1 INFO-VAX	Sun, 25 Nov 2001	Volume 2001 : Issue 656       Contents:7 "Why Great Companies Fail": Includes DEC (perhaps HP ?) # Complete Running PDP-11 for Sale!!!  Re: DEC C Error on GS v7.03  dnsquery4 Re: Gartner and IDC say HP will effectively kill DLT Re: Life After Alpha Re: MUTUAL BUSINESS PROPOSAL Re: MUTUAL BUSINESS PROPOSAL Re: Netware client for OpenVMS@ Re: Of Bogusity and Benchmarketeering (was Re: Life After Alpha)@ Re: Of Bogusity and Benchmarketeering (was Re: Life After Alpha)@ Re: Of Bogusity and Benchmarketeering (was Re: Life After Alpha)@ Re: Of Bogusity and Benchmarketeering (was Re: Life After Alpha)# Re: Problem with a customers system # Re: Problem with a customers system A Re: RMS file structure internals documentation freely available ? E Re: Special IPF-Inside Issue of Shannon Knows Compaq at www.tru64.org E Re: Special IPF-Inside Issue of Shannon Knows Compaq at www.tru64.org E Re: Special IPF-Inside Issue of Shannon Knows Compaq at www.tru64.org E Re: Special IPF-Inside Issue of Shannon Knows Compaq at www.tru64.org E Re: Special IPF-Inside Issue of Shannon Knows Compaq at www.tru64.org E Re: Special IPF-Inside Issue of Shannon Knows Compaq at www.tru64.org E Re: Special IPF-Inside Issue of Shannon Knows Compaq at www.tru64.org E Re: Special IPF-Inside Issue of Shannon Knows Compaq at www.tru64.org E Re: Special IPF-Inside Issue of Shannon Knows Compaq at www.tru64.org E Re: Special IPF-Inside Issue of Shannon Knows Compaq at www.tru64.org E Re: Special IPF-Inside Issue of Shannon Knows Compaq at www.tru64.org E Re: Special IPF-Inside Issue of Shannon Knows Compaq at www.tru64.org E Re: Special IPF-Inside Issue of Shannon Knows Compaq at www.tru64.org E Re: Special IPF-Inside Issue of Shannon Knows Compaq at www.tru64.org E Re: Special IPF-Inside Issue of Shannon Knows Compaq at www.tru64.org E Re: Special IPF-Inside Issue of Shannon Knows Compaq at www.tru64.org E Re: Special IPF-Inside Issue of Shannon Knows Compaq at www.tru64.org E Re: Special IPF-Inside Issue of Shannon Knows Compaq at www.tru64.org E Re: Special IPF-Inside Issue of Shannon Knows Compaq at www.tru64.org E Re: Special IPF-Inside Issue of Shannon Knows Compaq at www.tru64.org E Re: Special IPF-Inside Issue of Shannon Knows Compaq at www.tru64.org E Re: Special IPF-Inside Issue of Shannon Knows Compaq at www.tru64.org E Re: Special IPF-Inside Issue of Shannon Knows Compaq at www.tru64.org E Re: Special IPF-Inside Issue of Shannon Knows Compaq at www.tru64.org E Re: Special IPF-Inside Issue of Shannon Knows Compaq at www.tru64.org E Re: Special IPF-Inside Issue of Shannon Knows Compaq at www.tru64.org E Re: Special IPF-Inside Issue of Shannon Knows Compaq at www.tru64.org E Re: Special IPF-Inside Issue of Shannon Knows Compaq at www.tru64.org E Re: Special IPF-Inside Issue of Shannon Knows Compaq at www.tru64.org E Re: Special IPF-Inside Issue of Shannon Knows Compaq at www.tru64.org E Re: Special IPF-Inside Issue of Shannon Knows Compaq at www.tru64.org A Re: StongARM-Outside (was Re: Of Bogusity and Benchmarketeering ) A Re: StongARM-Outside (was Re: Of Bogusity and Benchmarketeering ) A Re: StongARM-Outside (was Re: Of Bogusity and Benchmarketeering ) ( Re: Tech Wasteland: HWP/CPQ Profitless ?( RE: Tech Wasteland: HWP/CPQ Profitless ?H Re: Terminal emulation quality, was: Re: Installing CC060 on VMS VAX 6.1H Re: Terminal emulation quality, was: Re: Installing CC060 on VMS VAX 6.19 Re: Tru64.org IPF Consolidation Survey Still Taking Votes + Re: Why Compaq is so in pushing Wintel crap   F ----------------------------------------------------------------------   Date: 25 Nov 2001 17:04:43 GMT) From: leslie@clio.rice.edu (Jerry Leslie) @ Subject: "Why Great Companies Fail": Includes DEC (perhaps HP ?)' Message-ID: <9tr8bb$pi3$1@joe.rice.edu> ' Keywords: reasons,companies,fail,dec,hp    From:   6    http://news.cnet.com/news/0-1003-201-7943716-0.html4    Why great companies fail -  Tech News -  CNET.com      By strategy+business     Special to CNET News.com "    November 25, 2001, 6:00 a.m. PT  7   "The inspiration for Clayton M. Christensen's seminal I    theory on disruptive technology came from watching Digital Equipment's G    fall in 1988. How could the management team that had been rightfully G    lauded for its brilliance by every popular business publication have     stumbled so badly?   F    As Digital's star fell, the business press blamed the ineptitude ofF    the company's management. But Christensen observed that every otherI    minicomputer company collapsed at the same time. Since no one colludes H    to fail, something more was at work. He concluded that the reason forF    the implosion of the minicomputer industry was not just the rise ofC    the personal computer, but what the PC represented: a disruptive D    technology to which the minicomputer companies could not respond.  ?    His theory of disruptive technology became the basis of "The H    Innovator's Dilemma: When New Technologies Cause Great Firms to Fail"I    (Harvard Business School Press, 1997). Named the best business book of G    that year by the Financial Times/Booz Allen Hamilton Global Business @    Book Awards, its influence was all the more potent because itH    prescribed no new management rules or standard solutions, but insteadE    confronted companies with Christensen's chilling vision of how the     world works.   I    A company can do all the right things--listen to its customers, invest I    in research and development, compete aggressively--and yet fall victim E    to a new technology or business plan that seemed, at first, almost     irrelevant.  	    [snip]   C    Businesses get blindsided because they focus on their best, most G    profitable customers and ignore other potential markets or customers C    seeking lower-cost products. This narrow view, Christensen says, H    ignores the fact that every market is characterized by three distinct    change trajectories:   I    o  Performance improvement that customers can readily use (that is, it (       matches their own changing needs).  <    o  Technology advances driven by sustaining technological       improvements.   B    o  New performance introduced by a disruptive technology, whichD       typically begins at a lower level of performance, but rapidly ?       improves until it meets the majority of customers' needs.   @    It is the tendency of all successful companies to match theirH    performance to their most demanding customers, exceeding the needs ofC    most of their customers, which creates an opening for disruptive     technologies, he says.   	    [snip]   E    At Hewlett-Packard, Dick Hackborn just seemed to have an intuition D    that you couldn't do disruptive and sustaining things in the sameI    organization. He was responsible for many of HP's moves into printers, %    workstations, scanners, and so on.   D    I fear that Carly Fiorina's intuition is exactly opposite: To cutD    costs, she's combining all of HP's computer businesses within oneI    organization--all the printer businesses in one, and so on. This might E    make short-term financial sense, but in the long run it undoes the D    architecture that enabled HP to dodge so many disruptive bullets.  	    [snip]   B    You can invest to create the new growth business while the coreF    business is still growing, because new business units don't need toG    get big fast. But when the core business stops growing, investing to H    create new growth businesses becomes impossible. To prop up the stockI    price, managers have to turn down the screws on everybody. That forces F    them to cancel all the projects that would lead to future growth inF    order to drop money to the bottom line. This is HP's dilemma today.G    Once a company's growth has stopped, the game as we have known it is     over. It's a scary thing..."   H The statement about "growth has stopped" could also apply to a company's6 viewpoint about a product, such as Compaq's about VMS.    4 --Jerry Leslie     (my opinions are strictly my own)   ------------------------------    Date: 25 Nov 2001 02:05:04 -0800. From: wintersystems@yahoo.com (Winter Systems), Subject: Complete Running PDP-11 for Sale!!!= Message-ID: <4db2fd92.0111250205.529f5210@posting.google.com>   A http://cgi.ebay.com/aw-cgi/eBayISAPI.dll?ViewItem&item=1298384370   # Complete Running PDP-11 for Sale!!! = eBay auction ends 9:00am Eastern on Monday, 26 November 2001.   0 Don't miss this opportunity to have your own ...  ; Some details below, but see the eBay listing for full info:   D This Digital PDP11/23 Plus has been in continuous use since the lateE 1970's and is still up and running. We have migrated the database off D of it and need to free up some space in the server room, take a loadF off of our air-conditioning unit, and save some electricity. We wantedF to try converting it to a LINUX-based web server as a pet project, butF don't have the time or energy to do it. The possible uses are endless:A Make it the newest member of your PDP-11 production environment.  D Scrap it for much needed parts or see what gold you can extract from
 this baby.  > Convert it to a liquor cabinet. Maybe you'll set a new trend. C Quench that burning desire that you've had since the 1970's to have  your own PDP-11.@ BID NOW AND BE THE ENVY OF YOUR FRIENDS!!! THE TALK OF YOUR USER GROUP!!!  F Here's the catch ... This is an all or nothing deal. Take what you can find, but take it all, please!    > This is a cash and carry deal. All equipment is located in the? Downtown Boston area. The buyer must arrange for all details of > pick-up or shipping, including packing. Cash or money order is@ expected at the time of pick-up within 30 days of auction close,% unless special arrangements are made.   F ALL EQUIPMENT IS SOLD AS IS. What you see is what you get. NO REFUNDS.  C We may ask you to sign a statement indicating that all data will be  erased and not shared.  C The Digital model number is PDP-11/23-B/E. I believe that the OS is 3 RSX 11M PLUS, but I don't know what version number. D Based upon information that I gathered from a couple of websites, itF seems that the processor was upgraded from KDF11-A M8186 LSI-11/23 CPUC to KDJ11-A M8192 LSI-11/73 CPU, 4KW cache. The latter is installed,  and I have the original.@ The Emulex fixed disk drive controller also appears to have been> upgraded TU0210401 Rev C to 002110402 Rev C (BR5904). Both areE installed. There are two multicolored ribbons coming out of the older A board and going to the large Control Data FDD. The newer board is D connected to a Control Data FDD 94166-182 PN 77774320 dated 88/5/31.E This is black and about 5 1/2" wide. It is installed in the slide out  chassis on the top. D There are loose blades including a printer adapter, Netcom KPV-1180, and M7504-DEQNA Ethernet. - The VT220 Digital terminals are version 2.1.  F Please do not take any of the above specs as law. I am unfamiliar withB PDP-11's and got as much information as I could with the resources
 available.   ------------------------------  % Date: Sun, 25 Nov 2001 11:10:36 +0100 2 From: martin@radiogaga.harz.de (Martin Vorlaender)$ Subject: Re: DEC C Error on GS v7.03; Message-ID: <3c00c39c.524144494f47414741@radiogaga.harz.de>   , Richard L. Dyson (rickdyson@home.com) wrote:H > I am working on getting GS v7.03 working for my OpenVMS systems.  I amI > using Compaq C v6.4-008 on an OpenVMS/Alpha v7.2-1 system.  I get quite L > a way into the build when I get the following errors that are entirely new > to me: [...] > > CC/NODEBUG/OPTIMIZE/DECC/PREFIX=ALL/NESTED_INCLUDE=PRIMARY -/ > /NAME=(AS_IS,SHORT)/DEFINE=("HAVE_MKSTEMP") - H > /INCLUDE=([.obj] ,[.src]) /OBJECT=[.obj]gdevpdfo.obj  [.src]gdevpdfo.cE > Assertion failure:  Deleting instruction with DefinesRoutineCtx set  > Instruction:  M > COD INSTRUCTION  BNE  : NEXT=0109B848, PREV=0109B498, LOCATOR={17224:1-30},  > S...F >          DEFINES_ROUTINE_CTX, OPCODE=66, NOT_IN_CURRENT_RTN, OP1=R1,
 > OP2=Targ... C > %GEM-F-ASSERTION, Deleting instruction with DefinesRoutineCtx set 1 > %TRACE-F-TRACEBACK, symbolic stack dump follows L >   image    module    routine             line      rel PC           abs PC ... ? >  DECC$COMPILER  GEM_FI_PEEP_ALPHA  GEM_FI_PEEP_APPLY_PEEPHOLE @ >                                           733 00000000000001C4 > 000000000063F804 ... $ > 	Does anyone have any suggestions?  B I haven't seen this before, either. But given the "PEEPHOLE" clue:. How does it go if you compile it /NOOPTIMIZE ?   cu,    Martin --  D                     | Martin Vorlaender    |    VMS & WNT programmer-   Smiert Spamionem  | work: mv@pdv-systeme.de D                     |       http://www.pdv-systeme.de/users/martinv/4                     | home: martin@radiogaga.harz.de   ------------------------------  % Date: Sun, 25 Nov 2001 13:21:01 -0500 - From: Michael Austin <miaustin@bellsouth.net>  Subject: dnsquery - Message-ID: <3C01368D.B72C3E34@bellsouth.net>   
 OpenVMS 7.2-1 
 TCPIP 5.0A  F Has anyone ported the unix dnsquery tool to OpemVMS.  It allows one to2 find out if a particular domain name is available.   Thanks Michael Austin   ------------------------------  # Date: Sun, 25 Nov 2001 17:30:34 GMT 1 From: "David J. Dachtera" <djesys.nospam@fsi.net> = Subject: Re: Gartner and IDC say HP will effectively kill DLT ' Message-ID: <3C012E32.40F2662D@fsi.net>    Tim Llewellyn wrote: >  > "David J. Dachtera" wrote: >  > > K > > Price points will be another issue. At $100+ (US) per cartridge, though A > > that has come down a bit, DLT-III and later DLT-IV were never  > > economical.  > 2 > Economical? How valuable is your employers data?  D This, like cost-to-acquire vs. TCO, is probably the longest standing battle for data integrity.  E Bean counters only understand the Cash Flow, Balance Sheet, P&L, etc. D They don't understand the value of IP, the contents of the database,G etc. since it tends to be nearly impossible to quantify, and has little 6 or no impact on the stock price, profits or dividends.   I usually word it these ways:   : 1. "How much money can you AFFORD to save (or not spend)?"  F 2. "That's the difference between what *CAN* be done and what *SHOULD*	 be done."   H A losing battle in both cases. If you can't punch it up on a calculator,8 lay it out on a spreadsheet, etc., they ain't list'nin'.   --   David J. Dachtera  dba DJE Systems  http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/    ------------------------------  # Date: Sun, 25 Nov 2001 16:02:37 GMT - From: Terry C Shannon <shannon@world.std.com>  Subject: Re: Life After Alpha C Message-ID: <Pine.SGI.4.30.0111251058080.5290-100000@world.std.com>   ( On Fri, 23 Nov 2001, David Froble wrote:   > Tim Llewellyn wrote: >  > >  > > Madman wrote:  > >  > > P > >>I stopped flying United some time ago because it was obvious that profit wasL > >>much more important than customer satisfaction - which brings us back to@ > >>Compaq!  I knew I could steer this back on topic....  -- Ian > >> > > H > > Did you read John McLeans recent post? It doesn't really appear that$ > > Compaq are interested in profit. > >  > > regards  > >  > >  >  > Nor customer satisfaction! >   C It may be a little-known fact, but CPQ does have folks in charge of B customer sat. Bo McBee (sp) at the top of the heap, Neil Davies in
 BCSG-land.  H These folks pay attention to the www.compaqworkinggroup.org survey. TheyJ also pay attention (and perhaps even *more* attention) to the surveys that4 Ken Farmer and I have been running at www.tru64.org.  F The current survey should be open for another week or ten days. Please) take the time to throw in your two cents.   G Ken and I will be compiling a couple more surveys, the content of which I will be influenced by the valuable feedback we've received from Bill Todd  and others in this newsgroup.   * Thanks for your participation and support.   terry s   H PS-- DISCLAIMER: I make no claims for the current level of customer sat,% especially WRT the IPF Consolidation.    ------------------------------  % Date: Sun, 25 Nov 2001 09:32:38 +0100 ( From: Paul Sture <paul.sture@bluewin.ch>% Subject: Re: MUTUAL BUSINESS PROPOSAL - Message-ID: <VA.000004ca.94f2ba67@bluewin.ch>   = In article <a_NL7.2884$Oh1.33902@insync>, Jerry Leslie wrote:t- > PIUS KOLADE (pkolade@totalise.co.uk) wrote:e > : ENGR. PIUS KOLADER > : 18 COWRIE HOUSEe > : LAGOS, NIGERIA.  >G [snip]  8 > The U.S. Secret Service has a web page for this fraud: > + >    http://www.treas.gov/usss/alert419.htmR >    USSS Operation 4-1-9  > K >   "...If you have been victimized by one of these schemes, please forwardwK >    appropriate written documentation to the United States Secret Service, I >    Financial Crimes Division, 950 H Street, NW, Washington, D.C. 20001,h7 >    or telephone (202) 406-5850, or contact by e-mail.f > H >    If you have received a letter, but have not lost any monies to thisA >    scheme, please fax a copy of that letter to (202) 406-5031."i > ( > This site is dedicated to such frauds: >  >     http://www.quatloos.com/ > P A well known fraud. A few years ago the Nigerian govt. was taking out full page - ads in the UK press to warn folks about this.r ___r
 Paul Sture Switzerlande   ------------------------------  + Date: Sun, 25 Nov 2001 12:59:23 +0000 (UTC)  From: david20@alpha2.mdx.ac.uk% Subject: Re: MUTUAL BUSINESS PROPOSAL + Message-ID: <9tqpva$1jv$2@aquila.mdx.ac.uk>p  X In article <VA.000004ca.94f2ba67@bluewin.ch>, Paul Sture <paul.sture@bluewin.ch> writes:> >In article <a_NL7.2884$Oh1.33902@insync>, Jerry Leslie wrote:. >> PIUS KOLADE (pkolade@totalise.co.uk) wrote: >> : ENGR. PIUS KOLADE >> : 18 COWRIE HOUSE >> : LAGOS, NIGERIA. >> >[snip]  >c9 >> The U.S. Secret Service has a web page for this fraud:r >> e, >>    http://www.treas.gov/usss/alert419.htm >>    USSS Operation 4-1-9 >> aL >>   "...If you have been victimized by one of these schemes, please forwardL >>    appropriate written documentation to the United States Secret Service,J >>    Financial Crimes Division, 950 H Street, NW, Washington, D.C. 20001,8 >>    or telephone (202) 406-5850, or contact by e-mail. >> nI >>    If you have received a letter, but have not lost any monies to thiseB >>    scheme, please fax a copy of that letter to (202) 406-5031." >> I) >> This site is dedicated to such frauds:r >> e >>     http://www.quatloos.com/n >>Q >A well known fraud. A few years ago the Nigerian govt. was taking out full page q. >ads in the UK press to warn folks about this. >___ >Paul Stureo >Switzerland >.  ) See http://home.rica.net/alphae/419coal/    ? for more information and how to report this in other countries.n    
 David Webb VMS and Unix team leader CCSS Middlesex University   ------------------------------  % Date: Sun, 25 Nov 2001 10:34:41 -0500o* From: Chuck Chopp <ChuckChopp@rtfmcsi.com>' Subject: Re: Netware client for OpenVMSq+ Message-ID: <3C010F91.41E01311@rtfmcsi.com>s   John Eisenschmidt wrote:   > It depends on the version of Netware you're running. Netware 6 has support for native everything - NFS, AFS, SMB, etc. If you're running that your best bet it to load NFS support and use VMS's support for NFS.a >w > If you're not running 6 (which most people aren't yet) you might be stuck. I'm pretty sure there is NFS support in earlier versions, but I doubt you'll get VMS to speak directly to Netware. The client for the Mac is even pretty marginal./ >  > Best of luck,  > John >aE > >>> "Mike Scott" <mscott_NOSPAM@axys.com> 11/22/2001 1:04:11 PM >>>aK > Here's a crazy one.  I'm looking for a Netware client that will run on mynL > OpenVMS 7.1 workstation.  All my research has turned up so far is a freaky; > solution for making a VMS box look like a Netware server.-   NFS is by far the best choice for allowing OpenVMS to access files on a remote NetWare server.  There are NFS servers that can run on NetWare v4.x, v5.x and v6.x, and there are several TCP/IP packages that run on OpenVMS v5.x, v6.x and v7.x% that provide NFS client capabilities.d   There was actually a product that ran on OpenVMS that emulated a NetWare386 system, which was basically a hybrid NetWare 2.x/3.x emulation.  It was not very reliable, was limited [I think] to only using the Ethernet_II frame type, and it didfa not allow the OpenVMs [client] to NetWare [server] type of connectivity that you are looking for.      Regards,   Chuckn -- Chuck Choppt  8 ChuckChopp@rtfmcsi.com            http://www.rtfmcsi.com0                                   ICQ # 22321532@ RTFM Consulting Services Inc.     864 801 2795 voice & voicemail2 103 Autumn Hill Road              864 801 2774 fax4 Greer, SC  29651                  800 774 0718 pager7                                   8007740718@skytel.come   ------------------------------   Date: 25 Nov 2001 15:37:28 GMT& From: peter@abbnm.com (Peter da Silva)I Subject: Re: Of Bogusity and Benchmarketeering (was Re: Life After Alpha)o% Message-ID: <9tr37o$16d@web.nmti.com>e  6 In article <20011123093908.24231.qmail@gacracker.org>,? Doc.Cypher  <Use-Author-Supplied-Address-Header@[127.1]> wrote:o9 > On 23 Nov 2001, peter@abbnm.com (Peter da Silva) wrote:n. > >In article <3BFD2635.561B567E@ecubics.com>,- > >emanuel stiebler  <emu@ecubics.com> wrote:h > >> David Froble wrote:K > >> > Why, when it comes to Intel and Microsoft, we're suppost to wait forB> > >> > some promised wonder, when everybody else 'has it now'?  K > >> Because you have to wait until THEY "invent" it . (VLIW, multitasking,r > >> ....) e  6 > >Any day now they'll have something like security...  - > I thought they had something like security.n  L I guess. In the same way a Chevy Corvette is something like a Navy Corvette.   -- e+  `-_-'   In hoc signo hack, Peter da Silva.oE   'U`    "A well-rounded geek should be able to geek about anything." L                                                        -- nicolai@esperi.org          Disclaimer: WWFD?   ------------------------------  # Date: Sun, 25 Nov 2001 16:27:05 GMTr- From: Terry C Shannon <shannon@world.std.com> I Subject: Re: Of Bogusity and Benchmarketeering (was Re: Life After Alpha)cC Message-ID: <Pine.SGI.4.30.0111251117280.5290-100000@world.std.com>c  % On Sat, 24 Nov 2001, Bill Todd wrote:e   >i: > Terry C Shannon <shannon@world.std.com> wrote in message@ > news:Pine.SGI.4.30.0111241206150.16466-100000@world.std.com... >a > ...r >cM > > I never heard that EV8 didn't work (in fact, the only thing that could orfH > > could not work would be a simulation as they sure as hell don't haveM > > bootable silicon) but I've heard from several senior technical types thath< > > EV8 posed some significant issues, SMT being among them. >nF > You *really* have to be careful when parsing Compaq statements.  ForN > example, few in the industry would question the fact that applying SMT to anL > architecture of any complexity would pose significant issues (just imagineN > the difficulty Itanic is going to have in this area) - but the real issue isM > whether these issues, which have been being closely examined for years now, 5 > were on-track with respect to being solved for EV8.o >eK > The answer I've heard consistently from the design team is that they wereeL > very much on track indeed, with no showstoppers in sight.  In other words,L > Compaq's statement about this, while not an outright lie, appears to be as0 > deliberately misleading as so many others are. >   J My opinion (based on discussions with CPQ folks) is that the EV8 folks hadG encountered no show-stoppers and may well have been on track. I'll givedC Compaq the benefit of the doubt when they claim that EV8 would havenG required flawless and timely execution. I'll also accept the claim thatpF EV8 would have required incremental design resources and money had the project gone forward.   D Regardless of what Compaq says about a "technical decision," I thinkC economics loomed large in the decision. As a complete architectural D redesign, EV8 would have been a difficult undertaking. And given theJ woefully limited market addressability of Alpha (sans Windows support as aF volume expansion solution) I believe Compaq did the math and concluded? that a continued. post-EV7 effort was not economically prudent.i  J Could I be wrong? Of course. But I haven't seen any concrete evidence thatG this is the case. Should such evidence materialize, I'd be the first touI say that I was wrong. Being wrong is one thing, refusing to admit same in.1 light of factual data is something else entirely!I   ------------------------------  # Date: Sun, 25 Nov 2001 16:33:34 GMTB- From: Terry C Shannon <shannon@world.std.com>vI Subject: Re: Of Bogusity and Benchmarketeering (was Re: Life After Alpha)tC Message-ID: <Pine.SGI.4.30.0111251128070.5290-100000@world.std.com>P  ( On Sat, 24 Nov 2001, David Froble wrote:   > Bill Todd wrote: > F > Well, in this one, Compaq could be absolutely right.  If there is noE > money to continue the project, then EV8 definitely won't be made toc4 > work, unless it's already working, which it's not. >o >yM > > The answer I've heard consistently from the design team is that they wereeN > > very much on track indeed, with no showstoppers in sight.  In other words,N > > Compaq's statement about this, while not an outright lie, appears to be as2 > > deliberately misleading as so many others are. >r< > Ah, slight correction.  One big showstopper.  R&D funding. >t  J Amen to that! FYI one of the biggest gaffes I've committed in SKC over theD past couple of years involves a write-up I did on Compaq's R&D. ThisJ write-up was based on a presentation delivered by a senior Compaq Research engineering manager.  G A couple of months after the writeup hit the street, I was chagrined to I learn that Compaq was quitely "rightsizing" its R&D programs and that thei- R&D workforce and budget were deeply slashed.o  H Many projects were tabled or killed... the Piranha CMP project being oneG of them. The research team at Bond University on Australia's Gold Coast.. (MilliCent, MediaVista, et al) was nuked. Etc.  5 Compare Compaq's R&D spend with that of Sun or IBM...    ------------------------------  # Date: Sun, 25 Nov 2001 16:45:20 GMT-* From: "Bill Todd" <billtodd@metrocast.net>I Subject: Re: Of Bogusity and Benchmarketeering (was Re: Life After Alpha)hA Message-ID: <Ae9M7.92132$qx2.5751781@bin5.nnrp.aus1.giganews.com>g  : "Terry C Shannon" <shannon@world.std.com> wrote in message= news:Pine.SGI.4.30.0111251117280.5290-100000@world.std.com...o   ....  F > Regardless of what Compaq says about a "technical decision," I thinkE > economics loomed large in the decision. As a complete architecturaleF > redesign, EV8 would have been a difficult undertaking. And given theL > woefully limited market addressability of Alpha (sans Windows support as aH > volume expansion solution) I believe Compaq did the math and concludedA > that a continued. post-EV7 effort was not economically prudent.h  I I suggest that Compaq seems to have missed the fact that *not* continuing-K efforts past EV7 *also* appears not to be economically prudent.  I.e., thatoE in toting up the numbers it neglected to evaluate the impact on salesdI (especially during the period when Alpha's operating systems were not yet-K available on Itanic) that the Alphacide would have.  Given that the cost oftK continued development, even as stated by Compaq, would have been relatively3E minor compared with the profits that Alpha had until the announcementrK returned, that appears to have been a decisive error (or indicates that ther8 decision was driven by issues other than economic ones).  J This, of course, even without considering the possibilities for increasingH Alpha's market penetration by demonstrating some corporate interest in aK platform that achieved its modest but profitable success in spite of ratherS than because of its owner.   > L > Could I be wrong? Of course. But I haven't seen any concrete evidence that > this is the case.e  K When choosing to break long-standing explicit commitments to its customers,OK the onus is on the *corporation* to prove its case rather than on others to L prove beyond any shadow of a doubt that it is faulty.  Compaq still owes itsL customers a thorough justification of its decision and detailed responses toH the legitimate questions raised about it, and shows no sign of providing either.    - bill   ------------------------------  % Date: Sun, 25 Nov 2001 08:47:26 +0000G From: Roy Omond <Roy@Omond.net> , Subject: Re: Problem with a customers system) Message-ID: <3C00B01F.C8EB0839@Omond.net>D   "David J. Dachtera" wrote:   > Paul Savage wrote: > >o
 > > Hi AllO > > Firstly let me tell you I have absolutly no idea about VMS systems, but....  > >  > > [... snip snip ...]  > H > I would recommend contacting a consulting software person in your areaJ > as this will likely not be solved via a Usenet newsgroup. Sound too muchD > like an application issue, and those can take man-years to resolve3 > without hands-on contact with the subject system.,  B Agreed.  I'm available Tuesdays and Thursdays (near Cambridge, butI willing to travel).  If you're near Bristol, you could try Tim Llewellyn.   	 Roy Omond- Blue Bubble Ltd.   ------------------------------  + Date: Sun, 25 Nov 2001 12:37:02 +0000 (UTC)i From: david20@alpha2.mdx.ac.uk, Subject: Re: Problem with a customers system+ Message-ID: <9tqole$1jv$1@aquila.mdx.ac.uk>0  h In article <9tot6a$3ds$1@newsg2.svr.pol.co.uk>, "Paul Savage" <paul@the-savages.freeserve.co.uk> writes: >Hi AlliL >Firstly let me tell you I have absolutly no idea about VMS systems, but.... > L >I work for a company that has recently aquired another company who producedI >a very clever software package on a VMS system. We have now had to startL >supporting these systems. >eM >One of these customers has recently had a serious power cut on their system. J >When thay have tried to restart the VMS system they have tried to enter aK >log off password, this didn't work correctly. They then tried to enter theSK >log on name and password. The system came back with file not found errors.d >b  H From the above I assume the system fully crashed due to the power outageF ie you didn't just lose the connections and your UPS battery backup or- generator didn't keep the VMS system running.yM I further assume the system has completed the reboot and you can login at theAC console (If not you need to ascertain where the reboot is failing).a  O If so what is this logoff password ? - the system would have lost all currentlyt$ running processes when it went down.  L I would check that all disks have been mounted correctly when the system was
 rebooted.  Do a t  
 show device do  J command to see all the disks. It maybe that the startup file had not been N updated last time changes were made to the system and it is not setup to mount all the disks.  M If all the disks are mounted correctly then assuming the error identifies the2N missing file try to find it on the disk. It maybe that the powercut has caused a disk problem.>    
 David Webb VMS and Unix team leader CCSS Middlesex University    J >My question is this, Is there a way of being able to start the VMS system >easily. >pK >Again I will apologise for maybe not giving a good enough description, and + >also for being a complete novice with VMS.  >l >Paul Savage >Software Engineer >s >    ------------------------------  % Date: Sun, 25 Nov 2001 11:24:22 +0100!( From: Paul Sture <paul.sture@bluewin.ch>J Subject: Re: RMS file structure internals documentation freely available ?- Message-ID: <VA.000004cc.95590456@bluewin.ch>   K In article <BrbL7.34376$xS6.59300@www.newsranger.com>, Simon Clubley wrote:t0 > On Wed, 21 Nov 2001 22:44:48 +0100, in article6 > <VA.000004c6.832e8c0f@bluewin.ch>, Paul Sture wrote: > >.Q > >I did manage to dig up a reference to an RMS-11 Internals manual in old DECUS eR > >SIGtapes. How to get hold of them in readable format I don't know (did I see a  > >mention of BRU in there?).  > >bQ > >http://www.ibiblio.org/pub/academic/computer-science/history/pdp-11/decus/11s1  > >01.html       > >n >  > Found it. :-)t > P > Noting that the document at the above URL is from the RSX SIG Spring Symposium4 > Collection for 1988, digging around further gives: > i > http://www.ibiblio.org/pub/academic/computer-science/history/pdp-11/rsx/decus/rsx88a/373310/rmsint2.docb > M Well, er, yes. Here's a question: Out of that 140 pages did you find anything6 of practical use?n ___c
 Paul Sture Switzerlandi   ------------------------------  # Date: Sun, 25 Nov 2001 07:02:02 GMTe& From: "Jeff Killeen" <Jeff@IDM-IO.com>N Subject: Re: Special IPF-Inside Issue of Shannon Knows Compaq at www.tru64.org7 Message-ID: <KH0M7.1281$zf.116224@typhoon2.gnilink.net>s  5 "Bill Todd" <billtodd@metrocast.net> wrote in messagef6 news:P70M7.35241$YD.3016334@news2.aus1.giganews.com...  5 > If there were any technical basis for your opinions   L What you have failed grasp is this isn't primarily a technical issue - it is. a business issue driven by market realities...   ------------------------------  # Date: Sun, 25 Nov 2001 07:08:43 GMTt& From: "Jeff Killeen" <Jeff@IDM-IO.com>N Subject: Re: Special IPF-Inside Issue of Shannon Knows Compaq at www.tru64.org7 Message-ID: <%N0M7.1284$zf.123228@typhoon2.gnilink.net>q  5 "Bill Todd" <billtodd@metrocast.net> wrote in message ; news:3f0M7.105913$2w.6435586@bin4.nnrp.aus1.giganews.com...   J > It has been the near-complete absence of any official Compaq response inK > *any* forum to substantive questions about the facts behind the June 25th@H > decision that has been surprising (or not, depending on one's existingG > impressions of Compaq 'responsiveness').  And people at his level arecJ > commonly found participating in forums like comp.arch on such subjects - atJ > least when they're comfortable putting their personal reputations on the3 > line rather than merely their corporate personae.:  L I would be extremely concerned about Compaq's judgement if they choose to doJ so given the level of professionalism demonstrated to date and the lack of: due diligence being given to today's business realities...   ------------------------------  # Date: Sun, 25 Nov 2001 08:22:32 GMTu* From: "Bill Todd" <billtodd@metrocast.net>N Subject: Re: Special IPF-Inside Issue of Shannon Knows Compaq at www.tru64.org< Message-ID: <cT1M7.35887$YD.3065558@news2.aus1.giganews.com>  1 "Jeff Killeen" <Jeff@IDM-IO.com> wrote in messageo1 news:KH0M7.1281$zf.116224@typhoon2.gnilink.net...c > 7 > "Bill Todd" <billtodd@metrocast.net> wrote in message 8 > news:P70M7.35241$YD.3016334@news2.aus1.giganews.com... >c7 > > If there were any technical basis for your opinionsa >oK > What you have failed grasp is this isn't primarily a technical issue - it. is0 > a business issue driven by market realities...  L What *you* have failed to grasp is that Compaq *made* this a technical issue> when it attempted to clothe its dubious business decision withG pseudo-technical justifications.  Had it had the sense not to, the main2L questions (though there still would have been plenty) would have simply beenF about its business competence or lack thereof, and questions about itsD integrity would at least have been limited to its abandonment of itsI long-standing and oft-reiterated 'commitments' rather than having also to B deal with its lying attempts at post-announcement rationalization.   - bill   ------------------------------   Date: 25 Nov 2001 08:34:11 GMT- From: djweath@attglobal.net (Dave Weatherall)pN Subject: Re: Special IPF-Inside Issue of Shannon Knows Compaq at www.tru64.org5 Message-ID: <DTiotGxQ0bj6-pn2-0CPgdyctKTxR@localhost>   E On Sun, 25 Nov 2001 07:02:02, "Jeff Killeen" <Jeff@IDM-IO.com> wrote:t   > 7 > "Bill Todd" <billtodd@metrocast.net> wrote in messages8 > news:P70M7.35241$YD.3016334@news2.aus1.giganews.com... > 7 > > If there were any technical basis for your opinionsu > N > What you have failed grasp is this isn't primarily a technical issue - it is0 > a business issue driven by market realities...  E Might be. But since when did DEC/ComPaq have a handle on its markets.r  B If your argument has merit, it is because the words 'Servers' and F 'Windows' are inextricably linked with the acronyms 'PC' and 'IA-32'. B VMS and Alphas are not in the same 'market' (well not exclusively F anyway). In that area, you ( and maybe the PC market) are right, 'goodF enough' applies. But that isn't the market/selling point/differential " that maintained VMS. We're spoilt.  F The argument you are seeking to make confirms the perception of ComPaq as a (failing) PC company.  E Isn't it ironic tho' that it's ComPaq reporting IA64 'sightings' and p0 not delivering systems. And there is no Plan B!!   --   Cheers - Dave.   ------------------------------  % Date: Sun, 25 Nov 2001 11:24:21 +0100e( From: Paul Sture <paul.sture@bluewin.ch>N Subject: Re: Special IPF-Inside Issue of Shannon Knows Compaq at www.tru64.org- Message-ID: <VA.000004cb.9559044c@bluewin.ch>e  L In article <W8UL7.1624$3n6.134716@typhoon1.gnilink.net>, Jeff Killeen wrote:   [snip]   > O > ....remember folks that is NOT chip to chip but server to server for the sameAH > dollars - PLEASE READ THAT AGAIN - IT IS SERVER TO SERVER FOR THE SAME" > DOLLARS THAT CUSTOMER LOOK AT... >oO For a brand new system, with all new applications, maybe. And then Sun and IBM  " et cetera are also in the bidding.  N What's missing from your equation is the fact that we customers out here have P many millions invested in applications software. Figures that make the price of # a top end server look like peanuts.o  M 20 or so years ago it was the norm to move to a different platform every few eE years and rewrite everything in the process, hopefully improving the .N application along the way. Little did we think that the applications we wrote O then might still be in use today (hence much of the Y2K problem - many assumed sF that a given application would be completely rewritten by that stage).  P But we've moved on from that model. If a change of chip means porting, it is no G longer true that customers only look at "server to server for the same e	 dollars".   M Want a real life example? We have an application which we considered porting 1N from VMS a couple of years ago. The estimated cost came in at several million N dollars, although the project leader told me it could easily run to much more C than that due to the application's complexity. So it stayed on VMS.0  N Force us to port it due to a different chip and yes, we might well spend tens P of millions doing that (think testing, testing, testing, as well as development O and more testing). However if we feel _forced_ to port by Compaq/HP decisions, oN you can bet a large chunk of that sum we are going to go Sun / IBM / whatever   else is available at that point.  N So from the perspective as a customer with a heavy investment in applications O which satisfy our business needs today I am saying that "IT IS MUCH MORE THAN   @ SERVER TO SERVER FOR THE SAME DOLLARS THAT CUSTOMERS LOOK AT..." ___t
 Paul StureE Neither a "3rd shift operator" nor a "bitter ex-Digital employee" :-)s   ------------------------------  % Date: Sun, 25 Nov 2001 12:25:06 +0100-1 From: John McLean <mcleanj@swissonline.delete.ch>cN Subject: Re: Special IPF-Inside Issue of Shannon Knows Compaq at www.tru64.org5 Message-ID: <3C00D512.12E3D70F@swissonline.delete.ch>    Jeff Killeen wrote:e >  ...   M > Allow me to suggest another line of thinking.  I believe BOTH HP and CompaqIL > basically have come to the same conclusion from different directions - theL > marketplace no longer values leading edge hardware.  That in fact hardwareL > is rapidly becoming something the market only values at commodity pricing.G > This perception is counter to the soul of both Compaq and Digital buttM > something the market is proving out.  For those of you who aren't long time M > Compaq customers Compaq became what it did be being the first in many cases N > to bring leading edge technology to PC's.  Compaq's business model was basedK > on the customer will pay a premium price for systems that push technologymL > edge.  Dell's overtaking of Compaq's PC business was very much a result ofF > Compaq's business model becoming invalid in the marketplace.  The PCG > marketplace switched from valuing premium technology to valuing pricetN > because the commodity technology was perceived as being good enough.  If youM > are Compaq once that happens to you on desktops and low end servers you laytN > awake at night wondering when Linux and Windows XX will eat your mid to high > end premium servers. > K > The 6/25/2001 wasn't about Intel controlling Compaq.  It was about CompaqiL > coming to believe the marketplace will no longer pay a premium for premiumI > hardware because the marketplace thinks that commodity hardware is goodnM > enough.  Remember Compaq isn't run by a hardware or software engineer theseAK > days - it is run by a former customer.  Once one sees the decision in thecG > context that premium hardware is not valued by the marketplace all of M > Compaq's decisions make sense.  Now one may disagree with basic premise bute5 > within the context of the premise it makes sense...     H Your hypothesis probably holds true in the PC and general low-end of theG business.  Anything that works is good enough.  I think though that you - have identified just one of several factors. p  - My list (and probably very incomplete) is ...e  F - late 90's were a boom time for PC sales because of Y2K issues.  I amH not aware of any PC company who acknowledged the temporary nature natureF of this, either at the time or since.  If they'd been smart they wouldF have identified this at the time and made only temporary expansions of their companies.    E - Year 2000 saw the IT industry riding the internet wave and ignoringeF the laws of "economic gravity" which say hot air will only support youF so far and then you have to start making real money.  While PCs shouldE have sold well - they must have because Comaq made a profit on them -uB the bigger sellers would have been web servers (which might be why@ Compaq's "Enterprise" segment almost doubled their 1999 profit).  E - Year 2001 has seen the internet wave (unsurprisingly) come crashing)E down taking the IT industry and the general economy with it.  IT gear.G was a basic need of every dot com, but with the collapse we see reducedMG demand and we see a market flooded with second-hand IT gear.  (I hadn'taF thought of it before, but perhaps the $697 million discrepancy betweenH the income in Compaq's original year 2000 accounts and the income in theF April 2001 revision of their segments could be "bad debts" from failed: dot com companies.  $421 million of the discrepancy was in "Enterprise".)  D - As you say, there is an increasing attitude, especially at the lowH end, that anything that does the job is good enough.  (Hell, I'm writingD this on a Windows98 system - it's far from perfect but I am familiarF with the bugs and the solutions.  I see no need to swap to a whole newG set of bugs...)  With the general economy in a slump most companies arerH asking the serious questions about the benefits of upgrading their PCS. F Do most of the staff really need 1.4Ghz systems for Word documents and! email ?  The answer has to be No.,  H - To some extent Microsoft have lowered expectations about windows-basedF operating systems through the highly publicised stability and securityE problems.   No wonder "near-enough is good-enough" for commercial and A personal comsumers; that does seem to be Micrsoft's attitude too.d  E - Also on Microsoft, IIRC, they have adjusted their charging/paymentstG system in order to preserve their income.  The market is so competitiverD that manufacturers have been forced to absorb some of these modified2 charges, thus reducing their margins even further.  B - PC manufacture and distribution is a labour-intensive business. H Assembly, transport, warehousing, sales distribution and (of course) allG cost effort, time and money.  The high-end doesn't suffer anywhere near.E as much for an initial sales, then unlike PCs there's every chance ofeB ongoing income from that sale in the form of maintenance, support,G software licensing and if you are lucky, permanent on-site support ....r    H If this posting makes you queston - yet again - the competence of CompaqE management then I know you are not alone.  Compaq's fixation with PCshG has got them into this mess.  They honestly cannot see any further thanq$ the PC on the desk in front of them.  C High-end systems such as VMS are far more stable, have few security-E problems, produce better initial profit, produce ongoing income ...  n  H (I get depressed when I keep stating the bleedin' obvious, trying to get3 a blind and deaf management team to take notice...)      John McLean:   ------------------------------  % Date: Sun, 25 Nov 2001 12:43:23 +0100o1 From: John McLean <mcleanj@swissonline.delete.ch>.N Subject: Re: Special IPF-Inside Issue of Shannon Knows Compaq at www.tru64.org5 Message-ID: <3C00D95B.6858861E@swissonline.delete.ch>w   Jeff Killeen wrote:  >   
 ...(snip)    I > Since Fenwick's name is now being mentioned in public I have heard fromwJ > enough Compaq sources for it to be super credible that what prompted theL > whole review of the viability of Alpha was Fenwick's work on the follow-onG > to the Wildfire.  Something that keeps getting lost in these internet-K > discussions is that systems, and not chips, is what customers buy.  It is2C > the end result system, such as Wildfire, that really delivers the J > performance.  It was Fenwick's team that came to the conclusion that theC > Alpha based follow-on to the Wildfire wouldn't offer a sufficient M > performance advantage over the IA64 servers that Compaq is also developing..  E You give the impression that only Wildfire boxes will be available inhF future and that is why future Alpha performance on WF was so critical.  C Sure Wildfires are nice boxes but are they really providing all the A benefits that Compaq said they would ?   Are many sites using thehG dynamic reallocation of resources ? Global memory ?  Sure the footprintpF is smaller but each instance of VMS requires its own system management- and that effort is not significantly changed.h  F In short, would the difficulties between the Alpha and the "follow-on"H to Wildfire be better solved by rethinking parts of the Wildfire conceptD rather than changing processor ?  Especially changing to a processorB whose capabilities can only be described right now as "wish-ware".     John McLeanA   ------------------------------  % Date: Sun, 25 Nov 2001 07:50:42 -0500f- From: JF Mezei <jfmezei.spamnot@videotron.ca> N Subject: Re: Special IPF-Inside Issue of Shannon Knows Compaq at www.tru64.org, Message-ID: <3C00E91F.54A290BA@videotron.ca>   Jeff Killeen wrote: N > What you have failed grasp is this isn't primarily a technical issue - it is0 > a business issue driven by market realities...  P Why then, as you said, was the Alpha murder supposedly initiated by a techy ????  J Had Alpha not been killed, and Compaq relegated IA64 boxes to small wintelI servers, then only HP would have been building "serious" servers based onaL IA64. And I doubt that IA64 would have gotten the volumes necessary to drive3 its cost of production down to make it "commodity".w  H Since it will take some time before the 8086 is dead and IA64 is used onM desktops, it will take an equal amount of time before IA64's costs become lowp enough for it to be commodity.  H If companies cease to even try to compete against intel because they areM convinced that anything INtel produces in the future , crap or not, will take L over the world, then there is a serious monopoly issue here. IA64 has yet toJ prove itself and companies have already abandonned competing products ???? Sorry, I call that anti-trust.  K How come AMD feels it can make a competing chip to IA64, but Alpha couldn't4T compete ? Where there is a will, there is a way. Compaq simply didn't have the will.  I Remember that prior to june 25, Compaq was saying that IA64 was a flawed,oN bloated architecture that would come back to haunt Intel whenever it would tryH to increase its performance to 1-beat the 8086 and 2-narrow the gap withN Alpha. What has changed so suddently to say that IA64's architecture will makeM i catch up to Alpha so fast Alpha should be murdered right away to donate thel1 enginers to Intel to quicken the pace even more ?     N And even if the gap narrows, is it wise to tell your customers you are killingN their platform years and years before IA64 has a chance of actually overtakingL Alpha ? Consider that to the installed base (which is the main customer baseF for Alpha since Compaq refuses to market these systems), any change inG architecture is a real pain in the ass. Why force customers into such ad- migration just because the gap is narrowing ?g  I Had Compaq continued development of Alpha until such a time that IA64 hadlN proven itself and had surpassed Alpha's performance, then the facts would haveF supported the decision to kill Alpha. But the way I seee it, Intel wasI struggling with IA64 and gave Compaq some seet incentive to not only killiN Alpha, but also move its serious systems over to IA64 prematurely to give IA64G a boost. This tells me that IA64 wasn't doing too well in the projectedfM marketplace and that Intel has to take active steps to force people to switchwK to IA64. Compaq was an easy target with a spineless accountant as its chiefl1 and a Microsoft-licker (Winkler) as a top aide.     M My hope is that the poor Digital engineers transplanted by force to the enemy N (Intel) will not fight for their ideas. Just present them and if rejected willI just accept it . This way, IA64 will not benefit from their expertise and W Intel engineers will continue their business as usual and IA64 will lag and be delayed.n   ------------------------------  % Date: Sun, 25 Nov 2001 08:54:23 -0500u( From: Bill Gunshannon <bill@cs.uofs.edu>N Subject: Re: Special IPF-Inside Issue of Shannon Knows Compaq at www.tru64.orgB Message-ID: <20011125085239.G13614-100000@server2.cs.scranton.edu>  - On Sun, 25 Nov 2001, Joann Difrancesco wrote:h   >tL > Mr. Fenwick is a senior system architect, not a marketing droid, hence his; > absence from this forum shouldn't be all that surprising.. >   J Do I take it then that you see Hoff as a marketing-droid??  (Or any of the, other Compaq worker-bees who frequent here!)   bill   -- FJ Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesD bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. University of Scranton   |> Scranton, Pennsylvania   |         #include <std.disclaimer.h>   ------------------------------  # Date: Sun, 25 Nov 2001 14:25:52 GMT65 From: "Joann Difrancesco" <joanndifrancesco@home.com>nN Subject: Re: Special IPF-Inside Issue of Shannon Knows Compaq at www.tru64.org> Message-ID: <Qb7M7.115694$pb4.70850492@news2.rdc2.tx.home.com>  1 "Jeff Killeen" <Jeff@IDM-IO.com> wrote in messaged2 news:xtTL7.1615$3n6.134522@typhoon1.gnilink.net... > H > "Bob Kaplow" <kaplow_r@eisner.encompasserve.org.mars> wrote in message/ > news:b+4fh7WSuUOP@eisner.encompasserve.org...t > >e > > >> Yep. David Fenwick. > >tK > > David was not part of the Alpha chip team, but was the guy who designede > thelL > > Turbolaser (8400) and Wildfire (GS320) platforms. As a user of both, I'mF > > most impressed with his teams work. I can't believe he'd rather be "intel > > inside". >lI > Since Fenwick's name is now being mentioned in public I have heard from J > enough Compaq sources for it to be super credible that what prompted theL > whole review of the viability of Alpha was Fenwick's work on the follow-onG > to the Wildfire.  Something that keeps getting lost in these internet0K > discussions is that systems, and not chips, is what customers buy.  It iseC > the end result system, such as Wildfire, that really delivers the.J > performance.  It was Fenwick's team that came to the conclusion that theC > Alpha based follow-on to the Wildfire wouldn't offer a sufficient A > performance advantage over the IA64 servers that Compaq is alsoi developing.  > % > I know enough of Fenwick to know...e >.6 > 1) He had the data and folks outside of Compaq don't >nD > 2) He looked at this from the standpoint of the end product system	 deliveredn3 > to the customer and not just esoteric chip issuesm >PJ > 3) There was no reason to expect he would fall on his sword and kill his9 > project i.e. there was nothing anti-Alpha about the guyk > I > ...remember folks that is NOT chip to chip but server to server for thed sameH > dollars - PLEASE READ THAT AGAIN - IT IS SERVER TO SERVER FOR THE SAME" > DOLLARS THAT CUSTOMER LOOK AT... >P  L Now why would those silly customers look beyond the chip? That would be likeL looking beyond the number of cupholders, color of the brush-guards, or cubic7 inch displacement of the engine in a brand new SUV? ;-}    ------------------------------  # Date: Sun, 25 Nov 2001 14:36:43 GMTw5 From: "Joann Difrancesco" <joanndifrancesco@home.com>nN Subject: Re: Special IPF-Inside Issue of Shannon Knows Compaq at www.tru64.org> Message-ID: <%l7M7.115699$pb4.70864634@news2.rdc2.tx.home.com>  5 "Bill Todd" <billtodd@metrocast.net> wrote in message 6 news:8xXL7.32383$YD.2870459@news2.aus1.giganews.com... >o3 > "Jeff Killeen" <Jeff@IDM-IO.com> wrote in messageE  L > > take the word of Mr. Fenwick.  Rather than pointless debate I will trustL > > readers to make up their own mind as to who is a more credible source... >hK > Then again, we have only Jeff's word on what the estimable Mr. Fenwick is  > supposed to have said,  H I don't know which sessions you attended at CETS2001 in Anaheim, but youI apparently missed the Monday session during which Mr. Fenwick spoke aboutoI Alpha, IPF, and the decision to go to an Intel-Inside strategy. Fenwick's L current project, the post-Marvel switched fabric and processor Blade server,J was designed to support multiple architectures (including Intel and Alpha)H right from the start; Fenwick and his team apparently pursued a strategy# that gave them maximum flexibility.e   ------------------------------  # Date: Sun, 25 Nov 2001 14:42:39 GMTn& From: "Jeff Killeen" <Jeff@IDM-IO.com>N Subject: Re: Special IPF-Inside Issue of Shannon Knows Compaq at www.tru64.org8 Message-ID: <zr7M7.2141$3n6.160342@typhoon1.gnilink.net>  > "John McLean" <mcleanj@swissonline.delete.ch> wrote in message/ news:3C00D95B.6858861E@swissonline.delete.ch...eE > Sure Wildfires are nice boxes but are they really providing all thepC > benefits that Compaq said they would ?   Are many sites using theu5 > dynamic reallocation of resources ? Global memory ?   G This is high end leading edge stuff today.  3 years from now it will beoI common run of the mill stuff.  With the Quick Blade and InfiniBand fabric L dynamic reallocation of resources will become mainstream.  Heck you are even% now seeing as a concept in PC land...a  % http://www.compaq.com/manage/rpm.htmlc  " ...that will become more robust...   ------------------------------  # Date: Sun, 25 Nov 2001 14:55:37 GMTS& From: "Jeff Killeen" <Jeff@IDM-IO.com>N Subject: Re: Special IPF-Inside Issue of Shannon Knows Compaq at www.tru64.org8 Message-ID: <JD7M7.2157$3n6.160031@typhoon1.gnilink.net>  : "JF Mezei" <jfmezei.spamnot@videotron.ca> wrote in message& news:3C00E91F.54A290BA@videotron.ca... > Jeff Killeen wrote: J > > What you have failed grasp is this isn't primarily a technical issue - it iss2 > > a business issue driven by market realities... >eG > Why then, as you said, was the Alpha murder supposedly initiated by an
 techy ????  . It was a senior engineering program manager...   ------------------------------  % Date: Sun, 25 Nov 2001 10:05:29 -0500e- From: JF Mezei <jfmezei.spamnot@videotron.ca>aN Subject: Re: Special IPF-Inside Issue of Shannon Knows Compaq at www.tru64.org, Message-ID: <3C0108AD.C2E83762@videotron.ca>   Joann Difrancesco wrote:L > was designed to support multiple architectures (including Intel and Alpha)J > right from the start; Fenwick and his team apparently pursued a strategy% > that gave them maximum flexibility.h  M Ahhh, perhaps this is why Alpha suddently lost much of its edge against IA64.pL If they had to compromise their wildfire systems design to the lowest commonM denominator to also accept IA64, then it would explain why the Alpha all of aiC sudden no longer had a huge edge over IA64 on wildfire type systemsa   ------------------------------  # Date: Sun, 25 Nov 2001 14:51:20 GMTo" From: Art Rice <arice@myhouse.org>N Subject: Re: Special IPF-Inside Issue of Shannon Knows Compaq at www.tru64.org9 Message-ID: <Iz7M7.951$h56.162176@paloalto-snr1.gtei.net>d   JF Mezei wrote:r   > Terry C Shannon wrote: >> >E >> > WARNING: 'Mixed Cluster' may not mean Vax+Alpha+itanic... It wasaC >> > used to mean Alpha+IPF, and when asked, the Q people passed ona >> > that one... >> > >> aE >> That would seem to be a reasonable assumption from a CPQ financialnI >> standpoint. Perhaps not so reasonable to the ~150K VAX users still out  >> there, though.b > I > What would be so difficult in allowing VAX to cluster with an IA64 ? IfmD > the VMS code on IA64 is to be common with that of Alpha, and alphaK > supports VMS as a fellow cluster member, I don't see why it would be suchn: > a big deal to allow IA64 to support VAX cluster members. > L Seems like that would be much easier than in our world.  Of course, even if I the Q kept Alpha,  It couldn't be mixed with the current MIPS processors oJ being used by Tandem.  Apparently, the Tandem engineers have wasted a lot K of time beginning to port to Alpha and now have to retrace their steps and rE begin porting to Itanium.  And guess who will have to pay for that...i   -- e Art Rice, Tandem Admin Special Data Processing Corp ----------------------------L All opinions are my own and do not reflect the views of the above mentioned 	 employer.T   ------------------------------  # Date: Sun, 25 Nov 2001 15:21:43 GMTr* From: "Bill Todd" <billtodd@metrocast.net>N Subject: Re: Special IPF-Inside Issue of Shannon Knows Compaq at www.tru64.org@ Message-ID: <b08M7.54525$8q.8866035@bin2.nnrp.aus1.giganews.com>  > "John McLean" <mcleanj@swissonline.delete.ch> wrote in message/ news:3C00D95B.6858861E@swissonline.delete.ch...t   ...9  H > In short, would the difficulties between the Alpha and the "follow-on"J > to Wildfire be better solved by rethinking parts of the Wildfire conceptF > rather than changing processor ?  Especially changing to a processorD > whose capabilities can only be described right now as "wish-ware".  I Jeff has no clue about this area.  The 'follow-on' to Wildfire is Marvel,tF which is specifically designed around EV7 and is still being touted byL Compaq as *the* place to be until their optimistic estimate of 2004 - 5 whenE Itanic supposedly magically grows hydro-foil water-wings and suddenlys catches up.y  J And there *are* no difficulties taking the next Alpha step (EV8) down thatK road, because EV8 was designed to present essentially the same interface to9H the surrounding box that EV7 did, so would slip right in and deliver its! improved performance immediately.o  G Itanic (right through the Madison/Deerfield members, which are McKinleypJ semi-clones), however, won't slip into a Marvel box - or anything like theI Marvel architecture - at all without added glue to replace what is on thesJ EV7 chip but is not on the Itanic chip.  There is no chip yet on the IntelE road map that can match EV7 in such areas (let alone EV8's additional E features), and thus - even leaving aside Itanic's major per-processornL performance deficit compared with Alphas that extends as far into the futureJ as either company had been willing to provide information - the suggestionE that any Itanic would have *ever* fitted into the Marvel architecture F anywhere nearly as well as, let alone better than, Alpha is ludicrous.  L Wildfire, in part because of its delay in reaching the market, turned out toC be a bit of a disappointment in areas such as memory access latency J (possibly a bit in bandwidth as well).  The on-chip features in EV7 shouldK dramatically alter that situation in Marvel.  The absence of those featuresmK on Itanic chips seems likely to make Itanic-crippled Marvel systems performnJ much more like Wildfires (but with slower processors than the then-current Alphas).  D And (to address some FUD from another portion of the discussion) theA suggestion that Itanic-based servers will be able to offer better H cost/performance metrics than Alpha-based servers holds no water either.G Not only will it take something like twice as many Itanic processors tokH provide equivalent performance (and server prices usually rise more thanK linearly with the number of processors, due to the added effort required tocJ glue them together effectively - *especially* lacking the kinds of on-chipH support that EV7 provides but Itanic does not), but the price differenceH between an N-processor Alpha box and an N-processor Itanic box (offeringK dramatically less capability) should be minimal for anything but very smallSL values of N, because processor prices (assuming that Itanic's ever *do* comeE down) just don't contribute much to the cost of a mid-range or larger  server.p  K So to suggest that users would not have been willing to pay a 'premium' foroG better technology is a red herring:  Alpha servers can and could in thedF future offer significantly better raw cost/performance at the hardwareH level - meaning that *Itanic* systems would be those demanding a premiumE (and making the additional advantages offered by the better operatingm' systems on Alpha platforms pure gravy).,  H In sum, Compaq's arguments in these areas don't stand up to scrutiny anyL better this time around than when they were presented last summer.  And theyE won't the next time it trots them out either unless and until it addsIE significant *new* information content that could alter the underlying 	 analysis.i   - bill   ------------------------------  # Date: Sun, 25 Nov 2001 15:30:53 GMT * From: "Bill Todd" <billtodd@metrocast.net>N Subject: Re: Special IPF-Inside Issue of Shannon Knows Compaq at www.tru64.orgA Message-ID: <N88M7.116892$dk.8572547@bin1.nnrp.aus1.giganews.com>o  @ "Joann Difrancesco" <joanndifrancesco@home.com> wrote in message8 news:%l7M7.115699$pb4.70864634@news2.rdc2.tx.home.com...   ...   J > I don't know which sessions you attended at CETS2001 in Anaheim, but youK > apparently missed the Monday session during which Mr. Fenwick spoke aboutoK > Alpha, IPF, and the decision to go to an Intel-Inside strategy. Fenwick's F > current project, the post-Marvel switched fabric and processor Blade server,.L > was designed to support multiple architectures (including Intel and Alpha)J > right from the start; Fenwick and his team apparently pursued a strategy% > that gave them maximum flexibility.K  G Ah - so perhaps Jeff was simply confused (no surprise there) and really I meant the *Marvel* follow-on rather than the *Wildfire* follow-on that hey was talking about.  F I have not heard much yet about post-Marvel plans, so can't comment inK detail (pointers to information would be appreciated).  It's certainly easyhE to understand that *if* one is given the edict from above (or somehowvI concludes from a brain-fart of one's own) that a server architecture must4L accommodate both Alpha and Itanic processing components, then the ability toH take advantage of Alpha's strengths is significantly crippled and use ofK Alpha processors may indeed not provide the performance boost that it could K in an architecture better tailored to take advantage of it.  That, however,mJ just calls the premise into question rather than validates the conclusion.   - bill   ------------------------------   Date: 25 Nov 2001 15:51:48 GMT& From: peter@abbnm.com (Peter da Silva)N Subject: Re: Special IPF-Inside Issue of Shannon Knows Compaq at www.tru64.org% Message-ID: <9tr42k$2n4@web.nmti.com>?  7 In article <VkXL7.1154$zf.107841@typhoon2.gnilink.net>, % Jeff Killeen <Jeff@IDM-IO.com> wrote:A7 > "Bill Todd" <billtodd@metrocast.net> wrote in message < > news:k7XL7.54536$uB.9422317@bin3.nnrp.aus1.giganews.com...N > > So exactly what did he go on the record as saying in this regard?  And how
 > > publicly?e  N > I believe he confirmed the closing gap at the Monday Q&A panel.  I know thisI > was going to be covered in his session on 9/11/2001 which was obviously 
 > canceled...   G I think it's obvious that the gap in performance between Alpha and IA64dH is closing... given the current performance if IA64 one would have to beH even more skeptical than I to assume that Intel won't do better in roundA two. On the other hand, there's room for an awful lot of closing.a   -- a+  `-_-'   In hoc signo hack, Peter da Silva.tE   'U`    "A well-rounded geek should be able to geek about anything."pL                                                        -- nicolai@esperi.org          Disclaimer: WWFD?   ------------------------------  # Date: Sun, 25 Nov 2001 16:16:46 GMT - From: Terry C Shannon <shannon@world.std.com>hN Subject: Re: Special IPF-Inside Issue of Shannon Knows Compaq at www.tru64.orgC Message-ID: <Pine.SGI.4.30.0111251109180.5290-100000@world.std.com>   % On Sat, 24 Nov 2001, Bill Todd wrote:    >aF > Bob Kaplow <kaplow_r@eisner.encompasserve.org.mars> wrote in message/ > news:b+4fh7WSuUOP@eisner.encompasserve.org...o@ > > In article <87y9kwyah1.fsf@prep.synonet.com>, Paul Repacholi! > <prep@prep.synonet.com> writes: 5 > > > Terry C Shannon <shannon@world.std.com> writes:s > > > , > > >> On 13 Nov 2001, Paul Repacholi wrote:F > > >> >BTW, does anyone know the name of 'the engineer' who said that6 > > >> >Alpha would lose it performance advantage yet? > > >l > > >> Yep. David Fenwick. > > >lE > > > This was the name I got on Thursday at the local version of thelB > > > IPF update. If anyone has an e-mail address, I would like toA > > > extend him the chance to stop his name being taken in vain.  > >w5 > > firstname.lastname@compaq.com usually works well.- > > K > > David was not part of the Alpha chip team, but was the guy who designedd > thejL > > Turbolaser (8400) and Wildfire (GS320) platforms. As a user of both, I'mM > > most impressed with his teams work. I can't believe he'd rather be "intelN > > inside". >fM > Unless the alternative were to be out of a job (because Compaq was going totI > kill Alpha come hell or high water and the only question was whether iti3 > would kill its server design team along with it).a >r  D David Fenwick would have no problem getting a job as a senior system6 architect with any enterprise IT vendor on the planet.    N > This is consistent with an early 'clarification' that the 'concerns' in thisI > area came from the server group rather than the Alpha design group.  ItnM > really doesn't take too much imagination to frame a question such as "Given N > the eventual creation on IA64 of EV7-like on-chip glue to enhance memory andG > MP performance, will the performance disparity between Alpha and IA64aL > shrink?" that can be answered in the affirmative by someone more concernedF > with how to put together a decent server at a reasonable price usingL > available components than about specific performance time-tables, customer) > migration grief, and corporate profits.e  H Note that the post-Marvel fabric and processor blade system was designedG from the get-go to support multiple CPU architectures. Alpha and HammerwJ disappeared from the equation right around the time Stimac and the rest of- the Houston Server Mafia jumped ship for RLX.      >mL > Funny how invisible David appears to have become since his name started toN > be specifically associated with this, though.  You'd think he'd be right outM > in front defending the details of the decision, given his apparent level ofhL > involvement - if indeed he came right out and said Alpha couldn't continueB > to compete effectively rather than something far more subject to > interpretation.d >t  J I presume that the comments made by Mr. Fenwick and others at CETS2001 andD on various concalls were influenced by Compaq Political Correctness.  H Having spent a lot of time at DECpaq, I am not suprised that Mr. FenwickC opted to design a platform that would not of necessity be tied to a I specific processor. In fact, early in the WildFire development cycle, onetJ faction within Compaq argued passionatel for an Alpha-and-IPF design. ThisH contingent lost the battle, but not the war. The 2H02 McKinley-based 32PC system from the ISSG will incorporate a lot of WildFire technology.)   ------------------------------  # Date: Sun, 25 Nov 2001 16:52:54 GMT - From: Terry C Shannon <shannon@world.std.com> N Subject: Re: Special IPF-Inside Issue of Shannon Knows Compaq at www.tru64.orgC Message-ID: <Pine.SGI.4.30.0111251144071.5290-100000@world.std.com>S  % On Sun, 25 Nov 2001, Bill Todd wrote:E  3 > "Jeff Killeen" <Jeff@IDM-IO.com> wrote in message 4 > news:xtTL7.1615$3n6.134522@typhoon1.gnilink.net... >t > ...e >fK > > Since Fenwick's name is now being mentioned in public I have heard fromgL > > enough Compaq sources for it to be super credible that what prompted theN > > whole review of the viability of Alpha was Fenwick's work on the follow-onI > > to the Wildfire.  Something that keeps getting lost in these internet4M > > discussions is that systems, and not chips, is what customers buy.  It isoE > > the end result system, such as Wildfire, that really delivers the0 > > performance. >wJ > Excellence in system design can certainly make an average processor lookL > good - witness SGI.  But of course it can make an excellent processor lookJ > far better - and can only make a sub-standard processor like Itanic look > marginally acceptable. >f> >   It was Fenwick's team that came to the conclusion that theE > > Alpha based follow-on to the Wildfire wouldn't offer a sufficienthC > > performance advantage over the IA64 servers that Compaq is alsoe
 > developing.u > 	 > Wrongo.  > N > There's no mention yet by Intel of any Itanic product that can hold a candleK > to EV7 in a large server, because of EV7's on-chip glue to enhance memoryp, > access and multi-processor configurations.   WRT EV7 this is true!s   > And even Compaq is stillK > contending that EV7 will be a winner that will easily hold the fort until-K > the (optimistic) 2004 -5 date (bet on 2005 - and then only because of thelM > massive infusion of talent Compaq gave up) when Compaq (but not to the bestCN > of my knowledge Intel) has said that the first Itanic with Alpha engineering) > influence should wallow into the ocean.t >hJ > So if any such decision was reached, it *definitely* didn't apply to theN > Wildfire follow-on (Marvel, with EV7s for its first engines), or to anythingI > else before the second half of this decade (the earliest one could haveiG > expected such on-chip glue from Intel *without* the Alpha engineering  > infusion).  & That's right. EV7 is not at isue here.    A > And since EV8 would have presented pretty much the same face tooG > the outside world that EV7 did (i.e., its enhancements were primarilygN > internal - e.g., SMT), there's every reason to believe that its improvementsM > would have merged transparently right into the same high-performance system M > architecture that EV7 pioneered (and that Itanic couldn't have even startedeF > to take advantage of for at least a year or two after EV8 appeared).    H Same external face. Yeah, to the best of my knowledge. But note that EV7G is the EV68 core with big L2 cache and the glueless SMP support via the 	 IO7 chip.e    > > > ' > > I know enough of Fenwick to know...n > >i8 > > 1) He had the data and folks outside of Compaq don't >eJ > But the Alpha engineering team certainly did, and without exception they3 > have said that Compaq's assertions were bullshit.n >   I Then either Fenwick is wrong or the Alpha engineering team is wrong. This,> might make for an interesting debate between the two factions.     > >aF > > 2) He looked at this from the standpoint of the end product system > deliverede5 > > to the customer and not just esoteric chip issuesr >eK > While the design of the wagon they pull is also important, the difference L > between a Clydesdale and a profusely-sweating pig as motive power does not  > exactly qualify as 'esoteric'. >e > >iL > > 3) There was no reason to expect he would fall on his sword and kill his; > > project i.e. there was nothing anti-Alpha about the guys > >oK > > ...remember folks that is NOT chip to chip but server to server for they > sameJ > > dollars - PLEASE READ THAT AGAIN - IT IS SERVER TO SERVER FOR THE SAME$ > > DOLLARS THAT CUSTOMER LOOK AT... > ? > And so far Alpha is 'way out in front in both performance andoN > cost-performance metrics.  That will continue with EV7 unless Compaq decidesI > to soak its captive Alpha customers even more than it does already (sayMN > moo).  That would have continued again with EV8, especially given that IntelK > would have had to compete against rather than take advantage of the Alpha  > team's expertise.a >dJ > The only area where the cost difference between an Alpha processor and aJ > smoking brick of death would make any noticeable difference (even if youI > ignore Alpha's significant per-processor performance advantage) is in adK > low-end 64-bit platform where no special high-performance multi-processor4K > system glue is required and commodity components can be used.  Of course,uM > Alpha hasn't made any money in that market segment for a long time if ever,uJ > and Itanic will have to compete there with sub-$1K SPARC boxes, its IA32M > brethren, and (one hopes) Hammer (all of which should have noticeably lowertI > per-processor costs than Itanic and *at least* equivalent performance).  >iM > So there is still no reason whatsoever to believe that Alpha would have hadtM > any problem for the foreseesable future in attracting customers away from aaM > cruise on the Itanic on the basis of both performance and cost-performance.-K > I suppose there remains a possibility that Fenwick knows something no oneeJ > else does - but as long as he keeps it to himself (which makes it ratherI > difficult to evaluate, let alone question) I suspect that opinions here. > won't change much. >l  G Alpha may well have had no difficulty attracting customers away from an8I exercise in rearranging the deck chairs on the Itanic, but note that saids@ Alpha customers would be limited to VMS and Tru64 staterooms, or* doule-bunks in the Linux steerage section.  C Marketing malfeasance and stupid strategy tricks severely curtailedlF Alpha's addressable market. Assuming flat VMS sales (actually down 6-7D percent in Q2 and Q3) and modest growth in Tru64 (say from 5 percentC market share to 6-8 percent) there wasn't a heck of a lot of growth  opportunity.   ------------------------------  # Date: Sun, 25 Nov 2001 16:57:16 GMTe- From: Terry C Shannon <shannon@world.std.com>gN Subject: Re: Special IPF-Inside Issue of Shannon Knows Compaq at www.tru64.orgC Message-ID: <Pine.SGI.4.30.0111251154260.5290-100000@world.std.com>d  $ On Sat, 24 Nov 2001, JF Mezei wrote:   > Jeff Killeen wrote:gH > > Given the choice between believing a self-admitted bitter ex-DigitalO > > employee and a the chief designer of the TurboLaser and the Wildfire I willu! > > take the word of Mr. Fenwick.m >vM > Mr Killeen, while the above argument does hold validity, you must also takebM > into consideration the possibility that Mr Fenwick has no choice by provide O > spin to support his employer's decision. Most Compaq employees have to do thewN > same, especially during presentation to customers where it becomes sometimesK > very apparent that the presenters are not saying what they truly believe.o > I > Since these wildfire boxes don't materialise overnight, I would be veryeM > curious to see what this same Mr Fenwick was saying about IA64 last year. IeP > wouldn't be surprised to have heard him blast IA64 as an inferior architecture; > that would have a lot of difficulty catching up to Alpha.     H In a private discussion that took place over a year ago Mr. Fenwick madeG some interesting observations about IPF. He clearly did NOT dismiss the  IPF threat out of hand.e   ------------------------------  # Date: Sun, 25 Nov 2001 17:00:49 GMTa- From: Terry C Shannon <shannon@world.std.com>rN Subject: Re: Special IPF-Inside Issue of Shannon Knows Compaq at www.tru64.orgC Message-ID: <Pine.SGI.4.30.0111251159150.5290-100000@world.std.com>r  % On Sun, 25 Nov 2001, Bill Todd wrote:o  J > It has been the near-complete absence of any official Compaq response inK > *any* forum to substantive questions about the facts behind the June 25thsH > decision that has been surprising (or not, depending on one's existingG > impressions of Compaq 'responsiveness').  And people at his level areeM > commonly found participating in forums like comp.arch on such subjects - ateJ > least when they're comfortable putting their personal reputations on the3 > line rather than merely their corporate personae.i >y  I Hmmm.. we appear to be in violent agreement on the subject of the lack oftE official Compaq response. That's one reason Ken and I are running theb surveys.   ------------------------------  # Date: Sun, 25 Nov 2001 17:09:27 GMT - From: Terry C Shannon <shannon@world.std.com> N Subject: Re: Special IPF-Inside Issue of Shannon Knows Compaq at www.tru64.orgC Message-ID: <Pine.SGI.4.30.0111251207050.5290-100000@world.std.com>   + On Sun, 25 Nov 2001, Bill Gunshannon wrote:w  / > On Sun, 25 Nov 2001, Joann Difrancesco wrote:s >r > >rN > > Mr. Fenwick is a senior system architect, not a marketing droid, hence his= > > absence from this forum shouldn't be all that surprising.c > >t > L > Do I take it then that you see Hoff as a marketing-droid??  (Or any of the. > other Compaq worker-bees who frequent here!) >f  J Nothing in Ms. DiFrancesco's assertion would seem to me to be an aspersionE against Hoff or any of the other CPQ technical folks who take it upon E themselves to participate in this forum... usually on their own time.lE Anyone who knows Hoff knows that Hoff is not a marketing droid. 'Nuffd said.t   ------------------------------  # Date: Sun, 25 Nov 2001 17:28:00 GMT * From: "Bill Todd" <billtodd@metrocast.net>N Subject: Re: Special IPF-Inside Issue of Shannon Knows Compaq at www.tru64.orgA Message-ID: <AS9M7.118138$dk.8672490@bin1.nnrp.aus1.giganews.com>8  : "Terry C Shannon" <shannon@world.std.com> wrote in message= news:Pine.SGI.4.30.0111251109180.5290-100000@world.std.com...b >0 >6' > On Sat, 24 Nov 2001, Bill Todd wrote:    ...   L > > Unless the alternative were to be out of a job (because Compaq was going toK > > kill Alpha come hell or high water and the only question was whether it 5 > > would kill its server design team along with it).' > >  >rF > David Fenwick would have no problem getting a job as a senior system8 > architect with any enterprise IT vendor on the planet.  K My point (perhaps poorly expressed) was that he might like his current job, L not that he wouldn't be able to find work elsewhere.  Though one should noteJ that the number of enterprise IT vendors may be shrinking (indeed, whetherG Compaq and even HP will continue to qualify is one of the issues here).u   ...   J > Having spent a lot of time at DECpaq, I am not suprised that Mr. FenwickE > opted to design a platform that would not of necessity be tied to aoK > specific processor. In fact, early in the WildFire development cycle, oneeL > faction within Compaq argued passionatel for an Alpha-and-IPF design. ThisJ > contingent lost the battle, but not the war. The 2H02 McKinley-based 32PE > system from the ISSG will incorporate a lot of WildFire technology.   F That could explain a lot, since I found it a bit hard to imagine how aI technically-competent individual could come up with this idea on his own. J It's easy to understand how a bean-counter could consider consolidation ofG server development to be a Good Thing, especially if such consolidationGJ moved the architecture toward 'industry-standard' components.  What's muchH harder to understand is how a corporation driven by profit motives could7 allow such a clueless individual to dictate its future..   - bill   ------------------------------  # Date: Sun, 25 Nov 2001 17:06:09 GMTr* From: "Bill Todd" <billtodd@metrocast.net>N Subject: Re: Special IPF-Inside Issue of Shannon Knows Compaq at www.tru64.orgA Message-ID: <5y9M7.117875$dk.8652050@bin1.nnrp.aus1.giganews.com>   3 "Peter da Silva" <peter@abbnm.com> wrote in messageA news:9tr42k$2n4@web.nmti.com...    ...o  I > I think it's obvious that the gap in performance between Alpha and IA64hJ > is closing... given the current performance if IA64 one would have to beJ > even more skeptical than I to assume that Intel won't do better in roundC > two. On the other hand, there's room for an awful lot of closing..  L Indeed there is.  And while if McKinley appears on schedule it should indeedI improve matters, when EV7 arrives shortly thereafter the gap will open upbG again - possibly even wider than it is today, given EV7's advantages in C on-chip glue support for greatly-improved memory-access latency and-K bandwidth plus far better multi-processor integration:  while these may noteI improve single-processor SPEC results dramatically, they have significantuF real-world performance consequences in many typical server situations.  = Then IA64 stagnates through Madison/Deerfield, which are justBK process-shrinks of McKinley with some added on-chip cache to take advantage L of the newly-available chip area.  While on the Alpha side EV8 was slated toK appear with SMT, which, again, could have been a major win in common servercE applications even if it had less effect on single-thread SPEC numbers L (though the enhanced issue-width would have helped some there as well).  NowB that Intel has the Alpha team it has at least a prayer of shippingG significant enhancements to IA64 during the second half of this decade; F without that team, who knows how long improving such a complex (thoughL unrewarding) architecture would have taken them - while that team was busily continuing to enhance Alpha?  I So I question whether it is indeed obvious that the performance gap wouldlK have closed (save for brief, unsustained periods), given IA64's and Alpha'stJ road maps as of June 25th:  in fact, I suspect it would have significantlyC widened with EV8 (and possibly even with EV7) for any save FP-stylekK applications (which typical server environments don't seem to entail).  AndrG there remains the question of how much of a liability IA64's far higheraG thirst for per-processor power and chip area would have been, all otheri performance issues aside.l  L Definitely a case of a silk purse and a sow's ear.  Why anyone still has anyC difficulty distinguishing between them is a complete mystery to me.    - bill   ------------------------------  # Date: Sun, 25 Nov 2001 17:15:21 GMT - From: Terry C Shannon <shannon@world.std.com> N Subject: Re: Special IPF-Inside Issue of Shannon Knows Compaq at www.tru64.orgC Message-ID: <Pine.SGI.4.30.0111251210270.5290-100000@world.std.com>   $ On Sun, 25 Nov 2001, JF Mezei wrote:   > Joann Difrancesco wrote:N > > was designed to support multiple architectures (including Intel and Alpha)L > > right from the start; Fenwick and his team apparently pursued a strategy' > > that gave them maximum flexibility.l >oO > Ahhh, perhaps this is why Alpha suddently lost much of its edge against IA64.,N > If they had to compromise their wildfire systems design to the lowest commonO > denominator to also accept IA64, then it would explain why the Alpha all of aaE > sudden no longer had a huge edge over IA64 on wildfire type systemsc >d  H The WildFire system design was not compromised by Intel-amenability. TheF lowest common denominator canard is totally indefensible WRT WildFire.H Marvel began as an Alpha-only architecture, although it is rumoured thatG IPF support is being reconsidered. (This would require grafting the IO7tH support chip to IPF, I'm not in a position to comment on the feasibility or desirability of doing this.)c  H It's the POST-MARVEL (Fire/Wind/Ice) switched fabric and processor bladeJ server that was to have offered "have it your way" processor support. If IF was designing a system targeted for a mid-decade rollout I suspect I'd2 build some flexibility into the architecture, too!   ------------------------------  # Date: Sun, 25 Nov 2001 17:13:12 GMTi1 From: "David J. Dachtera" <djesys.nospam@fsi.net>yN Subject: Re: Special IPF-Inside Issue of Shannon Knows Compaq at www.tru64.org' Message-ID: <3C012A21.4BE645A8@fsi.net>t   Terry C Shannon wrote: > ' > On Sun, 25 Nov 2001, Bill Todd wrote:y > L > > It has been the near-complete absence of any official Compaq response inM > > *any* forum to substantive questions about the facts behind the June 25thoJ > > decision that has been surprising (or not, depending on one's existingI > > impressions of Compaq 'responsiveness').  And people at his level are9O > > commonly found participating in forums like comp.arch on such subjects - atlL > > least when they're comfortable putting their personal reputations on the5 > > line rather than merely their corporate personae.a > >  > K > Hmmm.. we appear to be in violent agreement on the subject of the lack oftG > official Compaq response. That's one reason Ken and I are running theb
 > surveys.  D Not sure, but perhaps Bill was implying that "substantive" questions7 deserve "substantive" answers, not just more "official"n: poor-excuse-for-marketing-drivel, as we have seen to date.  . ...IMHO. ...and I'm probably wrong, to boot...   >>>BOOT DKA0   -- h David J. Dachtera. dba DJE Systemse http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/i   ------------------------------  % Date: Sun, 25 Nov 2001 11:24:31 -0500s( From: David Froble <davef@tsoft-inc.com>N Subject: Re: Special IPF-Inside Issue of Shannon Knows Compaq at www.tru64.org, Message-ID: <3C011B3F.4060602@tsoft-inc.com>   Jeff Killeen wrote:a  < > "JF Mezei" <jfmezei.spamnot@videotron.ca> wrote in message( > news:3C005864.45A208E2@videotron.ca... >  >>Jeff Killeen wrote:m >>G >>>Given the choice between believing a self-admitted bitter ex-Digital I >>>employee and a the chief designer of the TurboLaser and the Wildfire Ia >>>l > will >   >>>take the word of Mr. Fenwick. >>>eH >>Mr Killeen, while the above argument does hold validity, you must also >> > take > E >>into consideration the possibility that Mr Fenwick has no choice byd >>	 > provideu > * >>spin to support his employer's decision. >> > G > JF my understanding is he initiated the idea i.e. he wasn't doing thet > bidding of his masters...e  O And it's not possible that projecting that image isn't the idea of his masters?o    G Could you explain how being 'ex' and 'bitter' make anyone more or less v 'right' or 'wrong'?   E As for IA-64 being a business decision and not a technical decision, aH some really interesting thoughts occur.  In the worst case, IA-64 could < utterly fail, and then this business decision will show why G non-technical people should not make technical decisions.  If IA-64 is tE not the 'best' solution Compaq can offer to it's customers, they why nB should customers buy from Compaq?  That doesn't make for a decent  business decision.  C Maybe you should face one reality.  There are people in Compaq who mE advocated a particular corporate direction, and these people won the  E contest for control of the company.  Time will be the final judge of  7 their convictions.  I for one feel that they will fail.e   Dave     -- s4 David Froble                       Tel: 724-529-04504 Dave Froble Enterprises, Inc.      Fax: 724-529-0596> DFE Ultralights, Inc.              E-Mail: davef@tsoft-inc.com6 T-Soft, Inc.  170 Grimplin Road  Vanderbilt, PA  15486   ------------------------------  # Date: Sun, 25 Nov 2001 18:04:33 GMTh* From: "Bill Todd" <billtodd@metrocast.net>N Subject: Re: Special IPF-Inside Issue of Shannon Knows Compaq at www.tru64.org@ Message-ID: <RoaM7.59007$uB.9940601@bin3.nnrp.aus1.giganews.com>  : "Terry C Shannon" <shannon@world.std.com> wrote in message= news:Pine.SGI.4.30.0111251144071.5290-100000@world.std.com...    ...t  I > Alpha may well have had no difficulty attracting customers away from an K > exercise in rearranging the deck chairs on the Itanic, but note that saiddB > Alpha customers would be limited to VMS and Tru64 staterooms, or, > doule-bunks in the Linux steerage section.  C Well, since we're examining all possibilities here, do you have any K knowledge that Microsoft would refuse to resurrect Win64 on Alpha if Compaq K footed the bill?  Sure, it would make the Compaq leadership that let it get K axed look bad, but they already look bad enough to get summarily fired so aT little more wouldn't matter.  K Win64 was reportedly already running in field test 2+ years ago on Alpha aseH 64-bit Win2K (A.K.A. NT 5.0), and continued in development internally onI Alpha at least until Itanic platforms appeared.  Given that Windows XP isaL A.K.A. NT 5.1, the work required to get 64-bit XP running on Alpha should beJ trivial, at least compared with that required to port Windows to any other	 platform.a  G And the Achilles' heel of not being able to run IA32 binaries at nativeaK speed disappears in any comparison with IA64, since by all appearances IA32yJ code running on Alpha under FX!32 would significantly out-perform the same IA32 code running on IA64.  L I'll take a wild guess that getting XP running on Alpha (now that the 64-bitE Itanic version is supposedly complete, presumably including Microsoft L applications) would cost a lot less than porting VMS to Itanic.  Not that XPJ has yet shown itself to be anything like a 'must have' for an architectureF to succeed in Alpha's market segments, but if/when it does this option+ eliminates the problem you suggested above.l   - bill   ------------------------------  # Date: Sun, 25 Nov 2001 14:32:09 GMTp& From: "Jeff Killeen" <Jeff@IDM-IO.com>N Subject: Re: Special IPF-Inside Issue of Shannon Knows Compaq at www.tru64.org7 Message-ID: <Jh7M7.1526$zf.139410@typhoon2.gnilink.net>-  5 "Bill Todd" <billtodd@metrocast.net> wrote in messageo6 news:cT1M7.35887$YD.3065558@news2.aus1.giganews.com... > 3 > "Jeff Killeen" <Jeff@IDM-IO.com> wrote in messageeJ > > What you have failed grasp is this isn't primarily a technical issue - it > is2 > > a business issue driven by market realities... >oH > What *you* have failed to grasp is that Compaq *made* this a technical issuet@ > when it attempted to clothe its dubious business decision with" > pseudo-technical justifications.  / All that Compaq has ever said technically is...t  G 1) The performance gap between EV8 and its competitors would be closingi  7 2) That EV8 would require flawless execution to succeed   L ...Without their data you can't offer much other than opinions.  Compaq thenJ went on to say that given the gap _and_ the _risk_ of EV8 that they made aJ _business_ decision not to proceed with EV8.  You have always tended glossL over that last part of their statement.  But as you said previously said youK are not a "paper pusher" but one who tends to think in technical terms.  IfwE one does think in business terms what Compaq did looks like a classicnI business decision that was based on the _reward_ from EV8 not being worthDG _return_ given the required _investment_ to make it happen.  It was theiI technology _plus_ the risk _plus_ the investment that drove the decision. K Compaq never claimed it was a pure technology based decision alone - it wasiJ a classic risk/reward decision.  To put it another way if the risk and theK investment had been low I seriously doubt Compaq would have killed Alpha ons the gap issue alone.   ------------------------------  # Date: Sun, 25 Nov 2001 18:39:56 GMTp* From: "Bill Todd" <billtodd@metrocast.net>N Subject: Re: Special IPF-Inside Issue of Shannon Knows Compaq at www.tru64.orgA Message-ID: <0WaM7.107783$2w.6806721@bin4.nnrp.aus1.giganews.com>   1 "Jeff Killeen" <Jeff@IDM-IO.com> wrote in messagee1 news:Jh7M7.1526$zf.139410@typhoon2.gnilink.net...h   ...d  1 > All that Compaq has ever said technically is...  >aI > 1) The performance gap between EV8 and its competitors would be closingi  F No:  it stated specifically that the performance gap between Alpha andK *IA64* would be closing, which is a far more difficult assertion to support1) (since the opposite appears more likely).n   >r9 > 2) That EV8 would require flawless execution to succeedn  I If you examine that statement for content, it's by definition false:  the9K advantages offered by EV8 were substantial, and would clearly have given its- the ability to weather at least minor delays.    ...r   > IfG > one does think in business terms what Compaq did looks like a classiceK > business decision that was based on the _reward_ from EV8 not being worthi= > _return_ given the required _investment_ to make it happen.i  G Part of the reward for EV8 would have been avoidance of the significantcK decline in Alpha sales that now appears to be occurring.  There's reason touK suspect that merely continued sales at existing levels (let alone increasesmH that were almost certainly possible with even minimal effort) would haveK made completing EV8 development at least marginally justifiable financiallyy! *even if EV8 then never shipped*.-  J What Compaq did looks much more like a classic business blunder arrived atH by incompetents with no understanding of the underlying market who did aL half-assed job of cobbling up numbers (and other equally dubious assertions)K to rationalize it.  If they don't like that appearance, then it's their job2G to come up with a far more convincing explanation than anything they'vet offered so far.    - bill   ------------------------------  # Date: Sun, 25 Nov 2001 16:08:39 GMTp- From: Terry C Shannon <shannon@world.std.com>uJ Subject: Re: StongARM-Outside (was Re: Of Bogusity and Benchmarketeering )C Message-ID: <Pine.SGI.4.30.0111251103340.5290-100000@world.std.com>u  ! On 24 Nov 2001, Bob Kaplow wrote:i  x > In article <SdsL7.111151$pb4.66712766@news2.rdc2.tx.home.com>, "Joann Difrancesco" <joanndifrancesco@home.com> writes:P > > While Compaq's iPAQ Pocket PC is StrongARM-based, it's a bit late for CompaqM > > to get into the StrongARM manufacturing business. Not only doesn't Compaq0O > > have a fab in which to produce the microprocessor, the firm doesn't have and > > architectural license. > >tN > > Digital Equipment Corporation had both, they sold both to Intel as part of > > the Fab-6/Alpha deal.a >nL > Most likely at Compaq's (EPs) request in order to sell the last merger ... >r  H Could be, but unless you were a fly on the wall during the negotiations,& there's no way of knowing for certain.  C I have one data point on the Compaq acquisition: I did a consultingiB assignment for Compaq just three months before the acquisition wasJ announced. The nature and specifics of the questions that the Compaq execsG had re: Digital rendered it obvious to me that Compaq had only recently A started giving serious thought to the acquisition. Now I could be C completely off-base, but my gut feeling is that I didn't perform my ? walk-on role during the end game of a long, drawn-out series ofeI negotiations. Had such negotiations taken place, the questions would havew been very different!  
 Just my $0.02s   ------------------------------  # Date: Sun, 25 Nov 2001 17:13:32 GMTs* From: "Bill Todd" <billtodd@metrocast.net>J Subject: Re: StongARM-Outside (was Re: Of Bogusity and Benchmarketeering )A Message-ID: <0F9M7.117979$dk.8659835@bin1.nnrp.aus1.giganews.com>i  : "Terry C Shannon" <shannon@world.std.com> wrote in message= news:Pine.SGI.4.30.0111251103340.5290-100000@world.std.com...i   ...s  E > I have one data point on the Compaq acquisition: I did a consultingoD > assignment for Compaq just three months before the acquisition wasL > announced. The nature and specifics of the questions that the Compaq execsI > had re: Digital rendered it obvious to me that Compaq had only recentlytC > started giving serious thought to the acquisition. Now I could besE > completely off-base, but my gut feeling is that I didn't perform mysA > walk-on role during the end game of a long, drawn-out series of K > negotiations. Had such negotiations taken place, the questions would have. > been very different!  D That's hard to reconcile with the close 3-year relationship PfeifferI supposedly had with GQ Bob, unless all interactions took place at the CEOSK level (as they appear to have in the more recent HP/Compaq case) until latenK in the game (in which case the execs you talked with could have been fairlysJ clueless without obviating the possibility that earlier understandings had< been reached - not that I have any evidence that they were).   - bill   ------------------------------  # Date: Sun, 25 Nov 2001 18:03:45 GMTa- From: Terry C Shannon <shannon@world.std.com>nJ Subject: Re: StongARM-Outside (was Re: Of Bogusity and Benchmarketeering )D Message-ID: <Pine.SGI.4.30.0111251259530.10319-100000@world.std.com>  % On Sun, 25 Nov 2001, Bill Todd wrote:i   >h< > "Terry C Shannon" <shannon@world.std.com> wrote in message? > news:Pine.SGI.4.30.0111251103340.5290-100000@world.std.com...s >o > ...g >tG > > I have one data point on the Compaq acquisition: I did a consultingrF > > assignment for Compaq just three months before the acquisition wasN > > announced. The nature and specifics of the questions that the Compaq execsK > > had re: Digital rendered it obvious to me that Compaq had only recently/E > > started giving serious thought to the acquisition. Now I could be G > > completely off-base, but my gut feeling is that I didn't perform myeC > > walk-on role during the end game of a long, drawn-out series ofcM > > negotiations. Had such negotiations taken place, the questions would havee > > been very different! >uF > That's hard to reconcile with the close 3-year relationship PfeifferK > supposedly had with GQ Bob, unless all interactions took place at the CEOiM > level (as they appear to have in the more recent HP/Compaq case) until lateiM > in the game (in which case the execs you talked with could have been fairlynL > clueless without obviating the possibility that earlier understandings had> > been reached - not that I have any evidence that they were). >a  G Well, since neither Palmer nor Eckhard Pfeiffer were in the room when Io@ did my presentation, I can't really comment on the alleged closeI relationship between Palmer and Pfeiffer. The execs I spoke with may haveeJ been clueless, but they damn sure had purchasing authority else I wouldn'tF have been in Houston. The meeting was a fishing expedition on Compaq'sG part; the company most certainly was *not* conveying any information tod me.    ------------------------------  % Date: Sun, 25 Nov 2001 12:53:56 +0100a1 From: John McLean <mcleanj@swissonline.delete.ch>s1 Subject: Re: Tech Wasteland: HWP/CPQ Profitless ?m5 Message-ID: <3C00DBD4.B5038F83@swissonline.delete.ch>-   Jerry Leslie wrote:a > K > There's an interesting artice on Fortune's site about the Tech Wasteland, ; > caused by the commoditization of the computer industry...a > K >    http://www.fortune.com/indexw.jhtml?channel=artcol.jhtml&doc_id=204671d > J >   "To understand the implications of commoditization, all you have to doJ >    is look at what happened in the computer industry in the 1990s as theK >    Wintel PC architecture proliferated into every corner of the business.yF >    Today, for all practical purposes the PC business is the computerI >    industry. What we used to call mainframes now are enterprise serverscJ >    based, roughly, on hydra-like versions of Wintel PC designs. Even theF >    proprietary holdouts that make powerful Unix servers are buildingJ >    boxes that are, when you strip away marketing rhetoric and brand-nameJ >    bigotry, strikingly similar to each other and to Wintel machines. The= >    Unix offerings from Sun, HP, and IBM all contain similarrI >    microprocessors (or perhaps a bunch of them laced together), similartD >    architectures, and variations of the same operating system. TheG >    choices of IT managers in corporate America have been reduced to aiK >    handful of systems that are at heart very similar, and are thus easier)K >    to manage and integrate into systems that actually work. In short, thetB >    computer industry embraced the PC model--and its economies of7 >    scale--as the unifying principle of all computing.i >    . >    . >    .D >    Hence the pending HP-Compaq merger. CEOs Carly Fiorina and MikeG >    Capellas promise that the combined companies will get economies of:G >    scale and the critical mass to profitably provide sophisticated ITeG >    services. But the only real reason to make this deal is that theirnF >    mainstay hardware business is now, and forever will be, virtuallyJ >    profitless--thanks to Intel and Microsoft. If they do combine, HP andK >    Compaq will be the poster child of an industry that has become, in thenI >    words of Advanced Micro Devices Chairman Jerry Sanders, "an unstable = >    molecule"--one rendered even wobblier by the recession."f > 6 > --Jerry Leslie     (my opinions are strictly my own)    : Did you follow the "Compaq/HP" link from that page to URL   F http://www.fortune.com/indexw.jhtml?channel=artcol.jhtml&doc_id=204675  C It has some interesting - and largely negative - comments about thee merger.    A few extracts ...  G 'First, what the deal essentially does is increase HP's exposure to the C low-margin, high-competition, dwindling-demand PC industry (the new D entity would reclaim the No. 1 spot from Dell). "This seems like the? wrong battle to fight," writes Merrill Lynch analyst Kraemer.'    C '...the services angle is iffier. "Neither HP nor Compaq have builtrD themselves into services powerhouses," notes Kraemer, and they'll be3 competing against stalwarts in this area like IBM.'   F 'The bottom line for investors? There's a great risk that whatever canG go wrong with this messy combination, will. .... Explains one tech funddF manager who is shunning both stocks: "If they don't do this right, you, don't want to be anywhere near this puppy."'     John McLean<   ------------------------------  % Date: Sun, 25 Nov 2001 09:11:24 -0500d+ From: "Main, Kerry" <Kerry.Main@compaq.com> 1 Subject: RE: Tech Wasteland: HWP/CPQ Profitless ?eT Message-ID: <BE56C50EA024184DAF48F0B9A47F5CF4010D71BE@kaoexc01.americas.cpqcorp.net>   Jerry,  E Fwiw .. Imho, the author is correct that Customers are moving towards G standardized computing, but he is wrong to believe that the majority of ) them are based on Wintel technologies.=20   D "..What we used to call mainframes now are enterprise servers based,6 roughly, on hydra-like versions of Wintel PC designs."  = This author does not appear to have any large scale computingmG background. IDC report a month or two ago stated that the % of businessiF computing in the world done by mainframes or mainframe style computing is still at 60%.  = The Customers I deal with want to adopt new industry standardeF application and interoperability technologies that will integrate withF what they currently have (key issue) and yet still meet their requiredG RASS (reliability, availability, scalability and security) requirementsn for the future.=20  C I suspect this is true for almost all medium-large Customers today.   D The challenge is to define what are application and interoperability6 industry standards or in many cases defacto standards?  H The Customers I deal with are currently defining these to be applicationH interoperability technologies and protocols like LDAP, XML, J2EE, X.509,A X.500 (still going strong as a distributed repository with proven 7 background), TCPIP, HTTP(s), RMI/IIOP, IMAP4, POP3 etc.e  = Now, something to consider is that adopting industry standard,F technologies does NOT mean meeting future requirements all of a suddenD gets easy. Some of the application interoperability becomes somewhatH easier (new issues evolve as well), but the RASS requirements, in almost) all cases, becomes much more challenging.s  F These RASS requirements are escalating exponentially and even what wasB acceptable even a year ago, has changed dramatically. Security andB availability are good examples of requirements that have undergone% dramatic changes in the last year.=205  B This is why many large Customers today are looking at major serverD consolidation strategies. Of course, server consolidation strategiesC require even higher levels of RASS as any issues have a much higherr impact.e  < So, I suspect that in future, in order to meet Customer RASSC requirements, you will see most OS vendors will focus on optimizingeG their OS software for their chosen platform hardware (yes, proprietary,eE but all OS's maintained by a single vendor are proprietary), but will C adopt the "industry standard" application software technologies foro application interoperability.f  F Quote from attached - "In short, the computer industry embraced the PCC model--and its economies of scale--as the unifying principle of allN computing." end quoten  G Certainly not in any of the large shops I deal with. The PC model has a > place in most companies, but as "the unifying principle of all computing."? .. ROTFL !A   :-)o   Regards,  
 Kerry Main Senior Consultantr Compaq Canada Corp.w Professional Servicesm Voice: 613-592-4660s Fax  :  819-772-7036 Email: Kerry.Main@Compaq.com     -----Original Message-----0 From: Jerry Leslie [mailto:leslie@clio.rice.edu]  Sent: November 25, 2001 12:09 AM To: Info-VAX@Mvb.Saic.Come- Subject: Tech Wasteland: HWP/CPQ Profitless ?'    > There's an interesting artice on Fortune's site about the Tech
 Wasteland,9 caused by the commoditization of the computer industry...r   =20 J http://www.fortune.com/indexw.jhtml?channel=3Dartcol.jhtml&doc_id=3D20467= 1a  H   "To understand the implications of commoditization, all you have to doH    is look at what happened in the computer industry in the 1990s as the?    Wintel PC architecture proliferated into every corner of the-	 business.-D    Today, for all practical purposes the PC business is the computerG    industry. What we used to call mainframes now are enterprise serversCH    based, roughly, on hydra-like versions of Wintel PC designs. Even theD    proprietary holdouts that make powerful Unix servers are buildingH    boxes that are, when you strip away marketing rhetoric and brand-nameH    bigotry, strikingly similar to each other and to Wintel machines. The;    Unix offerings from Sun, HP, and IBM all contain similarsG    microprocessors (or perhaps a bunch of them laced together), similarsB    architectures, and variations of the same operating system. TheE    choices of IT managers in corporate America have been reduced to aiB    handful of systems that are at heart very similar, and are thus easierE    to manage and integrate into systems that actually work. In short,a the @    computer industry embraced the PC model--and its economies of5    scale--as the unifying principle of all computing.h    .    .    .B    Hence the pending HP-Compaq merger. CEOs Carly Fiorina and MikeE    Capellas promise that the combined companies will get economies ofvE    scale and the critical mass to profitably provide sophisticated ITyE    services. But the only real reason to make this deal is that theirrD    mainstay hardware business is now, and forever will be, virtuallyH    profitless--thanks to Intel and Microsoft. If they do combine, HP andE    Compaq will be the poster child of an industry that has become, inn the G    words of Advanced Micro Devices Chairman Jerry Sanders, "an unstable ;    molecule"--one rendered even wobblier by the recession."e      4 --Jerry Leslie     (my opinions are strictly my own)   ------------------------------   Date: 25 Nov 2001 15:06:18 GMT/ From: Thomas Dickey <dickey@saltmine.radix.net>oQ Subject: Re: Terminal emulation quality, was: Re: Installing CC060 on VMS VAX 6.11) Message-ID: <9tr1da$b1$1@news1.Radix.Net>@  H Simon Clubley <simon_clubley@remove_me.altavista.co.uk-earth.ufp> wrote:0 > On Fri, 23 Nov 2001 10:10:27 +0000, in article= > <3BFE2093.E06D15B@uk.thalesgroup.com>, Paul Williams wrote:M >> >>Simon Clubley wrote: >>> ? >>> Running vttest on the various emulators available is a verytF >>> revealing experience; even xterm itself is not a 100% VT emulator. >>F >>Is this just the support for double width, double height characters?  M > [The following is based on the xterm in Redhat Linux 6.2 and vttest version ( > 20000208, also running on Redhat 6.2.]  F That's very old (a few years).  The current xterm patch level is #164.  , > This is one of the issues. The others are:   > * No DRCS supporto  N do you use it?  (Other than that it would be interesting to implement, there'sL no reason to do that, since I've not seen any applications that rely on it).  I > [* Not an emulation issue, but an xterm issue: You cannot just grab thelE > scroll bar and move it up and down, you have to click in the scroll_ > bar area.]  = not an xterm issue either: it's a feature of the Xaw library.i  J > * Numeric keypad not mapped correctly. To be fair, you can get remap theF > keyboard to fix this, but it does not work correctly out of the box.J > I should also point out that you _do_ have to remap a DECterm to work inH > a Linux PC keyboard environment, but DECterm works out of the box on a > VAX/Alpha station.  J see above (if you complain about very old code, you don't accomplish much)  J > I mention this one, because I consider a VT emulator to be more than theN > ability to interpret escape sequences correctly, but the overall environmentL > that the emulator presents. This is why I include things like the keyboard; > mapping when deciding if something is a full VT emulator.i  M > This is probably more important if you use EVE/EDT (which I do) than if youx > use emacs.  N > I have some minor issues, but the above are the main ones. The minor issues:  L > * I haven't found a way to turn F1 into a Hold Screen key (Ie: pressing itO > toggles between pausing and resuming output and generating Xon/Xoff as well).)  N my keyboard has a ScrollLock key, which does what you appear to be describing.  O > * xterm is a vt220 emulator. I sometimes make use of the host writable status.M > line and from what I can see, this only became available in the VT320. I amoK > relying on not been able to find it in the VT220 reference manual to comegN > to that conclusion, as I have only used host writable status lines in recent > years.   > * Text will not blink.  H yes (that, and DRCS are noted in the manpage).  blink is distinct if you% set a different color for it, though.h  N > * xterm does not pass all of the tests. For example, test 1 does not displayH > a double border at the top of the screen. However, this has not caused  , I've not seen this with xterm - any version.  O > problems with any real application, and the other emulators I have tested are  > not 100% perfect either.   >>E >>> At times, I have wondered just how long it would take me to writei( >>> a GPLed 100% accurate VT emulator...  - (whether it is GPLed or not is irrelevant ;-)e   -- a= Thomas E. Dickey <dickey@radix.net> <dickey@herndon4.his.com>r http://dickey.his.comg ftp://dickey.his.com   ------------------------------   Date: 25 Nov 2001 15:08:02 GMT/ From: Thomas Dickey <dickey@saltmine.radix.net>aQ Subject: Re: Terminal emulation quality, was: Re: Installing CC060 on VMS VAX 6.1,) Message-ID: <9tr1gi$b1$2@news1.Radix.Net>r  - Paul Williams <flo@uk.thalesgroup.com> wrote:- > Simon Clubley wrote: >> 0> >> Running vttest on the various emulators available is a veryE >> revealing experience; even xterm itself is not a 100% VT emulator.)  H hmm - he's referring to XFree86 xterm, btw, so I'll respond accordingly.  F > Is this just the support for double width, double height characters?  8 it's workable, given some limitations on font selection.  D >> At times, I have wondered just how long it would take me to write' >> a GPLed 100% accurate VT emulator...o  I > I imagine the return on the investment would be low. Although there aresI > some mistakes in xterm's parser tables, I don't know of any that affectmI > real applications. How many applications do not run correctly on xterm?.   there may be mistakes -e? but it's been quite a while since anyone pointed out one to me.a   -- o= Thomas E. Dickey <dickey@radix.net> <dickey@herndon4.his.com>i http://dickey.his.coma ftp://dickey.his.com   ------------------------------   Date: 25 Nov 2001 15:46:07 GMT& From: peter@abbnm.com (Peter da Silva)B Subject: Re: Tru64.org IPF Consolidation Survey Still Taking Votes% Message-ID: <9tr3nv$23q@web.nmti.com>a  D Also, questions 7 and 8 need either more alternatives, or an "other" choice.a   -- e+  `-_-'   In hoc signo hack, Peter da Silva. E   'U`    "A well-rounded geek should be able to geek about anything."hL                                                        -- nicolai@esperi.org          Disclaimer: WWFD?   ------------------------------    Date: 25 Nov 2001 03:04:03 -0800) From: P.Young@unsw.EDU.AU (Patrick Young)-4 Subject: Re: Why Compaq is so in pushing Wintel crap= Message-ID: <55f85d77.0111250304.2936a9c8@posting.google.com>.  a JF Mezei <jfmezei.spamnot@videotron.ca> wrote in message news:<3C003B54.79BACD90@videotron.ca>...tI > While swearing at bad drivers during a bike ride, I came to realise why.  I I try to keep my turbo diesel 4WD HILUX pickup away from bike riders (anddH do a good job _most_ of the time, except when they decide "OH - the WALK; light is on - I'm now a pedestrian") but anyway...  :-) :-)q  H > Compaq seems so intent on pushing its unprofitable wintel crap at the 1 > expense of its profitable alpha based products.g   This is so *totally* true.  M I think I posted previously - we don't buy "Compaq PC", "HP PC", or any othernF brand name "PC". The lot goes in the trash bin after 2.5 to 3 years - 8 why would you want to spend any amount of money on it???  I The "PC" we do buy is "noname, cheap as chips" as it should be if it goesr in the bin so soon.h  ? Related: I still have an AS600 5 years old (OpenVMS) that runs  F systems my management would pour 5 PCs in to and replace every year to gain the same function.t   ------------------------------   End of INFO-VAX 2001.656 ************************