1 INFO-VAX	Sun, 24 Jun 2001	Volume 2001 : Issue 348       Contents: Re: Any people use KEA! 420 ? ( Re: Compaq's Alpha design team for sale?( Re: Compaq's Alpha design team for sale?( Re: Compaq's Alpha design team for sale?( Re: Compaq's Alpha design team for sale?( Re: Compaq's Alpha design team for sale?( Re: Compaq's Alpha design team for sale?( Re: Compaq's Alpha design team for sale?( Re: Compaq's Alpha design team for sale?( Re: Compaq's Alpha design team for sale?( Re: Compaq's Alpha design team for sale?( Re: Compaq's Alpha design team for sale?( Re: Compaq's Alpha design team for sale?( Re: Compaq's Alpha design team for sale?( Re: Compaq's Alpha design team for sale? Re: cool add Re: DSN/AES shutting down?9 Re: DSN/AES shutting down? - Use Mozilla browser instead?  Re: FORTRAN question Re: FORTRAN question Re: FORTRAN question" Re: FTP LIST problem. Help Please!' Gap in this forum between March and May ! Re: H-P EOLs PA-RISC Architecture ! PCSI_MENU 1.0 Part 1 ready for FT J Re: Reward for the first of the next 50 posts: which company shouldbuy VMS Re: sadness ( Re: Software to create PDF files on OVMS Re: VMS on IA64 (technical)  Re: VMS on IA64 (technical)  Re: VMS on IA64 (technical)  Re: VMS on IA64 (technical)  Re: VMS on IA64 (technical) ! Re: [OT] but now UK vs US English   F ----------------------------------------------------------------------  % Date: Sun, 24 Jun 2001 19:20:06 +0200 0 From: Didier Morandi <Didier.Morandi@Compaq.com>& Subject: Re: Any people use KEA! 420 ?* Message-ID: <3B362146.D7B3C884@Compaq.com>   > "Valdemir J. Santos" wrote:  >  > Hello all guys:  > F > Id like to know if is there any person in this group using KEA! 420 > for OpenVMS. > H > Id like learn how to create little macros to automatically connect in1 > the system, open and read data in files, etc...  > macro examples are welcome...  >  > Thank you in advance...   C I do. You do not need macros to "connect to a system". You create a H session icon, you save the session on your desktop, then when you doubleG click the icon, you are automatically at the username: prompt (which is < not the case with the EXTRA package, why, God only knows...)  G I have never programmed macros to "open and read data in files". Why do G you want to do so? Do you need PC - VMS communication? I will certainly F not recomment macros. I would program IP instead. On the Alpha, it's a child game, on a PC...?    D. --  B Compaq Computer (Schweiz) GmbH, Ueberlandstrasse 1, 8600 Dubendorf<      All opinions expressed are my own and not my employer's   ------------------------------    Date: 24 Jun 2001 05:53:43 -05009 From: Kilgallen@eisner.decus.org.nospam (Larry Kilgallen) 1 Subject: Re: Compaq's Alpha design team for sale? 3 Message-ID: <IjvRXWBwrVh4@eisner.encompasserve.org>    In article <rdeininger-2306012215460001@user-2ive69s.dialup.mindspring.com>, rdeininger@mindspring.com (Robert Deininger) writes: F > In article <CIEJLCMNHNNDLLOOGNJIOECFCNAA.tom@kednos.com>, Tom Linden > <tom@kednos.com> wrote:  > L >> Compaq certainly didn't buy Digital because of Alpha.  It was the service >> organization thatI >> they wanted.  I suspect that the cost to keep pace with Intel is high.  >> Never understood 3 >> why Digital didn't pursue the mips architecture.  > A > DEC used MIPS in some systems, and apparently found it lacking.   F I believe what they found lacking was the meeting of deadlines for newF MIPS chip versions once they had promised.  They wanted something theyC could control, and throw more resources at if it met their business C needs.  This problem would need to be considered if DEQ gave up the D design team.  The situation with Alpha FABing seems quite similar toD me.  There was a multi-year agreement for Intel to make Alpha chips.E But it turns out IBM is now making the fastest Alpha chips.  Somebody D who wants your business is doing a better job than someone who has a contractual committment.  @ Remember that Intel has been hanging around with Microsoft for a7 long time.  Microsoft - the folks alleged to have said:   # 	"Your problem is, you trusted us."    ------------------------------  % Date: Sun, 24 Jun 2001 11:57:01 +0100   From: Paul Sture <paul@sture.ch>1 Subject: Re: Compaq's Alpha design team for sale? + Message-ID: <VA.000003e3.065ba2c5@sture.ch>   I In article <1010623234059.38769A-100000@Ives.egh.com>, John Santos wrote:  > Newsgroups: comp.os.vms " > From: John Santos <JOHN@egh.com>4 > Subject: Re: Compaq's Alpha design team for sale? ' > Date: Sat, 23 Jun 2001 23:44:46 -0400  > ( > On 23 Jun 2001, Larry Kilgallen wrote: > t > > In article <dj0Z6.587$m6.648569@typhoon.ne.mediaone.net>, "Terry C. Shannon" <terryshannon@mediaone.net> writes: > > > 2 > > > <LBohan@dbc.spam_less..com> wrote in message8 > > > news:k3g7jtchp3tcg42t6hmb7oakbee8n75a93@4ax.com... > > >  > > >>= > > >> if the buyer is Intel, what would be their motivation? # > > >> why 2 64 bit architectures ?  > > > P > > > Well gee, if I was Intel and I was gonna buy Alpha (they had the chance toN > > > do so in 1990), I'd use some of the wizardry in Alpha to enhance the IPFP > > > family. I would not maintain two 64-bit architectures. Taking one more offQ > > > the playing field is a good thing for Intel, assuming the rumours are true.  > > I > > I though the previous settlement granted Intel the right to use Alpha = > > technology in their own chips via patent cross-licensing.  > 8 > I came across this while searching for something else: > 5 > http://www8.zdnet.com/eweek/news/1027/31ealpha.html  > ? > Makes your stomach hurt to see the same old negativity.  Then  > I looked closely at the date!  >  I honed in on this:   m "We will fully support 64-bit Intel processors [on Unix and Windows                   NT], and Alpha will be  > complementary to that,'' said Howard Elias, vice president of 6 Digital's NT Systems Business Unit, in Maynard, Mass."  A Ah, so Howard Elias was an NT man back in the dark days of "We'll  help you migrate to NT".  C Remember his recent promotion and joint letter (with Ric Marcello?) C on www.openvms.compaq.com? Oops - it's gone! I'm pretty sure it was  still there on Friday...  ? > Clearly the media was clueless at the time, and I don't think * > clue prices have dropped with the NASDQ. > < With the exception of websites well known here, the media is still largely clueless.  ___ 
 Paul Sture Switzerland    ------------------------------  % Date: Sun, 24 Jun 2001 11:57:03 +0100   From: Paul Sture <paul@sture.ch>1 Subject: Re: Compaq's Alpha design team for sale? + Message-ID: <VA.000003e4.065ba980@sture.ch>   < In article <3B351A17.8335FABD@virgin.net>, Alan Greig wrote:' > From: Alan Greig <a.greig@virgin.net>  > Newsgroups: comp.os.vms 3 > Subject: Re: Compaq's Alpha design team for sale? ' > Date: Sat, 23 Jun 2001 23:37:12 +0100  >  >  >  > JF Mezei wrote:  > R > > Consider that there is also a sizeable chunk of VAX still running. So you'd be4 > > asking ISVs to support both VAX, Alpha and IA64. > Q > This transition could be eased if IA64 based VMS servers could run 100% or near S > enough of Alpha binaries at a reasonable speed. Preferably without first manually N > VESTing. CompaqMigrate anyone? I wonder how feasable it would be to build anR > IA64/Alpha hybrid and let VMS schedule code across appropriate processors. MightS > even be possible to do this with an Alpha processors add-in card which could slot  > into a standard IA64 server. > R > I also wonder how hard it would be to add Alpha emulation to IA64 given that youS > now have the Alpha design team working for you and that the chip already emulates 8 > other architectures. I didn't say it had to be fast... > Q Interesting thought. Remember the Alpha NT translation utility (I forget its name D now)? Remember RSX emulation on VMS? Yes they were slow, but today's? multiple and faster cpus might make that area worth revisiting.   ? > I'm trying to read the best into these rumours at the moment.    I rather think we all are... ___ 
 Paul Sture Switzerland    ------------------------------  % Date: Sun, 24 Jun 2001 07:01:55 -0400 ) From: "Neil Rieck" <n.rieck@sympatico.ca> 1 Subject: Re: Compaq's Alpha design team for sale? 6 Message-ID: <tIjZ6.61$yy2.49786@news20.bellglobal.com>  6 "Christof Brass" <brass@infopuls.com> wrote in message& news:3B33CE62.A6DC1410@infopuls.com..." > LBohan@dbc.spam_less..com wrote: > @ > Some HW analysts came to the conclusion that EPIC is the wrongB > way to go because it needs a static (at compile time) prediction@ > about what will happen at runtime. If this new architecture is= > really the wrong way to go it would be a good idea to buy a = > mature architecture. I just read an article about the IA-64 ? > instruction set and the only thing I admire is the courage to ; > implement the division and the  transcendent mathematical ? > functions in SW. Is there a big difference in having an Alpha B > with a x86 emulation or an IA-64 with a slow IA-32 emulation? InA > both cases existing apps can be executed but also in both cases 1 > compiling these apps would be highly desirable.   H While it is true that Intel has abandoned their own projects in the pastJ (anybody remember the iAPX-432 cancelled in 1983?) I'm not sure they wouldJ cancel IA-64. Since IA-64 has been over-hyped and Alpha under-hyped (underK marketed?), Intel could easily "save face" by just letting Alpha technology = slowly sink into the history books with the "Avro Arrow" etc.   
 Neil Rieck Kitchener/Waterloo/Cambridge,  Ontario, Canada.! http://www3.sympatico.ca/n.rieck/    ------------------------------    Date: 24 Jun 2001 07:51:28 -05009 From: Kilgallen@eisner.decus.org.nospam (Larry Kilgallen) 1 Subject: Re: Compaq's Alpha design team for sale? 3 Message-ID: <j8oydEHY$V31@eisner.encompasserve.org>   b In article <tIjZ6.61$yy2.49786@news20.bellglobal.com>, "Neil Rieck" <n.rieck@sympatico.ca> writes:  J > While it is true that Intel has abandoned their own projects in the pastL > (anybody remember the iAPX-432 cancelled in 1983?) I'm not sure they wouldL > cancel IA-64. Since IA-64 has been over-hyped and Alpha under-hyped (underM > marketed?), Intel could easily "save face" by just letting Alpha technology ? > slowly sink into the history books with the "Avro Arrow" etc.   . Or by letting IA-64 sink if it cannot keep up.   ------------------------------  # Date: Sun, 24 Jun 2001 13:57:07 GMT  From: Dirk Munk <munk@home.nl>1 Subject: Re: Compaq's Alpha design team for sale? ' Message-ID: <3B35F132.CA840605@home.nl>   E If Compaq were to sell the Alpha and Alpha design team to Intel, this G could mean the end of the Alpha and every thing connected to it, unless % Intel should decide to drop the IA64.   B I don't believe for a minute that Intel would actively support twoE competing processor lines with their own chip sets. It is more likely E that the Alpha team would have to assist Intel to make the x86 / IA64  line more of a success.   E Let's see what happened to Intel the last couple of years. They can't @ deliver new processors etc. in time, the IA64 is years behind onD schedule. They had numerous bugs in processors and chip sets, recallI action after recall action. They lost quite a lot of ground to AMD, there C are many computer shops where you will not find any Intel processor  system anymore.   H And is the IA64 really that fast ? A 1 GHz Alpha 68 is a lot faster, andF the EV6x line of Alpha's is already in its mid-life. The EV7 is comingD quite soon (we hope). But the EV7 is a very different beast from theG previous designs. It is faster, but more importantly it is designed for I very efficient multi-cpu systems. As far as I know, Intel hasn't anything  like that planned.  C And the result for Compaq? Customer confidence in Alpha systems and I operating systems would be lost for good after their experiences with the C similar stupid actions from the late Digital. No one believes it is I possible to build really high performance computer systems if you have no G first hand knowledge of and influence on the processor design. And when D customers run away, they can't afford to design new Alpha systems orB operating systems. They would just end up as a mediocre Wintel box
 manufacturer.   I VMS and Tru64 on Intel?. Don't make me laugh. It would be a hell of a job H to do it, and I doubt if there would be any customers. Oh sure for us itF may be nice. We can buy a cheap PC and run VMS on it. But we like VMS,G managers don't know the difference between CP/M, Unix, MVS or VMS. They I don't care, and will buy what their colleagues buy, a Sun or HP. And what @ about the quality of PC hardware? Even something simple as a USBH interface card is loaded with bugs if we can believe Hoff. The idea of aE VMS crash due to crappy hardware every couple of days isn't what I am  looking forward to.   F So let's hope that Compaq has more sense than selling of the Alpha. ItF may generate money for the shareholders for now, but it can't generateA money in future anymore either. Some managers seem to forget that 4 customers pay you money, and you pay shareholders. .       Neil Rieck wrote:   G > Compaq's Alpha design team for sale? Let's hope this is only a rumor!  > ) > http://www.theinquirer.net/22060101.htm  >  > Neil Rieck > Kitchener/Waterloo/Cambridge,  > Ontario, Canada.# > http://www3.sympatico.ca/n.rieck/    ------------------------------  % Date: Sun, 24 Jun 2001 10:13:15 -0400 ) From: "Neil Rieck" <n.rieck@sympatico.ca> 1 Subject: Re: Compaq's Alpha design team for sale? 8 Message-ID: <RvmZ6.973$yy2.137877@news20.bellglobal.com>  F "Larry Kilgallen" <Kilgallen@eisner.decus.org.nospam> wrote in message- news:IjvRXWBwrVh4@eisner.encompasserve.org...  > H > I believe what they found lacking was the meeting of deadlines for newH > MIPS chip versions once they had promised.  They wanted something theyE > could control, and throw more resources at if it met their business E > needs.  This problem would need to be considered if DEQ gave up the F > design team.  The situation with Alpha FABing seems quite similar toF > me.  There was a multi-year agreement for Intel to make Alpha chips.G > But it turns out IBM is now making the fastest Alpha chips.  Somebody F > who wants your business is doing a better job than someone who has a > contractual committment. > B > Remember that Intel has been hanging around with Microsoft for a9 > long time.  Microsoft - the folks alleged to have said:  > $ > "Your problem is, you trusted us."  K I agree. I'd like to see IBM continue with Alpha development and production H because it wouldn't go over there to die. If Intel gets a hold of Alpha,I we'll need to trash our power-point-presentations of the Alpha roadmap to  2025.   H But consider this; what if the new owner was Microsoft? Most people knowK that Dave Cutler (and others) built some VMS functionality into Windows-NT. F Bill Gates started out on PDPs and Windows-NT ran better on Alpha thanL IA-32. When Compaq stopped developing Windows-NT on Alpha, maybe they forced Microsoft's hand.   
 Neil Rieck Kitchener/Waterloo/Cambridge,  Ontario, Canada.! http://www3.sympatico.ca/n.rieck/    ------------------------------  % Date: Sun, 24 Jun 2001 09:50:59 -0500 * From: "T. S. Murphy" <murphyts@swbell.net>1 Subject: Re: Compaq's Alpha design team for sale? 0 Message-ID: <o7nZ6.234$ZC6.315776@nnrp3.sbc.net>  4 "Neil Rieck" <n.rieck@sympatico.ca> wrote in message0 news:tIjZ6.61$yy2.49786@news20.bellglobal.com...  J > While it is true that Intel has abandoned their own projects in the pastL > (anybody remember the iAPX-432 cancelled in 1983?) I'm not sure they wouldL > cancel IA-64. Since IA-64 has been over-hyped and Alpha under-hyped (underB > marketed?), Intel could easily "save face" by just letting Alpha
 technology? > slowly sink into the history books with the "Avro Arrow" etc.   E I think you will find historically that Intel has very little problem L cancelling projects. There are so many high profile projects which have beenE cancelled, in addition to many more internal projects which have been K cancelled before the public knew about them. Remember: Intel was founded to H do DRAM and only DRAM, and completely exited that business fifteen yearsK ago. Intel is a pragmatic company, and won't let its own pride stand in itst way.  F Is Merced an exception? I don't think so. I think Intel still stronglyL believes that IA-64 is the best way to achieve their migration to 64 bits. IG think Alpha is out of the picture simply and solely because it does noteH support IA-32. I think Intel would be crazy to replace IA-64 with Alpha.   ------------------------------  % Date: Sun, 24 Jun 2001 10:03:03 -0500q* From: "T. S. Murphy" <murphyts@swbell.net>1 Subject: Re: Compaq's Alpha design team for sale?s+ Message-ID: <IinZ6.1$kT4.318@nnrp3.sbc.net>   . "Tom Linden" <tom@kednos.com> wrote in message3 news:CIEJLCMNHNNDLLOOGNJIIEDCCNAA.tom@kednos.com... C > I disagree.  Alpha did not advance the state of the art as far aseC > computer architecture is concerned.  Mips actually had a strongern > instruction set,B > and remember throughput is porportional to the product of freq * > instr_set_strength  E Um ... performance has nothing to do with "instruction set strength".dJ Assuming constant IPC's (and frequencies) a CISC machine will outperform aI RISC machine because some instructions do more. There is nothing stopping K you from making a killer IA-32 implementation at 200 MHz, or a really lousytC MIPS or Alpha implementation at 2 GHz. As it stands, IA-32 productsnF generally have higher clock speeds because they are primarily consumerG parts, and consumers value frequency as the primary indicator of speed.<   ------------------------------  + Date: Sun, 24 Jun 2001 09:38:36 -0600 (MDT)3" From: John Nebel <nebel@csdco.com>1 Subject: Re: Compaq's Alpha design team for sale? G Message-ID: <Pine.OSF.4.21.0106240930430.25500-100000@athena.csdco.com>0  F Right.  Anyone would be crazy to think Compaq would be able to stay inC business, let alone keep promises.  Who would buy wildfire machines.H knowing the hardware had a strictly limited life?  Who in the HPTC arena% would consider an Alpha-based system?t  
 John Nebel    % On Sun, 24 Jun 2001, Dirk Munk wrote:s  G > If Compaq were to sell the Alpha and Alpha design team to Intel, thisiI > could mean the end of the Alpha and every thing connected to it, unlessg' > Intel should decide to drop the IA64.r > D > I don't believe for a minute that Intel would actively support twoG > competing processor lines with their own chip sets. It is more likelynG > that the Alpha team would have to assist Intel to make the x86 / IA64n > line more of a success.p > G > Let's see what happened to Intel the last couple of years. They can'tiB > deliver new processors etc. in time, the IA64 is years behind onF > schedule. They had numerous bugs in processors and chip sets, recallK > action after recall action. They lost quite a lot of ground to AMD, there E > are many computer shops where you will not find any Intel processora > system anymore.2 > J > And is the IA64 really that fast ? A 1 GHz Alpha 68 is a lot faster, andH > the EV6x line of Alpha's is already in its mid-life. The EV7 is comingF > quite soon (we hope). But the EV7 is a very different beast from theI > previous designs. It is faster, but more importantly it is designed fortK > very efficient multi-cpu systems. As far as I know, Intel hasn't anything  > like that planned. > E > And the result for Compaq? Customer confidence in Alpha systems anddK > operating systems would be lost for good after their experiences with theeE > similar stupid actions from the late Digital. No one believes it issK > possible to build really high performance computer systems if you have nowI > first hand knowledge of and influence on the processor design. And wheneF > customers run away, they can't afford to design new Alpha systems orD > operating systems. They would just end up as a mediocre Wintel box > manufacturer.o > K > VMS and Tru64 on Intel?. Don't make me laugh. It would be a hell of a jobSJ > to do it, and I doubt if there would be any customers. Oh sure for us itH > may be nice. We can buy a cheap PC and run VMS on it. But we like VMS,I > managers don't know the difference between CP/M, Unix, MVS or VMS. TheyaK > don't care, and will buy what their colleagues buy, a Sun or HP. And what-B > about the quality of PC hardware? Even something simple as a USBJ > interface card is loaded with bugs if we can believe Hoff. The idea of aG > VMS crash due to crappy hardware every couple of days isn't what I am@ > looking forward to.t > H > So let's hope that Compaq has more sense than selling of the Alpha. ItH > may generate money for the shareholders for now, but it can't generateC > money in future anymore either. Some managers seem to forget that 6 > customers pay you money, and you pay shareholders. . >  >  >  > Neil Rieck wrote:a > I > > Compaq's Alpha design team for sale? Let's hope this is only a rumor!m > >e+ > > http://www.theinquirer.net/22060101.htm9 > >. > > Neil Rieck! > > Kitchener/Waterloo/Cambridge,l > > Ontario, Canada.% > > http://www3.sympatico.ca/n.rieck/t >  >    ------------------------------  % Date: Sun, 24 Jun 2001 18:31:37 +0200h) From: Christof Brass <brass@infopuls.com> 1 Subject: Re: Compaq's Alpha design team for sale?F, Message-ID: <3B3615E9.F15BAAC2@infopuls.com>   John Santos wrote: > ( > On 23 Jun 2001, Larry Kilgallen wrote: > t > > In article <dj0Z6.587$m6.648569@typhoon.ne.mediaone.net>, "Terry C. Shannon" <terryshannon@mediaone.net> writes: > > >l2 > > > <LBohan@dbc.spam_less..com> wrote in message8 > > > news:k3g7jtchp3tcg42t6hmb7oakbee8n75a93@4ax.com... > > >m > > >>= > > >> if the buyer is Intel, what would be their motivation? # > > >> why 2 64 bit architectures ?e > > >mP > > > Well gee, if I was Intel and I was gonna buy Alpha (they had the chance toN > > > do so in 1990), I'd use some of the wizardry in Alpha to enhance the IPFP > > > family. I would not maintain two 64-bit architectures. Taking one more offQ > > > the playing field is a good thing for Intel, assuming the rumours are true.o > >rI > > I though the previous settlement granted Intel the right to use Alpha = > > technology in their own chips via patent cross-licensing.c > 8 > I came across this while searching for something else: > 5 > http://www8.zdnet.com/eweek/news/1027/31ealpha.html  > ? > Makes your stomach hurt to see the same old negativity.  Then  > I looked closely at the date!  > ? > Clearly the media was clueless at the time, and I don't thinka* > clue prices have dropped with the NASDQ. >  > --
 > John Santos- > Evans Griffiths & Hart, Inc. > 781-861-0670 ext 539: Yes, the media was clueless - this is an especially stupid
 paragraph:  9 "Along with its porting of Digital Unix to Intel's 64-bit = architecture, Digital is co-developing with Microsoft Corp. a > common programming model that will help ensure that Alpha/Unix9 applications today will run on Merced, thereby preserving 
 investments."u  ? I wonder how helpful Micro$hit has been in porting Digital Unixs ...t   ------------------------------    Date: 24 Jun 2001 10:03:00 -07001 From: nothome@spammers.are.scum (Malcolm Dunnett).1 Subject: Re: Compaq's Alpha design team for sale?d, Message-ID: <pNUIa5X406sE@malvm5.mala.bc.ca>  7 In article <1010623234633.38769B-100000@Ives.egh.com>, t&     John Santos <JOHN@egh.com> writes: > J > Assuming this rumor is true (never assume anything), supporting both VAXK > and Alpha for me has been a non-issue.  Tons of VAX/DEC BASIC code, quitebJ > a bit of MACRO-32, some C, and thousands of DCL command files.  Once theI > teething problems of DEC BASIC were solved, and we bashed all our MACROlL > into the Alpha Macro-32 style, supporting both platforms has been trivial.  D    But you then assume that all the existing Alpha products would beH ported to a new architecture. There were certainly products on VAX whichE never got ported - many of the people who are still running VAXen areiH doing so for that very reason. I've heard BASIC almost didn't get portedI to Alpha - if it hadn't made it we'd still be a VAX shop ( or more likelyeG we'd now be running Solaris or something of that ilk ). Some stuff onlyuI exists because it's been VESTed ( eg FMS ). Will there be a VEST (AEST?) 12 for IA64 than can translate the translated images?  I    If Compaq doesn't commit on day 1 of the announcement of any IA64 portjH that absolutely everything that runs on Alpha today will be ported ( andF in a timely fashion ) then I think many of us will have lots to worry  about.   ------------------------------    Date: 24 Jun 2001 10:10:04 -07001 From: nothome@spammers.are.scum (Malcolm Dunnett)s1 Subject: Re: Compaq's Alpha design team for sale?l, Message-ID: <bchCBfHNPPGe@malvm5.mala.bc.ca>  - In article <3B3570C7.8E345A20@videotron.ca>, o2    JF Mezei <jfmezei.spamnot@videotron.ca> writes: > O > If Compaq were to announce that it would stop since single platform focus anduO > become a IT solutions company with multiple OS solutions on the same footing,nO > including VMS, Tru64, Linux, NT and even MVS and Solaris and make that a very I > public statement, I might feel more confident that something has really 1 > changed at Compaq and that the move is genuine.  > P    It seems to me that's basically what the transformation letter is saying. TheM question is can you believe they really mean it and are capable of pulling itnN off. Will they, for example, grow their software sales by marketing VMS better@ or by raising the prices they charge the existing customer base.   ------------------------------    Date: 24 Jun 2001 10:07:50 -07001 From: nothome@spammers.are.scum (Malcolm Dunnett)l1 Subject: Re: Compaq's Alpha design team for sale?t, Message-ID: <T+bxo$cAl0so@malvm5.mala.bc.ca>  - In article <3B35612D.21751AF0@videotron.ca>, i2    JF Mezei <jfmezei.spamnot@videotron.ca> writes:    M > performance than VAX and Digital was unwilling to lower the price of VAX torM > match the price/performance of the new Risc processors. Their answer was to"5 > temporarily adopt MIPS for Unix, and develop Alpha.t > F    Actually their first answer was to develop PRISM ( which would haveM included porting VMS ). Then they decided they didn't have the time/resources-G /talent/whatever to pull that off and scrapped it in favour of the MIPStK strategy ( along with a decision that VMS wouldn't get ported there ). ThatwK was when Digital started trumpeting the "VMS is dead, long live Unix" song.A   ------------------------------  # Date: Sun, 24 Jun 2001 13:36:44 GMT $ From: Scott Vieth <svieth@wi.rr.com> Subject: Re: cool add>) Message-ID: <3B35ED77.A055E264@wi.rr.com>o   It's there.g Look again....   -Scott  & "Brian Schenkenberger, VAXman-" wrote:  g > In article <n2IY6.21$rc5.1020@news.cpqcorp.net>, "Sue Skonetski" <susan.skonetski@compaq.com> writes:uG > >Normally I do not look at the prolient add's but this is a good one.i > >t6 > >> http://www.userfriendly.org/animation/compaq.html > >e > >h > >r >nN > I don't see any animation related to Compaq.  There's an add for some backupN > system -- which doesn't look all that impressive that I'd post a boast about& > it -- and a form to forward the URL. >  > --Q > VAXman- OpenVMS APE certification number: AAA-0001     VAXman(at)TMESIS(dot)COMr > Q > city, n., 1. a place where trees are cut down and streets are named after them.t   ------------------------------  % Date: Sun, 24 Jun 2001 11:57:05 +0100t  From: Paul Sture <paul@sture.ch># Subject: Re: DSN/AES shutting down?t+ Message-ID: <VA.000003e5.065bb22f@sture.ch>a  J In article <OFA43FCDDA.CB137FDB-ON80256A74.0048C479@qedi.quintiles.com>,   wrote: >  > F > Don't know Alan.  I read the same message (I think it was swapped on4 > Thursday night/Friday morning) and am not pleased.M > The web site referred to may contain much or all of the information held oneH > the DSN ITS database but it suffers from the Compaq corporate web-page > failing :  > JAVASCRIPT ERRORS. >eE Oh Spit! Add to that the fact that our corporate firewall filters outsM bits of Javascript deemed to be insecure (don't ask me, that's just what I'm eN told), I end up with loads of error boxes which usually need scrolling to hit F OK. It effectively means I have to browse on the PC and then ftp stuff to my VMS systems.  G BTW I visited www.compaq.com from my home VMS system the other day. It tL promptly tried to deliver half a dozen cookies. I'm pretty sure it didn't do that before.  J > ITS is a useful, presently supported, reliable method by which companiesM > with a service contract can use a character cell terminal to grab technicalaD > information, patches, coding examples and other stuff from Compaq. >  > And what's more  IT WORKS!!! >oJ Precisely. It got me out of a very large jam several years ago when I was L able to identify relevant ECOs and download them at a mission critical site.  M > Web sites that are, erm, "challenged" and which produce errors aren't worth' > the time they take to load.o > I The point they don't seem to get is that we are technical folks, seeking aK technical information. As such we do _not_ need dumbed down drop-menus and nL the like (and the scrolling required is extremely non productive). Fine for 5 the general public, but not for a technical audience.o  L > I'm just glad that Warren is there to feed and care for the VMS web site -F > that doesn't (to my knowledge) suffer the Compaq Corporate problems. > Thanks Warren. > ! I second that too. Thanks Warren./ ___.
 Paul Sture Switzerland    ------------------------------  % Date: Sun, 24 Jun 2001 11:57:06 +0100r  From: Paul Sture <paul@sture.ch>B Subject: Re: DSN/AES shutting down? - Use Mozilla browser instead?+ Message-ID: <VA.000003e7.065bb654@sture.ch>o  O In article <jLc9ONZL95hy@eisner.encompasserve.org>, Bradford J. Hamilton wrote:I9 > From: hamilton@encompasserve.org (Bradford J. Hamilton)h > Newsgroups: comp.os.vmshD > Subject: Re: DSN/AES shutting down? - Use Mozilla browser instead?" > Date: 22 Jun 2001 16:05:03 -0500 > 
 > Hi Paul, > C > Please check the recent posts in this NG regarding Mozilla 0.9.1.o > O > The "verdict" is still out, but it looks to be more useful tha NS3.03 is/was.i >AJ Thanks for the advice. Having seen problem reports here in the past I was 5 avoiding it, I guess now is the time to give it a go.i   > --Brad > Q > >In article <VA.000003de.0299e51b@sture.ch>, Paul Sture <paul@sture.ch> writes:h > <snip>L > > One bright note though - I don't need a username/password to get at the K > > database. That solves a long running problem for me as I had no direct  I > > access to AES-ITS dating back to security concerns of having a modem kM > > into our systems. It also infers that folks not on support contracts can '6 > > access and Hobbyists can now search this database. > > I > > I haven't tried it from a VMS system (I'm down to one monitor at the  J > > moment as believe it or not, one of them emitted a load bang and died B > > when viewing an MS word document!). Has anyone else tried from > > Netscape/Mozilla under VMS?t > > ___a > > Paul Sture > > Switzerland  > >  >3   ___ 
 Paul Sture Switzerland    ------------------------------   Date: 24 Jun 2001 09:07:31 GMT- From: djweath@attglobal.net (Dave Weatherall)C Subject: Re: FORTRAN questiona5 Message-ID: <DTiotGxQ0bj6-pn2-nW8NcQUdoTDs@localhost>u  F On Fri, 22 Jun 2001 18:14:39, Michael Austin <miaustin@bellsouth.net>  wrote:  G > Why would one code this (which does not work in the latest version of5- > FORTRAN... Original version was from 1990).p > F >         DATA IOATT,IOKIL,IORNE,IOWLB/"001420,"000012,"001020,"00040/2 > ...............................................^H > %F90-E-ERROR, Syntax error, found INTEGER_CONSTANT '12' when expecting
 > one of: , /s# > at line number 368 in file <name>  > + > Will this give the funtional equivalence?u > B >         DATA IOATT,IOKIL,IORNE,IOWLB/001420,000012,001020,00040/ > as this compiles correctly.o > D > And as you guessed, I am not a real programmer.. but I am a fairly( > decent hacker  (especially cut/paste).  C At a guess I would say that this code came originally from RSX and hD Fortran-IV-Plus. I could find similar stuff in my archives I'm sure.  A If you're going to change it, perhaps its time to forget the RSX IF history and either use the VMS symbols form $IODEF (IIRC) or at least 7 use the hexadecimal equivalents in the data statements.,            CALL GETADR(PR,TERMINAL).         CALL WTQIO(IOATT,LUN,1,PRI,ISB,PR,IDS)*         OPEN(UNIT=5,NAME='TI:',TYPE='NEW')  F This is very 'old-fashioned' and not even that necessay on VMS. Brings  back (incomplete) memories tho'.  pE I was just asking myself if IOATT (IO_ATTACH) still works. I seem to bD remember having to change all my stuff to $ASSIGN and the like. But & that was over a decade ago...  Enough.  E On the DO LOOP question. I usually try to eliminate the dreaded GOTO h' statement when escaping from a loop by n   	done = .false.s 	iterator = init_value  8 	do while (.not.done. and. iterator .lt. Max_iterations) 	     if (normal_case) then  		work 		iterator = iterator + step  $ 	    else if (terminating_case) then 		 done = .true.
 	    endif 	enddo  > Having said all that doesn't F90 have something more 'modern'?   -  Cheers - Dave.   ------------------------------  % Date: Sun, 24 Jun 2001 19:47:58 +0010i% From: paddy.o'brien@zzz.tg.nsw.gov.au" Subject: Re: FORTRAN questiond5 Message-ID: <01K55NQFWKJ6001W1S@tgmail.tg.nsw.gov.au>c   Michael,   Your first question.  C The data type is octal.  If you have a VAX, look at $ help fortran t compatibility octal.  O The correct coding now is O'nnnnn' or O"nnnnn".  I heartily suggest you change nL it to that. Compaq has a non-standard variant 'nnnnn'O but I suggest you do G not use it.  Change to anything else will change the semantics of your - program.   Your second question.e  L Your DO ... ENDDO code is *almost* similar.  The original code which Robert N pointed out must have used a GOTO 115 elsewhere in your program was not valid L Fortran even in F77.  Subject to Steve's comments, I would suggest that not O even a cross reference will tell you what happens.  WWIII could start if J has o4 a currently defined value which is less than NPARTS.  O Yes, the cross reference will highlight the possibly many places that the code mK could reach here, but on its arrival it will probably branch to the DO 115 eM line which is where this fun could start.  Read the VAX /machine listing (or d Alpha if you are fluent).n  M Check out the GOTO's and again if you have a VAX debug version, study how it 'O gets to that line and what it does if it comes from anywhere other than the DO oO statement.  I also think that the error message is not pinpointing the problem   line and should be changed.t  M Unlike the other posters, I would not suggest using the /old_f77 qualifier.  lK This is similar to the difference between using the old VAXC (K&R) and the ,I newer DECC (ANSI) compilers.  Correct what is wrong, throw the /standard oG qualifier (hmm, forget everytime that it tells you about using TABs --  N non-standard but doesn't hurt on VMS).  Yeah, you'll also get a lot of errors L if you use DEC structures.  These are harmless if on VAX, Alpha or even CVF @ but look for the ones that show traps that you have fallen into.  M Make sure that /warning keywords like unused and uninitialised are turned on.h  G Robert is, I believe, correct in that a few of the obsolescent feature  M declared in F90 were deleted in F95.  And again, as he said few vendors will sO remove them.  I forget the list off-hand, but I think some will not affect old tN codes.  IIRC, the PAUSE statement was one -- was it ever portable, and use of M REAL/DOUBLE indexing variables in DO loops -- most number crunchers knew the fK traps there and I doubt this "innovative" feature of F77 was used much.  I  O recently found one use in our applications, and immediately re-wrote the loop. eL Hollerith, I believe was another but that is likely to affect much old code.  M I understand that the committee is unlikely to *delete* much else, just keep aN them on the obsolescent list -- deletion of e.g. fixed form source will break  *a lot*.   Regards, Paddy   Paddy O'Brien, Transmission Development,e
 TransGrid, PO Box A1000, Sydney South,  NSW 2000, Australiaa   Tel:   +61 2 9284-3063 Fax:   +61 2 9284-3050& Email: paddy.o'brien@zzz.tg.nsw.gov.au  M Either "\'" or "\s" (to escape the apostrophe) seems to work for most people,l; but that little whizz-bang apostrophe gives me little spam.    ------------------------------  % Date: Sun, 24 Jun 2001 20:05:03 +0010u% From: paddy.o'brien@zzz.tg.nsw.gov.aua Subject: Re: FORTRAN questiond5 Message-ID: <01K55OCL6CZM001WNJ@tgmail.tg.nsw.gov.au>o   Dave Weatherall said:s   [snips]C  F >On the DO LOOP question. I usually try to eliminate the dreaded GOTO ( >statement when escaping from a loop by  >e >	done = .false. >	iterator = init_valuee >n9 >	do while (.not.done. and. iterator .lt. Max_iterations)y >	     if (normal_case) then >		worke >		iterator = iterator + stepq >A% >	    else if (terminating_case) thenh >		 done = .true.m >	    endift >	enddot >n? >Having said all that doesn't F90 have something more 'modern'?e   Yes it does.  L I have a distinct dislike for do while loops and so does F90/F95.  Although L this was introduced there, use of DO ... ENDDO and EXIT obviate pretty well K all uses.  Conversely, I do not have a total aversion to GOTO as sometimes  * judicious use can give more readable code.  N I ran some code which made extensive use of DO WHILE through the 7.4 compiler K using /annotate=all.  The comment it gave for that code was something like oF that it couldn't do anything with the code because the exits were too H complicated.  When I re-wrote it using an iterative DO and an exit, the K compiler (default level of 4) then told me that it had unrolled the loop 3  B times.  Code was larger, but think what the speed enhancement was.  L Your code could immediately get rid of the done variable by putting EXIT in  the ELSE IF.  O The other condition is slightly more complicated than mine because the "index" oM is conditionally incremented.  But O.K., put IF (iterator.ge.Max_iterations) lG EXIT at the end if the IF (normal_case) block and with an inifinite DO.   K I am off home now, but tomorrow I might look at what the compiler tells me o. with /annotate for these two code suggestions.   Regards, Paddy   Paddy O'Brien, Transmission Development,s
 TransGrid, PO Box A1000, Sydney South,  NSW 2000, Australia    Tel:   +61 2 9284-3063 Fax:   +61 2 9284-3050& Email: paddy.o'brien@zzz.tg.nsw.gov.au  M Either "\'" or "\s" (to escape the apostrophe) seems to work for most people,a; but that little whizz-bang apostrophe gives me little spam.n   ------------------------------  + Date: Sun, 24 Jun 2001 17:32:31 +0000 (UTC)w' From: david20@alpha1.mdx.ac.uk (D.Webb)d+ Subject: Re: FTP LIST problem. Help Please!g+ Message-ID: <9h587f$ka5$1@aquila.mdx.ac.uk>i  Y In article <1010622182951.38769E-100000@Ives.egh.com>, John Santos <JOHN@egh.com> writes:e% >On Fri, 22 Jun 2001, JF Mezei wrote:l >u >> "Richard D. Piccard" wrote:R >> > It can be configured to discard the semicolon and version number when you are >> tQ >> However, I am not sure, but aren't there 2 separate commands FTP commands sentm >> to a server ? >> lO >> One to get a parsable file list, and one to get a "text" file list. It seemspK >> that the command being sent by your FTP client asks for a text file lists# >> instead of a parsable file list.g >oJ >That would be nice, but AFAIK it doesn't exist.  There is absolutely *NO*J >standard for the info returned by an FTP DIR (or LS, which is a synonym). > I >I think it was originally intended to be only a human-readable directoryeJ >and was implemented by forking or spawning the server's regular directoryC >program.  (Which is why you need a copy of or link to LS in a /binrI >subdirectory if you are setting up a Unix anonymous FTP server.  Blech.)p >u    N ls is not an FTP protocol command. It is a common command in FTP clients whichO translate it into the commands LIST or NLST which are part of the FTP protocol. O Usually qualifiers to the ls command allow whichever of these is appropriate to  be sent.  D LIST returns a non-standardised listing which may include additional3 information such as file sizes, creation dates etc    I NLST just returns a list of filenames one per line and is meant to returnoG information which can be used by a program to further process the filesc automatically.  
 See RFC959 :-t     "  LIST (LIST)   N      This command causes a list to be sent from the server to the passive DTP.      If the pathnameN      specifies a directory or other group of files, the server should transfer%      a list of files in the specifiedrL      directory. If the pathname specifies a file then the server should send%      current information on the file.hM      A null argument implies the user's current working or default directory. "      The data transfer is over theB      data connection in type ASCII or type EBCDIC. (The user must      Page 33p  M      RFC 959                                                     October 1985    File Transfer Protocol  F      ensure that the TYPE is appropriately ASCII or EBCDIC). Since the      information on a file mayK      vary widely from system to system, this information may be hard to usee       automatically in a program,.      but may be quite useful to a human user.         NAME LIST (NLST)   K      This command causes a directory listing to be sent from server to userf      site. The pathname shouldO      specify a directory or other system-specific file group descriptor; a nulle      argument implies the.M      current directory. The server will return a stream of names of files ando      no other information. TheN      data will be transferred in ASCII or EBCDIC type over the data connection      as valid pathnameN      strings separated by <CRLF> or <NL>. (Again the user must ensure that the      TYPE is correct.)I      This command is intended to return information that can be used by a       program to further processfO      the files automatically. For example, in the implementation of a "multipleS      get" function.    "a      
 David Webb VMS and Unix team leader CCSS Middlesex University   ------------------------------  % Date: Sun, 24 Jun 2001 19:26:19 +0200/0 From: Didier Morandi <Didier.Morandi@Compaq.com>0 Subject: Gap in this forum between March and May* Message-ID: <3B3622BA.D2A80003@Compaq.com>  H Why is there a (big) gap between the 2nd of March and the 19th of May inE this forum? Where have the posts gone? And if they have expired (RIP) / why do the posts dated march are wtill present?@  F I am looking for the thread on the "know a number of records in a file
 in one line".e   Thanks,a   D. --  B Compaq Computer (Schweiz) GmbH, Ueberlandstrasse 1, 8600 Dubendorf<      All opinions expressed are my own and not my employer's   ------------------------------  # Date: Sun, 24 Jun 2001 13:22:01 GMT'1 From: CSABA  HARANGOZO   <csabah@zipworld.com.au>s* Subject: Re: H-P EOLs PA-RISC Architecture8 Message-ID: <ZPlZ6.238$DJ4.36606@nostril.pacific.net.au>  * Jerry Leslie <leslie@clio.rice.edu> wrote: > From:   E >    http://www.crn.com/Components/Search/Article.asp?ArticleID=27517 * >    HP Plans Exit From PA-RISC Technology  I >   "After 20 years, Hewlett-Packard has finalized its exit strategy from I >    the microprocessor business and hopes to train its channel to becomef, >    experts on Intel's 64-bit architecture.  G >    After about three more refreshes, HP plans to pull the plug on its I >    PA-RISC line and then walk away,with no regrets, says Duane Zitzner,r3 >    president of the vendor's computing business."0  B 	Yep, that's what some HP sales-droids were telling us a few weeks. 	back at my workplace. How is that saying goes; 	"One OS to rule them... etc." ? Now we can add a new line, + 	"And one Microprocessor to rule them all".o  ( 	Good luck to HP... they will need it... 						   Cheers,    Csabat  I    ----------------------------------------------------------------------!E    * Csaba I. Harangozo     |    'To err is human', said the hedgehog E    * csabah@zipworld.com.au |           as he dismounted a wirebrush. I    ----------------------------------------------------------------------i;    EARTH::AUSTRALIA:[SYDNEY]HARANGOZO.CSABA;1, delete? [N]:h   ------------------------------  % Date: Sun, 24 Jun 2001 19:46:28 +0200m0 From: Didier Morandi <Didier.Morandi@Compaq.com>* Subject: PCSI_MENU 1.0 Part 1 ready for FT* Message-ID: <3B362773.1DB31F19@Compaq.com>  H I have terminated the coding and testing for the menu driven part of the@ PCSI-MENU tool. I recall that part 1 is the menu driven PCSI kitF generation, part 2 is the VMSINSTALL to PCSI translator, and part 3 isE the training feature (from the errors catched during kit creation and  then installation).l  G I would be pleased to know how many of you are interested to field test 	 the tool.r  E A field test is not playing with the tool, then fixing here and there < some bugs, then keeping it for him/herself without feedback.  + A field test is a commitment to come to thehE http://groups.yahoo.com/group/vmsinstal forum (or better send mail to ! the list) to report bugs so that:i   o  I can fix them "my way"$ o  others can get the latest version/ o  there are not different versions everywhere.u  [ If you wish to become FT for PCSI_MENU Part 1, please send me mail at Didier.Morandi@gmx.cht   Thanks.o D.  9 PS: the spec has been posted here last Wednesday at 18:22r   ------------------------------  % Date: Sun, 24 Jun 2001 11:57:01 +0100   From: Paul Sture <paul@sture.ch>S Subject: Re: Reward for the first of the next 50 posts: which company shouldbuy VMS:+ Message-ID: <VA.000003e2.065ba199@sture.ch>w  9 In article <009FDF3E.EA2C4910@SendSpamHere.ORG>, VAXman-   Brian Schenkenberger wrote:e   [snip] > D > I read news with Madgoat NEWSRDR.  It's a textual newsreader and ID > find it much easier to read and less problematic than Netscape.  IE > suggest you give it a try.  The nicest feature with NEWSRDR is that2F > I can invoke an editor that I am both comfortable and familiar with  > using.  C Does NEWSRDR provide the ability to work offline, as in downloadingo0 a whole newsgroup's unread messages in one shot?  C I am coming from the angle of minimum dialup connection times here. A Netscape on VMS doesn't cut that, as it will grab the headers but A wants to be online for the text of each message I choose to read.@  B Experience tells me it is far cheaper to download the lot and hang up as soon as possible.g   ___h
 Paul Sture Switzerlandl   ------------------------------  % Date: Sun, 24 Jun 2001 06:53:51 -0400T) From: "Neil Rieck" <n.rieck@sympatico.ca>g Subject: Re: sadness6 Message-ID: <VAjZ6.56$yy2.47793@news20.bellglobal.com>  : "Michael Austin" <miaustin@bellsouth.net> wrote in message' news:3B33D6A4.F52E8C8F@bellsouth.net...  >MG > Unfortunately in order to be a CEO you must 1st) get a lobotomy, 2nd)o
 drink someE > nasty voodoo mixture that makes you become a NYSE zombie, 3rd) make)	 decisionseI > based on the information spewed forth from stock analysts lips.  All ofe thisJ > results in companies becoming "one trick ponies".     You need only look as farK > as Digital to find that this is true. You also start to hear garbage likee "weaE > must get back to our core competencies"... then once they outsourceN everything,lH > they then find out that what they outsourced (at great expense I might add...)rG > WAS their core competency, and believe it or not, the IT shop of mostiK > manufacturing plants is included here.  Only the workers who disigned theA stuffLG > (9 times out of 10) really understand how it works and how to fix it.. >. > my $.02USD worth.  >0 > Michael Austin > DBA Consultant >O  C Your 2 cents is worth much more. BTW, out sourcing 90% of our IS/ITeL functions is exactly what happened at my company more than 24 months ago andG it should be no surprise that the contractor flipped most of our officecF systems over to a large Wintel solution at great expense to us. For 18F months a small group of DEC loyalists have been lobbying management toK return to OpenVMS on Alpha (to replace aging VAXs as well as start some newuI projects) or Tru64 on Alpha (to replace aging SUN and/or HP machines) and G were just starting to make some head way when these "dump Alpha" rumorss started.  
 Neil Rieck Kitchener/Waterloo/Cambridge,i Ontario, Canada.! http://www3.sympatico.ca/n.rieck/A   ------------------------------  # Date: Sun, 24 Jun 2001 13:38:48 GMT $ From: Scott Vieth <svieth@wi.rr.com>1 Subject: Re: Software to create PDF files on OVMSt) Message-ID: <3B35EDF3.D9DF4DC1@wi.rr.com>    txt2pdf from Sanface www.sanface.come   Thomas Steuver wrote:t  M > Is there a package like Adobe Acrobat that runs on OVMS Alpha to create PDF  > files from text files? >e	 > Thanks,  > Thomas Steuver > Northern Kentucky University   ------------------------------  % Date: Sun, 24 Jun 2001 01:16:06 -0500 * From: "T. S. Murphy" <murphyts@swbell.net>$ Subject: Re: VMS on IA64 (technical)0 Message-ID: <IAfZ6.207$E97.208161@nnrp2.sbc.net>  ? "Robert Deininger" <rdeininger@mindspring.com> wrote in message_F news:rdeininger-2306012213110001@user-2ive69s.dialup.mindspring.com...  I > I disagree.  Many VMS applications ARE performance-critical.  And VAXes-- > don't run a very big percentage these days.1  F I'm almost certain that a majority of VMS systems in service are still VAX'es.o  I Seriously, what high performance applications are running on VMS? For therB relatively small market which Alpha does have for high performance: computing, most of the systems are running Linux or Tru64.  E > Though I might agree if you said VMS performance concerns are often H > system-wide, not simple CPU power.  Emulators generally give lousy CPUG > performance, but you might arrange to keep most of the underlying I/OrL > performance.  Except the traditional Intel-based system doesn't handle I/O# > terribly well in the first place.o  H Of course, from benchmarks posted on this newsgroup in the past, we knowL that a Celeron running Linux outperforms VMS/Alpha on disk I/O by one or twoJ orders of magnitude. A Pentium 4 system has a much better memory subsystem; than any Alpha, and the same I/O system as any Alpha (PCI).   K > I think JF was talking about _software_ reliability.  Alpha + VMS PALcodeeJ > contain some features that make VMS much easier to write.  Specifically,L > the VMS code base from VAX was easier to port because of these features inJ > alpha. Many software wheels did NOT have to be re-invented, implemented,I > and debugged.  The result was VMS on alpha sooner and better than woulds > otherwise have been the case.t  L While true, it is not clear than PALcode offers an advantage which cannot beL leveraged on a generic architecture with a software layer between the OS and the processor.   ------------------------------  % Date: Sun, 24 Jun 2001 01:26:27 -0500 * From: "T. S. Murphy" <murphyts@swbell.net>$ Subject: Re: VMS on IA64 (technical)0 Message-ID: <oKfZ6.208$E97.211776@nnrp2.sbc.net>  5 "Stanley F. Quayle" <stan@stanq.com> wrote in message-' news:3B3520A3.5539.2CBFC7F@localhost...   @ > By the way, Alpha emulators do exist for Intel platforms.  ButD > there's not a lot of incentive to do that, since you can still get) > Alphas.  That's not the case for VAXen.a  H Is there seriously an Alpha emulator for Intel (x86?) platforms? A quick  Google search turned up nothing.  K There is definitely an incentive for that from the point of view of cost. I.L bet a good Alpha emulator (i.e. doing dynamic code translation) running on aK Pentium 4 would be fast - at least, much faster than an equivalently pricedmK Alpha system. Since the I/O system of a PC and an Alpha system are so closelI (both have PCI), it would make much more sense to emulate an Alpha than a D VAX (whose I/O system is entirely alien to the PC's). Though from anF architecture point of view, emulating a RISC instruction set on a CISCB computer poses some interesting challenges as far as optimization.   ------------------------------  % Date: Sun, 24 Jun 2001 03:06:24 -0400s- From: JF Mezei <jfmezei.spamnot@videotron.ca>i$ Subject: Re: VMS on IA64 (technical), Message-ID: <3B35914D.526E9BB8@videotron.ca>   "T. S. Murphy" wrote:eH > I'm almost certain that a majority of VMS systems in service are still	 > VAX'es.9  I But now that VAXes are no longer sold, I suspect that VAX usres will haveiJ migrate. And with Alpha effectively dead if we are to beleive the rumours,M those VAX users won't have much VMS to migrate to in the next couple of years- until VMS on IA64 is available.   L Would you migrate from vax to Alpha today, knowing that Alpha is going to be7 phased out and you'll have to migrate to IA64 shortly ?t  3 The timing to kill Alpha is very bad in my opinion.    ------------------------------  % Date: Sun, 24 Jun 2001 17:41:52 +1000a/ From: "Phil Howell" <phowell@snowyhydro.com.au>=$ Subject: Re: VMS on IA64 (technical)3 Message-ID: <XQgZ6.10456$qJ4.432383@ozemail.com.au>t  5 "T. S. Murphy" <murphyts@swbell.net> wrote in messageo* news:IAfZ6.207$E97.208161@nnrp2.sbc.net...A > "Robert Deininger" <rdeininger@mindspring.com> wrote in messagenH > news:rdeininger-2306012213110001@user-2ive69s.dialup.mindspring.com... >dK > > I disagree.  Many VMS applications ARE performance-critical.  And VAXest/ > > don't run a very big percentage these days.E >OH > I'm almost certain that a majority of VMS systems in service are still	 > VAX'es.  >G I almost certain you are wrongJ The price of maintenance on VAXen made is cost effective to move to alphas many years ago Phil <rest snipped>   ------------------------------    Date: 24 Jun 2001 05:46:27 -05009 From: Kilgallen@eisner.decus.org.nospam (Larry Kilgallen)t$ Subject: Re: VMS on IA64 (technical)3 Message-ID: <888ZwKwXq2S3@eisner.encompasserve.org>    In article <rdeininger-2306012213110001@user-2ive69s.dialup.mindspring.com>, rdeininger@mindspring.com (Robert Deininger) writes:a  I > Actually, the hardware support is nearly all in PAL, not the underlyinguK > alpha.  Look at the Alpha Architecture Reference Manual.  Compare the PALtE > routines for VMS to the ones for Unix.  The differences were mostlyO2 > motivated by the need to make the VMS port easy.  D Which is no different than saying the instruction set of the VAX wasD designed to make writing VMS easy in the first place (which it was).7 Hardware without operating systems is pretty miserable.a  H > If Intel does buy Alpha, something like PALcode will find its way into: > future CPU designs if the folks at Intel have any sense.  C Possibly they are prohibited from having PALcode now due to patents.> since PALcode might be argued to not be part of "chip" design.   ------------------------------  % Date: Sun, 24 Jun 2001 11:57:06 +0100   From: Paul Sture <paul@sture.ch>* Subject: Re: [OT] but now UK vs US English+ Message-ID: <VA.000003e6.065bb437@sture.ch>.  = In article <01K548UMXMC2001VKN@tgmail.tg.nsw.gov.au>,  wrote: ' > From: paddy.o'brien@zzz.tg.nsw.gov.aut > Newsgroups: comp.os.vms1, > Subject: Re: [OT] but now UK vs US English' > Date: Sat, 23 Jun 2001 19:30:40 +0010i >  > Paul,  > K > Much as you, I got quite excited when I saw that OED was going on-line.  nC > This was about a year ago, then I too saw the subscription price.t > ? > I live with the funny spellings (and even pronunciations) of ?L > Merriam-Webster, but it is free and usually the words that I need to look K > up are pretty similar in even the US variant of English.  Since I am not  L > considered a slouch in vocabulary, my dictionary search is more often for  > the esoteric.  > M > I still prefer to spell colour, honour, centre, theatre, etc. in UK style,  H > but that is probably being taken over since I left.  In .au, they are K > vaccilating between the spellings.  If you do crosswords, you have to be hI > aware of UK and US variants.  The crossword compilers are happy to use r: > either whichever fits in with the pattern of the puzzle. > H > Yes, whoever runs the OED now is probably outside the would-have-been K > wishes of the Fowler brothers or Sir Ernest Gowers.  These are the names n( > that come to mind for the Concise OED. > G > Merriam-Webster claims that its service is within the spirit of Noah aL > Webster, and I'm sure a free OED would have been in the spirit of those I , > mentioned, and others previously involved. >sM I remember an entertaining discussion here from two or three years ago about dK UK vs US English. It frustrated me at the time that I couldn't find any UK rK English dictionaries online, although M-W did a decent job. My Concise OED  5 was a 3 hour flight away and it was very frustrating.r  M > A quick BTW, I subscribe to M-W's word-of-the-day -- a mail service.  Even rI > words that are my regular currency, I find interesting.  They give the fM > definition(s), an example use and most interesting of all the etymology -- eH > about a standard paragraph in length.  I highly reccommend it -- what  > about you Norm (?) > K Thanks for the tip, I enjoy that kind of stuff as well. Cryptic crosswords iJ are a useful tool to preserve my English skills, especially when learning   another language (Swiss German). __
 Paul Sture Switzerland1   ------------------------------   End of INFO-VAX 2001.348 ************************