >To this storage router is also connected with a SCSI cable a MSL5000 tape  >library with two DLT streamers. >My questions are: >aH >1. What are the name of DLT streamers (normal SCSI streamer name is for >example MKC200) displayed by. >$ show device	 >command.  >rG >2. Is it possible to operate tape library from OpenVMS command prompt.l/ >Are there any commands or additional software. , >I need commands performing operations like:  >- insert tape #4 in streamer #0! >- initialize tape in streamer #0 # >- use cleaning tape in streamer #1   K To control the robot you'll need a program. I wrote one to just do what you M want: control one or more drives and a robot via commandline. It is free, let, me know if you are interested.   Regards,    Christoph GartmannW   -- gE  Max-Planck-Institut fuer      Phone   : +49-761-5108-464   Fax: -452c  ImmunbiologieA  Postfach 1169                 Internet: gartmann@immunbio.mpg.de1  D-79011  Freiburg, Germanyo9                http://www.immunbio.mpg.de/home/menue.htmlt   ------------------------------  # Date: Fri, 25 Jul 2003 21:34:01 GMTa& From: jlsue <jefflsxxxz@sbcglobal.net>4 Subject: Re: HP World: Why Alpha's Omega Makes Sense8 Message-ID: <v273ivkljg566032fhvc973b7634mtga54@4ax.com>  H On Wed, 23 Jul 2003 13:56:02 -0400, JF Mezei <jfmezei.spamnot@istop.com> wrote:  
 >jlsue wrote:pJ >> it wouldn't be as nice as EV8, I'm confident that these people can (andJ >> will) create a product that will have great capabilities and longevity. >dM >People said that of IA64 5 years ago. People said that of Intel 2 years ago.d< >At one point, one can no longer keep talking in the future.  I Well, since we're just beginning to see real systems, I'll still wait andaK see.  I don't think it's all that long to wait.  The first release at whichf; OpenVMS full-release is aimed sounds to be a good platform.M   >hN >The impression I have is that EV7 was purposefully slowed down to quicken itsL >death. Not a smart move when you continue to talk about IA64 in the future.  @ I have no way to validate this "impression".  Without a bit more= substantial verification, it's just FUD from our competition.i   > J >> Not sure what you mean by "dismiss", but there are just no server-classG >> systems vendors - today - that I know of making plans for this chip.i >DM >Please remember Winkler's famous "windows will eat Unix's underbelly" words.dO >With this in mind, consider that the wintel market is slowly eroding sales andmM >margins from enterprise class systems.  The 64 bit 8086 from AMD isn't goingdJ >after IA64's throath, it is going after Intel's 8086 product line. If youK >reduce Intel's 8086 profits, where will the money come from to continue toa/ >subsidize the development of that IA64 thing ?b  J If/when this happens, we'll see.  At this point I don't see it as anythingK more than a possibility.  You can't expect people to discuss plans for somei3 eventuality that may just be somebody's pipe dream.p   > M >> Why would you expect it to be profitable yet?  No business-critical servergL >> market has been realized yet - it's just too early.  IIRC, the Alpha chipF >> was EV4 before most of us ever saw it in real business-use systems. > N >Did Digital spend 10 years spending megabucks before first commercial systemsN >became available on Alpha ? It seems to me that Alpha's design costed Digital> >peanuts compared to what Intel spent on the IA64 chip design.  J I don't know if it was 10 years, but it was quite a long time, with delays an everything.   ------------------------------  # Date: Fri, 25 Jul 2003 21:45:44 GMT-& From: jlsue <jefflsxxxz@sbcglobal.net>4 Subject: Re: HP World: Why Alpha's Omega Makes Sense8 Message-ID: <dk83iv8rjqnmehp3lct5pvap4mabsq7b6c@4ax.com>  B On 23 Jul 2003 20:43 CDT, carl@gerg.tamu.edu (Carl Perkins) wrote:  # >jefflsxxxz@sbcglobal.net writes...-L >}Why would you expect it to be profitable yet?  No business-critical serverK >}market has been realized yet - it's just too early.  IIRC, the Alpha chip(K >}was EV4 before most of us ever saw it in real business-use systems.  Whatr- >}is the current state of IA64?  EV2?  EV2.5?o >.; >The EVx designation does not mean what you think it means.s  C Actually, you're right, and I really did know that.  I was taking ae= short-cut in the discussion to make point.  Sorry 'bout that.t  H >There was no "EV3" or earlier except as purely developmental stuff. TheH >"EV4" was not only the first version that was available in systems, I'mI >pretty sure it was the first version that was *supposed* to be available8 >in systems.  K Yes, yes.  I apologize again for the short-cut.  I was thinking in terms of:K "chip generations" and using the term that seemed comparable to the current. discussion of IA64 generation.   >rF >Thus, in some loose sense, the current IA64 is equivalent to the EV56I >(21164A) - the shrunk version of the second generation of the processor.JI >(In both cases, ignoring all developmental test chips produced earlier.)jH >Of course it doesn't match exactly - the 21164 had pretty much the sameL >core as the 21064, but the Itanium 2 has a core that has had a considerableK >amount of redesign work done on it. You could say they skipped the EV5 and J >went directly to the EV6 by doing extesnsive redesign of the core insteadJ >of just doing a few tweaks and shrinking the thing - in that case the new: >version would be roughly equivalent to the EV67 (21264A).  E I'm not sure I agree with this progression.  Imo, the current IA64 is H really closer to the pre-21064/EV4 chip - i.e., the first one available,! albeit to only developers really.   I It's not a perfectly aligned comparison between these, but the release of G IA64 which is slated for wide-availability is, to me, somewhere between C 21064 and 21164.  Somethings are at or better than 21164, but otherC  features may be closer to 21064.  J In this sense, I believe that it's too early to tell what will happen with
 this chip.   ------------------------------  # Date: Fri, 25 Jul 2003 22:30:11 GMTe& From: jlsue <jefflsxxxz@sbcglobal.net>Y Subject: Re: Intel's had Dirty Bombs Anthrax and everything! (Was: Re: HP World: Why Alph.8 Message-ID: <cbb3ivg2tfbquv39ustoo7oojqg55sel4t@4ax.com>  H On Wed, 23 Jul 2003 18:27:52 -0400, "Bill Todd" <billtodd@metrocast.net> wrote: >n4 >"jlsue" <jefflsxxxz@sbcglobal.net> wrote in message3 >news:n6fthvkgjj86fl5iuac20dqqab9m1tk3he@4ax.com...AK >> On Fri, 18 Jul 2003 18:10:44 -0400, "Bill Todd" <billtodd@metrocast.net>t	 >> wrote:  >> >> >7 >> >"jlsue" <jefflsxxxz@sbcglobal.net> wrote in message 6 >> >news:h5aghvkavsukjcto77hkv3qsd1ciu3g7le@4ax.com... >> ry. >> >> >- >> >> < >> >> Right.  Somehow you feel they owe you an apology.(???) >> >J >> >No, I feel they owe those who made decisions based upon their specific >and2 >> >unequivocal 'commitments to Alpha' an apology. >>6 >> You can only claim this "requirement" for yourself. >mC >What part of "I feel that..." do you find difficult to understand?r  K The part where you start saying the rest of "those who made decisions basedtI upon ...".  That's the part where  you transfer your own unhappiness ontos the rest of the world.  K I'm working with many, many customers, and though nobody was entirely happye% about this, almost all have moved on.i   >  >  You don't get to H >> decide whether the way they did it was satisfactory for most of their
 >> customers.i >>L >But I *do* get to decide whether I think the behavior is worth making clearJ >to people who might be interested in or affected by it - so that they can >decide for themselves.   H Making predictions about supposed future behavior is nothing but FUD, no  matter how you like to couch it.   >PA >  Again, in the end, if you (or any other specific customer) areeK >> not satisfied, you are fully able to vote with your feet.  At that pointsL >> you have no other reason to continue this ranting, though, because you've- >> made your decision on how to deal with it.a >tL >Au contraire:  that would be a very self-centered attitude to take (sort ofH >like driving by an accident because it didn't affect me personally).  IL >don't know about you, but when I see something dramatically wrong I'm oftenC >motivated to try to correct it even if it doesn't really affect mecA >personally - and this *is* my 'decision on how to deal with it'.r  G Ah, so now you're saving the world from DEC/CPQ/HPQ?  I know that's nottJ what you're really saying, but you're just trying to rationalize continued0 harrassment based on some past dissapointment.    K It would be different if you stated your case, and then took a wait-and-seenJ attitude.  And if you're proved to be right, I'd even give you the "I told you so".  K It's the constant pounding, and using that to drive home that nobody can be J trusted anymore, so why try, that makes the effort so much waste.  This isI just my opinion, I'm not trying to brow-beat you with it.  But trust thatfH there are many, many people working to improve on the past.  Why detract; from their efforts before you see if they'll be successful?o     >>M >> Yeah, right.  They decide that the direction the company should go to meetoH >> business goals, and they incur the risks to take it there.  Hey, it's >theirK >> responsibility to keep the company profitable, if they don't do it, it's 8 >> their job on the line.  Armchair refs need not apply. >.K >I really don't have any problem with that characterization, if it will get.M >people to stop complaining about my efforts to make sure that the risks theyoI >took by throwing away a business with immense potential profit, breakingoE >commitments in the process, and then lying about it have appropriate.E >consequences.  Treachery, perfidy, and/or incompetence *should* have.G >consequences, after all:  that's the free-market philosophy in action..  B First of all, your point is based on lots of conjecture, one-sidedJ information, and a complete rejection of all counter arguments.  With suchD closed-mindedness, how do you really think this activity is helping?  F And the "consequences" that would befall them in this free-market is aK significant loss of customers.  That hasn't happened, so maybe, just maybe,r there's more to the story.   >>G >> There's a world of difference between saying "I don't believe in 'X' M >> because of past behavior", and ranting constantly, in excruciating detail,b+ >> that your beliefs and opinions are fact.o >oL >Ah, but they're mostly *not* opinions:  I document them carefully.  And youK >might note that they don't sprout up at random:  they're mostly offered in.M >rebuttal to some whack-a-mole who pops up thinking it may finally be safe tooI >start spewing the party-line lies again (or who starts reassuring people 3 >about cHumPaq's 'commitments' to VMS's future...).  >   I I commed you on your research and analysis.  Really I do.  However I just-H don't understand how you really think the public numbers really tell theJ whole story.  And even beyond that, take it to the extremes that you do in these conversations.  . In this scenario, it can never get any better.   ------------------------------  # Date: Fri, 25 Jul 2003 22:47:29 GMTU& From: jlsue <jefflsxxxz@sbcglobal.net>Y Subject: Re: Intel's had Dirty Bombs Anthrax and everything! (Was: Re: HP World: Why Alph(8 Message-ID: <oqc3ivg2a4ap3fpjljagq6c6rj1l8snshv@4ax.com>  F On 25 Jul 2003 06:40:57 GMT, "Dave Weatherall" <djweath@attglobal.net> wrote:   >> -M >> Yeah, right.  They decide that the direction the company should go to meetsN >> business goals, and they incur the risks to take it there.  Hey, it's theirK >> responsibility to keep the company profitable, if they don't do it, it's.8 >> their job on the line.  Armchair refs need not apply. >oD >Ok, then just accept Bill and the other people who've feel they've G >been let down, or other, stronger, emotions, as one of the risks. They 4 >will complain and seek to document their complaint.  K And they continue to complain, and complain, and complain, as if it's goings to change the past.r  H If you want to effect a positive change, that's not going to do it.  AndF besides, who in this forum is in any position to do anything about it.  G If I have to accept that they will complain about it, then they have togG accept that I'll continue to complain about their complaints... ;-) ;-)s   ------------------------------  # Date: Fri, 25 Jul 2003 23:15:19 GMT0& From: jlsue <jefflsxxxz@sbcglobal.net>Y Subject: Re: Intel's had Dirty Bombs Anthrax and everything! (Was: Re: HP World: Why Alphp8 Message-ID: <h7d3iv0dd82odeq3u9lplj4dv0pisdgbrq@4ax.com>  F On Tue, 22 Jul 2003 12:48:48 -0400, David Froble <davef@tsoft-inc.com> wrote:   >yP >Bill was upset about the end of Alpha, but it wasn't his product and his place M >to be more than upset.  However, the spin that Compaq attempted to place on  L >their decision was full of more holes than Swiss cheese.  Bill is entirely L >within his right to object to lying to customers and the public, and doing M >nothing more than exposing how ridiculous the story was, and the lies being .L >told.  Hey, if you don't want to see public reaction, don't make something R >public.  Compaq made public statements, and public comment on such statements is K >allowable, regardless of how much Compaq doesn't appriciate such comment. c- >Bill's story hasn't changes much in 2 years.p  K His story doesn't have to change.  That's no longer the point, imo.  He haseH every right feel the way he does and to state so.  But to color each andF every future comment/announcement based on something that happened twoI years ago is just not productive.  And it's just not accurate to continuev% to use that same out-dated yardstick.    >a >Some of the holes:f >tL >Some HP employees now publically embrace the questionable story put out by P >Compaq.  There may be more self interest for them doing so than the actions of M >others.  While I can understand such self interest, I can also recognize it.a  I Some of us don't embrace or non-embrace it.  Some (many?  most?) are justtF working to make a success of that busines in which we're living.  What3 might have been really just doesn't matter anymore.t   >fO >Compaq used statements from those who may have had differences with the Alpha rM >CPU developers to support it's story, but not one word from those Alpha CPU tQ >developers, the very ones who you'd expect to have the most relavent statements  M >to make.  Now there were some rumors of off-the-record statements from some aR >Alpha CPU developers, it's understandable if they fear to dispute the 'official' Q >statements of their employer, who can exert enormous financial pressure on such  O >employees.  I'd think that if the Alpha CPU developers agreed with the story, t+ >there would have not been any such rumors.i  H I'm sure there were also many, many VAX CPU developers who felt the sameI about Alpha.  There was a huge possibility that the VAX design could haveeK been improved significantly more than it was.  But once the company decided F Alpha was the future, VAX CPU improvements all but dried up (well, notD entirely, but long-term future work was stopped).  The last VAX chipH systems were vastly faster than the EV4x chips systems.  My VAX 7800 was? way faster than the DECsystem 7000 (or whatever that name was).c   >iP >Now these people who declared IA-64 surperior to Alpha, and on who's judgement Q >Compaq made a major business decision, they'll have a significant place in HP's tR >future plans I'd assume?  Can't lose such prestigeous, knowledgable, and capable  >folks, can they?   J Look, superior is judged on many levels.  More than architecture elegance,K and speed.  The desire for the business to continue investing in the entiresJ internal infrastructure to support the Alpha, and including the real majorK investment they'd have to make in marketting to "catch up" in market share, H may have just been outweighed by the easier route of using the IA64, andK tagging-along on the marketing of that large company.  And it would have toaK last some number of years for the Alpha investment to recover market share, F and begin paying back and making it worthwhile.  Many many people wereI calling it a dying chip already - even our most infamous Sun contributor.   A Taking all that in (and it's not the complete story), include theiJ Micro-Econ 101 concept of opportunity cost, and a past of least resistanceJ makes a lot of sense.  When it's said that IA64 is superior, or that AlphaD wouldn't keep up, that must be taken into account with the amount ofK investment the BOD was willing to make.  NOBODY in here knows what that is.y  F I don't believe for a minute that they are all so incompetent in their@ responsibility as a board member that this was never part of it.   >iQ >The latest Alpha, EV7, seems unable to keep up with IBM's Power series of CPUs,  O >in contrast to the past where it would match and many times surpassed the IBM oR >product.  Since there is similar concepts in both products, it's hard to imagine P >that such lag is because of the concepts in the Alpha design.  The most likely O >reason then is lack of effort to make EV7 the best it can be.  Since Alpha is oO >still being sold to customers, and since buyers of EV7 are some times looking eN >for maximum performance, it brings into question HP's honesty and commitment  >with such customers.i  H This is somewhat conjecture.  I have no idea how well it's keeping up as# it's not important to me anymore.  b   > K >Recently there have been statements about whether it would have been cost  Q >effective to continue with Alpha.  To me this is a statement that Compaq wasn't tR >interested in competing with it's competitors, and instead was more than willing O >to drop out of the competition.  I personally find this to be contrary to the nO >concept of running a business.  In the high tech field, you run a business to  O >provide the 'better mousetrap' in order to out-sale the competition.  If this pQ >was the case, why wouldn't Compaq just announce that they weren't interested in lP >developing CPUs?  Answer, because then they wouldn't appear to be competitive, 6 >and sales would go to those still willing to compete.  K There's nothing contrary to running a business when you decide that there's C a particular aspect of that business in which you no longer want toiH compete.  There are many different businesses, and chip-making isn't theK only one.  And Cpq wasn't the first company to take that tack - Sun doesn'taJ do their own chips, Tandem didn't (Mips based), there are others.  If theyI decided they wanted to concentrate our investments where we felt we addedo8 more value and distinction, that's perfectly reasonable.  G And I'm really kind of surprised by some of this business analysis thatcF contends that Alpha was such a winner for the company.  There were/areJ folks inside who called Alpha the sink-hole that never made a dime for theK company.  some of these people are very loud in extolling their opinions onrI this topic.  I wished they showed up in c.o.v, becasue you'd be surprisednG at how vitriolic THEY get - they make Bill Todd look positively docile.s     >w > L >Instead of trying to make fun of Bill, could his detractors just show some O >evidence of him being wrong?  Just address the issues?  Maybe name a customer oQ >that was promised that Alpha NT was the way to go, that was promised that Alpha .8 >had a future, who will trust the commitments of Compaq? >s  J I think there have been ample explanations of where he *may* be wrong.  OrJ where he just may be only seeing part of the story.  There is no evidence,C just as there is no real evidence (external) of how the business is- decided.  J They gave the answers they wanted to give.  They aren't willing to provideI more, and frankly, after two years, I don't see how you'd expect them to.AG There were many Alpha detractors inside who felt just a vehemently thatl( Alpha was bad business.  So who's right?   ------------------------------  % Date: Sat, 26 Jul 2003 01:25:50 -0400e* From: "Bill Todd" <billtodd@metrocast.net>Y Subject: Re: Intel's had Dirty Bombs Anthrax and everything! (Was: Re: HP World: Why Alphe2 Message-ID: <OpucnZsv2Ixxjb-iXTWJiQ@metrocast.net>  3 "jlsue" <jefflsxxxz@sbcglobal.net> wrote in messageo2 news:cbb3ivg2tfbquv39ustoo7oojqg55sel4t@4ax.com...J > On Wed, 23 Jul 2003 18:27:52 -0400, "Bill Todd" <billtodd@metrocast.net> > wrote: > >y6 > >"jlsue" <jefflsxxxz@sbcglobal.net> wrote in message5 > >news:n6fthvkgjj86fl5iuac20dqqab9m1tk3he@4ax.com... 4 > >> On Fri, 18 Jul 2003 18:10:44 -0400, "Bill Todd" <billtodd@metrocast.net> > >> wrote:: > >> > >> >9 > >> >"jlsue" <jefflsxxxz@sbcglobal.net> wrote in messagep8 > >> >news:h5aghvkavsukjcto77hkv3qsd1ciu3g7le@4ax.com... > >> ry.	 > >> >> >e > >> >> > > >> >> Right.  Somehow you feel they owe you an apology.(???) > >> >L > >> >No, I feel they owe those who made decisions based upon their specific > >and4 > >> >unequivocal 'commitments to Alpha' an apology. > >>8 > >> You can only claim this "requirement" for yourself. > >-E > >What part of "I feel that..." do you find difficult to understand?  >PG > The part where you start saying the rest of "those who made decisions  basedrK > upon ...".  That's the part where  you transfer your own unhappiness ontot > the rest of the world.  J Not at all:  I'm happy to let the rest of the world speak for itself.  AndL enough of them have voiced their displeasure to make me entirely comfortable2 with my assertion that an apology is long overdue.   >tG > I'm working with many, many customers, and though nobody was entirelys happyo' > about this, almost all have moved on.d  G Many to other vendors - though you're probably not 'working with' thosel particular people any more..   >r > >t > >  You don't get toeJ > >> decide whether the way they did it was satisfactory for most of their > >> customers.  > >-H > >But I *do* get to decide whether I think the behavior is worth making clearpL > >to people who might be interested in or affected by it - so that they can > >decide for themselves.n >eJ > Making predictions about supposed future behavior is nothing but FUD, no" > matter how you like to couch it.  E 'Fear, uncertainty, and doubt' is *precisely* the attitude with whicheF cHumPaq should be approached.  If you don't find Santayana's sentimentK persuasive, how about the more colloquial "Fool me once, shame on you; foole9 me twice, shame on me" proverb that David just mentioned?p  F Claiming that past behavior should not be considered in dealing with aK company is just plain stupid.  Such a claim coming from an employee of thatA: company might be considered something even less admirable.   >l > >eC > >  Again, in the end, if you (or any other specific customer) arelG > >> not satisfied, you are fully able to vote with your feet.  At that  pointwG > >> you have no other reason to continue this ranting, though, becauseS you've/ > >> made your decision on how to deal with it.a > >hK > >Au contraire:  that would be a very self-centered attitude to take (sortu ofJ > >like driving by an accident because it didn't affect me personally).  IH > >don't know about you, but when I see something dramatically wrong I'm oftenhE > >motivated to try to correct it even if it doesn't really affect me C > >personally - and this *is* my 'decision on how to deal with it'.r >eI > Ah, so now you're saving the world from DEC/CPQ/HPQ?  I know that's notc > what you're really saying,  K That is *precisely* what I'm saying - both because I think it's considerateeG to make sure newcomers are warned what they might expect, and because I G believe that some kinds of behavior should not be forgiven or forgotten H until some reasonable measure of restitution and atonement has occurred.  0  but you're just trying to rationalize continued0 > harrassment based on some past dissapointment. > @ > It would be different if you stated your case, and then took a wait-and-see > attitude.s  I 'Wait and see' and 'hope for the best' were the hallmarks of the approachtK that led to disappointment after disappointment.  Some people just lie downoI and become accustomed to that kind of treatment (see 'Stockholm syndrome' L for an explanation of why some actually may even start defending it); othersI eventually get sufficiently fed up that they take a more activist stance.   A   And if you're proved to be right, I'd even give you the "I toldt
 > you so".  L I don't choose to wait around to be proved right yet again:  I choose to try to force a different outcome.u   >rJ > It's the constant pounding, and using that to drive home that nobody can be > trusted anymore, so why try,  @ I've never claimed that *nobody* could be trusted, just cHumPaq.  & > that makes the effort so much waste.  I If *I* thought it were a waste, I wouldn't waste my time on it.  The fact I that the Alphacide appears to have cost cHumPaq a *net* loss in profit of K hundreds of millions of dollars annually, and the likelihood that this lossoI would have been far less if they had been allowed to lead their customers I like sheep without any opposition, makes me feel that the effort has beenu worthwhile.e  	   This iseK > just my opinion, I'm not trying to brow-beat you with it.  But trust that:= > there are many, many people working to improve on the past.p  K There have *always* been many, many good people working to improve VMS (andhL before mid-to-late 2001 Tru64 and Alpha as well).  And (except briefly underK Pfeiffer) it hasn't made a damn bit of difference to anything but technical I quality for close to a decade, because the people who make *decisions* in 2 the company weren't among them (and still aren't).  
   Why detract = > from their efforts before you see if they'll be successful?I  K Because they've had a decade to be successful already, and because the sameo& kinds of scumbags are still in charge.   >i >i > >>J > >> Yeah, right.  They decide that the direction the company should go to meetJ > >> business goals, and they incur the risks to take it there.  Hey, it's > >theirH > >> responsibility to keep the company profitable, if they don't do it, it's: > >> their job on the line.  Armchair refs need not apply. > >aI > >I really don't have any problem with that characterization, if it willn get J > >people to stop complaining about my efforts to make sure that the risks theyK > >took by throwing away a business with immense potential profit, breakingrG > >commitments in the process, and then lying about it have appropriate G > >consequences.  Treachery, perfidy, and/or incompetence *should* havehI > >consequences, after all:  that's the free-market philosophy in action.  >ND > First of all, your point is based on lots of conjecture, one-sidedL > information, and a complete rejection of all counter arguments.  With suchF > closed-mindedness, how do you really think this activity is helping? >.H > And the "consequences" that would befall them in this free-market is aF > significant loss of customers.  That hasn't happened, so maybe, just maybe, > there's more to the story.  J Let's see:  VMS systems went from $4 billion in annual revenue (June, 2000H statement by Rich Marcello, repeated in a March, 2001 slide presentationB *well* after the dot-com boom started going bust - and the limitedC information in the quarterly reports from Compaq corroborated VMS's.L resilience in the face of that bust right up to the Alphacide) to $2 billionJ in annual revenue (December, 2001 figure in a Compaq response to a GartnerE comment).  And that's for a system that *was* planned to be ported toiJ Itanic, so one can hardly imagine that Tru64 revenues ($3 billion annually' prior to the Alphacide) did any better.d  K Even recently, VMS system revenues were reportedly stated by Mark Gorham totH be only $2.5 - $3 billion, with $500 million annual profit (vs. the $800J million Marcello quoted).  And Mark is hardly likely to have *under*stated the figures.  K No 'significant loss of customers'?  Bullshit.  When you consider how largeeF a percentage of the remaining system revenue likely came from on-goingI service agreements for equipment no one was about to throw away before ityI required an upgrade rather than new-equipment purchases, the loss becomese even more staggering.s   - bill   ------------------------------  % Date: Fri, 25 Jul 2003 22:49:14 -0500a1 From: "David J. Dachtera" <djesys.nospam@fsi.net> Y Subject: Re: Intel's had Dirty Bombs Anthrax and everything! (Was: Re: HP World: Why Alpho' Message-ID: <3F21FA3A.7020D892@fsi.net>e   jlsue wrote: > H > On 25 Jul 2003 06:40:57 GMT, "Dave Weatherall" <djweath@attglobal.net> > wrote: >  > >>O > >> Yeah, right.  They decide that the direction the company should go to meet P > >> business goals, and they incur the risks to take it there.  Hey, it's theirM > >> responsibility to keep the company profitable, if they don't do it, it'si: > >> their job on the line.  Armchair refs need not apply. > >iE > >Ok, then just accept Bill and the other people who've feel they've=I > >been let down, or other, stronger, emotions, as one of the risks. TheyE6 > >will complain and seek to document their complaint. > M > And they continue to complain, and complain, and complain, as if it's goingu > to change the past.e  % ...or, perhaps, influence the future.t  E > If you want to effect a positive change, that's not going to do it.o  @ That *ALONE* will not. In order to effect change, one must firstH identify the need for change. Once that need is identified, alternativesH can be explored to satisfy the need. Of course, that assumes that one isH willing to accept and acknowledge the need to change, which behavior has) yet to be seen from - you know who/where.e   >  AndH > besides, who in this forum is in any position to do anything about it.  7 Time to revisit the "Sun and Solaris/x86" question, eh?.  I > If I have to accept that they will complain about it, then they have to I > accept that I'll continue to complain about their complaints... ;-) ;-)n  E In simile modo, if we have to accept that we will continue to be liediG to, then those responsible accept - tacitly, by continuing the behavior H that draws the protests - that the protests will continue so long as the, behavior which draws the protests continues.  H I've heard that "business decisions" malarkey before, too. Doesn't wash.H Hard to imagine how pissing off your customers is good for business, but hey - what do I know?s   -- S David J. Dachterae dba DJE Systemsi http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/n   ------------------------------  % Date: Fri, 25 Jul 2003 22:57:28 -0500n1 From: "David J. Dachtera" <djesys.nospam@fsi.net>eY Subject: Re: Intel's had Dirty Bombs Anthrax and everything! (Was: Re: HP World: Why Alph.& Message-ID: <3F21FC28.BAC9338@fsi.net>   jlsue wrote: > H > On Tue, 22 Jul 2003 12:48:48 -0400, David Froble <davef@tsoft-inc.com> > wrote: >  > >lQ > >Bill was upset about the end of Alpha, but it wasn't his product and his place N > >to be more than upset.  However, the spin that Compaq attempted to place onM > >their decision was full of more holes than Swiss cheese.  Bill is entirely.M > >within his right to object to lying to customers and the public, and doingDN > >nothing more than exposing how ridiculous the story was, and the lies beingM > >told.  Hey, if you don't want to see public reaction, don't make somethinglS > >public.  Compaq made public statements, and public comment on such statements is L > >allowable, regardless of how much Compaq doesn't appriciate such comment./ > >Bill's story hasn't changes much in 2 years.n > M > His story doesn't have to change.  That's no longer the point, imo.  He hasaJ > every right feel the way he does and to state so.  But to color each andH > every future comment/announcement based on something that happened twoK > years ago is just not productive.  And it's just not accurate to continuea' > to use that same out-dated yardstick.n  @ I have a major problem with that analogy: unless there's been anH official change issued by the bureau of standards, weights and measures,C then a yardstick is a yardstick, a metrestick is a metrestick, etc.p  E All in all, trust once broken is not easily won back, except from the B gullible. As the proverb goes, fool me once, shame on you; fool me twice, shame on me.   F Absent contrition, trust once lost may well never be recovered. If oneG cannot admit a failing, the concept of correction lacks credibility ande+ seems unwise (at the very least) to expect.w   --   David J. Dachtera  dba DJE Systemsa http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/d   ------------------------------  # Date: Fri, 25 Jul 2003 19:38:34 GMTl( From: "konabear" <maurert@ameritech.net>P Subject: Re: Legal status of EVE source code ?, was: RE: Creating an EVE o	r TPU@ Message-ID: <_GfUa.12696$Vx2.6313677@newssvr28.news.prodigy.com>  L Oh HP WANTS you to include TPU in LINUX.  Then in 5 years we can sue all the' LINUX users and distributes for $$$. ;)w  J Disclaimer:  All said tongue in cheek.  I don't know what my employer (HP) *really* wants with this issue.t   ToddG "Simon Clubley" <clubley@remove_me.eisner.decus.org-Earth.UFP> wrote in 5 message news:NCo3hOdm3KX4@eisner.encompasserve.org...g5 > In article <hLVRz$jsw9Dw@eisner.encompasserve.org>,mD clubley@remove_me.eisner.decus.org-Earth.UFP (Simon Clubley) writes: > >uJ > > The reason that I was not sure was that, because you are encouraged to modifyK > > EVE for your own needs, and are given the source code so that you _can_o modifyG > > it (unlike everything else in VMS unless you buy a source license),5 whether>K > > that means that all the usual restrictions on copying source code would  apply. > >sH > > I thought that the usual restrictions would still apply, but was not sure soa > > hence the question.d > >hL > > The implication is that a hobbyist could not create an implementation of TPUr+ > > on Linux and then run EVE on top of it.d > >  > H > It would be nice to get a ruling one way or another from HP on this so thatL > anybody considering a Linux TPU implementation in the future would know if. > they could take and use the EVE source code. > G > I don't think that you can, but who would be the best person in HP toi emailoB > in order to get HP's position on this so that we know for sure ? >h > Simon. >  > -- e= > Simon Clubley, clubley@remove_me.eisner.decus.org-Earth.UFPaD > VMS advocate: One who makes a Mac advocate look like a beginner at	 advocacy.t   ------------------------------  % Date: Fri, 25 Jul 2003 14:16:34 -0400m* From: "Stanley F. Quayle" <stan@stanq.com>; Subject: Re: Migrate obsolete VAX/VMS SW to OpenVMS Itaniumi. Message-ID: <3F213BC2.30135.8B4CAA1@localhost>  0 On 25 Jul 2003 at 9:56, David J. Dachtera wrote:I > See my response to Didier. Getting that last 5% on board will be key to  > regrowing the OpenVMS market.s > G > Kind of like putting out 95% of a fire. The remaining 5% can rekindlei2 > and once again you have a roaring conflagration.  : I don't think it's "key" -- we need to be developing more D applications.  Take that 5% and DOUBLE IT -- it only becomes 10% of  the market.e  D In reference to your example, fighting the last 5% of a fire is the A cheapest, because the equipment and personnel are already on the tE scene.  In the case of migrating those last 5% of ancient VMS users, cB the costs are staggering -- for some migrations, the costs can be @ millions of dollars.  HP shouldn't be footing the bill for that!  C The phrase "point of diminishing returns" is completely applicable . here.D  E I agree that HP and the VMS user community can do more to help those e@ users.  And Didier's task force is a good step.  But there will ; always be some small number of people who cannot be helped.e  
 --Stan Quayle. Quayle Consulting Inc.  
 ----------C Stanley F. Quayle, P.E. N8SQ  +1 614-868-1363  Fax: +1 614 868-1671-1 8572 North Spring Ct. NW, Pickerington, OH  431475= Preferred address:  stan@stanq.com       http://www.stanq.comH   ------------------------------  % Date: Fri, 25 Jul 2003 18:19:36 -0400a* From: "Stanley F. Quayle" <stan@stanq.com>; Subject: Re: Migrate obsolete VAX/VMS SW to OpenVMS Itanium@. Message-ID: <3F2174B8.31559.9934B21@localhost>  + On 25 Jul 2003 at 12:29, Bob Koehler wrote: E >    I thought an updated version had been released to expand this toeF >    later versions.  Of course, you may have code from 1.x running on >    your VAX.   From release notes:a  E - DECmigrate (VEST) 1.1A can migrate code that runs on VAX V5.4-2 if oE it was linked on V4.0 or later.  Images linked after V5.4-2 may fail t= (or not).  Requires Alpha V6.1 or later on the target system.   B - OpenVMS Migration Software for VAX to Alpha (OMSVA) (previously F known as DECmigrate) can migrate code that runs on VAX V7.3 if it was A linked on V4.0 or later.  Images linked after V7.3 [not yet, but hE check the roadmaps] may fail (or not).  Requires Alpha V6.2 or later i on the target system.c  # The release notes for OMSVA are at:   F http://h71000.www7.hp.com/openvms/products/omsva/omsva_012_release_not es.pdf    F Note that NO ATTEMPT was made to migrate from versions prior to V4.0. " How many people does this affect??  
 --Stan Quaylen Quayle Consulting Inc.  
 ----------C Stanley F. Quayle, P.E. N8SQ  +1 614-868-1363  Fax: +1 614 868-1671e1 8572 North Spring Ct. NW, Pickerington, OH  43147 = Preferred address:  stan@stanq.com       http://www.stanq.com    ------------------------------  % Date: Fri, 25 Jul 2003 17:14:23 -0500d1 From: "David J. Dachtera" <djesys.nospam@fsi.net> ; Subject: Re: Migrate obsolete VAX/VMS SW to OpenVMS Itanium ' Message-ID: <3F21ABBF.BBBC241B@fsi.net>    "Stanley F. Quayle" wrote: > 2 > On 25 Jul 2003 at 9:56, David J. Dachtera wrote:K > > See my response to Didier. Getting that last 5% on board will be key to ! > > regrowing the OpenVMS market.t > >eI > > Kind of like putting out 95% of a fire. The remaining 5% can rekindle 4 > > and once again you have a roaring conflagration. > ; > I don't think it's "key" -- we need to be developing morevE > applications.  Take that 5% and DOUBLE IT -- it only becomes 10% ofa
 > the market.   D BUT - take that same 5% and raise it to the second power, it becomes 25%, the third power: 125%, ...t  $ Rather like nuclear fission, get it?  ? > In reference to your example, fighting the last 5% of a fire 1  G ...unless you're trying to keep warm. Put tinder and kindling on top ofr' still-glowing embers, and what happens?:  H THAT's the analogy I was going for: taking the barely still-alive (i.e.,$ VMS) and making it vital once again.   -- l David J. Dachterae dba DJE Systemst http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/e   ------------------------------  % Date: Fri, 25 Jul 2003 23:02:21 -0500m1 From: "David J. Dachtera" <djesys.nospam@fsi.net>j; Subject: Re: Migrate obsolete VAX/VMS SW to OpenVMS Itaniumv' Message-ID: <3F21FD4D.B80F6CA6@fsi.net>r   "Stanley F. Quayle" wrote: > - > On 25 Jul 2003 at 12:29, Bob Koehler wrote:)G > >    I thought an updated version had been released to expand this to H > >    later versions.  Of course, you may have code from 1.x running on > >    your VAX. >  > From release notes:u > F > - DECmigrate (VEST) 1.1A can migrate code that runs on VAX V5.4-2 ifF > it was linked on V4.0 or later.  Images linked after V5.4-2 may fail? > (or not).  Requires Alpha V6.1 or later on the target system.r > C > - OpenVMS Migration Software for VAX to Alpha (OMSVA) (previouslymG > known as DECmigrate) can migrate code that runs on VAX V7.3 if it wasaB > linked on V4.0 or later.  Images linked after V7.3 [not yet, butF > check the roadmaps] may fail (or not).  Requires Alpha V6.2 or later > on the target system.d > % > The release notes for OMSVA are at:i > H > http://h71000.www7.hp.com/openvms/products/omsva/omsva_012_release_not > es.pdf > G > Note that NO ATTEMPT was made to migrate from versions prior to V4.0.s$ > How many people does this affect??  H I have a better question: How many more VMS sites can we afford to lose?  * Kinda puts things back in perspective, eh?   -- t David J. Dachterae dba DJE Systemss http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/    ------------------------------  # Date: Fri, 25 Jul 2003 21:04:43 GMTe& From: jlsue <jefflsxxxz@sbcglobal.net>' Subject: Re: New OpenVMS-to-Itanium FAQa8 Message-ID: <8s63ivgvbkmancgsv13n1dedktd43o9frj@4ax.com>  H On Tue, 22 Jul 2003 14:32:24 -0400, sol gongola <sol@adldata.com> wrote:   >It's a roman numerall >f Friggin ferrinners.        ;-) ;-)    ------------------------------  + Date: Fri, 25 Jul 2003 21:42:25 +0000 (UTC) 7 From: moroney@world.std.spaamtrap.com (Michael Moroney)n) Subject: Re: NEWBIE: DECWindows Bit Depthn( Message-ID: <bfs881$s2l$1@pcls4.std.com>  G Semi-related:  What is the highest resolution now available on VMS, ando what card supports that size?A --   -Mikep   ------------------------------   Date: 25 Jul 2003 17:48 CDTs' From: carl@gerg.tamu.edu (Carl Perkins) ) Subject: Re: NEWBIE: DECWindows Bit Depthm- Message-ID: <25JUL200317480433@gerg.tamu.edu>w  ' Tom Rymes <tomnews@rymes.net> writes...y' }In article <3f20cf9e$1@news.euriware>,t8 } "Philippe Bocher" <philippe.bocher@euriware.fr> wrote: } = }> "Tom Rymes" <tomnews@rymes.net> a crit dans le message debB }> news:tomnews-C23664.22100224072003@news.comcast.giganews.com... }> > Hi folks, }> > }  }[snip]  }  }> > Any help appreciated....h }> > }> > }> > Thank you,y }> > }> > Tom	 }> Hello,b }>  1 }>     edit sys$manager:decw$device_config_gz.com, }>   }> modify what you want... }> h }>     @decw$startup init. }>     @decw$startup restart }> w }> HTH }> Philippet } 5 }Thank you! I'll give it a try once I get to work....e }  }Tom   First, don't try that.  ; The supported way is to edit DECW$PRIVATE_SERVER_SETUP.COM.pE In there you can specify the resolution, refresh rate, and bit depth.hC (And various other things.)  The section i there for my XP900 lookse
 like this:    $ decw$server_page_file== 160000 $ decw$xsize_in_pixels == 1280 $ decw$ysize_in_pixels == 10246 $ Define/Exec/System/NoLog decw$server_refresh_rate 75B $ Define/Exec/System/NoLog decw$server_pixel_depth 8 !8, 16, or 24  > Second, you may find it too slow at bit depths greater than 8.   --- Carl   ------------------------------  # Date: Fri, 25 Jul 2003 23:30:56 GMTn# From: hoff@hp.nospam (Hoff Hoffman)') Subject: Re: NEWBIE: DECWindows Bit Depth 0 Message-ID: <Q4jUa.927$EI5.755@news.cpqcorp.net>  b In article <bfs881$s2l$1@pcls4.std.com>, moroney@world.std.spaamtrap.com (Michael Moroney) writes:H :Semi-related:  What is the highest resolution now available on VMS, and :what card supports that size?  C   The HP 3X-PBXGG-AB (Radeon 7500) AGP card, AFAIK, at 2048 x 1536?   G     http://h18002.www1.hp.com/alphaserver/products/graphics/radeon.htmlh  G   Check the support for the card, as there are (or at least were) some tH   (at least initial) platform restrictions around OpenVMS Alpha support G   for this graphics controller, and I don't know off-hand if there is acD   PCI version that (presently) has OpenVMS Alpha support.  (And mostG   Alpha systems do not have an AGP slot -- but since you indicated onlynH   "VMS" in your question and (unfortunately) did not indicate a specificI   architecture or specific platform, this answer might be too generic...)-  G   Please check the AlphaServer and AlphaStation hardware compatibility bI   lists for details of supported peripherals on specific Alpha platforms,(H   of course.  (I expect you know the location of these support matrixes,D   but for those that do not, please see the URL in the OpenVMS FAQ.)  I   Other high-end cards include the PowerStorm 350 and 3DLabs Oxygen VX1, t-   with both having support for 1920 x 1200.     -   PowerStorm 4D20 gets as far as 1600 x 1200.   J   There may well be some other moderate-resolution cards around (and that G   also have OpenVMS Alpha support, of course), this list has various of.G   the higher-resolution (and supported) graphics cards I am immediately    aware of.o  N  ---------------------------- #include <rtfaq.h> -----------------------------K     For additional, please see the OpenVMS FAQ -- www.hp.com/go/openvms/faqiN  --------------------------- pure personal opinion ---------------------------E         Hoff (Stephen) Hoffman   OpenVMS Engineering   hoff[at]hp.comh   ------------------------------  + Date: Sat, 26 Jul 2003 03:54:02 +0000 (UTC)t7 From: moroney@world.std.spaamtrap.com (Michael Moroney)a) Subject: Re: NEWBIE: DECWindows Bit Depthp( Message-ID: <bfsu0q$2n6$2@pcls4.std.com>  % hoff@hp.nospam (Hoff Hoffman) writes:5  c >In article <bfs881$s2l$1@pcls4.std.com>, moroney@world.std.spaamtrap.com (Michael Moroney) writes:yI >:Semi-related:  What is the highest resolution now available on VMS, andv >:what card supports that size?J  D >  The HP 3X-PBXGG-AB (Radeon 7500) AGP card, AFAIK, at 2048 x 1536?  H >    http://h18002.www1.hp.com/alphaserver/products/graphics/radeon.html  I I don't see VMS there anywhere so I figure it's not supported (and won't lI work -- no driver)  Anyway, forgot there are AGP alphas.  I'm looking fors PCI. --   -Mike    ------------------------------  # Date: Fri, 25 Jul 2003 19:26:23 GMTt# From: "John Smith" <a@nonymous.com>s: Subject: Re: OpenVMS homepage and 'sales.liveperson.net' ?H Message-ID: <zvfUa.39097$Ii1.16600@news02.bloor.is.net.cable.rogers.com>  : "Jonathan Boswell" <jsb@ost.cdrh.fda.gov> wrote in message* news:3F217041.FF309EF4@ost.cdrh.fda.gov... > warren sander wrote:E > > This is because we are increasingly getting customer requests forr help in ; > > 'buying' stuff. And since we 'sell stuff' it seems likev > A > Resurrect the Digital Electronic Store and make it easy to buy.  Gratuitous useF > of javascript is IMHO not the way to make it easy for anybody to buy	 anything.cB > Other vendors have lost a lot of business from me over the years
 because of > this.  > E > And while you're at it, bring back the DECdirect catalog!  (I know,  I'miD > dreaming.)  Didn't the DECdirect catalog have one of the highest $
 volumes ofE > any catalog?  Turn the tables on Michael Dell.  That reminds me.  It amC > constantly getting junk mail at home from Dell even though I havec neverSD > purchased anything from his company.  Do I get any glossy catalogs from HP? > Nope.v  2 But you'd only want the VMS server catalog anyway.   ------------------------------  % Date: Fri, 25 Jul 2003 16:22:35 -0400r- From: Jonathan Boswell <jsb@ost.cdrh.fda.gov>o: Subject: Re: OpenVMS homepage and 'sales.liveperson.net' ?0 Message-ID: <3F21918B.713FFB5D@ost.cdrh.fda.gov>   John Smith wrote:m4 > But you'd only want the VMS server catalog anyway.  I I don't remember ever seeing one of those.  There was from time to time aiL hardware edition, a software edition, a full-line catalog, and I think I sawO network editions and workstation editions.  (I can't lay my hands on an exampleA! of the latter two at the moment.)     - JB    ------------------------------    Date: 25 Jul 2003 16:18:02 -0500; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler)h: Subject: Re: OpenVMS homepage and 'sales.liveperson.net' ?3 Message-ID: <YV6pM3mgMXoz@eisner.encompasserve.org>B  ` In article <3F217041.FF309EF4@ost.cdrh.fda.gov>, Jonathan Boswell <jsb@ost.cdrh.fda.gov> writes:  Q > Resurrect the Digital Electronic Store and make it easy to buy.  Gratuitous usevQ > of javascript is IMHO not the way to make it easy for anybody to buy anything. sM > Other vendors have lost a lot of business from me over the years because ofI > this.e > ; > And while you're at it, bring back the DECdirect catalog!n  ,    Gee, a SOC would be nice early next year.   ------------------------------    Date: 25 Jul 2003 18:08:12 -0500- From: Kilgallen@SpamCop.net (Larry Kilgallen)o: Subject: Re: OpenVMS homepage and 'sales.liveperson.net' ?3 Message-ID: <MQiIqkENQoWq@eisner.encompasserve.org>   ` In article <3F217041.FF309EF4@ost.cdrh.fda.gov>, Jonathan Boswell <jsb@ost.cdrh.fda.gov> writes: > warren sander wrote:L >> This is because we are increasingly getting customer requests for help in: >> 'buying' stuff. And since we 'sell stuff' it seems like > A > Resurrect the Digital Electronic Store and make it easy to buy.   G We bought something from it last month (well, BusinessLink).  There waskH a time period where you had to write to them to ask that your account be
 preserved.   ------------------------------  % Date: Fri, 25 Jul 2003 10:53:14 -0700 # From: "Tom Linden" <tom@kednos.com>c+ Subject: RE: Packed decimal arithmetic in Cc9 Message-ID: <CIEJLCMNHNNDLLOOGNJIMEEMHKAA.tom@kednos.com>    >-----Original Message-----s7 >From: Chris Scheers [mailto:chris@applied-synergy.com] % >Sent: Friday, July 25, 2003 10:34 AMh >To: Info-VAX@Mvb.Saic.Com, >Subject: Re: Packed decimal arithmetic in C >  >t >Bob Koehler wrote:  >>< >> In article <3F206A93.27CF7D9B@applied-synergy.com>, Chris, >Scheers <chris@applied-synergy.com> writes: >> >> >> > 1) An integer is not appropriate for tracking currency as >dollars, sincer< >> > an amount such as $1.35 can not be represented exactly. >> >H >> > 2) A scaled integer consists of two values: the value and where the" >> > decimal point is (the scale). >>L >>    If you do things in fixed point (a PL/I built-in type and a capabilityJ >>    of all compute hardware), you can do everything to $0.0001, which isF >>    common in finance.  You do have to get a large enough integer toJ >>    stuff your million dollars in (signed 32 bits scaled to $0.0001 willC >>    hold about $200000.0000), but most systems also have extendedtE >>    arithmetic instructions, carry bits, etc., that are needed so ar+ >>    compiler can readily do this for you.  >e >eA >Correct.  "scaled integer" and "fixed point" are the same thing.  >t$ These terms have no meaining in PL/I  L To be precise, PL/I has data type FIXED [BINARY |  DECIMAL] and the precisonJ may include a scale factor.  When referring to data types you should avoid thenH word INTEGER.  With DECIMAL, scale indicates the position of the Decimal point, with BINARY, the binary point.  $ thus you would declare it as follows  G declare X FIXED DECIMAL(p,q), 1 <= p <= 31 p >= q >= 0   Note these areq digitsH         Y FIXED BINARY (p,q)  1 <= p <= 31 p >= q >= -31  These are bits0 Note that binary may have negative scale factor.  * See page 3-6 of the PL/I Reference Manual,A ftp://freja.kednos.com/pub/PLIDOCS/ZIP/PDF/PLI_REF_MANUAL_PDF.ZIP         H >-----------------------------------------------------------------------% >Chris Scheers, Applied Synergy, Inc.e >tC >Voice: 817-237-3360            Internet: chris@applied-synergy.com  >  Fax: 817-237-3074 >t >---' >Incoming mail is certified Virus Free.m; >Checked by AVG anti-virus system (http://www.grisoft.com).yA >Version: 6.0.502 / Virus Database: 300 - Release Date: 7/18/2003e >l ---d& Outgoing mail is certified Virus Free.: Checked by AVG anti-virus system (http://www.grisoft.com).@ Version: 6.0.502 / Virus Database: 300 - Release Date: 7/18/2003   ------------------------------  # Date: Fri, 25 Jul 2003 19:29:32 GMTt# From: "John Smith" <a@nonymous.com> + Subject: Re: Packed decimal arithmetic in CnH Message-ID: <wyfUa.39113$Ii1.37648@news02.bloor.is.net.cable.rogers.com>  < "Chris Scheers" <chris@applied-synergy.com> wrote in message- news:3F216D66.9E734C3C@applied-synergy.com...I > David Froble wrote:e > >o > > Chris Scheers wrote: > >s% > > > briggs@encompasserve.org wrote:o > > >s8 > > >>In article <3F1EE1FB.FE8CBFC4@istop.com>, JF Mezei# <jfmezei.spamnot@istop.com> writes:  > > >> > > >>>John Reagan wrote:  > > >>>tE > > >>>>If they were only that smart...  I've seen dozens of examplese	 of people @ > > >>>>using binary floating to track currency (and each one is broken).  I'veF > > >>>>even seen them in languages that provided scaled integers like BASIC. > > >>>>F > > >>>Interesting since there are often standards on exactly how many decimals to'E > > >>>use when specifying currency and conversion factors, as in theo case ofxA > > >>>international SWIFT transactions. With floating point, yous can't specify howl? > > >>>many decimals to consider during arithmetic, only during- display. > > >>> A > > >>Ahhh.  The point being that floating point considers _more_c digits@ > > >>in some cases than the standards specify and thus gets the answer > > >>"wrong" for that reason? > > >> > > >>Interesting. > > >>D > > >>So if you were going to use floating point correctly in such aD > > >>case, one appropriate approach would be to use it to implement scaledD > > >>integers (make sure your integers fit into the mantissa, round toD > > >>nearest integer after multiplication or division and track the, > > >>decimal scale factor by hand).  Right? > > >>A > > >>I once used double precision floating point in exactly that> manner. C > > >>It made sense since I had compiler and hardware support for Ds floating? > > >>arithmetic but no compiler or hardware support for 64 bit> integer>C > > >>arithmetic.  32 bit integers were too small, but the mantissa>A > > >>on 64 bit floats was big enough for my purposes.  If I werec? > > >>re-implementing today, I'd be using 64 bit native integer. arithmetic.Y > > >> > > >>        John Briggs" > > >> > > >  > > > A > > > To try to summarize the issues:  (Hopefully I will get this  mostly9 > > > right.  I apologize if I am rehashing the obvious.)  > > >nD > > > For now, we are only considering financial applications, i.e.,A > > > applications that track currency and need to do so exactly.p > > >e% > > > The first problem is exactness:f > > > ? > > > 1) An integer is not appropriate for tracking currency asc dollars, since= > > > an amount such as $1.35 can not be represented exactly.k > > >hE > > > 2) A scaled integer consists of two values: the value and where  the A > > > decimal point is (the scale).  This is what you are used to  seeing on a'F > > > calculator display.  A scaled integer represents $1.35 as 135s2, i.e.,gF > > > 135 with the decimal point shifted left 2 (between the 1 and 3). ThisB > > > is also known as tracking pennies.  Sometimes a scale larger than two isaF > > > required.  Quite often, compound interest calculations require a larger; > > > scale.  Also, different currencies may have different 
 requirements.o > > >t: > > > A scaled integer meets the requirement of exactness. > > >l@ > > > 3) A floating point number is not appropriate for currency calculationsE > > > because a floating point number can not exactly represent 1.35.s GettingtD > > > technical, the problem is that the base used in floating point numbers F > > > is a power of two whereas currency calculations need a base that is aD > > > power of ten.  In base 2, 1/10 is a repeating fraction and can not be > > > represented accurately.o > > >p> > > > A floating point number does not meet the requirement of
 exactness. > > > C > > > 3) Comparing floating point and scaled integer, it may becomer obviousnE > > > that a scaled integer is really a floating point number that isn base 10yE > > > instead of base 2.  They are really implementations of the same' thing,< > > > but scaled integers are more appropriate for financial
 applications.u > > > " > > > The second problem is range: > > > B > > > 1) Using a scaled integers, some calculations can exceed the range  > > > available in 32 bits.b > > > = > > > 2) Decimal arithmetic is one solution to this.  Decimal 
 arithmeticF > > > tracks the number as a string of decimal digits.  (Again, like aD > > > calculator.)  Decimal numbers are scaled integers, but are not= > > > restricted to 32 bits.  Generally, a decimal arithmeticr implementation( > > > allows up to (at least) 31 digits. > > >c? > > > 3) Floating point is another solution to this problem.  A  floating pointD > > > number is implemented as some number of bits for the power and	 some bitsn@ > > > for the mantissa.  The mantissa is the numeric part of the
 value.  IfE > > > the power is held at zero, the mantissa is the integer value ofs the @ > > > number.  (Ignoring normalization.)  Basically, you use the floating& > > > point number to hold an integer. > > >tC > > > What's the point to this?  Well, a 64 bit or 128 bit floating  pointa? > > > value has more than 32 bits of mantissa.  This allows thep floating pointD > > > value to hold an integer larger than 32 bits.  You can also do math on D > > > this number with floating point instructions.  Just manage the powersC > > > appropriately.  If this is done correctly, the floating pointt value A > > > becomes exact for our purposes.  The extra bits give us thet range weD > > > need.  You still need to maintain an external scale.  You have	 created ayE > > > scaled integer with the integer part stored in a floating pointn value. > > >  > > >o > > > Summary: > > > D > > > Either an extended precision floating point value or a decimal variableD > > > can be used to provide a scaled integer with more than 32 bits of > > > precision. > > >oA > > > COBOL and PL/I have built in support for decimal variables.t > > >cE > > > The Alpha does not have hardware support for decimal variables,  but doesA > > > have hardware support for extended precision floating pointy
 values, so@ > > > code can be more efficient.  But code libraries need to be
 written to> > > > manipulate the floating point values as scaled integers. > > >n > > > Summary #2:  > > >-B > > > Native 64 bit integer support in the Alpha makes much of the	 precedingp3 > > > discussion moot, if you are writing new code.m > > >d" > > > Let the discussion continue! > >o > > Ok.- > >2F > > I've found double precision floating point to be entirely adequate for storingdF > > monetary amounts.  I'll note that I've only used it for US$.  I'll	 also notecC > > that $1,000,000.00 is a rather large monetary amount for any ofh
 the work I'verD > > done, and $100 million is more than ever required.  I'll concede that there are= > > places where much more magnitude is required, while stilld requiring precisenessd > > to the penny.t > >n= > > As for calculations, that's a bit different.  To do so, ah programmer mustoF > > routinely perform a rounding calculation after, or as part of, any calculation,A > > or group of calculations.  Sure, there are values that cannotr
 exactly beF > > represented in a floating point number, but, what's the difference between 1.35C > > and 1.34999999999999?  If an amount is added to this value, andi
 the result is D > > rounded to 2 digits after the decimal point, that's adequate for countingA > > dollars and cents.  Multiply it by 100, and still, what's thei
 meaningfulA > > difference between 134.99999999999 and 135?  Once rounded, itp becomes 135. >u >e9 > For any single operation, the difference is negligible.  >cE > But you need to realize that every time you round, you introduce an F > error term.  Basically an error term says that the result is correct +/-n > some amount. >mB > If you do error analysis, you will find that (in most cases) anyF > operation between numbers with error terms generates a result with aB > larger error term.  Do enough operations, and the error term can become > significant. >.E > Since an error term is a +/- range, it may be close to zero and nottB > affect the results.  But every now and them, all the errors willD > accumulate to one side or the other and the results will be off by realA > amounts.  This is where you can be off by a $100,000 on a larged project. > B > The problem with errors is that it can be extremely difficult to predictn" > when they are going to bite you. > A > You avoid this by using exact numbers.  With exact numbers, theu error ? > term is zero.  This is where scaled integers (fixed point) ort decimal  > values help you. >iF > Think about this:  If you need to PROVE that your code will generateD > correct results, which would you rather use: floating point values or > decimal values?o >dD > What if your company has to indemnify your customer for any losses from > errors in calculations?      Concur.l  - See my earlier post for a real-world example.n  > We had to decide whose calculations were in error and why. The0 difference was a bit over $50,000 in real money.   ------------------------------    Date: 25 Jul 2003 16:16:49 -0500; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler)e+ Subject: Re: Packed decimal arithmetic in Ci3 Message-ID: <ZC95agrTYYbi@eisner.encompasserve.org>i  e In article <3F2169F4.6EFE5A7B@applied-synergy.com>, Chris Scheers <chris@applied-synergy.com> writes:J  B > Correct.  "scaled integer" and "fixed point" are the same thing.  F    I don't know if you'll find this defined anywhere definitively, butF    to me "fixed point" means you or the compiler remember the locationG    of the decimal point.  "scaled integer" means you track the scale in>    a separate memory location.  E    You can add two fixed point numbers IFF they have the same decimaleC    point location just by doing an integer add instruction.  To addlA    two scaled integers you first rescale to the higher of the two>    scales, then do integer add.       As examples:.  F    TOPS-20 had an elapsed time clock in the format day.fraction being E    18.18 in a 36 bit word.  You had to know it was 18.18 in order to kD    format the time, but time calculations were just ordinary 1 word     integer instructions.  >    On older spacecraft computers, there were no floating point?    instructions.  One word held the scale and one word held thew?    scaled value and a bunch of arithmetic routines were writtenh%    to deal with such scaled integers.    ------------------------------  % Date: Fri, 25 Jul 2003 23:02:42 +0100n* From: "John Travell" <john@travell.uk.net>+ Subject: Re: Packed decimal arithmetic in Cn9 Message-ID: <bfs9e5$ieqqm$1@ID-120847.news.uni-berlin.de>   H "Bob Koehler" <koehler@eisner.nospam.encompasserve.org> wrote in message- news:i3HJb1zYn5B7@eisner.encompasserve.org...o > >n
 > > 0.2 what?. > >iI > > 0.2 pennies doesn't need to be expressed precisely.  We are operatingc# > > on whole pennies by assumption.  >nC >    When you sell millions of copies of a product you need to workoH >    units out to less than the whole penny level.  A change of .01 centB >    over a million units is $10K, and you will answer to the nextD >    level of management if you screw up that much since that may be- >    close to the size of their annual bonus.d >s  H For the benefit of us Brit's, to whom a penny is part of our currency, aF coin valued at precisely 100th part of one pound sterling, can someone4 explain EXACTLY what America refers to as a 'penny'.     -- John Travell  VMS crashdump expertise for hire john@jomatech.comi +44-(0)23-92552229 http://www.jomatech.com/       ---/& Outgoing mail is certified Virus Free.: Checked by AVG anti-virus system (http://www.grisoft.com).A Version: 6.0.500 / Virus Database: 298 - Release Date: 10/07/2003r   ------------------------------    Date: 25 Jul 2003 18:12:32 -0500- From: Kilgallen@SpamCop.net (Larry Kilgallen):+ Subject: Re: Packed decimal arithmetic in Ca3 Message-ID: <C0RlVnjo0YnT@eisner.encompasserve.org>   f In article <bfs9e5$ieqqm$1@ID-120847.news.uni-berlin.de>, "John Travell" <john@travell.uk.net> writes:  J > For the benefit of us Brit's, to whom a penny is part of our currency, aH > coin valued at precisely 100th part of one pound sterling, can someone6 > explain EXACTLY what America refers to as a 'penny'.  F In the US, "penny" is a colloquial term for the coin whose actual nameC is a "cent".  It has the value of precisely 100th part of a dollar.o  E Now you know how we in the US feel about Australian dollars, CanadianpG dollars, etc.  Those countries should follow the example of "dinar" andhL "franc" by using a word from their own native language, rather than ours :-)   ------------------------------  % Date: Fri, 25 Jul 2003 13:52:16 -0400 * From: JF Mezei <jfmezei.spamnot@istop.com>, Subject: Re: Sun Micro Profit, Revenues Fall) Message-ID: <3F216E4C.9EB1897F@istop.com>    Fred Kleinsorge wrote:M > revenue by fiscal 2001.  To support that kind of revenue stream and growth,cL > they over time built up a sales force, engineering, bricks and mortar, and5 > general overhead needed to deliver their product.  g  M If McNealy is smart, he will dispose of excess mortar and bricks to bring his L infrastructure back in line with realistic sales/growth. He should have beenK aware that the .COM era was a big bubble, and while reaping the benefits ofoI that frenzy, he should have been aware that any growth during that period,G would have been temporary. This means get some Atco trailers instead ofi building a new office tower.  U Nortel has had no problem of disposing of tons of office space when the bubble burst.c  L Out of curiosity, with the supposedly succesful merger of HP and Compaq, hasJ Carly been able to dispose of any significant amount of infrastrcuture andJ office space ? Or is the combined company still pretty much a total sum of' whatever both had prior to the merger ?E  I > Now the short term problem.  Sun's market cap is about $13 billion, andVN > reportedly they have nearly $6 billlion in cash, and are losing $2 billion aI > year (currently).  They are a publicly held company that has not paid a.B > dividend, so stockholders need to see an increasing stock price.  J Unless you are Microsoft. (but heck, Microsoft realised it needed to startM paying dividends since its own employees were starting to starve because part@N of their salaries was handed out as stock options which yielded no real revenu to employees).  M > speculation on future earnings.  If the stock dips too low, someone will betG > very tempted to take over Sun using Suns own cash for the bulk of theVK > purchase - and perhaps dissasemble them for the parts - or perhaps to tryyL > and shore up their own position (say, like AMD or even IBM seeing a chance > to deep-six a competetor).  K I am not so sure IBM would do this. I think that IBM likes Sun. I think SunuM hurts HP far more than it hurts IBM. If IBM were to buy Sun, it would have toeK contend with the tripple trouble of AIX, Linux and Solaris,  as well as thee< FTC being awakened once again to the issue of IBM monopoly.   I > I'd say the odds are 50-50 that someone takes over Sun, they merge witheE > someone, or Sun looks to aquisitions to get into other parts of thea( > business - maybe a software-only play.  L If Sun is the target of a takeover, I suspect some Japanese firm such as NEC) would likely be the buyer, not IBM or HP.o  L > I would not be suprised to see Sun stock tumble under $3.00 and the sharks > to start circling.  N But if Sun has one quarter with a positive outlook, you'll see its stock priceJ rise significantly. When expectations are purposefully set low, it becomesJ quite easy to make the stock price go up with a few positive comments from	 McNealy. a  N I suspect McNealy would have a plan to announce some good news the minute they2 hear someone is seriously considering buying them.   ------------------------------  % Date: Fri, 25 Jul 2003 11:10:36 -0700i& From: Greg Cagle <gregc@gregcagle.com>, Subject: Re: Sun Micro Profit, Revenues Fall/ Message-ID: <vi2skubdtasd80@corp.supernews.com>    JF Mezei wrote:   N > Out of curiosity, with the supposedly succesful merger of HP and Compaq, hasL > Carly been able to dispose of any significant amount of infrastrcuture andL > office space ? Or is the combined company still pretty much a total sum of) > whatever both had prior to the merger ?a  D There has been a lot of consolidation of field offices, usually into@ Compaq-owned space instead of HP-leased space. The Mountain View? sales office has moved from leased space into a Compaq buildingo@ across the street from HP's Cupertino site, for example. I don'tF believe there have been any major site closure outside of the MayfieldC site, but it wouldn't surprise me at all to see something like thate in the very near future.   - Greg -- o
 Greg Cagle gregc at gregcagle dot com   ------------------------------  # Date: Fri, 25 Jul 2003 18:40:41 GMTC9 From: "Fred Kleinsorge" <my-last-name@stardotzko.dec.com> , Subject: Re: Sun Micro Profit, Revenues Fall0 Message-ID: <JQeUa.908$6o5.850@news.cpqcorp.net>  5 "Bill Todd" <billtodd@metrocast.net> wrote in messageH, news:lhSdnTd-dMHi-7yiXTWJhA@metrocast.net... >2F > "Fred Kleinsorge" <my-last-name@stardotzko.dec.com> wrote in message, > news:06dUa.889$dh5.211@news.cpqcorp.net... >M > ...u > . >   Sun's market cap is about $13 billion, andF > > reportedly they have nearly $6 billlion in cash, and are losing $2 billion  > ae > > year (currently).h >tF > 'Are losing $2 billion a year' (even qualified by 'currently') to me impliesyK > an on-going drain, whereas my impression was that almost all of that loss J > this year was the write-down of the value of an acquisition (rather than > some recurring problem). >o  E Mea Culpa.  They narrowed their loss from $520m/yr to $250m/yr if yourK exclude the non-cash "imparement" of loss of goodwill on aquisitions - so InH guess that's a "good" trend.  I'll amend what I said and guess that theyI only need perhaps a boost of $4-5 billion/yr in revenue to start treadingn+ water.  Everything else pretty much stands.t   > ...t >cG > > I would not be suprised to see Sun stock tumble under $3.00 and then sharks > > to start circling. >eC > Possibly, but they didn't strike when that happened not long ago.o >p  
 We shall see.o   ------------------------------  # Date: Fri, 25 Jul 2003 18:52:21 GMT.9 From: "Fred Kleinsorge" <my-last-name@stardotzko.dec.com>6, Subject: Re: Sun Micro Profit, Revenues Fall/ Message-ID: <F%eUa.910$wn5.14@news.cpqcorp.net>:  7 "JF Mezei" <jfmezei.spamnot@istop.com> wrote in message-# news:3F216E4C.9EB1897F@istop.com...u > Fred Kleinsorge wrote:G > > revenue by fiscal 2001.  To support that kind of revenue stream ands growth,/J > > they over time built up a sales force, engineering, bricks and mortar, andF5 > > general overhead needed to deliver their product.. >cK > If McNealy is smart, he will dispose of excess mortar and bricks to bringl hiskI > infrastructure back in line with realistic sales/growth. He should haveo beenJ > aware that the .COM era was a big bubble, and while reaping the benefits ofK > that frenzy, he should have been aware that any growth during that periodmI > would have been temporary. This means get some Atco trailers instead ofa > building a new office tower. > I > Nortel has had no problem of disposing of tons of office space when the 
 bubble burst.a >r  I Bricks and mortar house people, get rid of the people, you don't need theeB office.  Get rid of the office and keep the people... tough to do.  J > Out of curiosity, with the supposedly succesful merger of HP and Compaq, haseL > Carly been able to dispose of any significant amount of infrastrcuture andL > office space ? Or is the combined company still pretty much a total sum of) > whatever both had prior to the merger ?n >   D I believe that the answer is yes.  There is ongoing consolidation of: facilities.  You can probably dig it out of an SEC filing.  K > > Now the short term problem.  Sun's market cap is about $13 billion, and F > > reportedly they have nearly $6 billlion in cash, and are losing $2	 billion ayK > > year (currently).  They are a publicly held company that has not paid agD > > dividend, so stockholders need to see an increasing stock price. >vL > Unless you are Microsoft. (but heck, Microsoft realised it needed to startJ > paying dividends since its own employees were starting to starve because partI > of their salaries was handed out as stock options which yielded no realr revenu > to employees). >   H Yeah, well it was always a curiosity to me - how MS was able to give lowK salaries but with options also make lots of millionaires.  I think paying anA dividend will help stablize MS stock from lots of short term wild I fluctuations.  I think giving employees stock instead of options... well,aL it's certainly an interesting idea.  I think a lot less millionaires will be made.s  L > > speculation on future earnings.  If the stock dips too low, someone will beI > > very tempted to take over Sun using Suns own cash for the bulk of theaI > > purchase - and perhaps dissasemble them for the parts - or perhaps to  tryaG > > and shore up their own position (say, like AMD or even IBM seeing a  chance > > to deep-six a competetor). >tI > I am not so sure IBM would do this. I think that IBM likes Sun. I thinkl SunrL > hurts HP far more than it hurts IBM. If IBM were to buy Sun, it would have toI > contend with the tripple trouble of AIX, Linux and Solaris,  as well as@ theo= > FTC being awakened once again to the issue of IBM monopoly.  >6K > > I'd say the odds are 50-50 that someone takes over Sun, they merge with0G > > someone, or Sun looks to aquisitions to get into other parts of theg* > > business - maybe a software-only play. >-J > If Sun is the target of a takeover, I suspect some Japanese firm such as NEC@+ > would likely be the buyer, not IBM or HP.u >c   You may well be right.  G > > I would not be suprised to see Sun stock tumble under $3.00 and the  sharks > > to start circling. >aJ > But if Sun has one quarter with a positive outlook, you'll see its stock price-L > rise significantly. When expectations are purposefully set low, it becomesL > quite easy to make the stock price go up with a few positive comments from
 > McNealy. >u  K I think expectations are set pretty low at this point.  Last year they lostsL 4 cents a share or so in the first quarter, so even the $12 million (0 cents2 a share) they made this quarter would look "good".  F The bad news in McNealy and Co have telegraphed their intentions to doJ significant price reductions - price is the only tool they have to competeK with - they don't have the performance.  So what does that do to the bottomyI line if your already losing a quarter billion a year (before write offs).h  K > I suspect McNealy would have a plan to announce some good news the minutei they4 > hear someone is seriously considering buying them.   Really?    ------------------------------   Date: 25 Jul 2003 23:44:59 GMT2 From: "Zane H. Healy" <healyzh@shell1.aracnet.com> Subject: Re: The PDP lives+ Message-ID: <bfsfdr0h9i@enews3.newsguy.com>.  < Bob Koehler <koehler@eisner.nospam.encompasserve.org> wrote:b > In article <bfprlq0g3i@enews3.newsguy.com>, "Zane H. Healy" <healyzh@shell1.aracnet.com> writes:M >> Hmmm...  So, a Mentec M11 isn't a real PDP-11 since it wasn't designed and$F >> built by Dec?  Development of the PDP-11 family didn't end with theC >> PDP-11/93 /94, that was just the end of DEC's development of it.-  H >    OK, I was not aware that Mentec had implemented any new systems.  ID >    thought they had shipped one update to RSX and were maintaining >    existing systems.  I They've shipped several since taking over, the M11 is the current model. pD It's an FPGA implementation of the J11, and looks pretty cool, 1-4MbH onboard, as well as 4 DLV11J compatible serial lines.  What would reallyJ rock is if they'd come out with a board with ethernet and SCSI on the sameJ board instead of having to access it through the Q-Bus.  Still the currentL move seems to be more along the lines of running under emulation, and if youJ have Q-Bus or Unibus widgets that you *must* retain, to connect them via aL PCI-to-QBus or PCI-to-Unibus converter that the emulator is able to access. J If all you care about is the software running it under E11 on a fast PC isD "wicked fast" (I don't have any experience with the PCI converters).  K As for the OS's they've shipped more than just one update.  They've shippedcI 1 update for RT-11 (5.7) and RSTS/E (10.1A), and they've done 1-2 updatesaC for each of the RSX versions they sell (RSX-11S, RSX-11M, RSX-11M+, L MicroRSX, VAX RSX).  For the layered products I think the only thing they've* really done were Y2k updates where needed.   			Zaneu   ------------------------------  # Date: Sat, 26 Jul 2003 00:53:44 GMTe2 From: "William Hymen" <t18_pilot@hotmail.spam.com> Subject: UCX portsG Message-ID: <sikUa.120278$Io.10223053@newsread2.prod.itd.earthlink.net>   8 I have a vendor application (TrueChange) which will work; with specific port numbers.  My newbie system administratorl5 claims the application will create it's own ports (?)h  ; Thats a new one on me, I though *he* had to create/allocatet. a range of ports *for* the application to use.  C If my client piece talks to the license manager on port 12345 usingl  * ADCLMHOST :== computer.bankdomain.com:1234H and my server piece broadcasts using an installed image license manager: adclmd -t 1234  F I think my system admin is responsible for creating ucx port 1234 (??)   How would he do this in UCX?   Thanks in advance -    Bill   ------------------------------  % Date: Fri, 25 Jul 2003 22:42:13 -0400a* From: JF Mezei <jfmezei.spamnot@istop.com> Subject: Re: UCX ports) Message-ID: <3F21EA7A.9A180A84@istop.com>H   William Hymen wrote: > : > I have a vendor application (TrueChange) which will work= > with specific port numbers.  My newbie system administratorh7 > claims the application will create it's own ports (?)   D Yes, a server side application can create its own port , You requireL privileges to create a known port (below 1024). How to do this is documented in the programming manual.  M The UCX/TCPIP services database is useful to start a piece of software when adL call comes in. (or in the case of TCPIP service 5.3, start the software when> the TCPIP stack starts (and restart the software if it fails).   ------------------------------  % Date: Fri, 25 Jul 2003 15:19:39 -0400h- From: Jonathan Boswell <jsb@ost.cdrh.fda.gov>D3 Subject: Re: VMS and Intel, yet another question...g0 Message-ID: <3F2182CB.AF7DB072@ost.cdrh.fda.gov>   Didier Morandi wrote:s> > http://www.intel.com/ebusiness/pdf/prod/itanium/ds022001.pdf > C > When you open the document, you do not find anymore matches. Why?e  O My supposition is that you have Acrobat running as a plug-in for your browser. cM Are you by chance accidentally searching with the browser Find in Page ratherjM than the plug-in's binoculars button?  (The one-and-only OpenVMS reference is " just above the bottom line on P3.)    - JBm   ------------------------------   End of INFO-VAX 2003.409 ************************