1 INFO-VAX	Fri, 28 Apr 2006	Volume 2006 : Issue 235       Contents: Capturing console activity8 Re: DCL versus Unix CLIs, was: Re: File output like Unix8 Re: DCL versus Unix CLIs, was: Re: File output like Unix8 Re: DCL versus Unix CLIs, was: Re: File output like Unix8 Re: DCL versus Unix CLIs, was: Re: File output like Unix8 Re: DCL versus Unix CLIs, was: Re: File output like Unix8 Re: DCL versus Unix CLIs, was: Re: File output like Unix Re: decnet vs decnet over IP Re: decnet vs decnet over IP% Re: Exclude a disk from MSCP serving?  Looking for DEC/Compaq FE $ Need help with stalled printer queue( Re: Need help with stalled printer queue( Re: Need help with stalled printer queue( Re: Need help with stalled printer queue Re: OT: FAA goes Linux- Re: OT: Intels quickens cadence for new 8086s - Re: OT: Intels quickens cadence for new 8086s $ Re: Semi-OT: Changes coming at Intel$ Re: Semi-OT: Changes coming at Intel$ Re: Semi-OT: Changes coming at Intel$ Re: Semi-OT: Changes coming at Intel$ Re: Semi-OT: Changes coming at Intel  F ----------------------------------------------------------------------    Date: 28 Apr 2006 10:45:30 -0700 From: glen.thompson@gmail.com # Subject: Capturing console activity B Message-ID: <1146246330.715631.92760@u72g2000cwu.googlegroups.com>  ( System: OpenVMS 7.2 on a Vaxstation 4000  G Our system console is a Decwriter printing terminal because the log was G required many years ago by our internal audit department.  We'd like to F get rid of the Decwriter and replace it with one of our spare VTs.  IsD there anyway to automatically capture all the activity on the systemG console?  I'd like an electronic version of what is currently on paper, 5 including both operator input and output displayed.      Thanks,  glen   ------------------------------  % Date: Fri, 28 Apr 2006 01:57:33 -0400 ' From: Dave Froble <davef@tsoft-inc.com> A Subject: Re: DCL versus Unix CLIs, was: Re: File output like Unix 9 Message-ID: <I96dnem1DfuTM8zZnZ2dnUVZ_vOdnZ2d@libcom.com>    Bob Harris wrote: 1 > In article <8smdnd5WQ-WVsszZRVn-pA@libcom.com>, + >  Dave Froble <davef@tsoft-inc.com> wrote:  >  >> Simon Clubley wrote:  >>N >>> I guess I was just been too subtle. I wasn't referring to invisible files,P >>> but DCL's default processing. If you omit a filename from a target filespec,8 >>> then the filename is taken from the source filespec. >>> N >>> So unless it has changed in recent versions, the following is what happens/ >>> when you try to rename simon.test to .test:  >>>  >>> $ create simon.test     
 >>>  Exit ! >>> $ rename/log simon.test .test M >>> %RENAME-I-RENAMED, {deleted}SIMON.TEST;1 renamed to {deleted}SIMON.TEST;1  >>> . >>> You have to use tricks to rename the file.H >> What you say is true.  Statistics are also true, but can easily lie, $ >> based upon how they're presented. >>H >> So you present this as something not easily done in DCL.  I'll agree L >> with you, but I could get there if I needed to.  What I'll state though, J >> is that the default DCL behavior is exactly what I'd expect, and want,  >> it to do. >>H >> I have no use for filespecs without the filename part.  The converse H >> would be more acceptable, though I prefer both.  I wouldn't want the  >> Unix behavior.  >>G >> Now tell me all the common things you do with filespecs without the   >> filename part?  >>K >> Taking a behavior that by default works in a manner most useful to most  E >> users, then stating that it's deficient because the default isn't  L >> something that in most cases is harmful, isn't a good argument, at least  >> with me.  >>K >> So in Unix, if I don't specify the filename part of the filespec, and I  G >> have multiple files with the same extension, then I can easily lose  I >> files.  I'm assuming that the renamed file, or the one currently with  H >> that name, will end up deleted since Unix doesn't have versions.  At = >> best the operation would fail.  What's so good about that?  > < > Firstly, on *NIX it is _NOT_ ".extension".  It is ".name". > E > Technically the characters following a name are only considered an  C > extension if some application decides it wants to treat it as an  > > extension (the C compiler for example).  But if the program C > processing the file doesn't care, then *NIX doesn't care.  Never   > did. > D > The earliest convention for .named files was for personalization, @ > preferences, config type files.  The LOGIN.COM equivalent for D > example would be .profile for the bourne shell (.login and .cshrc @ > for the C-Shell; and a bunch of different names for different @ > shells).  Other applications and utilities may also allow for D > customizations such as your editor might have a .exrc, or .emacs, D > or .vimrc file.  The debuger might have a .gdbinit file.  A spell ; > checking utility might keep your personal word list in a   > .ispell_english file.  > A > Another use for .named files is as temporary files, such as an  D > editor intermediate work file, often useful for recovering from a C > crash, and useful as a flag in case 2 users (or you from another  ( > window) attempt to edit the same file. > B > The directory listing utility (ls) maintained the convention of E > not displaying any file name that started with a period, unless -a  E > was included.  And shell wildcard expansion ignore files beginning  B > with a period unless you explicitly specify a leading period in  > the wildcard expression. > @ > This has the effect of hiding .named files, unless explicitly  > asked for. > B > Now there are other utilities that do not exclude .named files, 4 > such as backup utilities, the find command, etc... > E > It is in the configuration file arena that most *NIX users working  @ > on OpenVMS find the need/desire to have .named files, as they A > often times find favorite commands and utilities from the *NIX  A > world which have been ported to OpenVMS and these commands and  ) > utilities look for .named config files.  > 0 >                                     Bob Harris= >                                     And I'm not a trolling! @ >                                     12 Years OpenVMS includingE >                                     PATHWORKS for OpenVMS developer = >                                     14 years *NIX including > >                                     Tru64 UNIX developer and@ >                                     the first version of those@ >                                     radar weather maps you see< >                                     on the Weather ChannelB >                                     And my favorite color is ...D >                                     oops! too much information :-)  + Thanks for the information and perspective.    --  4 David Froble                       Tel: 724-529-0450> Dave Froble Enterprises, Inc.      E-Mail: davef@tsoft-inc.com DFE Ultralights, Inc.  170 Grimplin Road  Vanderbilt, PA  15486    ------------------------------  % Date: Fri, 28 Apr 2006 10:25:13 -0400 3 From: "Richard B. Gilbert" <rgilbert88@comcast.net> A Subject: Re: DCL versus Unix CLIs, was: Re: File output like Unix : Message-ID: <zridncpKKpZXuM_ZnZ2dnUVZ_tKdnZ2d@comcast.com>  
 AEF wrote: > Simon Clubley wrote: > a >>In article <1145989404.330424.164290@g10g2000cwb.googlegroups.com>, bob@instantwhip.com writes:  >>B >>>why do users want convuluted unix commands like pipe grep slurp= >>>gulp ... when they use common english makes sense commands  >>>like COPY on vms? >>>  >>: >>Pipe is a DCL command. Unix uses redirection characters. >>I >>Those Unix commands are generally more powerful than the VMS equivalent ; >>and naming commands using English words only goes so far.  >>J >>For example, what English verbs and qualifiers would you use to describeI >>the functionality that ghostscript offers and by the time that you have O >>finished have you really gained anything over the current approach (as people 0 >>still have to learn what your qualifiers do) ? >>) >>Two examples of more powerful commands: C >>	grep (the Unix version of search) can do full regular expression  >>	pattern matching. >  > A > Can it do something like $ SEARCH FILE.TYP WORD1,WORD2,... with C > /MATCH=(one of OR, NOR, AND, NAND, ...) Maybe it can ... I'm just & > asking but I think the answer is no. >   6 You are right about that.  But, OTOH, can SEARCH find:5 1. a record that "begins with" a string or a pattern? 3 2. a record that "ends with" a string or a pattern? I 3. find a record containing one or more spaces followed by exactly three  I decimal digits, then one upper case alphabetic character and then one or   more spaces?????  A SEARCH and grep have some small overlap in function but they are  G different tools.  One is better than the other only if you specify the   job you want it to do!  G Regular expressions can be a real pain and I usually have to go to the  ? book for anything complicated but the power is worth the price.   F I ported Gnu grep to VMS a few years ago.  It was largely a matter of D adding missing include files and function declarations to get the C G compiler to shut up its complaining.  Ditto for gawk.  I've found both  ) tools to be very handy from time to time.    ------------------------------  # Date: Fri, 28 Apr 2006 16:05:38 GMT , From: Hoff Hoffman <hoff-remove-this@hp.com>A Subject: Re: DCL versus Unix CLIs, was: Re: File output like Unix 2 Message-ID: <m3r4g.6963$mM1.1588@news.cpqcorp.net>   Dave Froble wrote: > Simon Clubley wrote: > G >> I guess I was just been too subtle. I wasn't referring to invisible  M >> files, but DCL's default processing. If you omit a filename from a target  A >> filespec, then the filename is taken from the source filespec.   E    Extension-based file typing is for kiddie-toy file systems -- and  H yes, I do include those file systems of OpenVMS, Windows and Linux/Unix F and most every other current operating system available here.  It's a I hack, err, longstanding semi-consistent convention.  It's ugly.  It just  A doesn't work.    This extension-based naming is arcane, obscure,  @ ambiguous and altogether a very poor solution to a real problem.  =    FTP runs afoul of this, as do about a gazillion different  I applications.  What's a .DAT file, really?  Sequential?  Binary?  What's  B a .COM file?  OpenVMS thinks it is a DCL procedure, while Windows H usually assumes it is an executable.  A null filename is a hidden file? G   Gimme a break!  We couldn't implement hidden files through any other   means?  G    (If OpenVMS had that particular hack, err, longstanding convention,  @ and if Unix had file system APIs with actual selective filename E processing mechanisms or had truly hidden files, then the Unix folks  I would be justifiably laughing.  Instead, we have discussions of the same  G sort of hidden files that are grafted onto the file systems of Windows   and Linux/Unix.)  @    Unfortunately, this whole particular hack, err, longstanding H semi-consistent convention become so ingrained in how folks think about E filenames, that most folks don't think about it as the massive hack,  E err, longstanding semi-convention that it is -- until and unless the  F users slam into one of the vast number of corner cases that exist, or E into the implicit ambiguity of the scheme.  Instead, I regularly see  H discussions of "re-arraigning the deck-chairs" of this particular hack, H err, the tweaking this long-standing semi-consistent convention, rather D than actually addressing the actual underlying problem of correctly D identifying the data and the data structures stored within the file.  G    It's broke, folks.  Just because we've been using this scheme for a  I couple of decades now doesn't mean that it is an appropriate scheme, nor  I one that we can't do (far) better given even a modicum of thought around   the underlying problems here.   I    One approach is to get rid of the antiquated hierarchical file system   scheme entirely:  F    http://h20325.www2.hp.com/blogs/hoffman/archive/2006/02/03/739.html   ------------------------------    Date: 28 Apr 2006 11:59:29 -0500B From: clubley@remove_me.eisner.decus.org-Earth.UFP (Simon Clubley)A Subject: Re: DCL versus Unix CLIs, was: Re: File output like Unix 3 Message-ID: <VMLbJY7O172j@eisner.encompasserve.org>   a In article <m3r4g.6963$mM1.1588@news.cpqcorp.net>, Hoff Hoffman <hoff-remove-this@hp.com> writes:  > I >    (If OpenVMS had that particular hack, err, longstanding convention,  B > and if Unix had file system APIs with actual selective filename G > processing mechanisms or had truly hidden files, then the Unix folks  K > would be justifiably laughing.  Instead, we have discussions of the same  I > sort of hidden files that are grafted onto the file systems of Windows   > and Linux/Unix.) >    [snip]   > I >    It's broke, folks.  Just because we've been using this scheme for a  K > couple of decades now doesn't mean that it is an appropriate scheme, nor  K > one that we can't do (far) better given even a modicum of thought around   > the underlying problems here.  >   G Well I guess we now know what you think about filenames prefixed with a I dot. :-) (But don't forget that to Unix the filename/filetype combination D is just a string of characters - as with ODS-5 now, file.tar.gz is a perfectly valid Unix filename).   L Would you like to offer any comments on editing past line wrap boundaries or) tab completion of directories/filenames ?    Simon.   --  ; Simon Clubley, clubley@remove_me.eisner.decus.org-Earth.UFP K If Google's motto is "don't be evil", then how did we get Google Groups 2 ?    ------------------------------  % Date: Fri, 28 Apr 2006 13:20:06 -0400  From: BobH <bobh@x.y> A Subject: Re: DCL versus Unix CLIs, was: Re: File output like Unix % Message-ID: <89s4g.34$9R.28@fe04.lga>    Hoff Hoffman wrote:  > Dave Froble wrote: >  >> Simon Clubley wrote:  >>H >>> I guess I was just been too subtle. I wasn't referring to invisible G >>> files, but DCL's default processing. If you omit a filename from a  I >>> target filespec, then the filename is taken from the source filespec.  >  > K >   Extension-based file typing is for kiddie-toy file systems -- and yes,  I > I do include those file systems of OpenVMS, Windows and Linux/Unix and  J > most every other current operating system available here.  It's a hack, E > err, longstanding semi-consistent convention.  It's ugly.  It just  C > doesn't work.    This extension-based naming is arcane, obscure,  B > ambiguous and altogether a very poor solution to a real problem.  4 How do you feel about the way the Mac OSs handle it?  H >   It's broke, folks.  Just because we've been using this scheme for a K > couple of decades now doesn't mean that it is an appropriate scheme, nor  K > one that we can't do (far) better given even a modicum of thought around   > the underlying problems here.   H Actually rather a bit more than that.  PS/8 (before it was OS/8) had it H in 1970, and that was copied from the 36 bit systems no doubt.  I would 3 guess we can trace it back a good 40 years minimum.   
 Bob Hassinger    ------------------------------   Date: 28 Apr 2006 17:43:51 GMT( From: bill@cs.uofs.edu (Bill Gunshannon)A Subject: Re: DCL versus Unix CLIs, was: Re: File output like Unix , Message-ID: <4bf2inF10vp5cU1@individual.net>  2 In article <m3r4g.6963$mM1.1588@news.cpqcorp.net>,/ 	Hoff Hoffman <hoff-remove-this@hp.com> writes:  > G >    Extension-based file typing is for kiddie-toy file systems -- and  J > yes, I do include those file systems of OpenVMS, Windows and Linux/Unix @ > and most every other current operating system available here.   G Why include Unix?  In Unix, no filename (or extension) has any implicit F meaning.  Applications may impart meaning, but the OS just don't care.J (Hint: ever "cat"ed a binary file to your terminal with disastrous results or opened a directory in vi?)    bill   --  J Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesD bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. University of Scranton   |A Scranton, Pennsylvania   |         #include <std.disclaimer.h>       ------------------------------    Date: 28 Apr 2006 03:37:54 -0700 From: etmsreec@yahoo.co.uk% Subject: Re: decnet vs decnet over IP B Message-ID: <1146220674.432565.17310@i40g2000cwc.googlegroups.com>  G Yup, used it myself to migrate a client from a multi-protocol WAN to an E IP only WAN, as well as allowing DECnet over a TCP/IP-only link for a  former employer.  C Disadvantages: Need to make sure that the IP address for the remote F side is registered, plus you need to rejig proxies so that they're not' using the DECnet phase IV style naming. E Advantages : Network manager doesn't know that you're using DECnet on F his IP-only network so you can do remote file activities, set host etcE etc and all the other DECnet niceties without worrying whether you'll  get "found out".E Also prevents corruption of savesets and indexed files because you're # using DECnet between nodes, not IP.   F Bandwidth and latency are often something of a red-herring, since it'sF the same network.  Makes no difference to the bits on the bits of wireE whether they're running in IP packets or DECnet packets.  IP networks B can come up with some funny routes though where instead of sendingG traffic down a single link between two sites on a corporate WAN they go G via sites 4, 5 and 6 because that was the one with least traffic on the - last time that routing updates were received.    Steve    ------------------------------  % Date: Fri, 28 Apr 2006 08:18:53 -0400 % From: Javier <javier@invalid.invalid> % Subject: Re: decnet vs decnet over IP / Message-ID: <e2t14n$13pa$1@stationair.kjsl.com>    Chuck Aaron wrote: > Questions: > = > What is the advantage of running DECNET over IP vs. DECNET?   > Others already started to go over the pros vs. cons of either.  G I'll mention that there's another option, the MultiNet's DECnot (or IP  8 services for DECnet or whatever it's called these days).  H This emulates DECnet from the application point of view, but uses IP as H the underlying protocol. On networks where IP is preferred or mandated, ; it's an excellent alternative for tunneling DECnet over IP.    -jav   ------------------------------  # Date: Fri, 28 Apr 2006 11:54:24 GMT 5 From: rdeininger@mindspringdot.com (Robert Deininger) . Subject: Re: Exclude a disk from MSCP serving?[ Message-ID: <rdeininger-2804060754270001@dialup-4.233.152.205.dial1.manchester1.level3.net>   D In article <e2rb7l$cch$1@online.de>, helbig@astro.multiCLOTHESvax.de1 (Phillip Helbig---remove CLOTHES to reply) wrote:   I >In article <000301c66922$9551d310$994614ac@domina.fom>, "Rudolf Wingert"  ><win@fom.fgan.de> writes:   > 	 >> Hello,  >>   >> Robert Deininger wrotes:  >>   >> >>>K >> The satellite will need a local DOSD (Dump Off System Disk), at least in H >> the first implementation.  And most folks will want a local page/swapH >> disk.  Here "local" means not MSCP-served.  SAN-based disks are fine. >> <<< >>  I >> In case of this I have the question, is it possible to exclude disk(s) K >> from MSCP serving? AFAIK you can only enable/disable MSCP for all disks!  > F >I think "not MSCP-served" here means that the node in question has a H >direct connection to the disk, not that it is not MSCP-served to other  >nodes.   H That's right.  The satellite has to have a direct connection to the DOSD" disk; an MSCP connection won't do.  C Whether the satellite later serves the disk via MSCP is irrelevant.    Sorry for the confusion.   ------------------------------   Date: 28 Apr 2006 07:50:06 GMT> From: Help Wanted <operations@REMOVETHIS.cdswest.dnsalias.net>" Subject: Looking for DEC/Compaq FE3 Message-ID: <Xns97B387F0A9D7878200@204.153.244.170>   J Looking for a DEC/Compaq FE in the SOCAL area for field and support role.   6 Respond to: operations@REMOVETHIS.cdswest.dnsalias.net   ------------------------------    Date: 28 Apr 2006 07:35:29 -0700$ From: "AEF" <spamsink2001@yahoo.com>- Subject: Need help with stalled printer queue C Message-ID: <1146234929.370802.115470@v46g2000cwv.googlegroups.com>   D I once set up printer queues on two VAX systems (MicroVAX 3100 ModelE 95, Model 80, both running patched VMS v6.2) using TCPware v5.3-3 and E everything worked fine. Just recently they stopped working. The print F job goes into the "Processing" state for 60 seconds and then the queueF and job stall. I checked with the Windows admin and he said nothing isA different on the printserver. Actually, it may have been added to F "Active Directory" domain, but I use host tables on my VAX systems andB it is pingable. I checked with one of our network guys and he says nothing has changed. Any ideas?   
 $ Q HP4200/fu D Server queue NODEX_HP4200, stalled, on NODEX::, mounted form DEFAULTE   /BASE_PRIORITY=4 /DEFAULT=(FEED,FORM=DEFAULT) /LIBRARY=HP4200DEVCTL , /OWNER=[SYSTEM] /PROCESSOR=TCPWARE_VMSLPRSMB-   /PROTECTION=(S:M,O:D,G:R,W:S) /RETAIN=ERROR   4   Entry  Jobname         Username     Blocks  Status4   -----  -------         --------     ------  ------5      18  LOGIN           SYSTEM            2  Stalled @          Submitted 28-APR-2006 09:41:41.06 /NOFLAG /FORM=DEFAULT
 /PRIORITY=100 B          File: _DSA0:[VMS$COMMON.SYSMGR]LOGIN.COM;18 (processing)) /NOFLAG    In TCPware:   $ The currently configured queues are:  G OpenVMS Print Queue             Remote Host       Remote Printer    OvR  Fmt G -------------------             -----------       --------------    ---  --- F NODEX_HP4200                    OCEPRINTSERVER    HP4200_SYSADMIN   NO VMS .   Additional Qualifiers: /LIBRARY=HP4200DEVCTL  D Other systems can print fine on the printer (I can print from my PC,? other [non-VAX] users have no problems), meaning the problem is F specific to the VAX systems. But this has happened on two VAX systems,< meaning the problem is likely somewhere else. I can ping theB OCEPRINTSERVER (a Windows box) from the problem VAX systems. But IE cannot print. The printer is an HP Laserjet 4200n. I tried setting up E another queue pointing to a different printer on the same printserver C but get the same problem. Also, the Windows Admin told me he saw no F signs of any job from my VAX systems on the printserver. So it appearsD there is a problem with the symbiont talking to the printserver. But what could it be?   D Can anyone help me? Suggest what to look at? I'm happy to answer any helpful questions. Thanks!   AEF    ------------------------------    Date: 28 Apr 2006 07:43:06 -0700 From: etmsreec@yahoo.co.uk1 Subject: Re: Need help with stalled printer queue C Message-ID: <1146235386.432008.247700@i39g2000cwa.googlegroups.com>   G If it's been added to the Active Directory domain, does the PrintServer E know what the VAXen are (their IP addresses for example) and does the + Printserver know to allow the jobs through? E Does the Active Directory know to allow the print jobs from the VAXen  in using LPR/LPD? D Has the addition of the printers to the active directory removed the+ LPR/LPD functionality from the printserver?    ------------------------------    Date: 28 Apr 2006 08:26:29 -0700$ From: "AEF" <spamsink2001@yahoo.com>1 Subject: Re: Need help with stalled printer queue C Message-ID: <1146237989.722964.113760@u72g2000cwu.googlegroups.com>    etmsreec@yahoo.co.uk wrote: I > If it's been added to the Active Directory domain, does the PrintServer G > know what the VAXen are (their IP addresses for example) and does the - > Printserver know to allow the jobs through? G > Does the Active Directory know to allow the print jobs from the VAXen  > in using LPR/LPD? F > Has the addition of the printers to the active directory removed the- > LPR/LPD functionality from the printserver?    (Probably not)**4   C I didn't know AD would have such an effect. Actually, I didn't even F know until now that the printserver was added to AD. Fortunately, thisG is not a production issue. It just makes it easier for me to print from  the VAX systems.  E I will check with the Windows admin as soon as he has time. Thanks, I F think this is probably the problem. I will report back when I find out more.   ' Thanks for your very prompt response!!!    AEF    ------------------------------  # Date: Fri, 28 Apr 2006 16:51:00 GMT , From: Hoff Hoffman <hoff-remove-this@hp.com>1 Subject: Re: Need help with stalled printer queue 2 Message-ID: <UJr4g.6966$7R1.4091@news.cpqcorp.net>   etmsreec@yahoo.co.uk wrote:   F > Has the addition of the printers to the active directory removed the- > LPR/LPD functionality from the printserver?   D    I've never seen Microsoft Active Directory (AD) get tangled with I LPR/LPD.  (That's not to state such could not happen, just that I've not  F seen that particular train-wreck arise -- and we've certainly had our > share of intentionally and unintentionally "creative" network  configurations here inside HP.)   E    My first suspicion would be that TCPWARE_VMSLPRSMB is getting its  G knickers in a twist, or is going toes-up; err, an unexpected condition  E is arising within the local symbiont's run-time environment, and the  F symbiont is accordingly demonstrating unspecified untoward behaviour. E This could include the symbiont, or associated parts of the IP stack.   G    My secondary assumption would involve a potential problem somewhere  I within the OpenVMS VAX queue manager or related components, or somewhere  H within the network stack.  Accordingly, I'd look at the process and (if I present) at the process state, and (if present) at any logs or debugging  D mechanisms available for this symbiont, and I'd look for applicable H updates from Process and at the application of the mandatory and of any , queue-related ECOs from OpenVMS Engineering.  I    I have had direct success with connections with LPR/LPD and DCPS (via  H TCP/IP Services, however), and would tend to use DCPS for printing to a & LaserJet 4200 or LaserJet 4300 series.  G    You could eliminate DNS and DHCP use as a test, assuming you have a  I fixed address.  (I tend to not configure the printer to use DHCP.)  This  C would largely eliminate any potential involvement of AD, of course.    ------------------------------    Date: 28 Apr 2006 08:53:50 -0700 From: bob@instantwhip.com  Subject: Re: OT: FAA goes Linux C Message-ID: <1146239630.110719.139670@e56g2000cwe.googlegroups.com>   A no cost in reliability and security?  compared to vms clustering, A security and cert counts, time will show how reliable linux is to C vms ... I don't want to be on the plane or planes when it fails ...    ------------------------------    Date: 28 Apr 2006 05:28:37 -0700 From: bob@instantwhip.com 6 Subject: Re: OT: Intels quickens cadence for new 8086sA Message-ID: <1146227317.031317.4040@j33g2000cwa.googlegroups.com>   F thee is only so many tricks you can devise for the x86 boat anchor ...= when they run out soon, goodbye amd and itnel if they abandon  itanium/alpha ...    ------------------------------  % Date: Fri, 28 Apr 2006 07:31:56 -0600 6 From: "Michael D. Ober" <obermd.@.alum.mit.edu.nospam>6 Subject: Re: OT: Intels quickens cadence for new 8086s- Message-ID: <KNo4g.5$MA5.741@news.uswest.net>      H This rumor (the x86 architecture is dead) keeps coming up.  Usually fromF people who have a vested interested in a competing processor.  AMD hasB already proved that the x86 architecture can be extended to 64 bit: computing.  Whether Intel can keep up with AMD, who knows.  
 Mike Ober.  & <bob@instantwhip.com> wrote in message; news:1146227317.031317.4040@j33g2000cwa.googlegroups.com... H > thee is only so many tricks you can devise for the x86 boat anchor ...? > when they run out soon, goodbye amd and itnel if they abandon  > itanium/alpha ...  >  >    ------------------------------  % Date: Fri, 28 Apr 2006 09:07:56 +0200 + From: Karsten Nyblad <nospam@nospam.nospam> - Subject: Re: Semi-OT: Changes coming at Intel = Message-ID: <4451bf3a$0$60779$157c6196@dreader1.cybercity.dk>    JF Mezei wrote:  > GreyCloud wrote: > J >>"We are taking a tight look at spending and structure," Otellini said atD >>the chipmaker's shareholder meeting. "We are going to restructure,I >>resize and repurpose Intel to adjust to the business realities of today  >>and tomorrow." >  >  > C > Yep. But at this point in time, all we can expect is for Intel to I > quietly shift more resources from IA64 to the 8086. And that means that D > odds of more delays in IA64 increase, and each iteration will have > fewwer new features.  I Net necessarily.  Intel has enough money to finance both processors, and  - Intel can use only so many developers on x86.   B If I was CEO, I would ask why Intel can have better semiconductor D technologies than AMD, yet AMD has been capable of producing faster I chips,  It seems like the parts of Intel that design microprocessors are  H not good enough.  That problem does not look like a problem that can be I solved by throwing more people at it.  It looks like Intel does not have    the right people to work on x86.  C Besides, Intel has developed many x86 cores that never came to the  D market.  Intel may say that they want to introduce a new core every H second year, but it I would not be surprised if half of them never make  it to the market.    ------------------------------  % Date: Fri, 28 Apr 2006 09:55:03 +0100 * From: "Richard Brodie" <R.Brodie@rl.ac.uk>- Subject: Re: Semi-OT: Changes coming at Intel 2 Message-ID: <e2sl97$v8s$1@blackmamba.itd.rl.ac.uk>  9 "Karsten Nyblad" <nospam@nospam.nospam> wrote in message  7 news:4451bf3a$0$60779$157c6196@dreader1.cybercity.dk...   [ > If I was CEO, I would ask why Intel can have better semiconductor technologies than AMD,  X > yet AMD has been capable of producing faster chips,  It seems like the parts of Intel V > that design microprocessors are not good enough.  That problem does not look like a Z > problem that can be solved by throwing more people at it.  It looks like Intel does not ' > have the right people to work on x86.   D Intel have had rough times lately, and holding back their own x86-64E architecture didn't do them any good. However, all the reports of the E  new Core architecture I have seen have been overwhelmingly positive.   This year could be interesting.    ------------------------------  % Date: Fri, 28 Apr 2006 07:32:09 -0400 ) From: "Neil Rieck" <n.rieck@sympatico.ca> - Subject: Re: Semi-OT: Changes coming at Intel 8 Message-ID: <J2n4g.4164$fx.376919@news20.bellglobal.com>  6 "Richard Brodie" <R.Brodie@rl.ac.uk> wrote in message , news:e2sl97$v8s$1@blackmamba.itd.rl.ac.uk... >  [...snip...] > E > Intel has had rough times lately, and holding back their own x86-64 G > architecture didn't do them any good. However, all the reports of the F > new Core architecture I have seen have been overwhelmingly positive.! > This year could be interesting.   L Five months ago I read a new book called "The Pentium Chronicles" which was K more about how Intel developed the Pentium than the chip itself. That book  J talked about things happening at many North American companies like "more J decisions being made by MBAs" and "marketing people controlling allocated I funds (and the technical people developing chips required to beg and bow  # before them to get at those funds).   A One thing that sticks out in my mind was the development of SIMD  J instructions like MMX and SSE. The technical people thought it would be a I good idea to develop these technologies but marketing did not agree. The  M conflict was resolved  when engineers convinced marketing that AMD was going  K to develop this idea first, and it was better to be a leader rather than a  	 follower.   M With regards to x86-64 I can only assume that a similar conflict existed but  ) marketing won the argument (eg. delayed).   	     #####   G On a related note, in the 1960s General Motors would alternate the top  L position (the term CEO did not yet exist) between Engineering and Financial J every year (or two?). GM had more than 50% of the market share and the US C government was considering legislation to treat them as a monopoly.   L I think it was in 1970 when "the bean counters" final got control of things J so "the wrench heads" would never run things again. This is also the time F when GM started to do sneaky things like quietly substituting smaller J transmissions etc. Well here we are today and GM has approximately 25% of D the market share, is in financial trouble, can't meet their pension J obligations, etc, In an interview on "60 Minutes" a few weeks back one GM L executive (I can't remember his name but he was 74 years old and had a mind F like a steel trap) made a statement in passing about getting the bean , counters out of the "100% control" position.  H Look around the world at the really successful technology companies and F you'll see the original technical people still at the top (Microsoft, I Oracle, RIM, etc).  On the flip side you'll see companies like (Digital,  L Apple, Lotus,  Nortel, etc.) driven down by non-technical people who should  have known better.  H p.s. I'm not advocating "kill all the bean counters"; technology people K running companies need to defer many thing to financial and legal experts;  ; just don't let them take over 100% control of your company.   	     #####   J Getting back to Intel (as well as many North American companies) reducing K the ratio of "chiefs to indians" is a good start. Getting technical people  = back into the decision making process of technical companies  $ (fundamentalism?) is also desirable.  
 Neil Rieck Kitchener/Waterloo/Cambridge,  Ontario, Canada.8 http://www3.sympatico.ca/n.rieck/links/cool_openvms.html: http://www3.sympatico.ca/n.rieck/links/openvms_demos.html    ------------------------------  % Date: Fri, 28 Apr 2006 07:39:21 -0400 ) From: "Neil Rieck" <n.rieck@sympatico.ca> - Subject: Re: Semi-OT: Changes coming at Intel 8 Message-ID: <z9n4g.4167$fx.377340@news20.bellglobal.com>  K "David J. Dachtera" <djesys.nospam@NeOaSrPtAhMlNiOnWk.net> wrote in message  > J > In a nutshell, Intel's looking at $9 billion in profits this year versusH > $12 billion last year. A major re-org is expected, including "a 90-day? > structural reorganization of 'nonperforming business units.'"  > I When ever bean counters are slashing, they usually assume they are doing  I things to satisfy the investors. With such a mindset it is possible that  M someone will preferentially choose profitable x86 (and x86-64) rather than a   much less profitable Itanium.   
 Neil Rieck Kitchener/Waterloo/Cambridge,  Ontario, Canada.8 http://www3.sympatico.ca/n.rieck/links/cool_openvms.html: http://www3.sympatico.ca/n.rieck/links/openvms_demos.html    ------------------------------  % Date: Fri, 28 Apr 2006 10:36:14 -0400 3 From: "Richard B. Gilbert" <rgilbert88@comcast.net> - Subject: Re: Semi-OT: Changes coming at Intel 0 Message-ID: <Aq6dnQZU06bDtc_ZRVn-jQ@comcast.com>   Alan Greig wrote:    >  >  > Neil Rieck wrote:  >  >>F >> When ever bean counters are slashing, they usually assume they are D >> doing things to satisfy the investors. With such a mindset it is H >> possible that someone will preferentially choose profitable x86 (and 6 >> x86-64) rather than a much less profitable Itanium. >  > K > An ex DEC employee told me that he met Bob Palmer not long after he took  J > over DEC. Palmer asked those present "What is our main product?". Loads K > of answers from the assembled underlings but nobody got it "right" until  K > Palmer proudly announced that Digital's main product was the share price.  >  > Says it all really.  > I > However Palmer destroyed his main product as much as he did everything   > else.   E If you think your share price is your "main product" you are doomed.  3 Managing for the next quarters earnings is madness.   C Producing a product the customers really need and doing it better,  + faster, and cheaper is the road to success.    ------------------------------   End of INFO-VAX 2006.235 ************************