1 INFO-VAX	Thu, 07 Jun 2001	Volume 2001 : Issue 313       Contents:, Re: "Talk Baq" to Compaq - or perhaps to Sun, Re: "Talk Baq" to Compaq - or perhaps to Sun, Re: "Talk Baq" to Compaq - or perhaps to Sun Re: $QIO and TCP/IP server Re: $QIO and TCP/IP server Re: $QIO and TCP/IP server# Re: 10 vs 100 Mb ethernet on xp1000 ; Re: A New Twist on OpenVMS Marketing... and Media Packaging  Re: alphastation 200 4/233; Re: Availability Manager vs DECAMDS Collector Modules query ; Re: Availability Manager vs DECAMDS Collector Modules query " Re: Byte Order on Alpha ProcessorsH Re: Can you format sent VMS mail( was: Can you format recieved VMS mail)% Re: Dec-net ethernet address on a vax  DECAMDS threshold "classes"  Diamond Forum agenda Re: disabling floppy drive Re: DSN Link/ esa12000fc and ema12000 differences/experiences 3 Re: esa12000fc and ema12000 differences/experiences 3 Re: esa12000fc and ema12000 differences/experiences 2 Re: Hydro, was (OT) Current Microsoft V DOJ status2 Re: Hydro, was (OT) Current Microsoft V DOJ status2 Re: Hydro, was (OT) Current Microsoft V DOJ status2 Re: Hydro, was (OT) Current Microsoft V DOJ status2 Re: Hydro, was (OT) Current Microsoft V DOJ status2 Re: Hydro, was (OT) Current Microsoft V DOJ status2 Re: Hydro, was (OT) Current Microsoft V DOJ status Re: I demand your respect  RE: RE: Is it just me? RE: RE: Is it just me?# OpenVMS Diamond Forum in California  Re: OT - demise of ML770 Re: OT - demise of ML770 OT: RE: Is it just me? Re: select() on non-sockets  Re: subject _OPEN_VMS  RE: The future of VMS  RE: The future of VMS  Re: The future of VMS  Re: The future of VMS  Re: The future of VMS  Re: The future of VMS  Re: The future of VMS  Re: The future of VMS  Re: The future of VMS  Re: The future of VMS  Re: The future of VMS  Re: The future of VMS  RE: The future of VMS  Re: The future of VMS  Re: The future of VMS ; RE: UPDATE - finding the dec-net ethernet address using vms ; Re: UPDATE - finding the dec-net ethernet address using vms ; RE: UPDATE - finding the dec-net ethernet address using vms A Re: UPDATE - finding the dec-net ethernet address using vms 5.5-2 8 Use of SYSMAN to keep password file in sync on 8 systems< Re: Use of SYSMAN to keep password file in sync on 8 systems
 vax/vms cd Re: VMS 7.2-1 and UCX 4.2  Re: VMS 7.2-1 and UCX 4.2  Re: VMS 7.2-1 and UCX 4.2   Re: Weblogic/BEA On OpenVMS XML?  RE: Weblogic/BEA On OpenVMS XML?  Re: Weblogic/BEA On OpenVMS XML?  Re: Weblogic/BEA On OpenVMS XML?= Re: Who gets blamed for faulty software (was: A New Twist...) ) Re: [OT] Current Microsoft v DOJ status ? ) Re: [OT] Current Microsoft v DOJ status ? ) Re: [OT] Current Microsoft v DOJ status ? ) Re: [OT] Current Microsoft v DOJ status ? ) Re: [OT] Current Microsoft v DOJ status ? ) Re: [OT] Current Microsoft v DOJ status ? ) Re: [OT] Current Microsoft v DOJ status ? ) Re: [OT] Current Microsoft v DOJ status ? ) Re: [OT] Current Microsoft v DOJ status ? ) Re: [OT] Current Microsoft v DOJ status ? ) Re: [OT] Current Microsoft v DOJ status ? ) Re: [OT] Current Microsoft v DOJ status ? ) Re: [OT] Current Microsoft v DOJ status ? ) Re: [OT] Current Microsoft v DOJ status ? ) Re: [OT] Current Microsoft v DOJ status ?   F ----------------------------------------------------------------------  # Date: Wed, 06 Jun 2001 18:20:34 GMT 4 From: "Terry C. Shannon" <terryshannon@mediaone.net>5 Subject: Re: "Talk Baq" to Compaq - or perhaps to Sun = Message-ID: <SvuT6.23063$zl5.7501700@typhoon.ne.mediaone.net>   : "Tim Llewellyn" <tim.llewellyn@bbc.co.uk> wrote in message# news:3B1E52B1.5DFD7DAF@bbc.co.uk...    > F > How true, I sadly remember the day when we benchmarked starting WordC > on an Alpha Multia and a 133 MHz Pentium. Pentium had Word on the H > screen before the Multia had barely responded to the click (approx 1.5 secs). > G > Of course, M$ should and could have ported MS Office to alpha native.   G But of course! And had the then-CTO of Digital negotiated a decent deal I rather than snapping at the first squatulent offer Microsoft threw on the K table, Digital would have received client-side Intel apps parity for Alpha.   K Sadly, the former CTO did not do this. A smart engineer he might have been, E but the guy was IMHO totally clueless as a businessperson. A shame he , managed to bail out on a platinum parachute.   ------------------------------  % Date: Wed, 06 Jun 2001 15:22:09 -0400 - From: JF Mezei <jfmezei.spamnot@videotron.ca> 5 Subject: Re: "Talk Baq" to Compaq - or perhaps to Sun , Message-ID: <3B1E82DD.65EB0B4F@videotron.ca>   andrew harrison wrote:@ > Alpha NT was unsucessfull not because NT was trash but because? > Digital and laterly Compaq failed to get the ISV's to port to = > NT/Alpha and failed to conclusively demonstrate that it was 4 > worth paying a serious premium for a Alpha/NT box.  L No, Digital and Compaq failed to realise that Alpha was not worth charging aD premium for anymore, now that the 8086 had caught up with the Alpha.  J When Alpha was first released, it was supposed to make Digital competitiveN with Sun and HP because it was felt that the price performance for VAX was notL good enough (again, because Digital didn't want to lower the price of VAX toM match its value compared to more modern chips). So there was no reason to put J Alpha at a premium at that point since Sun and HP were already leaders and< Digital should have been trying to regain market leadership.  K But the 8086 quickly caught up (geez, I wonder why ? :-) and then it was no J longer the Suns and HPs that were the real competitors, but the DOS/NovellH game boxes (and later NT). Again, Digital failed to acknowledge this andQ maintained a price penalty on the Alpha to ensure it did not compete against PCs.   M The whole problem is Digital and Compaq putting artificial price barrriers on M the Alpha because they don't want it to eat on PC sales (or don't want people X to buy lots of small cheap Alphas instead of that one big multi million dollar monster).   ------------------------------  % Date: Thu, 07 Jun 2001 01:24:23 +0200 ) From: Christof Brass <brass@infopuls.com> 5 Subject: Re: "Talk Baq" to Compaq - or perhaps to Sun , Message-ID: <3B1EBBA7.B894BE17@infopuls.com>   JF Mezei wrote:  >  > Christof Brass wrote: D > > Think about the following: there are companies which want to run@ > > their apps on VMS, there are others that want to run them onB > > Tru64, and - unfortunately - a few companies still continue to > > use Slowaris.  > O > Ask yourself: of the remaining VMS sites, how many are actually growing their 8 > VMS infrastructure by adding new applications on VMS ?  ? The SWX Swiss Exchange the offcial developer of the first fully @ electronical stock exchange trading system which is also used by7 the second biggest European bourse the German bourse in @ Frankfurt is constantly expanding their customer range and is in: a process to turn one part of their business into a system= deliverer for fully electronical stock exhanges systems. This = will not only increase the number of systems running VMS this > will substantially increase the number of customers using VMS.  > BTW one reason (from what I know the only one) for the failure9 of the merger between the bourse in Frankfurt the biggest ? European stock exchange the LSE (London Stock Exchange) was the @ fact that the Germans insisted in using their superior VMS order< book system while the LSE wanted to continue to use their OM> system (which may also run on VMS as OM has a strong VMS line)? but the system isn't fully electronical AFAIK the settlement is @ done manually. As a sidenote the Swedish OM has just started the? second unfriendly takeover attempt against the LSE in a plan to * expand their trading system customer base.  O > It seems to me that of the remaining VMS sites, any growth is to accomodate a K > growth in existing application usage, not by adding new applications. New / > applications are usually added on NT or Unix.   > I don't have the insight to argue against but I expect this to= be a temporary phaenomenon as major sites and companies using  Slowaris have major problems.    N > Why would a customer struggle and fight against Compaq to put an applicationM > which does not fit in Compaq's view of what VMS is supposed to be running ? F > (eg: customers have gotten the hints from Compaq that outside of the3 > identified niches, VMS doesn't have much future).   > There are always people which don't delegate thinking to their
 suppliers :-)   K > Sun on the other hand is out to grab as much market share as possible and O > widen the applications available on its platform, including Star Office which A > hopes to compete against Compaq's/Microsoft's Office bloatware.   ? Star office is crap also. Stupid, inconsistent C++ programming. < People have looked into the sources and I don't know if they= recovered from sickness yet. SUN is the Micro$hit in the UNIX = area: telling lies to customers, marketing vaporware etc.. If @ the economical world is ruled on the long term by that practices? I understand why intelligent and successful people left the EDP 7 industry. I for one don't like to do business that way.    ------------------------------  % Date: Wed, 06 Jun 2001 14:49:10 -0400 - From: JF Mezei <jfmezei.spamnot@videotron.ca> # Subject: Re: $QIO and TCP/IP server , Message-ID: <3B1E7B24.547A64D8@videotron.ca>  & "Brian Schenkenberger, VAXman-" wrote:A > >- a "multi-thread" server within a single server process (AST) I > >- the "fork" method to have one process for read/write with one client  > >($creprc) > >   J > I don't do Pascal but I prefer the multi-threaded approach.  It's really > quite simple to implement:  M However, if you use multi threading, you need to deal with concurrency issues K such as writing to a log file. Many things that are easy on a single thread 7 program become a hassle on multi threaded applications.   G With a single thread AST driven system, you can control the delivery of 5 "logic" because AST's are queued one after the other.   L I guess it depends on the level of "interactivity" you need. If transactionsM are quick "ask-and-get-an-answer types, then threads may not be worth it. But I if you're implementing a POP server for instance where you really need to L serve multiple clients at the same time, threads are the best solution. WithN an AST driven single thread, sending data to multiple clients at the same timeJ becomes an interesting challenge (you can't do a single QIO to send that 4& megabyte email message to the client).   ------------------------------  % Date: Wed, 06 Jun 2001 12:20:48 -0700 ! From: Shane.F.Smith@healthnet.com # Subject: Re: $QIO and TCP/IP server D Message-ID: <OFCD958D3B.77A1E363-ON88256A63.006827C1@foundation.com>  H I have a hybrid solution, designed because of a third party package thatJ couldn't hack multithreading. I use a central multithreaded server, with aH pool of "slave" processes doing the actual work, all connected by TCP/IPK channels. The central server takes incoming calls and acts as a switchboard F to route the tasks to whichever slave is free. The slave processes areF started during initialization and are re-usable, so there's no process0 creation overhead incurred during a transaction.  D If you make the various processes keep an eye on each other, this isI extremely robust. So far the only live problem ours hasn't recovered from D on its own is a machine reboot, and systartup takes care of that. OfA course, this can pose interesting problems in itself. Once during J development I messed up the shutdown mechanism, and it just refused to die by any other means...    Shane           A koehler@encompasserve.org (Bob Koehler) on 06/06/2001 06:30:28 AM   9 Please respond to koehler@encompasserve.org (Bob Koehler)    To:   Info-VAX@Mvb.Saic.Com  cc:   $ Subject:  Re: $QIO and TCP/IP server    = In article <cddefe9c.0106060105.3bad51bb@posting.google.com>, + joelleauffret@lucent.com (jauffret) writes:   F > After reading a lot of questions/answers I know there is two ways to > implement the server. @ > - a "multi-thread" server within a single server process (AST)H > - the "fork" method to have one process for read/write with one client > ($creprc)   B Process creation is expensive (slow) compared to the alternatives.1 Avoid it when there are appropriate alternatives.   F ----------------------------------------------------------------------? Bob Koehler                     | Computer Sciences Corporation = NASA GSFC Flight Software       | Federal Sector, Civil Group E                                 | please remove ".aspm" when replying    ------------------------------  $ Date: Wed, 6 Jun 2001 21:59:29 +0100, From: "Richard Maher" <maher_rj@hotmail.c*m># Subject: Re: $QIO and TCP/IP server 3 Message-ID: <9fm5gr$930$1@plutonium.btinternet.com>    Hi,   E > I have to write a TCP/IP server, using $QIO interface and accepting - > multiple client connections simultaneously. ( > The environment is OpenVms 7.1/Pascal.  ( You don't say whether it's VAX or Alpha?  ; How many concurrent users do you have? As stated above, you = obviously need multiple connections to a single process and a A threading mechanism to go with it, but I just thought I'd mention A the INETD / Auxilliary Server in case you'd be happy to live with  one VMS process per user.   D > I read examples in UCX$EXAMPLES, but I just find examples with one	 > client.   C If you're using Alpha then try and get your hands on a VAX and look B at sys$examples:db_server.mar. Unfortunately the revisionists haveA seen fit not to port this Ab Fab example of DECnet multi-threaded ? task-to-task programming to Alpha. (Ok. There is a C version on ! Alpha if it's of any use to you.)   : I know you want TCP/IP examples and not DECnet but the ASTC threading concepts are almost identical and IMHO well worth a look! 0 Some of the notable exceptions to the above are:  C 1) Your _BG: devices don't have associated mailboxes so you have to = handle connection failures in *every* $qio AST on the device.   A 2) Out-of-Band/Interrupt data handling is totally different. (And  limited to 1 byte with TCP/IP)  @ 3) You have to accept a connection before you can reject it with TCP/IP.   @ 4) The stream i/o nature of TCP/IP sockets makes variable length= records, and M:M message interaction, a real pain. DECnet has ! those beautiful record boundries!   < 5) TCP/IP has io$m_readattn. With DECnet you have to do zero byte reads.   = So you could do worse than combine the $qio "syntax" from the > one connection TCP/IP example with the threading "design" fromD the DECnet example. No? Oh well, it was just a thought. I just can't> understand why anyone in their right mind would go through the< development contortions necessary for C/pthreads/cma! I mean= take mutexing for example! What's that all about? Surely they < were having a laugh? (What is the Kernel thread limit on VMS> these days and how long did it take to get that limit over 1?)  < Anyway, once you've looked into the problem in detail you'll0 probably start asking yourself question such as:  : . How do I authorize user access to my server? Will I have6 to duplicate what loginout.exe does? Oh no! what about! auditing and intrusion detection?   ; . My ASTs must be short and fast (and not to mention that a ; lot of layered products, notably Rdb, aren't designed to be = called from AST level) so I'll need an additional worker pool & of processes to perform the real work.  > . How can I grow/shrink the size of this worker pool as client demand varies?  ; . How will the Communication server talk to and control the ! Execution Server processing pool?   2 . How much and How long is this all gonna take????  B If you're now thinking along the lines of "It's all too hard!" and5 "Why hasn't someone sorted this out before?" Then you : might want to have a look at the TIER3 Functional Overview below.  = If you would like to know about TIER3 more then just mail me.   9 If you'd like to know why Compaq didn't buy TIER3 and yet @ continues to invest /squander /fritter away your license fees on" Bridgeworks then please mail them.  @ One thing that isn't in the functional overview (yet) is support@ for TIP (Transaction Internet Protocol) which will give you true8 heterogeneous 2PC between anything controlled by MTS/DTC< and anything controlled by DECdtm! It's still on the drawing board but it's firming up!   Regards Richard Mahera  = PS. IMHO (and Microsoft's - Have you noticed their sea change 7 away from threads and back to Asynchronous Callbacks in0 W2K?) threads just won't scale!e     TIER3s   Functional Overview.    K Tier3 is the client/server system that enables remote client access to yourEJ server system resources. As the middleware residing between the client andK server components of your application, Tier3 regulates and schedules client J requests, allocates the resources necessary to satisfy those requests, and< channels the resulting output back to the requesting client.  G Similar to a Remote Procedure Call, Tier3 allows a remote client (eg: arK Powerbuilder program on a Windows PC) to access your 3GL routine running on D a VMS server. But unlike other RPC implementations, Tier3 offers the following benefits: -   K  Transparent Multi-Threading. Tier3 provides an independent multi-threaded I communication server for each application. This relieves your server code 8 from the need to perform its own threading and mutexing.  K  Application Based Tuning & Configuration. Because every Tier3 applicationnJ has its own communication server, inter-application resource contention isE reduced, and each application on a given node can be tuned/configured- independently.  F  Persistent Network Connection. Once accepted, the network connectionG between client and communication server is maintained until either your G client or your server code asks specifically for it to be dropped. ThisMI removes the overhead of having to obtain a channel, request a connection, F and pass authorization, each time a client needs to access your server application.  F  Secure Client/Server Applications. In addition to ensuring that onlyI authorized clients can access your application, Tier3 also identifies thePG username that the client is entitled to assume on the server node. ThisiE allows your server application to perform additional security checks,cJ perform charge-back accounting, or maintain an audit trail of who has been modifying a database.e  I  Re-usable Execution Servers. Tier3 execution servers are re-usable on aMF transactional basis, with your development team controlling how long aH transaction, or server affinity, will be maintained. This is achieved byE effectively supplying your 3GL RECEIVE routine with a full-duplex, or + conversational, pipe as its only parameter.e  D  Dynamic Execution Server Creation. As client demand dictates, eachG applications communication server will grow/shrink the execution server K processing pool, in accordance with the parameters specified by your systems( manager in the Tier3 Configuration File.    Contd. . . L  Preservation of Existing Investment. Tier3 server development is as simpleK as creating a shareable image containing the six subroutines or User ActionvF Routines that Tier3 will activate on your behalf during the life of anL execution server. Your UARs can be written in any 3GL, and have the completeD VMS execution environment at their disposal. You are free to spawn aL sub-process, perform inter-process communication, or your server applicationE can become a client of another Tier3 application on a different node.n  I  This ability to re-use/incorporate existing 3GL code into your new Tier3aI applications, coupled with the fact that your development staff will find G very little difference between developing Tier3 server applications andBL developing any other VMS application, means that your existing IT investment
 is preserved.t  L  No Tier3 Specific Client Software Required. By not layering another API onI top of the standard Socket or System Service interface, Tier3 allows youreI client developers unrestricted access to the underlying network protocol.aI Performance features such as asynchronous SENDs and Interrupt/Out-of-Bandt' functionality are all available to you.   E In summary, once your organization has decided it needs the security,vJ performance, scalability and functionality of a three-tiered client/server5 architecture, then your organization will need TIER3.h  5 TIER3 is a registered trademark of Tier3 Software Ltda    4 jauffret <joelleauffret@lucent.com> wrote in message7 news:cddefe9c.0106060105.3bad51bb@posting.google.com...: > Hi,0E > I have to write a TCP/IP server, using $QIO interface and acceptingk- > multiple client connections simultaneously.D( > The environment is OpenVms 7.1/Pascal.D > I read examples in UCX$EXAMPLES, but I just find examples with one	 > client.u > F > After reading a lot of questions/answers I know there is two ways to > implement the server.s@ > - a "multi-thread" server within a single server process (AST)H > - the "fork" method to have one process for read/write with one client > ($creprc)  >r? > I just want to know if one solution is better than the other, 7 > and if somebody has a good example to implement this.s >t >a > Best regards - Jollei   ------------------------------   Date: 6 JUN 2001 18:27:46 GMTX+ From: Dave Greenwood <greenwoodde@ornl.gov> , Subject: Re: 10 vs 100 Mb ethernet on xp10001 Message-ID: <6JUN01.18274627@feda01.fed.ornl.gov>e  F Since I originated this thread and we now have a successful resolution/ I thought I'd report that.  I wrote previously:n  H >                                            [CSC] told him to down loadL > EWDRIVER_D3500BA_721-1R (assuming I wrote that done correctly).  It didn't > solve the problem.  G The actual name is SYS$EWDRIVER_DE500BA_V72-1R.EXE and it did solve thewE problem - once we renamed it correctly.  For whatever reason - eitherdG faulty CSC instructions or our faulty hearing - the file was originallytL named SYS$EWDRIVER.EXE.  It should have been named SYS$EWDRIVER_DE500BA.EXE.9 The XP1000 is now running 100 Mbit FDX. (Hurrah! Hurrah!)c  F Btw, apparently the imbedded DE500-BA in the XP1000 is not exactly theE same as the add-on DE500-BA and the original VMS support wasn't quiter right for the imbedded version.s  2 Thanks to those - especially Hoff - who responded. Dave --------------9 Dave Greenwood                Email: Greenwoodde@ORNL.GOV H Oak Ridge National Lab        %STD-W-DISCLAIMER, I only speak for myself   ------------------------------  $ Date: Wed, 6 Jun 2001 14:17:37 -0400. From: "Kenneth Randell" <kenr@datametrics.com>D Subject: Re: A New Twist on OpenVMS Marketing... and Media Packaging+ Message-ID: <9fls2h$d1e$1@bob.news.rcn.net>t  ; IIRC, I thought Brian had this on his web site at one time.a   Ken Randell   = Terry C. Shannon <terryshannon@mediaone.net> wrote in message 7 news:eTsT6.23036$zl5.7473788@typhoon.ne.mediaone.net...o >e? > "andrew harrison" <andrew.nospam@uk.sun.com> wrote in messager& > news:3B1E48C0.608D3F5E@uk.sun.com... > >r > > JF Mezei wrote:r > > >o > > > "Terry C. Shannon" wrote:iI > > > > If I was King, er, Capellas For a Day, my first edict would be to  find > thelF > > > > corporate graphics person responsible for the gaffe. Once said person > wasbG > > > > found, I would sentence him or her to a day of education on then
 > products > > > > that Compaq sells. > > >oK > > > Probably done by an outside firm. I doubt that Compaq has Macintoshes  to > do& > > > serious artwork inhouse. :-) :-) > > >n > >i< > > They do. Only a MAC user could have been involved in the7 > > he famous BSOD incident where Compaqs Art dept usedsB > > a picture of a Compaq Intel box with a very nice blue colouredA > > screen with small white letters on it in a Compaq ad campaignO > > for Windows2000. > >o2 > > Or perhaps you were refering to this incident. > >t > K > No, I was referring to the VMS V73 packaging, but thanks for reminding mem ofL > one of the truly classic Great Moments in Compaq Marketing Stupidity. Wish IaG > had a JPEG or GIF of the BSOD, it would be a wonderful addition to myd > presentations! >  >' >a   ------------------------------  $ Date: Wed, 6 Jun 2001 18:44:40 -0500% From: "Rich Jordan" <rjordan@mcs.net>n# Subject: Re: alphastation 200 4/233 4 Message-ID: <3ezT6.7713$j02.115538@news.goodnet.com>  . Brian Schenkenberger, VAXman- wrote in message' <009FD0A8.56D791A9@SendSpamHere.ORG>...o > L >For those looking, look for the Flash ROM chip with the designation 28F020.K >If you have a half-flash Alpha, there shoud be an empty socket of the samesK >size in the general vicinity of the existing 28F020.  The part you need ise? >only a few bucks and had been available from JDR MicroDevices.     E A couple years ago when I tried to update mine, JDR did not carry thehK correct brand.  The ones they did have would not work, as the updater couldtL not program them.  I ended up ordering from another company to get AMD brandL flash chips that were useable; I believe I was told at the time that AMD andG IBM brand chips were the ones that would work; best bet is to match thei' brand you have installed in your AS200.r   Rich Jordant rjordan@mcs.neth   ------------------------------  # Date: Wed, 06 Jun 2001 20:32:28 GMT ) From: rob.buxton@wcc.govt.nz (Rob Buxton)bD Subject: Re: Availability Manager vs DECAMDS Collector Modules query0 Message-ID: <3b1e9256.75110403@news.wcc.govt.nz>   Larry,  C No, there is an Availability Manager Data Analyser that runs on VMS2D on Alpha only. I've not tested it yet, I run the Data Analyser on my NT Workstation.   F My original issue with Availability manager was that there was no data) collector module for VAXes. There is now.i  E Think I'll throw Availability Manager Data Collector on to one of therC VAX test boxes and see if I determine any difference between it and, DECamds.   Rob.  F On 6 Jun 2001 11:50:27 -0500, Kilgallen@eisner.decus.org.nospam (Larry Kilgallen) wrote:n  g >In article <3B1E4292.DE8474C4@clarityconnect.com>, "Mark D. Jilson" <jilly@clarityconnect.com> writes:nE >> The data collectors that run on the VMS system are essentially the K >> same.  But Availability Manager is the future thus if you have to make aeH >> choice then going with AvailMan is the preferred path as it will save& >> you a migration step in the future. >t= >I thought Availability Manager required a Microsoft machine.e >y! >If so, it is not in _my_ future.c   ------------------------------  # Date: Wed, 06 Jun 2001 20:36:56 GMTe) From: rob.buxton@wcc.govt.nz (Rob Buxton)lD Subject: Re: Availability Manager vs DECAMDS Collector Modules query0 Message-ID: <3b1e936b.75387110@news.wcc.govt.nz>   Jilly,  D Thanks, didn't think there was but there's nothing I can find in the@ document sets that say why you should use one and not the other.* There's stuff on running the two together!  E Also looks like some of the features don't really come into their ownnD until you upgrade to VMS 7.3. Bit of a way off here as I've not even: seen the kits yet! Takes a while to get this far south....  F I'll install the Availability Manager on one of our test VAXes and see if I can see any differences.i   Rob.          2 On Wed, 06 Jun 2001 14:43:15 GMT, "Mark D. Jilson"! <jilly@clarityconnect.com> wrote:s  C >The data collectors that run on the VMS system are essentially thepI >same.  But Availability Manager is the future thus if you have to make a F >choice then going with AvailMan is the preferred path as it will save$ >you a migration step in the future. >r >Rob Buxton wrote: >> n
 >> Hi All, >> o4 >> In the past I've used DECAmds to monitor Systems.6 >> Saw pointers to the new Availability Manager stuff. >> oH >> I've loaded the Windows Availability Manager Data analyser on my P.C.D >> It can see all of the VMS Systems, everything seems to be workingI >> fine. A mixture of DECamds 7.2 & 7.3A, VMS 7.2 (Vax) and 7.2-1 (Alpha)e >> r >> The question is:oB >> What's the difference between the Data Collector as supplied by- >> DECamds 7.3A and Availability Manager 2.0.1 >> s8 >> Why would I install one over the other at this point? >>   >> Rob.e >e >-- E >Jilly	- Working from Home in the Chemung River Valley - Lockwood, NY 1 >	- jilly@clarityconnect.com			- Brett Bodine fanl/ >	- Mark.Jilson@Compaq.com			- since 1975 or soh- >	- http://www.jilly.baka.com               -    ------------------------------  $ Date: Wed, 6 Jun 2001 12:39:40 -06004 From: "Michael D. Ober" <mdo.@.wakeassoc.com.nospam>+ Subject: Re: Byte Order on Alpha Processors-3 Message-ID: <WNuT6.1339$Dv6.145648@news.uswest.net>   1 I need to manually convert, which is why I asked.G -- Thanks,f
 Mike Ober.  ? "Hoff Hoffman" <hoffman@xdelta.zko.dec.nospam> wrote in messagel- news:0CsT6.1105$fi2.29041@news.cpqcorp.net...-J > In article <XahT6.721$1Z3.58150@newsread1.prod.itd.earthlink.net>, "Mike% Ober" <obermd.@.alum.mit.edu> writes::I > :I need to convert Alpha WORD and LONG to their Intel PIII equivalants.t The J > :interface is an ASCII stream.  Does anyone either have code (C or BASICK > :preferred) the will make this conversion or information on byte order onO thes- > :Alpha itself.  I have the PIII byte order.o >lH >   As has been discussed elsewhere, the Intel Pentium serues processorsH >   use the little-endian byte order, as do the Alpha platforms that areF >   running OpenVMS.  (Certain Alpha microprocessors can be configured >   for big-endian >'I >   That said, this particular problem has been solved many times before.nH >   Some of the typical solutions include htonl/htons and ntohl/ntohs inF >   the standard C library, and the XDR (eXternal Data Representation)H >   libraries.  One of the more recently introduced network formats, andF >   one that is seeing extensive press and marketing coverage, is XML. >tH >   Or you could simply assume you can use little-endian directly and/orD >   use your own conversion routines as your own "network format"... > ( >  ---------------------------- #include' <rtfaq.h> -----------------------------eL >       For additional, please see the OpenVMS FAQ -- www.openvms.compaq.com, >  --------------------------- pure personal# opinion --------------------------- 1 >    Hoff (Stephen) Hoffman   OpenVMS Engineering  hoffman#xdelta.zko.dec.com >b   ------------------------------  % Date: Wed, 06 Jun 2001 15:02:24 -0400t- From: JF Mezei <jfmezei.spamnot@videotron.ca>oQ Subject: Re: Can you format sent VMS mail( was: Can you format recieved VMS mail)l, Message-ID: <3B1E7E3D.8CD94332@videotron.ca>   "Jay E. Morris" wrote:K > I, and my remote users, would love this.  And at one time we were lookingdM > into what would be required to do this.   And then the Air Force instituted  > Barrier Reef.  Firewall.  K In that case, the only truly safe way to send your documents to ensure thateI the client doesn't mangle them during printing would be PDF. PDF also hastK built-in compression to a certain extent, so the overhead of the PDF format'T may be compensated by the compression. It would also allow you to have nicer output.  K You could write a nice postscript wrapper that includes your corporate logoyK ect, and then includes the ascii text , splitting pages properly. You could I then pass this through ghostscript or some other utility to generate PDF.?   ------------------------------  $ Date: Wed, 6 Jun 2001 14:44:35 -0400; From: "Brian Tillman" <tillman_brian@notnoone.notnohow.com> . Subject: Re: Dec-net ethernet address on a vax$ Message-ID: <3b1e7a34$1@news.si.com>  C >  I am new to VMS !!! My question is how do u find out the DEC-NET ( >ethernet address of a MicroVax 3100-30.  J Do you mean the hardware NIC or the modified Ethernet address?  The former' can be found with (for DECnet Phase IV)   ( $ mcr ncp show line xxxx characteristics  J where "xxxx" is your Ethernet line descriptor (ISA-0, FQA-0, MFA-0, etc.). For example:   $ ncp sho line fqa-0 char     8 Line Volatile Characteristics as of  6-JUN-2001 14:42:45   Line = FQA-0   Receive buffers          = 12E! Controller               = normalf# Protocol                 = Etherneto Service timer            = 4000d8 Hardware address         = 08-00-2B-B0-9D-3E        <--- Device buffer size       = 1498a Requested TRT            = 7987o Valid transmission time  = 2621  Restricted token timeout = 1000l Ring purger enable       = off, NIF target               = 00-00-00-00-00-00, SIF configuration target = 00-00-00-00-00-00, SIF operation target     = 00-00-00-00-00-00, Echo target              = 00-00-00-00-00-00 Echo data                = 00r Echo length              = 0  < The modified Ethernet address can be calculated.  It will beH AA-00-01-00-xx-yy, where xx-yy will be formed by multiplying your DECnetL area by 1024 and adding the node number, converting to hex and reversing theK two halves.  For example, for a node whose DECnet address is 2.17, the last  two octets would be -11-08 (2*1024+17=2065=%x0811).  I (I may be a little off on the first three octets of the address.  Someoner here will correct me if I am.) --A Brian Tillman                   Internet: tillman_brian at si.comIA Smiths Aerospace                          tillman at swdev.si.comt= 3290 Patterson Ave. SE, MS      Addresses modified to preventa< Grand Rapids, MI 49512-1991     SPAM.  Replace "at" with "@"8        This opinion doesn't represent that of my company   ------------------------------   Date: 6 JUN 2001 20:19:05 GMTc4 From: karcher@thuria.waisman.wisc.edu (Carl Karcher)$ Subject: DECAMDS threshold "classes"5 Message-ID: <6JUN01.20190532@thuria.waisman.wisc.edu>a  I While we're on the subject of AMDS here's something I've always wondered:d  D In the customizable event threshold values (AMDS$THRESHOLD_DEFS.DAT)G there are 4 "classes" of values defined. I assume these are for variousuC processor performance levels. For example: there are four different B threshold values that trigger the event "Buffered IO rate is high"< (HIBIOR) - 20, 80, 150 and 200 for classes 1-4 respectively.  A What I'd like to know is how to set the "class" for ALL the eventg thresholds?n  H I'm running AMDS V7.3A on an Alphaserver 1200/533 and all of the defaultC threshold values are class 1 which I think is way too low. I'd liketD to change the threshold values to class 2 or 3. Is there a way to do5 that without changing every threshold value manually?h   There's an interesting logical:r  K  "AM_NATIVECLASSES" [exec] = "AMDS$AM_SYSTEM:AMDS$AM_NATIVECLASSES_SHR.EXE"t  , However that sharable image is not present.    ------------------------------  $ Date: Wed, 6 Jun 2001 15:41:01 -04002 From: "Sue Skonetski" <susan.skonetski@compaq.com> Subject: Diamond Forum agendat3 Message-ID: <fHvT6.1123$fi2.29233@news.cpqcorp.net>l  !  Here is the agenda for your use.1   Compaq's OpenVMS Diamond Forum   Compaq Customer Centre   Silicon Valley  & 10600 Ridgeview Court  Building CAC13   Cupertino, CA  95014      
 June 26, 2001    Agenda  %   8:00   Sign-in and buffet breakfastg     8:35   Welcome to Compaq!o  >                         Jim Burdick, Director ~ Western Region  )   8:45   OpenVMS Strategies and Directiono  C                         Rich Marcello, VP and General Manager, High' Performance Systemsb  "   9:45          eBusiness Strategy  ?                         Wendy Herman, OpenVMS eBusiness Manager   " 10:15   Swiss Stock Exchange Video  
 10:20   Breaki  9 10:30            AlphaServer Product Update and Directione  >             Cathy Stockwell, Manager  ~ AlphaServer Technology Communications  . 11:30   Mission Critical Services - Case Study  D                         West Coast Mission Critical Services Manager  
 12:00   Lunchr   12:45   Server Consolidation  (                         Mark Santanocito  5 1:30            OpenVMS SAN for eBusiness Integrationb  F                         TBD,             Enterprise Storage Consultant  #   2:45            Customer Dialoguee     3:00   Close   ------------------------------  $ Date: Wed, 6 Jun 2001 14:03:17 -04005 From: "Fred Kleinsorge" <kleinsorge@star.zko.dec.com>q# Subject: Re: disabling floppy drivee3 Message-ID: <gfuT6.1115$fi2.29223@news.cpqcorp.net>s  I Won't work.  The floppy controller is in the junk IO and the drive itselfh/ isn't "sensed" in any way by autoconfiguration.   J To not load the floppy, you can edit sys$config.dat and remove (or corruptI in non-fatal ways) the records that define the floppy (there are several,-- because of severl methods on some platforms).o  G But if you are booting the CD, the only way to do it is to create a 3rd0H party boot floppy (yes, a floppy) which is a normal VMS disk, but with aG rooted directory structure (i.e. like a system disk).  On that disk you4K would include a CONFIG.DAT file which would have an appropriately corrupted J record for the floppy (like a bogus driver name) which would then overrideL the sys$config.dat information.  You would then do a 3rd party boot, and theI floppy would be read during the boot, but would not configure a driver atm runtime.  H Did this for a problem when IntraServer was first getting 3rd party boot working.      - William_Bochnik@acml.com wrote in message ...a >nA >how about brute force - pull the cable from the floppy drive (orn$ >pull the controller if not onboard) >  >s > $ >                    Fred Kleinsorge; >                    <kleinsorge@star.zk                To:n Info-VAX@Mvb.Saic.Com ; >                    o.dec.com>                         cc:mJ >                                               Subject:     Re: disabling floppy drive( >                    06/05/2001 03:46 PM >l >p   >m >s >e@ >Yes, but it's not pretty.  But if you can mount the system disk	 >and edit/) >sys$config,dat, then it's fairly simple.C >b >y< >Herb Wong wrote in message <3b1a4d6d_3@news.pcmagic.net>...? >>I have a ALPHA server 1000. O believe there is a problem with  >the floppygB >>drive controller . When I run the console 'TEST" command, it get
 >errors onA >>the drive. I have tried replacing the drive and cable, there isn
 >still the@ >>problem. I believe there is a floppy drive controller problem. >When I tryu >yor> >>install VMS 6.x thry VMS 7.x, I get a bugcheck dump when the >install? >>procedure trys to poll the floppy drive. The error message is- >I/O? >>inconsistant database. However if I install TRUX64 or LINUX I  >have noA >>problems. My question is: Is there anyway to disable the floppy- >so that VMS  >>does not try to check for it?? >> >>Thanks >> >> >> >> >M >i >f >e >p >oG >______________________________________________________________________  >m; >The information contained in this transmission may containr= >privileged and confidential information and is intended onlyl= >for the use of the person(s) named above. If you are not thel= >intended recipient,  or an employee or agent responsible for @ >delivering this message to the intended recipient,  any review,A >dissemination, distribution or duplication of this communication ? >is strictly prohibited. If you are not the intended recipient,sB >please contact the sender immediately by reply e-mail and destroy$ >all copies of the original message. >    ------------------------------  # Date: Thu, 07 Jun 2001 00:21:14 GMT ) From: rob.buxton@wcc.govt.nz (Rob Buxton)  Subject: Re: DSN Linkt0 Message-ID: <3b1ec7fd.88845272@news.wcc.govt.nz>  6 There's also WIS which is the WEB based variant of DSN  @ You connect to this via a browser, login using a Compaq supplied Username and password.  F You have access to the ITS Database, useful for trying to find answers to known problems.  7 You can log calls and keep a progress on them via this.   < As for DSN you need to have a Software Maintenance Contract.  @ Here in New Zealand, the local WIS Database is over the ditch inA Australia. Not sure where your local DSN or WIS service would be.u  D Go back to your local support people and push them a bit harder. YouE may need to log a Software Call to get the Service set up. That's the0 way I did it here.      E On Wed, 06 Jun 2001 15:34:12 +0100, steven.reece@quintiles.com wrote:J   >0 >Kenneth asked:e >>>>L >I saw many advise here to ask people look for information in DSN Link, whatJ >is it and how can I access or subscribed it? I have call my local support >ine' >HK and he seems know nothing about it.- ><<< >b( >DSNlink = Digital Services Network link > G >It provides a method for Compaq to talk to you and you to talk to them=J >where talk might be conducted over the internet or over a modem and phone >line between you and them.oH >The talk itself can be you logging a fault call for a machine which youD >have on maintenance or receiving information from Compaq on serviceJ >offerings, show-stopping crashes and anything else that they think may beC >useful for the customer base but that is inappropriate to put in a0H >newsgroup.  Compaq can also login to your systems using DSNlink so thatD >they can investigate crashes, study system performance etc for you. >yL >Another part of the DSNlink offering is the ITS database which I believe isH >a subset of the information which Compaq have internally in their STARSK >database.  The information includes examples of coding, patches (which can,L >be downloaded or, if they're too large, sent by tape), useful tips etc etc. >aL >In the UK the requirement is that you have at least one node on maintenanceK >with Compaq.  I think this has to be the gateway node (the machine at youruB >site which actually talks to Compaq) but this may be a gray line. >s >Steve.  >n   ------------------------------  % Date: Wed, 06 Jun 2001 19:02:07 -0500m. From: Lyndon Bartels <lbartels@pressenter.com>8 Subject: esa12000fc and ema12000 differences/experiences. Message-ID: <3B1E7E2F.24BF3D41@pressenter.com>  B I'm looking at the differences between the ESA based storage works2 hardware and the EMA based storage works hardware.  G In particular, I'm investigating the esa12000fc and the ema12000. (URLSu below.)e  A http://www.compaq.com/products/storageworks/ma8kema12k/index.htmlw   andn  A http://www.compaq.com/products/storageworks/esa12000fc/index.htmli    F Has anybody had experience with either of this storage solutions? WhatC can you tell me about reliability? Flexibility? Expandability? ETC.   F I'm toying with getting the newer EMA solution, but I'm a bit skiddishE about jumping into relatively new hardware, when the ESA and the goodd7 ol' Storageworks SBBs have run so well, for so long....s     Thanks in advance,   Lyndon       -- PG My opinions are mine and mine alone. They seldom align with those of myi	 employer.u    H The only good thing about putting the cart before the horse is you don't have to look at the horse's butt.   ------------------------------  % Date: Wed, 06 Jun 2001 20:32:39 -0400v( From: Hamlyn Mootoo <univms@bigfoot.com>< Subject: Re: esa12000fc and ema12000 differences/experiences+ Message-ID: <3B1ECBA7.9211EEFB@bigfoot.com>n  L No problems to report using the 1" drives in MA8000 configs.  I think you'llJ like 'em better.  I always hated the cheesy plactic old SBB's.  But that's7 only one opinion.  And hot-damn are those HSG80's fast.d   HM   Lyndon Bartels wrote:o  D > I'm looking at the differences between the ESA based storage works4 > hardware and the EMA based storage works hardware. >oI > In particular, I'm investigating the esa12000fc and the ema12000. (URLSs	 > below.)  >sC > http://www.compaq.com/products/storageworks/ma8kema12k/index.html  >r > andm > C > http://www.compaq.com/products/storageworks/esa12000fc/index.htmli >aH > Has anybody had experience with either of this storage solutions? WhatE > can you tell me about reliability? Flexibility? Expandability? ETC.m > H > I'm toying with getting the newer EMA solution, but I'm a bit skiddishG > about jumping into relatively new hardware, when the ESA and the goodo9 > ol' Storageworks SBBs have run so well, for so long....t >o > Thanks in advance, >i > Lyndon >n > --I > My opinions are mine and mine alone. They seldom align with those of myf > employer.  >yJ > The only good thing about putting the cart before the horse is you don't > have > to look at the horse's butt.   ------------------------------  # Date: Thu, 07 Jun 2001 04:18:42 GMTe$ From: Scott Vieth <svieth@wi.rr.com>< Subject: Re: esa12000fc and ema12000 differences/experiences) Message-ID: <3B1F011A.60638106@wi.rr.com>p  L A big difference between the ESA12000 and EMA12000 is the speed and capacity  
 of the disks.r  H The ESA12000 setup will never go faster than 10,000 rpm.  And there will never be drivesyF bigger than 36 GB.  And the maximum number of supported devices is 72.  D The EMA12000 is supposed to host up to 84 disks with an upcoming ACS release.4 (That's six buses with 14 disks on each bus! Zowie!)  I The ESA12000 and RA8000 are already old news.  But after having installedt threerG of these arrays I can say they're pretty rock-solid.  And the 36GB, 10k  drives are easy to find  on the market at a good price.  L The next array I buy will definitely be an EMA12000.  We'll be moving to the
 1" drives.  G The ESA12000 that I have connected to my IDX system has produced a HUGEp increase0 in i/o throughput over the HSJ50s we were using.   -Scott, fc plumber ;^)   Lyndon Bartels wrote:n  D > I'm looking at the differences between the ESA based storage works4 > hardware and the EMA based storage works hardware. >eI > In particular, I'm investigating the esa12000fc and the ema12000. (URLSl	 > below.)t > C > http://www.compaq.com/products/storageworks/ma8kema12k/index.htmlh >s > anda > C > http://www.compaq.com/products/storageworks/esa12000fc/index.htmlU >0H > Has anybody had experience with either of this storage solutions? WhatE > can you tell me about reliability? Flexibility? Expandability? ETC.. >.H > I'm toying with getting the newer EMA solution, but I'm a bit skiddishG > about jumping into relatively new hardware, when the ESA and the good 9 > ol' Storageworks SBBs have run so well, for so long....o >  > Thanks in advance, >  > Lyndon >r > --I > My opinions are mine and mine alone. They seldom align with those of myS > employer.E >BJ > The only good thing about putting the cart before the horse is you don't > have > to look at the horse's butt.   ------------------------------  % Date: Wed, 06 Jun 2001 14:57:33 -0400$- From: JF Mezei <jfmezei.spamnot@videotron.ca>D; Subject: Re: Hydro, was (OT) Current Microsoft V DOJ statusH, Message-ID: <3B1E7D1A.97910EED@videotron.ca>  & paddy.o'brien@zzz.tg.nsw.gov.au wrote:E > In Wales there are two sites Dinorwic and Ffestinniog which are nottF > classed as hydro stations, but "pumped storage" (another, Foyers, inK > Scotland).  The former was built in a granite mountain when the authority:I > was govt.  Private would never even consider such a venture, but that'so > another hobby-horse.  N Don't they have some pumping schemes in the Snowy project as well ? (closer toK your home). When I visited the area in 1997, I was suprised to be told thatkM the peak time for power in Sydney is the morning, whereas here in Canada, ourIK peak consumption is early evening. I guess it is an air conditioning versus M heating issue ! (Here is Qubec, electric heating of homes is quite popular).6  K I know that they are able to pump water from Lake Jindabyne up to some pondmJ way up in the mountains before that water then drops down to Murray-1 in aM very impressive series of very large white "pipes" that come out from the topwL of the mountain and goes down to the station. (Reminiscent of some scenes inK the TV series "V" quite a few years ago). There are over 50km of tunnels to K take water that would normally flow into the pacific and redirect it acrosshL the continental divide to flow in the murry river that empties near Adelaide in the southern ocean.   ------------------------------  % Date: Wed, 06 Jun 2001 15:29:38 -0400)- From: JF Mezei <jfmezei.spamnot@videotron.ca>u; Subject: Re: Hydro, was (OT) Current Microsoft V DOJ statusr, Message-ID: <3B1E849E.B7EE20D5@videotron.ca>   Robert Deininger wrote: G > There's another consideration, which is keeping the river flow fairly H > consistent.  There'd be trouble if they pumped so much water uphill to  > significantly reduce the flow.  E Or divert rivers completely. Residents/farmers downstream of the Lake M Jindabyne in Australia had to fight to retain a bit of water flow because thehM Snowy project diverted all the water from that river "up" to the Murray river K on the other side of the continental divide not only to produce electricity6> but also to irrigate the desert between the alps and adelaide.  M However, if you live in an aera with winter/snow, vast variations in the flowtL and level of rivers/lakes are a fact of life. The problem with Hydro is thatH water flow is the least in winter when you most need electricity. So theN "pumping" is actually a question of maintaining water levels in the reservoirsB during summer/automn so that you have enough water for the winter.   ------------------------------  % Date: Wed, 06 Jun 2001 17:02:26 +0100r  From: steven.reece@quintiles.com; Subject: Re: Hydro, was (OT) Current Microsoft V DOJ status2H Message-ID: <OF1C9C2ED8.DE92DD0B-ON80256A63.0057E44A@qedi.quintiles.com>  E I would imagine that there's some sort of clutch assembly between the H turbine and the generator so that the water can still be allowed to flow' even if the generators aren't required..  K Anyway, what about a plate with some chips on it to help the fishies? :-)))i   Steve.   Robert Deininger wrote:" >>>iJ I tend to think of the big hydro plants.  I shudder to think what it would= take to pump water back up into the Hoover dam, for instance.   E There's another consideration, which is keeping the river flow fairlydF consistent.  There'd be trouble if they pumped so much water uphill toG significantly reduce the flow.  At some times of year, they have to let E water down even if they don't need the power, for example to help the  fishies. <<<a   ------------------------------  % Date: Wed, 06 Jun 2001 16:23:16 -0400a- From: JF Mezei <jfmezei.spamnot@videotron.ca>n; Subject: Re: Hydro, was (OT) Current Microsoft V DOJ status , Message-ID: <3B1E912C.613D3735@videotron.ca>  ! steven.reece@quintiles.com wrote:l > G > I would imagine that there's some sort of clutch assembly between the J > turbine and the generator so that the water can still be allowed to flow) > even if the generators aren't required.   N Simple: just don't electrify the magnets in the generator, no electricity willF be generated ! But you still need to control the speed of the turnine.   ------------------------------  $ Date: Wed, 6 Jun 2001 17:01:09 -0700* From: "Jack Peacock" <peacock@simconv.com>; Subject: Re: Hydro, was (OT) Current Microsoft V DOJ status-4 Message-ID: <hxzT6.3972$Ib.425946@news1.primary.net>  ? "Robert Deininger" <rdeininger@mindspring.com> wrote in messagerF news:rdeininger-0606011150570001@user-2ivec9c.dialup.mindspring.com...H > Thanks for the explanation.  I suspect that our US electricity markets areoG > not quite "free" enough to make pumping work on a large scale.  There  mayt= > be some smaller plants that do this that I've not heard of.- >aF > I tend to think of the big hydro plants.  I shudder to think what it would ? > take to pump water back up into the Hoover dam, for instance.g > A Not *at* Hoover Dam but not too far away just such a pump storageeH generation system is being constructed for peak power for Las Vegas.  It> was approved by the Utilities Commission a few months back andG construction has begun (according to the local paper).  The system will H pump water uphill at night and use it to generate power in the afternoonC peak demand hours.  A private company is building it.  As Nevada is H power plant friendly, unlike California, and new area plants are already> approved or under construction there should be enough off hourA generating capacity to make it worthwhile.  The water will remaino> underground so evaporation loss will theoretically be minimal.  D By the way, the intakes at Hoover dam are located in the towers thatF stick out from the dam itself, into the lake.  Water is taken near theH top of the lake to get an additional 300ft or so drop.  I've been in theB underground area where the intake pipes go to the generators, it'sF impressive (and very damp, under the water table).  These are separate@ from the spillways, which are caverns in the side of the canyon.    Jack Peacock"   ------------------------------  $ Date: Wed, 6 Jun 2001 17:09:20 -0700* From: "Jack Peacock" <peacock@simconv.com>; Subject: Re: Hydro, was (OT) Current Microsoft V DOJ statuss4 Message-ID: <YEzT6.3973$Ib.426018@news1.primary.net>  : "JF Mezei" <jfmezei.spamnot@videotron.ca> wrote in message& news:3B1E849E.B7EE20D5@videotron.ca...F > However, if you live in an aera with winter/snow, vast variations in the flowF > and level of rivers/lakes are a fact of life. The problem with Hydro is that C > water flow is the least in winter when you most need electricity.  >uF Depends on where you live.  In the U.S. Southwest demand skyrockets inH the summer.  Here in Las Vegas we also have a bit more of a demand afterH dark, and more in winter than summer, for the lights of the casino signsF (trivia: the lights on top of the Luxor, the pyramid, can be seen fromC low earth orbit even against the backdrop of the rest of the town).b   Jack Peacock   ------------------------------  $ Date: Wed, 6 Jun 2001 22:48:03 -0400  From: John Santos <JOHN@egh.com>; Subject: Re: Hydro, was (OT) Current Microsoft V DOJ status 5 Message-ID: <1010606223739.4229A-100000@Ives.egh.com>   % On 6 Jun 2001, Larry Kilgallen wrote:   K > Twenty years ago there was a pump storage plant in western Massachusetts,i$ > but I believe it is now shut down.  E It was still operating a couple of years ago.  It is on the DeerfielduD River, very near the former Rowe reactor, now dismantled.  (Rowe wasI the 1st commercial nuke in the US, and the first one to be decommissionedp! and dismantled due to age, IIRC.)r  = Keeping the water flow on the Deerfield under control is veryv? important because white-water rafting is now one of the biggestr? businesses in the area!  The pumped storage plant uses a dam ont@ the river as the low reservoir, and a pond, possibly artificial,< high on a ridge, as the high reservoir.  When too much water< comes down from the top, they need to release water from the> dam, but they can't do it when people are rafting on the river> below (a big surge could kill someone).  However, they have to< maintain at least a minimal flow on the river during raftingA times (daytime during the summer), even when they would rather be>= storing water at the dam to pump uphill that night.  This all_< depends on the recent weather and the forecast, so there are< lots of scheduling difficulties. I wonder if they use VMS???   -- i John Santoso Evans Griffiths & Hart, Inc. 781-861-0670 ext 539   ------------------------------  % Date: Wed, 06 Jun 2001 11:49:32 -0700n! From: Shane.F.Smith@Healthnet.comi" Subject: Re: I demand your respectD Message-ID: <OF6F2BBF1C.EF4E7434-ON88256A63.006745DE@foundation.com>  B What an excellent way to make sure nobody has any respect for him.   Shaneh  . [Foul mouthed, insignificant rantings snipped]   ------------------------------  % Date: Wed, 06 Jun 2001 16:57:59 -0500g* From: WILLIAM WEBB <WWEBB1@email.usps.gov> Subject: RE: RE: Is it just me?M- Message-ID: <0033000000978200000002L002*@MHS>    =0ANo, it's not just you.g  H I've suggested OpenVMS golf shirts for the "office casual" environment.=     WWWebb   > -----Original Message-----1 > From: Info-VAX-Request@Mvb.Saic.Com at INTERNETs( > Sent: Wednesday, June 06, 2001 3:51 PMF > To: Webb, William W - Raleigh, NC; Info-VAX@Mvb.Saic.Com at INTERNET! > Subject: OT: RE: Is it just me?o >s >nB > And for me, they keep eyeing my pocket sized OpenVMS compass and > clock that I keep on my desk.n >t? > I also would like to see more of the Compaq OpenVMS marketingh= > merchandise on the "Cool OpenVMS Mechandise" site for sale.  >a? > Afterall, they have this stuff (T-shirts, Galaxy Balls, etc.) C > in stock with various OpenVMS Ambassadors at one time or another.hD > Why not advertise, stock and sell it?  .... Like we would want for$ > the other stuff they carry <grin>. >> > :) jck > John Koska > Matthew Bender & Co., Inc. -$ >   A Member of the LexisNexis Group > 1275 Broadways > Albany, NY  12204c > USAl > 518-487-3255 > JKoska@lexisnexis.com  >h, > "I post personal opinion only, and all the, > disclaimers one could imagine apply.  That* > includes, I speak for myself only and my, > views in no way represent my employer(s)." >  > > -----Original Message-----, > > From: Koloth [mailto:koloth@tmisnet.com]) > > Sent: Tuesday, June 05, 2001 10:36 PMm > > To: Info-VAX@Mvb.Saic.Comt > > Subject: Is it just me?o > >m > >a@ > > Is it just me or do other people think that the CD holder in > > the OpenVMSa4 > > V7.3 distribution would also work for music CDs? > >tH > > Is it possible to order just the CD case with its, "Rich Corinthian=  < > > Leather," cover?  Several people were eyeing my CD case.8 > > Could be a better marketing gimic than the umbrella. > > 7 > > "OpenVMS mean you have time to listen to the music"i > >a > > Regardsm > >a > > Cass Witkowski > >A >=   ------------------------------  % Date: Wed, 06 Jun 2001 18:17:34 -0700,! From: Tom Linden <tom@kednos.com>  Subject: RE: RE: Is it just me? 9 Message-ID: <CIEJLCMNHNNDLLOOGNJIOEFECLAA.tom@kednos.com>t  B If you are going to do Golf shirts, what about green fees as well?   Tome Pebble Beach
 California   > -----Original Message-----3 > From: WILLIAM WEBB [mailto:WWEBB1@email.usps.gov] ( > Sent: Wednesday, June 06, 2001 2:58 PM > To: Info-VAX@Mvb.Saic.Com0! > Subject: RE: RE: Is it just me?r >  >  >  > No, it's not just you. > I > I've suggested OpenVMS golf shirts for the "office casual" environment.n >  > WWWebb >  > > -----Original Message-----3 > > From: Info-VAX-Request@Mvb.Saic.Com at INTERNETs* > > Sent: Wednesday, June 06, 2001 3:51 PMH > > To: Webb, William W - Raleigh, NC; Info-VAX@Mvb.Saic.Com at INTERNET# > > Subject: OT: RE: Is it just me?o > >a > >1D > > And for me, they keep eyeing my pocket sized OpenVMS compass and! > > clock that I keep on my desk.b > >aA > > I also would like to see more of the Compaq OpenVMS marketinge? > > merchandise on the "Cool OpenVMS Mechandise" site for sale.  > >gA > > Afterall, they have this stuff (T-shirts, Galaxy Balls, etc.) E > > in stock with various OpenVMS Ambassadors at one time or another.tF > > Why not advertise, stock and sell it?  .... Like we would want for& > > the other stuff they carry <grin>. > >e
 > > :) jck > > John Koska  > > Matthew Bender & Co., Inc. -& > >   A Member of the LexisNexis Group > > 1275 Broadwayh > > Albany, NY  12204a > > USAo > > 518-487-3255 > > JKoska@lexisnexis.com- > >-. > > "I post personal opinion only, and all the. > > disclaimers one could imagine apply.  That, > > includes, I speak for myself only and my. > > views in no way represent my employer(s)." > >n  > > > -----Original Message-----. > > > From: Koloth [mailto:koloth@tmisnet.com]+ > > > Sent: Tuesday, June 05, 2001 10:36 PMl > > > To: Info-VAX@Mvb.Saic.Comy > > > Subject: Is it just me?  > > >S > > >aB > > > Is it just me or do other people think that the CD holder in > > > the OpenVMSA6 > > > V7.3 distribution would also work for music CDs? > > >sI > > > Is it possible to order just the CD case with its, "Rich Corinthianm> > > > Leather," cover?  Several people were eyeing my CD case.: > > > Could be a better marketing gimic than the umbrella. > > >c9 > > > "OpenVMS mean you have time to listen to the music"  > > >n
 > > > Regardsi > > >t > > > Cass Witkowski > > >t > >  >    ------------------------------  $ Date: Wed, 6 Jun 2001 15:15:36 -04002 From: "Sue Skonetski" <susan.skonetski@compaq.com>, Subject: OpenVMS Diamond Forum in California3 Message-ID: <rjvT6.1122$fi2.29367@news.cpqcorp.net>     Dear Newsgroup,  K Attached is an invitation to the Diamond Forum in California.  I think that E I have gotten anything that might be microsoft out of the invitation. F Please keep in mind that this is a high level session, not a technicalJ session.  Rich Marcello will be the keynote.  The next message will be theL agenda.  Please keep in mind that Dale Howard the contact  my new supervisor and be kind ;;.)   Suer  * __________________________________________" Come to the OpenVMS Diamond Forum!  G You are invited to join an exclusive audience to hear about the "Future + Directions and Strategy of Compaq OpenVMS".d   Compaq Customer Center  % 10600 Ridgeview Court  Building CAC13    Cupertino, California  95014      E Beginning with breakfast, this event will share business strategy andhA product plans, and re-introduce OpenVMS customers to the exciting B developments and long-term commitment to this recognized leader inH available, scaleable environments.  Here you will be afforded the uniqueH opportunity to meet with Compaq OpenVMS Management to ask questions that$ will assist you in your IT planning.       Tuesday, June 26th, 2001   8 a.m. - 3 p.m.w  * As seating is limited, please R.S.V.P. to:    Dale Howard at (603-884-7300) or   dale.howard@compaq.com   ------------------------------    Date: 07 Jun 2001 01:59:30 +0800, From: Paul Repacholi <prep@prep.synonet.com>! Subject: Re: OT - demise of ML770 - Message-ID: <87lmn51ocd.fsf@prep.synonet.com>b  6 "Terry C. Shannon" <terryshannon@mediaone.net> writes:  < > > Just out of my curiousity, when was the ML770 officiallyF > > 'retired'?  I believe that Compaq signed up to re-badge the Unisys" > > product only last > June 2000.  - > > Was this due to poor sales, poor support?/  F > It was in fact retired last month. I'd venture to guess that no moreE > than a dozen were sold. Nowhere near enough to justify the care anduC > feeding effort.  You want 32 Intel processors? Buy a four-pack oftA > ProLiant 8x00s and cluster 'em. This solution would save you 30uF > percent over the cost of a Miss Piggy (internal name for the ML770).  6 About 3 weeks ago as I remember. Sales? 7 in the UK...  E > The ML770 suffered some serious scalability problems, especially inn > the I/O department.i  B > And Compaq is working on its own McKinley-based 32 (or more) wayD > system.  Makes sense: there aren't many apps certified for WindozeD > Enterprise Deluxe Premium Envoy Class, or whatever they're callingD > the top-o-the-line Windoze these days. What's more, the Unisys boxF > was going to be upgraded to Itanium, prudent customers will wait forF > McKinley. And Compaq is developing its own 32-way (or more) McKinley> > box, replete with WildFire technology and other Good Things.  A It will be interesting to see how fast off the block big piggy ish. compared to the DS20s getting the 1GHz Alphas.   --  < Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda.t@                                              West Australia 6076. Raw, Cooked or Well-done, it's all half baked.H Spam-To: uce@ftc.gov,enforcement@sec.gov,sness@fcc.gov,hfurchtg@fcc.gov,#   mpowell@fcc.gov,gtristan@fcc.gov g   ------------------------------  $ Date: Wed, 6 Jun 2001 14:34:55 -0400. From: "Kenneth Randell" <kenr@datametrics.com>! Subject: Re: OT - demise of ML770n+ Message-ID: <9flt2v$hln$1@bob.news.rcn.net>   I The Unisys web site claims that Hitachi and ICL are also re-selling thesehJ boxes.  I thought DELL was as well.  Some 350+ of the boxes have been soldJ so far.  I also saw where Unisys is adding about 200 folks to support this Datacenter effort.  K If the ramp-up to Datacenter has been 'much slower than expected', how many L were expected to be sold by now?  Does this box have VAX 9xxx written on it?   Ken Randellp  = Terry C. Shannon <terryshannon@mediaone.net> wrote in message 7 news:UQPS6.21567$zl5.6607784@typhoon.ne.mediaone.net...r >c > > ! > > Assunto: OT - demise of ML770  > >  > >0J > > Just out of my curiousity, when was the ML770 officially 'retired'?  IJ > > believe that Compaq signed up to re-badge the Unisys product only last > June	 > > 2000.d > >n- > > Was this due to poor sales, poor support?t > K > It was in fact retired last month. I'd venture to guess that no more thand a F > dozen were sold. Nowhere near enough to justify the care and feeding effort. E > You want 32 Intel processors? Buy a four-pack of ProLiant 8x00s and  clusternL > 'em. This solution would save you 30 percent over the cost of a Miss Piggy  > (internal name for the ML770). >hI > The ML770 suffered some serious scalability problems, especially in thev I/Oo
 > department.c >lJ > And Compaq is working on its own McKinley-based 32 (or more) way system.F > Makes sense: there aren't many apps certified for Windoze Enterprise DeluxeE > Premium Envoy Class, or whatever they're calling the top-o-the-linei WindozeaE > these days. What's more, the Unisys box was going to be upgraded to  Itanium,L > prudent customers will wait for McKinley. And Compaq is developing its ownK > 32-way (or more) McKinley box, replete with WildFire technology and other- > Good Things. >a >  >, >    ------------------------------  % Date: Wed, 06 Jun 2001 16:52:44 -0400d> From: "Koska, John C. (LNG-MBC)" <John.C.Koska@lexisnexis.com> Subject: OT: RE: Is it just me?AK Message-ID: <3D35AD137AAAD411A6BA0008C7B1B12D6DC0EE@MBCALBEXC03.BENDER.COM>m  A And for me, they keep eyeing my pocket sized OpenVMS compass and l clock that I keep on my desk.h  > I also would like to see more of the Compaq OpenVMS marketing ; merchandise on the "Cool OpenVMS Mechandise" site for sale.e  = Afterall, they have this stuff (T-shirts, Galaxy Balls, etc.)tB in stock with various OpenVMS Ambassadors at one time or another. B Why not advertise, stock and sell it?  .... Like we would want for" the other stuff they carry <grin>.   :) jck
 John Koska Matthew Bender & Co., Inc. -"   A Member of the LexisNexis Group
 1275 Broadwayo Albany, NY  12204f USA, 518-487-3255 JKoska@lexisnexis.comf  * "I post personal opinion only, and all the* disclaimers one could imagine apply.  That( includes, I speak for myself only and my* views in no way represent my employer(s)."   > -----Original Message-----* > From: Koloth [mailto:koloth@tmisnet.com]' > Sent: Tuesday, June 05, 2001 10:36 PMr > To: Info-VAX@Mvb.Saic.Coml > Subject: Is it just me?l >  > ? > Is it just me or do other people think that the CD holder in n
 > the OpenVMSa2 > V7.3 distribution would also work for music CDs? > E > Is it possible to order just the CD case with its, "Rich Corinthiant: > Leather," cover?  Several people were eyeing my CD case.6 > Could be a better marketing gimic than the umbrella. > 5 > "OpenVMS mean you have time to listen to the music"2 > 	 > Regards7 >  > Cass Witkowski >    ------------------------------  % Date: Wed, 06 Jun 2001 20:02:24 -0500s1 From: Robert Alan Byer <byer@mail.ourservers.net>m$ Subject: Re: select() on non-sockets3 Message-ID: <3B1E8C50.3B154A6C@mail.ourservers.net>e   Edward Brocklesby wrote: >  > Hi,m > H > is it possible to call select() on a non-socket FD (eg, stdin) on VMS?D > I have some software which does this (an IRC client), but it seems# > it fails with 'invalid argument'.m > + > if not, is there a better way to do this?  > 
 >     -larne-t  E Check my port of Freeciv v1.11.4 as I recall that Roger Tucker put ina6 some OpenVMS specific code to get around this problem.  4 You can get my Freeciv v1.11.4 for OpenVMS kit at...  ) 	http://www.ourservers.net/openvms_ports/i  1 The file you want to look at is [.SERVER]SERNET.C    Hope this helps.   -- o  @  +------------------+--------------------------+---------------+@  | Robert Alan Byer | byer@mail.ourservers.net | ICQ #65926579 |@  +------------------+--------------------------+---------------+@  | Send an E-mail request to obtain a copy of my PGP key.      |@  +-------------------------------------------------------------+@  | "It is by caffeine alone I set my mind in motion.  It is by |@  |  cans of cola the thoughts aquire speed, the hands aquire   |@  |  shakes, the shakes become a warning.  It is by caffeine    |@  |  alone I set my mind in motion."                            |@  +-------------------------------------------------------------+   ------------------------------  % Date: Wed, 06 Jun 2001 20:16:01 +0200 2 From: "Haimo G. Zobernig" <Haimo.Zobernig@cern.ch> Subject: Re: subject _OPEN_VMS' Message-ID: <3B1E7361.8336ED28@cern.ch>    David Mathog wrote:s > K > Today we have OpenSSL, OpenSSH, OpenBSD, OpenPGP, etc. all of which share-F > the features that they are free to use and the source code is freelyM > available.  Obviously neither one of these descriptions apply to _Open_VMS.o > G > I've really got to hand it to the DEC marketeers of the days of yore.JL > These folks couldn't advertise their way out of a paper sack.  Yet somehowP > they managed to smudge the company's bread and butter product with a name thatM > was silly and pointless from the get go, and one that only grows worse with M > age.  Now, years later any but the cognoscenti would assume that OpenVMS ise5 > an open source project - which it certainly is not.  > I > Perhaps Compaq should rename it "ClosedVMS".   That would be relativelyiK > descriptive of both VMS security and Compaq's current mind set concerningsJ > the product.  And there's something nicely Orwellian about "OpenVMS" and# > "ClosedVMS" being the same thing!n > 
 > Regards, >  > David Mathog > mathog@caltech.edu@ > Manager, sequence analysis facility, biology division, CaltechL > **************************************************************************L > *                                RIP VMS                                 *L > **************************************************************************  
 "ClosedVMS"? r  @ That reminds me of the first edition of the IDS for Alpha we gotG together with our first DEC 3000 / 400 Alpha Workstations, which I haver in front of me.r8 ("VMS for Alpha Platforms Internals and Data Structures"A   Preliminary Edition, Volume 1, Order number EY-L466E-P1, 1992 )i   In the Preface, it states:= Digital Equipment Corporation has announced a new 64-bit RISC  architectureC and VMS support for it. At the time of writing, the architecture is  knownsC by its internal code name, Alpha, and does not yet have an officialsF trademarked name. As a stand-in for an official name, this publication uses               +---------+u the adjective |censored.|h               +---------+  ...   
 +--------+4 |Censored| VMS is ported from VAX/VMS Version 5.4-2.
 +--------+ ...g    7 (there is a rectangular frame drawn around "censored." a4 My ascii rendition will only look ok in fixed fonts)  G The whole book, about 1 inch thick , maybe 300-400 pages, has thousandsa= of such references to "censored" VMS ( and also to "censored"nD architecture, "censored" processor, "censored" instructions etc.). IF always thought it was VMS Engineering's reaction to tacking the "Open"C onto VMS, and to the "AXP" moniker that marketing gave to the Alpha-
 processor.E I'm hoping this book will become a greatly sought-after rarity on theo book market...   -- nG Haimo G. Zobernig                                  tel   +41 22 7677361dG Univ. of Wisconsin                                 fax   +41 22 7678370o c/o CERN Bld. 32 R-A02G 1211 Geneve 23, Switzerland                email Haimo.Zobernig@cern.ch-   ------------------------------   Date: 6 Jun 2001 14:12:38 -0500B9 From: Kilgallen@eisner.decus.org.nospam (Larry Kilgallen)g Subject: RE: The future of VMS3 Message-ID: <EgaagymRG+qS@eisner.encompasserve.org>t  z In article <3B55D7F383B0D31197D9009027541CBF0D9D1EFC@cmiexch1.cmi.itds.com>, Christopher Smith <csmith@amdocs.com> writes: >  >> -----Original Message-----e* >> From: Kilgallen@eisner.decus.org.nospam >  >> In article A >> <3B55D7F383B0D31197D9009027541CBF0D9D1EFB@cmiexch1.cmi.itds.co 4 >> m>, Christopher Smith <csmith@amdocs.com> writes: > I >> > The point is that if a programmer writes comments into the code, thel >> > declaration:e >> > f5 >> > function blarg (A, B: boolean) return boolean iso >> > V( >> > can be condensed to something like: >> > hE >> > --Blarg is a function which does... it accepts... and returns...i# >> >  bool func blarg (A, B: bool):r >> > r? >> > If you're going to write the comment anyway, the previous a >> is just extra
 >> > work. > I >> Certainly that is an example of a poor comment.  Comments should _not_ E >> parrot what can be determined from the source.  I would expect thecD >> comment for the Blarg function to explain why it is not using the@ >> standard Bla function for the project.  I would also expect a? >> better name for the function, but perhaps it is a well-knownn >> term in the problem domain. > H > Or maybe it's a poor example of a comment. :)  What I mean is that theH > comment should say what your function does to its arguments and (or atH > least) why, what to expect out of it, and what (not) to do with it, if > appropriate. i  ; My first priority would be to have some of that informationo= incorporated into the name of the function, so I know it evenr. when I am looking at the module that calls it:   	Blarg 	  vs. 	Blarg_Thread_Safe  > Of course what I _really_ want is for the person who wrote the original to have called it:    	Blarg_Thread_Unsafe  + but I realize that is too much to hope for.m  D The comment can then explain anything further than needs explaining,? and in some languages that might include the data types for thee
 arguments.  M > I do think it's appropriate to give more information than necessary in youreH > comments, because most people reading your code will speak your nativeL > language more fluently than whatever programming language it's in.  If youA > can help them to avoid some 3:00 am errors, so much the better.,  F Given that as the goal, I would suggest that the information describedC is not "more information than necessary".  Certainly, there will bee@ _some_ readers for which some of the information is superfluous,; such as those who have read it before.  They can take theiro  consolation in feeling superior.   ------------------------------  % Date: Wed, 06 Jun 2001 13:40:39 -0500 + From: Christopher Smith <csmith@amdocs.com>T Subject: RE: The future of VMSL Message-ID: <3B55D7F383B0D31197D9009027541CBF0D9D1EFD@cmiexch1.cmi.itds.com>   > -----Original Message-----) > From: Kilgallen@eisner.decus.org.nospam   
 > In article e@ > <3B55D7F383B0D31197D9009027541CBF0D9D1EFC@cmiexch1.cmi.itds.co3 > m>, Christopher Smith <csmith@amdocs.com> writes:h  = > My first priority would be to have some of that information ? > incorporated into the name of the function, so I know it even 0 > when I am looking at the module that calls it:   > 	Blarg > 	  vs. > 	Blarg_Thread_Safe  @ > Of course what I _really_ want is for the person who wrote the > original to have called it:i   > 	Blarg_Thread_Unsafe  - > but I realize that is too much to hope for.-  H Fair enough, as long as you don't start tacking senseless prefixes on toC accomplish the goal... For instance, Blarg_Thread_Safe is fine, butn% m_pfulBlarg is probably worthless. :)   > > > I do think it's appropriate to give more information than  > necessary in your ? > > comments, because most people reading your code will speak a
 > your native1> > > language more fluently than whatever programming language  > it's in.  If youC > > can help them to avoid some 3:00 am errors, so much the better.l  H > Given that as the goal, I would suggest that the information describedE > is not "more information than necessary".  Certainly, there will be,B > _some_ readers for which some of the information is superfluous,= > such as those who have read it before.  They can take theire" > consolation in feeling superior.  D True enough.  It's not as if people are forced to read the comments.   Regards,   Chriss  ! Christopher Smith, Perl Developer- Amdocs - Champaign, IL   /usr/bin/perl -e '? print((~"\x95\xc4\xe3"^"Just Another Perl Hacker.")."\x08!\n");u 'c   ------------------------------  % Date: Wed, 06 Jun 2001 15:06:51 -0400e- From: JF Mezei <jfmezei.spamnot@videotron.ca>d Subject: Re: The future of VMS, Message-ID: <3B1E7F48.19389E30@videotron.ca>   "Main, Kerry" wrote:M > Similarily, keeping an IT infrastructure up and running on a 24x7 basis and F > fixing things before they impact the business and complex middlewareM > programming designed to be highly scalable, reliable, take advantage of therM > local platforms clustering features etc requires a whole different level of_K > support than a simple college programmer (VB?) or admin person who has noe > real experience.    G Unfortunatly, Bill Gates convinced the industry that it is easier to benH tolerant of faults than to be fault tolerant. Case in point, a large GSMB wireless company in Canada who though that having its web site andM email-to-from-phones services down for a whole week was quite acceptable, andnF 6 months later, their web site still says that if you have a mac, they apoligize but it won't work.  L (And these people upgraded from NT to a serious system : SUN). I guess theirN windows teenage weenies had to learn how to operate stuff on Unix *after* they went into production.)   ------------------------------  % Date: Thu, 07 Jun 2001 00:15:28 +0200,) From: Christof Brass <brass@infopuls.com>- Subject: Re: The future of VMS, Message-ID: <3B1EAB80.AEFBE9CA@infopuls.com>   Robert Deininger wrote:8 >  > In articleC > <3B55D7F383B0D31197D9009027541CBF0D9D1EEC@cmiexch1.cmi.itds.com>,e. > Christopher Smith <csmith@amdocs.com> wrote: > I > > Well, Robert Deininger has suggested Ada, which is not my preference,rO > > although it would work.  I tend to prefer Mr. Wirth's more modern languagesn0 > > to a DOD project based on his previous work. > @ > There's really not much DOD left in Ada these days.  The Ada95% > standardization work was very open.v > 4 > What don't you like about Ada, just to be curious?  ? I don't have much practice with Ada but I even don't regard thew? history as problem. At least we could conclude that the DoD hadd> utmost quality in several aspects in mind: interoperability of; different platforms i.e. a very well defined standard whichu6 excludes platform dependency to a degree that could be6 questionable (e.g. the inherent task model), good self= documentation, safety and a good programming style. We should ; keep in mind that Ada was very well supported by DEC in thes 80ies.< The only problem I associate with Ada is "a camel is a horse= designed by a committee" and the PL/1 syndrom. But I recently-? looked at the oo extension introduced in Ada95 and I found themo@ quite useful. Ada or Modula-3 are my favourite candidates from a@ perspective a broad acceptance. Also I have some other languages "in petto".p  G > > Perhaps we could come up with a list of constructs which most ofteni# > > contribute to code malfunction?p > L > See the C reference manual?  Partly kidding.  I'll let Christof supply the > heavy ammunition here.  :-)C  
 Thanks :-)  M > > I'm sure I missed some things.  This is what comes to mind at the moment.  > > O > > So the implementation the above "features" in the language would need to be.M > > tightly controlled (or absent all together for some).  For those featurespM > > above which are useful, I would consider a separate module, where you can,K > > specifically include them, perhaps even on a per-item, and per-functionuI > > basis. (eg. "I want to do manual memory allocation in this function")t > E > Have you read the Ada Rationale?  Interesting to follow some of theh( > reasoning that went into the language. > A > > > I recently read a proposal to replace "unsafe" by "only fors) > > > restricted use" in the area of PLs.  > >eN > > If you mean in practice, moving all of the "unsafe" features into a moduleM > > that could be ignored by most people, then I think it's a very good idea.tM > > As I said above, it may be even better if we can selectively include theme& > > one at a time in single functions. > K > Ada does this, for example unchecked_deallocation and uncheck_conversion. H > When a module uses these, you have a red flag to be careful around it. >  > -- > Robert Deininger > rdeininger@mindspring.com   ? Basic question to the Ada expert: Is it possible to dynamicallye@ load code like in Java or some very few other languages/systems?> This is especially interesting in an oo framework environment.   ------------------------------  % Date: Thu, 07 Jun 2001 00:40:44 +0200 ) From: Christof Brass <brass@infopuls.com>f Subject: Re: The future of VMS, Message-ID: <3B1EB16C.80E2B330@infopuls.com>   Christopher Smith wrote:   [SNIP]  8 Sorry, have forgotten one thing: the point about the X11: interface is a no-issue because this is basically language7 independent. If this exists - fine; if not, we write an- interface. End of story.   ------------------------------  % Date: Thu, 07 Jun 2001 00:36:46 +0200e) From: Christof Brass <brass@infopuls.com>r Subject: Re: The future of VMS, Message-ID: <3B1EB07E.CCF470CA@infopuls.com>   Christopher Smith wrote: >  > > -----Original Message-----4 > > From: Christof Brass [mailto:brass@infopuls.com] >  > > Christopher Smith wrote: > A > > > Undefined behavior. :)  (Seriously, "undefined" is not good  > > enough.  An ? > > > error should be emitted in all cases where behavior can't  > > be "defined.") > ) > > ??? Could you please give an example?o > < > Here's something I dug up just now.  It's C, of course. :) > I > ..actually it's an extension to C, I believe.  The whole document is atg+ > http://www.lysator.liu.se/c/restrict.html  > 5 > 1  Figure 7  Restricted pointer function parametersf > 2    float x[100]; > 3    float *c; > 4n< > 5    void f3(int n, float * restrict a, float * const b) { > 6       int i;  > 7        for ( i=0; i<n; i++ )" > 8            a[i] = b[i] + c[i]; > 9    } > 10   void g3(void) {  > 11       float d[100], e[100];? > 12       c = x; f3(100,   d,    e); /* Behavior defined.   */.? > 13              f3( 50,   d, d+50); /* Behavior defined.   */o? > 14              f3( 99, d+1,    d); /* Behavior undefined. */ ? > 15       c = d; f3( 99, d+1,    e); /* Behavior undefined. */o? > 16              f3( 99,   e,  d+1); /* Behavior defined.   */l > 17   } > 18< > 19 Two of the calls shown in g3 result in aliasing that isA > 20 inconsistent with the restrict qualifier, and their behaviormA > 21 is undefined.  Note that it is permitted for c to point intoX; > 22 the array associated with b. Note also that, for theseuA > 23 purposes, the ``array'' associated with a particular pointerMA > 24 means only that portion of an array object which is actually % > 25 referenced through that pointer.    Thanks, I got the picture.  M > Note that this document says that lines 14 and 15 will not behave properly,nM > however, it doesn't specify in this case that the compiler should return anmI > error and refuse to accept them.  There are a few things of this nature-# > floating around, especially in C.2  = This is in deed very evil and that shouldn't of course not inu; any programming language. Sometimes there is weaker form of58 this: you can write it but it will behave differently on5 different platforms but at least equal (i.e. defined,n deterministic) on one platform.   @ > > > I'm sure I missed some things.  This is what comes to mind > > at the moment. > B > > I would add a bunch of features wrt basic software engineering= > > like separate compilation and type safe linking - both ist= > > missing in C/C++. The make system for C/C++ is completelyu: > > rotten. The development enviroment must ensure withoutA > > intervention of the programmer that the generated executablesi > > are technically correct. > M > I was actually going to suggest that the linker be made "smarter" than most-J > are.  That slipped my mind before I wrote it down, but there are severalN > problems that I've seen which could have been avoided if the linking process > was a bit more intelligent.o  
 Fine with me.   B > > The language/RTS must offer dynamic loading like Java. And, ofD > > course, the language must be oo, although a hybrid solution like" > > Java or Modula-3 is very good. >  > Can't argue with that.   :-)   D > > Your proposal is close to what is done in modern Wirth languagesC > > like Oberon or Modula-3, the development of the latter has been-B > > completely funded by DEC: functions of restricted use are only> > > available with a special declaration at the beginning of a > > module.i > K > I noticed, myself, that it was turning out to be very similar.  You never-+ > know, though, until you think it through.h  = We should then conclude that your proposal has to be rated asP high quality work :-)r  > > > May I suggest not to define a new PL - there are more than> > > enough to choose from - but instead to decide which one toC > > support. The language should be adequate to the quality of VMS' D > > design,  architecture and implementation. This would be then the- > > language of choice to write apps for VMS.y > M > Modula-3 is probably a very good choice.  The question is, how does one getiN > people to use it?  Another issue comes up with porting to other systems.  IfK > there's a decent cross-platform Modula-3 compiler, I'd like to know aboutoJ > it.  If you could get the same app to run on VMS, Beos, MacOS, etc, etc,J > that would likely be useful.  Also, do you know whether the VMS Modula-3F > compiler can use X11 for graphics -- what about the "cross-platform": > equivalent?  (People like to have graphical interfaces.) > 
 > Regards, >  > Christ > # > Christopher Smith, Perl Developerg > Amdocs - Champaign, IL >  > /usr/bin/perl -e 'A > print((~"\x95\xc4\xe3"^"Just Another Perl Hacker.")."\x08!\n");* > '*  $ These are a good but hard questions:< - should we aim for portability from VMS to other platforms?8 - how do we get people to use the language(s) of choice?  ? I personally think that VMS is unique and therefore any attemptn> to make portability inherent will lower the quality of the VMS= version - at least wrt VMS integration. I'm not talking aboutr< (trivial) best practices like layering and modularisation. I= think about using the unique VMS API (QIO, AST, RMS). I thinkY> the best what can happend to VMS is that high quality apps are> developed or ported to VMS that take full advantage of the VMS> unique features. The idea is to attract *users* or *customers*7 to use VMS instead to offer good programs also on othern
 platforms.  8 I'm personally sure that there is and will be only small? percentage of programmers who are willing to concentrate on thed7 real interesting programming tasks instead of mastering @ unecessary problems of PLs that aren't designed or don't meet up; to date quality criteria. The question is therefore to findu< exactly this minority and to create a community or something; like that which includes these people and the customer/user0= base. What interests me most in that respect: how many peoplea= are there of this type (including those that don't know yet)?n; And is this number high enough to achieve what we need (alle# necessary apps and enough payment)?"   ------------------------------  % Date: Thu, 07 Jun 2001 00:42:28 +0200c) From: Christof Brass <brass@infopuls.com>g Subject: Re: The future of VMS, Message-ID: <3B1EB1D4.DE1B4B18@infopuls.com>   Chris Casey wrote: > $ > Christopher Smith wrote in messageE > <3B55D7F383B0D31197D9009027541CBF0D9D1EF9@cmiexch1.cmi.itds.com>...E > >> -----Original Message----- G > >> From: rdeininger@mindspring.com [mailto:rdeininger@mindspring.com]- > >- > >> In articleeF > >> <3B55D7F383B0D31197D9009027541CBF0D9D1EEC@cmiexch1.cmi.itds.com>,1 > >> Christopher Smith <csmith@amdocs.com> wrote:e > >-A > >> > although it would work.  I tend to prefer Mr. Wirth's more3 > >> modern languagesF3 > >> > to a DOD project based on his previous work.s > J > PMFJI but if you want a higly scaleable and portable language you should2 > take a look at Mumps and its newer hybrid Cache.F > This even works on billy boxes as well as providing high performance7 > database systems on real boxes and operating systems.pK > It includes many modes including object based, SQL based or just good olda > scripting.E > It's database elements have proven time and again to be the fastestcN > operating in the VMS environment and, if it wasn't for good ol DEC trying to: > flog RDB (now Oracle) it would have been their flagship. > 3 > Not trying to sell you anything, just my opinion.o  ? Sorry to say this again: I looked into MUMPS programs and foundC8 them awful - combining the worst aspects of BASIC, C and assembly language.   ------------------------------  % Date: Thu, 07 Jun 2001 00:50:55 +0200t) From: Christof Brass <brass@infopuls.com>s Subject: Re: The future of VMS, Message-ID: <3B1EB3CF.561590E0@infopuls.com>   Robert Deininger wrote:f >  > In articleC > <3B55D7F383B0D31197D9009027541CBF0D9D1EF9@cmiexch1.cmi.itds.com>,e. > Christopher Smith <csmith@amdocs.com> wrote: > D > > > There's really not much DOD left in Ada these days.  The Ada95) > > > standardization work was very open.t > >oI > > Maybe I should look over it again, then.  I do still have a couple oft, > > reference books, which I should dig out. > >p8 > > > What don't you like about Ada, just to be curious? > >ID > > It's been a while, but here are my opinions from when I used it: > >aJ > > I like the Pascal-style syntax, and I like that it's been applied to a9 > > language with a modern design, OO type features, etc.l > >hE > > I dislike that the designers chose to make you type in the entireuO > > hierarchical path to a given procedure when it's not in the current module.yP > > (Ok, there is the 'use' command, which I had some trouble with since I could2 > > only get it to 'use' one module at a time. :)) > H > If long multi-part procedure names become a pain, you can shorten them" > with some renaming declarations.  5 The trend in high quality programming is towards full > declaration, long names etc., and strongly against renaming or< aliasing because this makes the same objects available under/ different names which is overall not desirable.a  ; Some verbosity of all well designed languages (look at Java @ which is very verbose for a C style language) is a result of the? bad experiences with operator based languages like C/C++, Lisp,e= APL or PERL. The basic contest: how high is the percentage of@; random texts that the compiler/interpreter regards as valid5: programs? The idea of verbosity is to some extend to avoid@ stupid mistakes and make the programs easy readable. Do you like to read RegExps?  J > If your packages are well organized, a USE clause for the package should< > give you a good collection of procedures with short names. > D > Ada style takes some getting used to.  Trying to do things the (C,/ > Fortran) was leads to unnecessary discomfort.  >  > > I don't particularly7 > > appreciate many of the "self-documenting" features._ > B > I'm not sure I'm following you.  Self documenting?  You mean theA > redundancy, for example the near-total duplication of procedure ? > declarations and definitions?  I look at that as double-entryaD > bookkeeping.  It lets the compiler cross-check, even with separate > compilation. > K > Ada is more verbose than many others, at least superficially.  In a large H > program, a lot of Ada is written at a very high level.  You may end upL > with fewer, smaller modules than in older languages.  But each module willK > seem verbose.  C++, properly used, is probably as expressive.  But it has  > other problems.  > ) > I'm probably missing your point here...o >  > > I do write my owneM > > documentation, and I do it fairly well.  A self-documenting language willoN > > just slow me down, because I'll end up writing all the documentation twice  > > -- once for the compiler. :) > I > AFAIK, the compiler really _uses_ all that information.  Except for the H > comments of course.  Have you found an Ada compiler that forces you to > write comments?e2 > I'd love to inflict it on some of my co-workers. > E > Someone mentioned earlier that C and C++ would be help by a smarter\K > linker.  With Ada, you get that smarter linker, if by another name.  Once-L > Ada signs off on an executable, you should not have any linker-type errors4 > left.  (Unless you are mixing in non-Ada modules.) > / > > I don't need two and three character names, N > > but fifteen is excessive.  That's not a big problem, and I could live with > > it anyway. > J > Ada doesn't force long names on you.  The longest reserved word is what,D > "procedure"?  (I don't have all 60-odd at my fingertips, so that's  > probably NOT the longest one.) > J > > Looking around, it seems that several of the problems I'd had with the, > > language are non-existent at this point. > > K > > > > Perhaps we could come up with a list of constructs which most ofteni' > > > > contribute to code malfunction?a > >b< > > > See the C reference manual?  Partly kidding.  I'll let > > > Christof supply theh! > > > heavy ammunition here.  :-)o > >n7 > > Oh, they're all in there. :)  See my previous list.n > > I > > > Have you read the Ada Rationale?  Interesting to follow some of theg, > > > reasoning that went into the language. > >d > > Nope, do you have a link?S >  > http://www.adapower.com/. > http://www.adapower.com/rationale/index.html > ; > > > Ada does this, for example unchecked_deallocation and- > > > uncheck_conversion. L > > > When a module uses these, you have a red flag to be careful around it. > >TP > > It seems that in theory Ada is very similar to Modula-3, which Christoff andH > > I have both mentioned.  It would be interesting to see a comparison. > K > I haven't had time to study Modula-3, but my impression is that there wase > some borrowing of ideas. >  > -- > Robert Deininger > rdeininger@mindspring.como  ' Additional questions to the Ada expert:e - Has Ada a GC?n= - Has Ada95 a proper type test for dynamically typed objects?'   ------------------------------  % Date: Thu, 07 Jun 2001 01:04:44 +0200 ) From: Christof Brass <brass@infopuls.com>l Subject: Re: The future of VMS, Message-ID: <3B1EB70C.6877DAB9@infopuls.com>   Christopher Smith wrote: >  > > -----Original Message-----F > > From: rdeininger@mindspring.com [mailto:rdeininger@mindspring.com] >  > > In articleE > > <3B55D7F383B0D31197D9009027541CBF0D9D1EF9@cmiexch1.cmi.itds.com>,p0 > > Christopher Smith <csmith@amdocs.com> wrote: > F > > Ada style takes some getting used to.  Trying to do things the (C,1 > > Fortran) was leads to unnecessary discomfort.d > N > I have noticed that.  It's easier to think of Ada as a Pascal-alike languageI > and program it as if you were writing Pascal.  That works better -- noty > quite perfectly, though.  ? Ada has been stronly influenced by other languages - especially  Pascal.d? As a sidenote to Robert Deiningers "I haven't had time to studyr? Modula-3, but my impression is that there was some borrowing of > ideas." PLs development is a circus of borrowing of ideas. But7 some people are smarter and some are more interested in @ design/quality than in market share, similarity to existing ones> etc.. Sometimes the latter and the former are the same and you> get really good results, i.e. really good languages. Alas, you< need some time to adapt. It's like in the OS arena: a person@ only acquainted with Windoze or UNIX (which should never be used= anyway because it's a collection of designless pieces of coder< fragments) has real problems to understand VMS. I just had a> discussion with a programmer and former sys admin who admitted> that VMS was terra incognita since they phased it our anly six= months after joining the university at that time. This person?< has really no clue about VMS. How could we expect someone to* like something what is completely unknown?   > > > I don't particularly9 > > > appreciate many of the "self-documenting" features.  > D > > I'm not sure I'm following you.  Self documenting?  You mean the > L > Any language which is incredibly verbose in its statements is said by someK > to be "self-documenting."  That's crap, of course.  No language is, but IcN > get the impression that all of the functions/procedures were named in such aK > way that it may be simple for a non-programmer to understand what's goingeJ > on.  That's not a bad thing, of course, but the extra verbosity required, > also necessitates extra typing on my part.  > If you could use an ultra modern development environment there? might be a chance to copy something instead of re-typing it ;-)s  ? About 50% of the quality of a PL is the style introduced by its-= designers and continued/maintained by its community. To writer= easy understandable SW is to a very high degree a question ofr; style. Ada supports and even sometimes enforces this style.i  A > > Ada is more verbose than many others, at least superficially.e > >  In a largeXJ > > program, a lot of Ada is written at a very high level.  You may end up= > > with fewer, smaller modules than in older languages.  But  > > each module willA > > seem verbose.  C++, properly used, is probably as expressive.  > >  But it hasr > > other problems.u > G > That's exactly what I mean.  I find that it takes about 1.5 times the L > characters to express something in Ada properly.  No big thing in general,$ > but it can get annoying sometimes. > + > > I'm probably missing your point here...i > F > The point is that if a programmer writes comments into the code, the > declaration: > 2 > function blarg (A, B: boolean) return boolean is > % > can be condensed to something like:v > B > --Blarg is a function which does... it accepts... and returns...  >  bool func blarg (A, B: bool): > I > If you're going to write the comment anyway, the previous is just extrau > work.t  3 But there is one important rule (like in good operai? productions): never duplicate the program in the comment (never ? show on stage what is already clear with the music), i.e. never = comment what is obvious from what is written - always comment = about overall structure or what is going to happen or what iso> the idea behind. The comment should contain hints which aren't obvious from what is written.   ; > > AFAIK, the compiler really _uses_ all that information.  > > Except for theJ > > comments of course.  Have you found an Ada compiler that forces you to > > write comments?.4 > > I'd love to inflict it on some of my co-workers. > N > No, but I've seen people who could use it.  Maybe look into intercal.  It at. > least requires them to ask nicely sometimes. > > > > Ada doesn't force long names on you.  The longest reserved > > word is what,pF > > "procedure"?  (I don't have all 60-odd at my fingertips, so that's" > > probably NOT the longest one.) > J > What about in the standard procedures and functions? (What does Ada call
 > them? :) > 
 > Regards, >  > Chriso > # > Christopher Smith, Perl Developeri > Amdocs - Champaign, IL >  > /usr/bin/perl -e 'A > print((~"\x95\xc4\xe3"^"Just Another Perl Hacker.")."\x08!\n");  > 'i   ------------------------------  % Date: Thu, 07 Jun 2001 01:07:33 +0200e) From: Christof Brass <brass@infopuls.com>  Subject: Re: The future of VMS, Message-ID: <3B1EB7B5.F762FD48@infopuls.com>   Larry Kilgallen wrote: > | > In article <3B55D7F383B0D31197D9009027541CBF0D9D1EFB@cmiexch1.cmi.itds.com>, Christopher Smith <csmith@amdocs.com> writes: > >> -----Original Message-----.G > >> From: rdeininger@mindspring.com [mailto:rdeininger@mindspring.com]o > >i > >> In articletF > >> <3B55D7F383B0D31197D9009027541CBF0D9D1EF9@cmiexch1.cmi.itds.com>,1 > >> Christopher Smith <csmith@amdocs.com> wrote:t > >pG > >> Ada style takes some getting used to.  Trying to do things the (C,s2 > >> Fortran) was leads to unnecessary discomfort. > >nP > > I have noticed that.  It's easier to think of Ada as a Pascal-alike languageK > > and program it as if you were writing Pascal.  That works better -- nott > > quite perfectly, though. > F > This is true of just about any language.  Long before there was Ada,B > somebody said "It is possible to write FORTRAN in any language".@ > The unspoken lesson is that because something is possible does@ > not mean it is a good idea.  Any language should be programmed* > in a style appropriate to that language. > N > > Any language which is incredibly verbose in its statements is said by someM > > to be "self-documenting."  That's crap, of course.  No language is, but IiP > > get the impression that all of the functions/procedures were named in such aM > > way that it may be simple for a non-programmer to understand what's going L > > on.  That's not a bad thing, of course, but the extra verbosity required. > > also necessitates extra typing on my part. > C > Consider somebody new to the language, or to your project, tryingnF > to understand your code while you are on vacation (or "holiday" :-).G > They are going to read that line _many_ more times than you typed it. G > Anyone writing Ada code should spend a _lot_ more time thinking aboutdB > it than typing it, and so far as I can see that applies to other > languages as well. > H > > The point is that if a programmer writes comments into the code, the > > declaration: > >s4 > > function blarg (A, B: boolean) return boolean is > > ' > > can be condensed to something like:p > > D > > --Blarg is a function which does... it accepts... and returns..." > >  bool func blarg (A, B: bool): > >cK > > If you're going to write the comment anyway, the previous is just extra 	 > > work.s > H > Certainly that is an example of a poor comment.  Comments should _not_D > parrot what can be determined from the source.  I would expect theC > comment for the Blarg function to explain why it is not using thee? > standard Bla function for the project.  I would also expect aP> > better name for the function, but perhaps it is a well-known > term in the problem domain.u > ? > >> Ada doesn't force long names on you.  The longest reservedh > >> word is what,G > >> "procedure"?  (I don't have all 60-odd at my fingertips, so that's # > >> probably NOT the longest one.)c > > L > > What about in the standard procedures and functions? (What does Ada call > > them? :) > E > There are standard "packages" (similar to modules if you don't lookn) > too closely), containing "subprograms":a > ; >         http://www.ada-auth.org/~acats/arm-html/RM-A.html  > I > For your project there may be heavily used "packages" that would not behK > "standard" due to only being of interest to those working in your problem 	 > domain.  > H > While Generic_Complex_Elementary_Functions is certainly a keyboardful,I > anyone really using it will take advantage of renaming or "use" clausesrG > to access the package.  Within that package the longest function name I > I can find is "Arctanh", and I doubt that even specialists in the fieldo > would want it shorter. > E > By the way, Ada does let you have an Arctanh for floating point andrH > a different Arctanh for long floating point (if your machine has such)G > and Ada decides which one you mean based on the type of arguments you K > feed it.  And no, there is no Arctanh that accepts two Booleans, althoughaD > you could define one if you thought you knew what it should do :-)  / But then you would need a very good comment ...(   ------------------------------  $ Date: Thu, 7 Jun 2001 00:25:24 +0100. From: "Chris Casey" <chris.casey@ntlworld.com> Subject: Re: The future of VMSC Message-ID: <F%yT6.14445$fs6.1033355@news6-win.server.ntlworld.com>B  C Christof Brass wrote in message <3B1EB1D4.DE1B4B18@infopuls.com>...n  @ >Sorry to say this again: I looked into MUMPS programs and found9 >them awful - combining the worst aspects of BASIC, C andr >assembly language.c  I I didn't think that you had actually said that before in a literate form.3@ I would be interested in your reasoning behind these statements.   ------------------------------   Date: 6 Jun 2001 20:27:55 -0500w9 From: Kilgallen@eisner.decus.org.nospam (Larry Kilgallen)  Subject: Re: The future of VMS3 Message-ID: <c0RnMXbVRUa1@eisner.encompasserve.org>y  X In article <3B1EB3CF.561590E0@infopuls.com>, Christof Brass <brass@infopuls.com> writes:   > - Has Ada a GC?   F The Ada standard treats garbage collection as an implementation detailB and neither requires it nor prohibits it.  The marketplace currentH state is that compiler implementers are not providing garbage collectionG and say their customers are not demanding it.  The obvious exception isc2 Ada compilers that target a Java bytecode machine.  ? > - Has Ada95 a proper type test for dynamically typed objects?    Yes.   ------------------------------  % Date: Wed, 06 Jun 2001 19:28:30 -0500o+ From: "Main, Kerry" <Kerry.Main@compaq.com>o Subject: RE: The future of VMSR Message-ID: <DC4745D1A85CA04180C83CDC706A9D180D966C@cthexc02.americas.cpqcorp.net>   JF -  K >>> Unfortunatly, Bill Gates convinced the industry that it is easier to beo1 tolerant of faults than to be fault tolerant. >>>-  E While this has been true in the past when there was relatively littlenJ competition, I would suggest that the days of not providing high levels of& service are rapidly coming to a close.  7 Bottom line is that there is just to much competition. d  I There are some very hungry competitors who are ready to quickly take your ? business out of the hands of those lower service level vendors.p  G The dot.bomb or flop com market is a good example where only those witht' serious offerings are going to survive.    Should be interesting.   :-)h   Regardso  
 Kerry Main Senior Consultanta Compaq Canada Inc. Professional Services  Voice: 613-592-4660p Fax  :  819-772-7036 Email: Kerry.Main@Compaq.com     -----Original Message-----4 From: JF Mezei [mailto:jfmezei.spamnot@videotron.ca] Sent: June 6, 2001 3:07 PM To: Info-VAX@Mvb.Saic.Comw Subject: Re: The future of VMS     "Main, Kerry" wrote:I > Similarily, keeping an IT infrastructure up and running on a 24x7 basisi and'F > fixing things before they impact the business and complex middlewareI > programming designed to be highly scalable, reliable, take advantage of  therJ > local platforms clustering features etc requires a whole different level ofK > support than a simple college programmer (VB?) or admin person who has nos > real experience.    G Unfortunatly, Bill Gates convinced the industry that it is easier to bedH tolerant of faults than to be fault tolerant. Case in point, a large GSMB wireless company in Canada who though that having its web site andI email-to-from-phones services down for a whole week was quite acceptable,c andeF 6 months later, their web site still says that if you have a mac, they apoligize but it won't work.  L (And these people upgraded from NT to a serious system : SUN). I guess theirI windows teenage weenies had to learn how to operate stuff on Unix *after*E they went into production.)   ------------------------------  % Date: Wed, 06 Jun 2001 17:31:32 -0700t! From: Shane.F.Smith@healthnet.comy Subject: Re: The future of VMSD Message-ID: <OFC16F447C.764E5DE0-ON88256A64.000202CD@foundation.com>  K Someone recently put out the quote about "you can write FORTRAN code in anyeH language". (And it's true, I know this from experience...) Well, you canH write crap code in any language too. The most important component in anyH piece of quality code is programmer discipline. Sure, you can use singleH character variable names in C and leave out the comments, but you can doH equivalently stupid things in any language. A disciplined and determinedH programmer can produce sources (ie commented code) you can understand inH any language I know of. If anyone has written a language you /can't/ putK comments in, they should be caught and beaten liberally with a wet haddock,r7 then locked away somewhere they can't do any more harm.   H This shouldn't be turned into a "my language is better than yours" thingD here, I'm not addressing that at all. We all know some languages areI inherently easier to follow than others, we just can't all agree on which  ones... ;-)t   Shanee          = Christof Brass <brass@infopuls.com> on 06/06/2001 03:50:55 PMr  5 Please respond to Christof Brass <brass@infopuls.com>    To:   Info-VAX@Mvb.Saic.Com  cc:c   Subject:  Re: The future of VMS      Robert Deininger wrote:h >C > In articleC > <3B55D7F383B0D31197D9009027541CBF0D9D1EF9@cmiexch1.cmi.itds.com>,k. > Christopher Smith <csmith@amdocs.com> wrote: >aD > > > There's really not much DOD left in Ada these days.  The Ada95) > > > standardization work was very open.  > > I > > Maybe I should look over it again, then.  I do still have a couple of , > > reference books, which I should dig out. > > 8 > > > What don't you like about Ada, just to be curious? > >oD > > It's been a while, but here are my opinions from when I used it: > >yJ > > I like the Pascal-style syntax, and I like that it's been applied to a9 > > language with a modern design, OO type features, etc.e > > E > > I dislike that the designers chose to make you type in the entiredG > > hierarchical path to a given procedure when it's not in the currentt module.lJ > > (Ok, there is the 'use' command, which I had some trouble with since I could 2 > > only get it to 'use' one module at a time. :)) >-H > If long multi-part procedure names become a pain, you can shorten them" > with some renaming declarations.  5 The trend in high quality programming is towards fullh> declaration, long names etc., and strongly against renaming or< aliasing because this makes the same objects available under/ different names which is overall not desirable.e  ; Some verbosity of all well designed languages (look at Java @ which is very verbose for a C style language) is a result of the? bad experiences with operator based languages like C/C++, Lisp, = APL or PERL. The basic contest: how high is the percentage oft; random texts that the compiler/interpreter regards as valida: programs? The idea of verbosity is to some extend to avoid@ stupid mistakes and make the programs easy readable. Do you like to read RegExps?  J > If your packages are well organized, a USE clause for the package should< > give you a good collection of procedures with short names. >-D > Ada style takes some getting used to.  Trying to do things the (C,/ > Fortran) was leads to unnecessary discomfort.R >T > > I don't particularly7 > > appreciate many of the "self-documenting" features.  >tB > I'm not sure I'm following you.  Self documenting?  You mean theA > redundancy, for example the near-total duplication of procedure ? > declarations and definitions?  I look at that as double-entrypD > bookkeeping.  It lets the compiler cross-check, even with separate > compilation. >uK > Ada is more verbose than many others, at least superficially.  In a largeiH > program, a lot of Ada is written at a very high level.  You may end upG > with fewer, smaller modules than in older languages.  But each modulea willK > seem verbose.  C++, properly used, is probably as expressive.  But it hase > other problems.s >e) > I'm probably missing your point here...r >e > > I do write my own H > > documentation, and I do it fairly well.  A self-documenting language willH > > just slow me down, because I'll end up writing all the documentation twice   > > -- once for the compiler. :) >iI > AFAIK, the compiler really _uses_ all that information.  Except for the H > comments of course.  Have you found an Ada compiler that forces you to > write comments?y2 > I'd love to inflict it on some of my co-workers. >'E > Someone mentioned earlier that C and C++ would be help by a smarter(K > linker.  With Ada, you get that smarter linker, if by another name.  OnceeE > Ada signs off on an executable, you should not have any linker-typed errors4 > left.  (Unless you are mixing in non-Ada modules.) >d/ > > I don't need two and three character names,iI > > but fifteen is excessive.  That's not a big problem, and I could liveo with > > it anyway. >oJ > Ada doesn't force long names on you.  The longest reserved word is what,D > "procedure"?  (I don't have all 60-odd at my fingertips, so that's  > probably NOT the longest one.) >'J > > Looking around, it seems that several of the problems I'd had with the, > > language are non-existent at this point. > >sK > > > > Perhaps we could come up with a list of constructs which most oftenn' > > > > contribute to code malfunction?a > > < > > > See the C reference manual?  Partly kidding.  I'll let > > > Christof supply ther! > > > heavy ammunition here.  :-)e > >n7 > > Oh, they're all in there. :)  See my previous list.s > >eI > > > Have you read the Ada Rationale?  Interesting to follow some of the , > > > reasoning that went into the language. > >e > > Nope, do you have a link?t >  > http://www.adapower.com/. > http://www.adapower.com/rationale/index.html >d; > > > Ada does this, for example unchecked_deallocation and. > > > uncheck_conversion.aH > > > When a module uses these, you have a red flag to be careful around it.m > >sB > > It seems that in theory Ada is very similar to Modula-3, which
 Christoff andiH > > I have both mentioned.  It would be interesting to see a comparison. >gK > I haven't had time to study Modula-3, but my impression is that there wasa > some borrowing of ideas. >l > -- > Robert Deininger > rdeininger@mindspring.com   ' Additional questions to the Ada expert:r - Has Ada a GC?h= - Has Ada95 a proper type test for dynamically typed objects?    ------------------------------  % Date: Wed, 06 Jun 2001 22:44:34 -0500e1 From: "David J. Dachtera" <djesys.nospam@fsi.net>  Subject: Re: The future of VMS' Message-ID: <3B1EF8A2.D26496BB@fsi.net>p   Chris Casey wrote: > E > Christof Brass wrote in message <3B1EB1D4.DE1B4B18@infopuls.com>...) > B > >Sorry to say this again: I looked into MUMPS programs and found; > >them awful - combining the worst aspects of BASIC, C and. > >assembly language., > K > I didn't think that you had actually said that before in a literate form.sB > I would be interested in your reasoning behind these statements.  0 If you've ever seen MUMPS, how can you ask that?  6 I'd say it combines the worst of C, assembler and APL.   -- e David J. Dachtera  dba DJE Systems  http://www.djesys.com/  : Unofficial Affordable OpenVMS Home Page and Message Board: http://www.djesys.com/vms/soho/a  F This *IS* an OpenVMS-related newsgroup. So, a certain bias in postings is to be expected.  @ Feel free to exercise your rights of free speech and expression.  F However, attacks against individual posters, or groups of posters, are strongly discouraged.o   ------------------------------  % Date: Wed, 06 Jun 2001 12:50:55 -0500u* From: WILLIAM WEBB <WWEBB1@email.usps.gov>D Subject: RE: UPDATE - finding the dec-net ethernet address using vms- Message-ID: <0033000000952124000002L042*@MHS>o  . =0APlease Do Not Throw Sausage Pizza Away. :^)   WWWebb   > -----Original Message-----1 > From: Info-VAX-Request@Mvb.Saic.Com at INTERNETe) > Sent: Wednesday, June 06, 2001 11:56 AMpF > To: Webb, William W - Raleigh, NC; Info-VAX@Mvb.Saic.Com at INTERNETF > Subject: RE: UPDATE - finding the dec-net ethernet address using vms >i > = > In article <3B1E448A.BD4082C3@fsi.net>, "David J. Dachtera"o9 > <djesys.nospam@fsi.net> writes: >"Brian Schenkenberger,- > VAXman-" wrote:e > >> [snip] A > >> The following command will set him the information he seeks:u > >>7 > >> $ MCR NCL SHOW NODE 0 CSMA-CD STATION * ALL STATUSf. > >>                ^^^^^^--- this is optional > >>0 > >> $ MCR NCL SHOW CSMA-CD STATION * ALL STATUS > >> > >> Compared with Phase-IV< > >>. > >> $ MCR NCP SHOW KNOWN LINE CHARACTERISTICS > >>* > >> Gee, it's only two characters longer! > >eE > >....but *FAR* less straight forward - don't even get me started on1B > >"intuitive", whatever the hell *THAT* word is supposed to mean! > ><= > >To someone who was "brought up" on "lines" and "circuits",t > what the helli= > >is a "CSMA-CD STATION"? Some new Micro$hit bloatware? ...ay
 > sun desktop H > >machine? ...the latest arcade game? ...a service/supply facility for=   > >motor vehicles? > > = > >(Yes, I *KNOW* what a "CSMA-CD STATION" is - *NOW*! I took 
 > the time tol' > >look it up circa. six(6) years ago.)o >a? > The CSMA-CD plays into the 1st (physical) and 2nd (Data Link)p > layers ofi > the OSI stack. >g >e > >  > >> [snip]l< > >> Read/study "Computer Networks" Andrew S. Tanenbaum.  My > copy is from myA: > >> MSEE days (copyright 1981) but I'm sure it's still in > print in some Nth 
 > >> edition.e > >hH > >This book documents DEC's NCL? ...and fills in the gaps where the DE= Cu' > >docset and on-line HELP are lacking?n >g? > No.  It covers DECnet (phase IV in my 1981 ed.), Arpanet (IP)o > and SNA.  It? > discusses these network implementations in terms of the OSI 7t > layer model.= > I'm aware of newer versions of this same text (albeit I not  > thumbed throughu) > it myself) which cover the OSI as well.  >p? > The idea is to get your head into the seven layer mindset andj > knowing just > a bit about networking.l >o > > < > >> >Dnet-OSI is strictly for the most esoteric of techies, > and those who ares> > >> >willing to study the mystical, darkly secret, multi-line > incantations ofk0 > >> >NCL which replace the simple, (relatively) > easy-to-understand, one-line > >> >commands of NCP. > >> >B > >> >Not for the faint of heart, and certainly not for newbies... > >>? > >> If you were weaned on Phase-V, you might find NCP commands  > just as distaste-c	 > >> ful.  > >hD > >How many of us fall under that heading? Fair few, I should think. > >  >u= > I've been struggling through RFCs for the past few weeks to  > gleen certaine= > protocol functionality from them.  Phase-V documentation is  > pure nirvana > in comparison. >r > --4 > VAXman- OpenVMS APE certification number: AAA-0001 > VAXman(at)TMESIS(dot)COM > ? > city, n., 1. a place where trees are cut down and streets are  > named after them.n >=   ------------------------------  % Date: Wed, 06 Jun 2001 14:31:36 -0500h1 From: "David J. Dachtera" <djesys.nospam@fsi.net>iD Subject: Re: UPDATE - finding the dec-net ethernet address using vms' Message-ID: <3B1E8518.3DEE2F8D@fsi.net>,   WILLIAM WEBB wrote:  > - > Please Do Not Throw Sausage Pizza Away. :^)   H *THAT*'s a good one. I went to a Cisco class last summer where they cameH up with a couple good chuckles for the layer names. I wasn't fast enough* to write 'em down, though. I like yours...  A ...and yeah, Brian, I am familiar with the OSI seven layer model.    -- t David J. Dachterar dba DJE Systems/ http://www.djesys.com/  : Unofficial Affordable OpenVMS Home Page and Message Board: http://www.djesys.com/vms/soho/i  F This *IS* an OpenVMS-related newsgroup. So, a certain bias in postings is to be expected.  @ Feel free to exercise your rights of free speech and expression.  F However, attacks against individual posters, or groups of posters, are strongly discouraged.g   ------------------------------  % Date: Wed, 06 Jun 2001 11:42:57 -0700v! From: Shane.F.Smith@Healthnet.comlD Subject: RE: UPDATE - finding the dec-net ethernet address using vmsD Message-ID: <OFFCDCA66B.2122B527-ON88256A63.00667991@foundation.com>  K Sorry, nowhere near as memorable as the one my friends worked out during my F student days. Unfortunately though, if I post that one I'd probably be% arrested for public indecency.... ;-)l   Shaneo          > WILLIAM WEBB <WWEBB1@email.usps.gov> on 06/06/2001 10:50:55 AM  6 Please respond to WILLIAM WEBB <WWEBB1@email.usps.gov>   To:   Info-VAX@Mvb.Saic.Com2 cc:+  E Subject:  RE: UPDATE - finding the dec-net ethernet address using vmst      + Please Do Not Throw Sausage Pizza Away. :^)w   WWWebb   > -----Original Message-----1 > From: Info-VAX-Request@Mvb.Saic.Com at INTERNETo) > Sent: Wednesday, June 06, 2001 11:56 AMdF > To: Webb, William W - Raleigh, NC; Info-VAX@Mvb.Saic.Com at INTERNETF > Subject: RE: UPDATE - finding the dec-net ethernet address using vms >u >y= > In article <3B1E448A.BD4082C3@fsi.net>, "David J. Dachtera"n9 > <djesys.nospam@fsi.net> writes: >"Brian Schenkenberger,- > VAXman-" wrote:a > >> [snip]1A > >> The following command will set him the information he seeks:l > >>7 > >> $ MCR NCL SHOW NODE 0 CSMA-CD STATION * ALL STATUSR. > >>                ^^^^^^--- this is optional > >>0 > >> $ MCR NCL SHOW CSMA-CD STATION * ALL STATUS > >> > >> Compared with Phase-IV  > >>. > >> $ MCR NCP SHOW KNOWN LINE CHARACTERISTICS > >>* > >> Gee, it's only two characters longer! > >rE > >....but *FAR* less straight forward - don't even get me started onrB > >"intuitive", whatever the hell *THAT* word is supposed to mean! > >e= > >To someone who was "brought up" on "lines" and "circuits",e > what the hell = > >is a "CSMA-CD STATION"? Some new Micro$hit bloatware? ...ap
 > sun desktop G > >machine? ...the latest arcade game? ...a service/supply facility for  > >motor vehicles? > >y= > >(Yes, I *KNOW* what a "CSMA-CD STATION" is - *NOW*! I tooks
 > the time toD' > >look it up circa. six(6) years ago.)a >i? > The CSMA-CD plays into the 1st (physical) and 2nd (Data Link)d > layers of  > the OSI stack. >b >  > >a > >> [snip]t< > >> Read/study "Computer Networks" Andrew S. Tanenbaum.  My > copy is from my : > >> MSEE days (copyright 1981) but I'm sure it's still in > print in some Nthn
 > >> edition.f > >vH > >This book documents DEC's NCL? ...and fills in the gaps where the DEC' > >docset and on-line HELP are lacking?r >u? > No.  It covers DECnet (phase IV in my 1981 ed.), Arpanet (IP)t > and SNA.  It? > discusses these network implementations in terms of the OSI 7m > layer model.= > I'm aware of newer versions of this same text (albeit I nots > thumbed throughb) > it myself) which cover the OSI as well.  >n? > The idea is to get your head into the seven layer mindset and  > knowing just > a bit about networking.  >i > >r< > >> >Dnet-OSI is strictly for the most esoteric of techies, > and those who ares> > >> >willing to study the mystical, darkly secret, multi-line > incantations ofc0 > >> >NCL which replace the simple, (relatively) > easy-to-understand, one-line > >> >commands of NCP. > >> >B > >> >Not for the faint of heart, and certainly not for newbies... > >>? > >> If you were weaned on Phase-V, you might find NCP commandse > just as distaste-e	 > >> ful.  > >aD > >How many of us fall under that heading? Fair few, I should think. > >  >p= > I've been struggling through RFCs for the past few weeks toh > gleen certainp= > protocol functionality from them.  Phase-V documentation ish > pure nirvana > in comparison. >a > --4 > VAXman- OpenVMS APE certification number: AAA-0001 > VAXman(at)TMESIS(dot)COM >t? > city, n., 1. a place where trees are cut down and streets are  > named after them.  >    ------------------------------  # Date: Wed, 06 Jun 2001 19:44:04 GMTi2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman)J Subject: Re: UPDATE - finding the dec-net ethernet address using vms 5.5-23 Message-ID: <8KvT6.1124$fi2.29269@news.cpqcorp.net>r  p In article <009FD20E.F4D5E07D@SendSpamHere.ORG>, system@SendSpamHere.ORG (Brian Schenkenberger, VAXman-) writes:\ :In article <3B1E448A.BD4082C3@fsi.net>, "David J. Dachtera" <djesys.nospam@fsi.net> writes:( :>"Brian Schenkenberger, VAXman-" wrote:
 :>> [snip]@ :>> The following command will set him the information he seeks:  L   The OpenVMS FAQ has general information on obtaining the Ethernet hardwareJ   address, in the section "How to determine the network hardware address?"  J :>To someone who was "brought up" on "lines" and "circuits", what the hell :>is a "CSMA-CD STATION"?   "   Well, that is what OSI calls it.  H :>(Yes, I *KNOW* what a "CSMA-CD STATION" is - *NOW*! I took the time to& :>look it up circa. six(6) years ago.)  B   And for those that don't...  Carrier-Sense Multiple-Access with A   Collision Detection.  This is the classic Aloha-style Ethernet tA   networking design, in other words.  The stations all sense the nA   carrier before transmitting, and they all also back-off (with anC   "random" back-off timer) and retransmit on transmit collisions...'D   The use of the CD scheme also establishes such related constructs B   as the maximum length of the cabling and the minimum packet sizeC   permissible -- you cannot have multiple packets in-flight on the  B   wire; each packet transmission must last long enough to blanket B   all stations on the cable.  Thus shorter packet sizes would meanG   correspondingly shorter cabling plants, or a failure of the collision D   detection scheme and thus the potential for undetected collisions.  H :The CSMA-CD plays into the 1st (physical) and 2nd (Data Link) layers of :the OSI stack.     1   It is the wire-level interface, in other words.s  
 :>> [snip]K :>> Read/study "Computer Networks" Andrew S. Tanenbaum.  My copy is from myoK :>> MSEE days (copyright 1981) but I'm sure it's still in print in some Nthl :>> edition. :>G :>This book documents DEC's NCL? ...and fills in the gaps where the DECr& :>docset and on-line HELP are lacking?  E   There are two editions of this book around.  IIRC (and it's been a eF   few years since I've found and read my copies), the DECnet coverage H   that was found in the first edition was replaced by OSI in the second.G   Both editions of the books are quite good, even with what has changedt-   ("much") since the original publications...u  N  ---------------------------- #include <rtfaq.h> -----------------------------N       For additional, please see the OpenVMS FAQ -- www.openvms.compaq.com    N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------  % Date: Wed, 06 Jun 2001 21:06:06 -0400c$ From: Norman Woo <nwoo@videotron.ca>A Subject: Use of SYSMAN to keep password file in sync on 8 systemsr8 Message-ID: <4aktht43ijudqghfnjj7k8cv0tk3ju599a@4ax.com>   Hi folks  B We have 8 NON-CLUSTERED systems (4 x Compaq Alphas running OpenVMSC 7.1-2 and 4 x Digital VAX running VMS 5.5-2) distributed throughoutpD the country.  Users can basically log onto any of these 8 systems toA do their work.  Their accounts are captive in the sense that theyy@ cannot get at the system command prompt.  Once they log onto theF system, they are into the application.  The application is the same onF all 8 systems.  One of the issues we need to resolve is when the usersD change their passwords when logging onto anyone of the 8 systems, weF would like to have this password updated on the other 7 systems.  This> way, when the user logs onto another system, they just need toC remember the new password and not have to worry which systems stillx have their old passwords.a  E We looked at the VAX Clustering solution but found it was too costly.uF We also looked at Compaq DCE but DCE only works on OpenVMS 7.0 and up.E We also looked at third party solution (from Ideous and Symantec) butt& were also costly (We have 1000+ users)  E Doing some digging into the manuals, we saw a potential: SYSMAN.  Anyi7 comments as to this solution? Or any other suggestions?a   Thanks   ------------------------------  # Date: Thu, 07 Jun 2001 04:21:19 GMTB$ From: Scott Vieth <svieth@wi.rr.com>E Subject: Re: Use of SYSMAN to keep password file in sync on 8 systems ) Message-ID: <3B1F01B8.CE3B4001@wi.rr.com>A  J You should check out the external authentication that is built-into recent  H versions of VMS.  That would solve your problem (after you upgrade those VAXen from 5.5-2).  
 -Scott :^)   Norman Woo wrote:n  
 > Hi folks > D > We have 8 NON-CLUSTERED systems (4 x Compaq Alphas running OpenVMSE > 7.1-2 and 4 x Digital VAX running VMS 5.5-2) distributed throughoutkF > the country.  Users can basically log onto any of these 8 systems toC > do their work.  Their accounts are captive in the sense that theytB > cannot get at the system command prompt.  Once they log onto theH > system, they are into the application.  The application is the same onH > all 8 systems.  One of the issues we need to resolve is when the usersF > change their passwords when logging onto anyone of the 8 systems, weH > would like to have this password updated on the other 7 systems.  This@ > way, when the user logs onto another system, they just need toE > remember the new password and not have to worry which systems stillk > have their old passwords.e > G > We looked at the VAX Clustering solution but found it was too costly.oH > We also looked at Compaq DCE but DCE only works on OpenVMS 7.0 and up.G > We also looked at third party solution (from Ideous and Symantec) but ( > were also costly (We have 1000+ users) >lG > Doing some digging into the manuals, we saw a potential: SYSMAN.  Anyn9 > comments as to this solution? Or any other suggestions?  >e > Thanks   ------------------------------  % Date: Thu, 07 Jun 2001 13:03:27 +1000e) From: Hang Sit Chan <hs.chan@Unsw.edu.au>  Subject: vax/vms cd * Message-ID: <3B1EEEFF.B39103B@Unsw.edu.au>   Hi,wH I am trying to mount a VAX/VMS 7.2 cd on a RedHat Linux (pc) machine.  I) was trying to read and transfer the files_E to my old vaxstation to do an upgrade. (I have just found my RRD40 cd G drive on my vaxstation is dead.)   But I have no idea how to mount thisaF cd on a linux machine.  Does anybody know whether linux could read vms* cd at all?  Could you please let know how? Many thanks.   ------------------------------  # Date: Wed, 06 Jun 2001 18:04:10 GMTr8 From: hammond@not@peek.ppb.cpqcorp.net (Charlie Hammond)" Subject: Re: VMS 7.2-1 and UCX 4.23 Message-ID: <uguT6.1116$fi2.29243@news.cpqcorp.net>d  ^ In article <31886a6f.0106060949.7e6ec7da@posting.google.com>, janicmx@yahoo.com (Mike) writes:1 >Is UCX 4.2 supported under VMS 7.2-1  on Alpha ?a  D I don't know what is "official policy", but our networking engineersA encourage upgrading to TCP/IP V5.0A (known as "V5.0-10" to PCSI).i  F The OpenVMS V7.2-1 upgrade will allow you to keep UCX 4.2 if you wish,@ but this does NOT imply support, nor that it will work properly.  # If possible, I suggest you upgrade.t   -- SK     Charlie Hammond -- Compaq Computer Corporation -- Pompano Beach  FL USAyH        (hammond@not@peek.ppb.cpqcorp.net -- remove "@not" when replying)J       All opinions expressed are my own and not necessarily my employer's.   ------------------------------  $ Date: Wed, 6 Jun 2001 14:27:32 -0400. From: "Kenneth Randell" <kenr@datametrics.com>" Subject: Re: VMS 7.2-1 and UCX 4.2+ Message-ID: <9flsl3$fqb$1@bob.news.rcn.net>p  E If you believe the documentation, item 3.15.2 on page 3-10 of the 7.3o Release Notes says this:  G DIGITAL TCP/IP Services for OpenVMS Version 4.2 (UCX) is not support ona' OpenVMS Version 7.0 and later releases.t  K This same quote was in the field test release notes; I submitted a question-L about it as to whether or not the '7.0' was the intended version; the answer came back that it was.   Ken Randellc  C Charlie Hammond <hammond@not@peek.ppb.cpqcorp.net> wrote in message-- news:uguT6.1116$fi2.29243@news.cpqcorp.net...e? > In article <31886a6f.0106060949.7e6ec7da@posting.google.com>,i  janicmx@yahoo.com (Mike) writes:3 > >Is UCX 4.2 supported under VMS 7.2-1  on Alpha ?, >nF > I don't know what is "official policy", but our networking engineersC > encourage upgrading to TCP/IP V5.0A (known as "V5.0-10" to PCSI).e >tH > The OpenVMS V7.2-1 upgrade will allow you to keep UCX 4.2 if you wish,B > but this does NOT imply support, nor that it will work properly. >g% > If possible, I suggest you upgrade.  >> > --I >     Charlie Hammond -- Compaq Computer Corporation -- Pompano Beach  FLN USAOJ >        (hammond@not@peek.ppb.cpqcorp.net -- remove "@not" when replying)L >       All opinions expressed are my own and not necessarily my employer's. >m   ------------------------------  # Date: Wed, 06 Jun 2001 19:59:06 GMTS2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman)" Subject: Re: VMS 7.2-1 and UCX 4.23 Message-ID: <eYvT6.1126$fi2.29020@news.cpqcorp.net>v  ^ In article <31886a6f.0106060949.7e6ec7da@posting.google.com>, janicmx@yahoo.com (Mike) writes:0 :Is UCX 4.2 supported under VMS 7.2-1 on Alpha ?  =   Nope.  This is not, will not, and has never been supported.   B   TCP/IP V5.1 (with ECO) is prefered on OpenVMS V7.1 (V7.1-*) and D   later, and TCP/IP V5.0 (or later) is required on OpenVMS V7.2 and ;   later, and TCP/IP V5.0A (with ECO) is prefered over V5.0.t  A   Here is my (rough) understanding (all mistakes here are mine!):(  9   TCP/IP Services V5.1 (with ECO) is the current release..  F   TCP/IP Services V4.2 will be supported through the end of June 2001 *   on the OpenVMS V7.1 and V7.1-* releases.  G   V5.0 (and later) are the ONLY versions that are available for OpenVMS    V7.2 and later.      Before July 2001:s)     V4.0: PVS on OpenVMS VAX V5.5-2 only.tJ     V4.2: PVS on OpenVMS VAX V6.2, and on OpenVMS Alpha V6.2 and V6.2-1H*,G           and on OpenVMS VAX and Alpha V7.1, and OpenVMS Alpha V7.1-1H*e           and V7.1-2.nL     V5.0A and later: the minimum versions for use on OpenVMS V7.2 and later.     Starting July 2001: )     V4.0: PVS on OpenVMS VAX V5.5-2 only.iN     V4.2: PVS on OpenVMS VAX V6.2 and on OpenVMS Alpha V6.2 and V6.2-1H* only.L     V5.0A and later: the minimum versions for use on OpenVMS V7.1 and later.   	--I     PVS: Prior Version Support.c    N  ---------------------------- #include <rtfaq.h> -----------------------------N       For additional, please see the OpenVMS FAQ -- www.openvms.compaq.com    N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------  % Date: Wed, 06 Jun 2001 13:35:00 -0500 % From: Bruce Stephens <hbs@beasys.com>u) Subject: Re: Weblogic/BEA On OpenVMS XML?.* Message-ID: <3B1E77D4.3083BB13@beasys.com>  + Please check out our platform support page:e  , http://www.weblogic.com/platforms/index.html   and further to    0 http://www.weblogic.com/platforms/vms/index.html  ! All of our docs are available on:-" http://edocs.beasys.com/index.html  9 For VMS, look at the information for Weblogic Server V5.1 - http://www.weblogic.com/docs51/resources.html-   and further at: > http://www.weblogic.com/docs51/install/OpenVMS_WLSinstall.html  $ you can download and kick the tires:   http://commerce.beasys.com/SoftwareProductDetailWLS?SWName=WebLogic+Server+Evaluation+Software&SWVersion=Version+5.1.0+SP9&SWPlatform=For+Compaq+OpenVMS+7.2   ------------------------------  % Date: Wed, 06 Jun 2001 19:43:09 -0500w+ From: "Main, Kerry" <Kerry.Main@COMPAQ.com>s) Subject: RE: Weblogic/BEA On OpenVMS XML? R Message-ID: <DC4745D1A85CA04180C83CDC706A9D180D966D@cthexc02.americas.cpqcorp.net>   Alan,s  D Like all large companies, the amount of feedback one gets in any oneE particular area really depends on the expertise of the person you are  talking to./  I Should every individual in a large company know a great deal about all ofe its partners ?   Sure - in an ideal world.   G The reality is that this is not the case and incidents where any vendor,@ employee gives out incorrect information needs to be corrected.   J However, all large companies have this problem ie. keeping their field andK support center employees up to date with all of the changes to all of their  partner solutions.  H >>> You can tell I'm a bit tetchy after fighting fast ethernet switches,G service contracts and (absence of) Alpha literature this last couple ofi weeks can't you :-) <<<   B Hey, the vocal Customers are the ones that get changes or problems corrected. f  I From a vendor perspective, the absolute worst thing that can happen is anuK incident happens, and the Customer says nothing - they just go away and thee problem does not get corrected.M  = So, keep up the good work and force issues to get corrected !e   :-)    Regards,  
 Kerry Main Senior Consultant  Compaq Canada Inc. Professional Services- Voice: 613-592-4660- Fax  :  819-772-7036 Email: Kerry.Main@Compaq.com     -----Original Message-----, From: Alan Greig [mailto:a.greig@virgin.net] Sent: June 6, 2001 10:01 AM  To: Info-VAX@Mvb.Saic.ComD) Subject: Re: Weblogic/BEA On OpenVMS XML?u    1 On Wed, 06 Jun 2001 06:56:42 -0500, "Main, Kerry"e <Kerry.Main@COMPAQ.com> wrote:  I >Actually, the BEA story with OpenVMS and Tru64 UNIX is much better these>5 >days .. check out the platform support pages on BEA:n   Kerry,  D A customer says they just talked to BEA sales (which did not inspireE confidence) and you reply "it is much better these days". What? SinceL yesterday perhaps?  C As well as product availability both Compaq and third party vendorsPE must be aware of their VMS products if they are to sell/support them.eB Sounds like local BEA folks like local Compaq folks need some more
 education.  D You can tell I'm a bit tetchy after fighting fast ethernet switches,D service contracts and (absence of) Alpha literature this last couple of weeks can't you :-)  < Still at least kuff now has some URLs to point BEA folks at.  5 >http://www.bea.com/products/messageq/platforms.shtml.5 >http://www.bea.com/products/tuxedo/platforms_6.shtmly > K >Key is that, contrary to some other OS platform providers, Compaq does not G >offer its own competing product in the middleware space, so it makes ae >pretty good fit.  >  >:-) >k	 >Regards,s >  >Kerry Maint >Senior Consultant >Compaq Canada Inc.a >Professional Services >Voice: 613-592-4660 >Fax  :  819-772-7036m >Email: Kerry.Main@Compaq.com9 >  >  >-----Original Message-----n/ >From: Kuff@Tessco.Com [mailto:Kuff@Tessco.Com]k >Sent: June 6, 2001 5:24 AMf >To: Info-VAX@Mvb.Saic.Com* >Subject: Re: Weblogic/BEA On OpenVMS XML? >u >  >  >  >e >DK >   A short update... we talked to the local BEA sales office and indicatedhJ >that we wanted to purchase one or more modules for OpenVMS.  We wanted toI >know what modules we would need, how they installed and workd on OpenVMS " >(requirements etc) and pricing.   >l > G >   We were rather surpirsed to hear that the best they could say afternI >four days of conversations with the local sales guy and sales support ony+ >the West Cost was that it worked on VMS.  s >hE >                     Issue                                  Response E >                     -----                                  -------- > >   We asked for release notes and install guides         NONEG >   We aksed for pre-sales support for OpenVMS            Still looking G >   OpenVMS Product/Development Manager                   Still Looking B >   Reference sites                                       Not Sure >? >iG >   One sales fellow said 'how about if we get Compaq to say it runs on D >OpenVMS' our response was 'how about we ask Compaq who your product5 >manager is since no one at BEA can find him/her'... n >mJ >   In summary, BEA seems unable to produce anyone internal that has seen,H >heard of, or manages the OpenVMS segment. Most disturbing is what wouldK >have happened if we had bought something and needed support.  Not at all a-
 >good sign.  , >  >  >1 >In articleiE ><A082CBA49B3F9AD6.7F8801D5B676C39C.E983D06FD2941F70@lp.airnews.net>,T" >Kuff@Tessco.Com (Hal Kuff) wrote: >sA >> We have a partner that is implementing Weblogic Server on NT..t >>  I >> They wish to send/receive XML files (with the proper serialisation andr- >> recovery-delivery assurances) via SSL ....l >> 1H >> Is anyone out there running the middle-ware pieces of BEA/Weblogic onH >> OpenVMS and exchanging files with another system? Is it compatible in some! >> way with RTR or IBM MessageQ ?  >> :G >> Anyone using an XML parser in production... The Compaq freeware one?a >> Someone elses?i   -- Alan   ------------------------------  # Date: Thu, 07 Jun 2001 04:48:35 GMTs; From: "Curtis Rempel" <vmsguy.no.spam.here.please@home.com>a) Subject: Re: Weblogic/BEA On OpenVMS XML?t< Message-ID: <DIDT6.112678$Ub.1283957@news1.rdc1.ab.home.com>  H And today also, the Compaq/BEA consultant we have on site for an OpenVMSJ high availability failover engagement informed us that in his conversationJ with BEA today he was told that OpenVMS is going to be desupported soon so get off it.d   Ahem, most unnerving.e  J Can anybody refute this claim, somebody ELSE from BEA for example?  I tend2 to think he may have been talking to a marketroid.  	 Please???'    2 "Alan Greig" <a.greig@virgin.net> wrote in message2 news:6dcsht07hj0bs6u9i36o9le6ucq85tsoko@4ax.com...3 > On Wed, 06 Jun 2001 06:56:42 -0500, "Main, Kerry"c  > <Kerry.Main@COMPAQ.com> wrote: > K > >Actually, the BEA story with OpenVMS and Tru64 UNIX is much better these 7 > >days .. check out the platform support pages on BEA:o >  > Kerry, >sF > A customer says they just talked to BEA sales (which did not inspireG > confidence) and you reply "it is much better these days". What? SinceM > yesterday perhaps? >pE > As well as product availability both Compaq and third party vendorshG > must be aware of their VMS products if they are to sell/support them.oD > Sounds like local BEA folks like local Compaq folks need some more > education. >8F > You can tell I'm a bit tetchy after fighting fast ethernet switches,F > service contracts and (absence of) Alpha literature this last couple > of weeks can't you :-) >a> > Still at least kuff now has some URLs to point BEA folks at. >i7 > >http://www.bea.com/products/messageq/platforms.shtmlp7 > >http://www.bea.com/products/tuxedo/platforms_6.shtml  > >oI > >Key is that, contrary to some other OS platform providers, Compaq doese not,I > >offer its own competing product in the middleware space, so it makes au > >pretty good fit.- > >- > >:-) > >e > >Regards,- > >-
 > >Kerry Main- > >Senior Consultant > >Compaq Canada Inc.e > >Professional Services > >Voice: 613-592-4660 > >Fax  :  819-772-7036e > >Email: Kerry.Main@Compaq.com- > >- > >  > >-----Original Message-----M1 > >From: Kuff@Tessco.Com [mailto:Kuff@Tessco.Com]- > >Sent: June 6, 2001 5:24 AMe > >To: Info-VAX@Mvb.Saic.Com, > >Subject: Re: Weblogic/BEA On OpenVMS XML? > >: > >o > >M > >  > >p > >rC > >   A short update... we talked to the local BEA sales office andk	 indicated.L > >that we wanted to purchase one or more modules for OpenVMS.  We wanted toK > >know what modules we would need, how they installed and workd on OpenVMSp" > >(requirements etc) and pricing. > >n > >yI > >   We were rather surpirsed to hear that the best they could say aftereK > >four days of conversations with the local sales guy and sales support ont+ > >the West Cost was that it worked on VMS.h > >lG > >                     Issue                                  ResponselG > >                     -----                                  --------t@ > >   We asked for release notes and install guides         NONEI > >   We aksed for pre-sales support for OpenVMS            Still lookinguI > >   OpenVMS Product/Development Manager                   Still LookingtD > >   Reference sites                                       Not Sure > >n > >fI > >   One sales fellow said 'how about if we get Compaq to say it runs onnF > >OpenVMS' our response was 'how about we ask Compaq who your product6 > >manager is since no one at BEA can find him/her'... > > L > >   In summary, BEA seems unable to produce anyone internal that has seen,J > >heard of, or manages the OpenVMS segment. Most disturbing is what wouldK > >have happened if we had bought something and needed support.  Not at allt as
 > >good sign.> > >j > >R > >e
 > >In articleeG > ><A082CBA49B3F9AD6.7F8801D5B676C39C.E983D06FD2941F70@lp.airnews.net>,.$ > >Kuff@Tessco.Com (Hal Kuff) wrote: > >aC > >> We have a partner that is implementing Weblogic Server on NT..w > >>K > >> They wish to send/receive XML files (with the proper serialisation and / > >> recovery-delivery assurances) via SSL ....  > >>J > >> Is anyone out there running the middle-ware pieces of BEA/Weblogic onJ > >> OpenVMS and exchanging files with another system? Is it compatible in some# > >> way with RTR or IBM MessageQ ?o > >>I > >> Anyone using an XML parser in production... The Compaq freeware one?g > >> Someone elses?a >p > -- > Alan   ------------------------------  % Date: Thu, 07 Jun 2001 01:46:26 -0400 - From: JF Mezei <jfmezei.spamnot@videotron.ca>s) Subject: Re: Weblogic/BEA On OpenVMS XML? , Message-ID: <3B1F152F.65BC783E@videotron.ca>   Curtis Rempel wrote: > J > And today also, the Compaq/BEA consultant we have on site for an OpenVMSL > high availability failover engagement informed us that in his conversationL > with BEA today he was told that OpenVMS is going to be desupported soon so
 > get off it.l  M I have absolutely no information. However, if one person states an experiencekG that the sales droids don't know about the product existing on VMS, andkD another person reports that one employee told him that VMS was to beN decomissioned, then, considering the history of VMS and its still absence fromE mainstream marketing, I would immediatly consider that product on theh endangered species list.  M There are still quite a few products that are to be decommissioned as part of L decisions taken during the palmer years. And Compaq has little motivation toK try to convince such vendors to reverse their plans if Compaq beleives that N customers will just migrate to a Compaq-NT based solution once the old VMS oneL is decommissioned. A prime example is the ST400 software that connects banksM to the SWIFT fund transfer network. Decision to decommission ST400 takes someuL years ago, and it will actually happen in one or two years, and Compaq seemsN perfectly comfortable with this because they beleive that they will be able toJ keep the ST400 customers by selling them wintel based solutions (SWIFT now% supports NT as a strategic platform).a  M VMS management have their hands tied. They have limited manpower and budgets.1N And they decided to focus on a few niches and that is all they can afford. TheN rest is on auto-pilot and if the rest falls on the wayside, it is no surprise.  U If VMS management succeeded in keeping Oracle on VMS, then that is a big acheivement.a  B But the days of VMS being seen as a great platform due to software availability are long gone.u   ------------------------------  % Date: Wed, 06 Jun 2001 12:36:24 -0700l! From: Shane.F.Smith@healthnet.com F Subject: Re: Who gets blamed for faulty software (was: A New Twist...)D Message-ID: <OF7DDAA61E.18705048-ON88256A63.006B3809@foundation.com>  J IMHO, you should never rely /purely/ on testing based on reading the code.G Informed testing is invaluable, but I don't consider it thorough unless2G it's supplimented with black box testing too.  I once knew a tester who7C would include tests like "what happens if I accidentally sit on theyI keyboard?" - not much got by him.  I once knew a tester who would include K stuff like "what happens if I accidentally sit on the keyboard?" - not muchn got by him.a   Shane           J Kilgallen@eisner.decus.org.nospam (Larry Kilgallen) on 06/06/2001 09:38:36 AM  E Please respond to Kilgallen@eisner.decus.org.nospam (Larry Kilgallen)n   To:   Info-VAX@Mvb.Saic.Come cc:   G Subject:  Re: Who gets blamed for faulty software (was: A New Twist...)r    J In article <9flidf$ag0@gap.cco.caltech.edu>, mathog@seqaxp.bio.caltech.edu (David Mathog) writes:5 > In article <y3NndjnxEqJl@eisner.encompasserve.org>,M; Kilgallen@eisner.decus.org.nospam (Larry Kilgallen) writes:hG >>In article <9fl2b5$ahc$2@bob.news.rcn.net>, jmfbahciv@aol.com writes:g >>? >>> Ptui.  Don't people _test_ their software anymore?  You canP? >>> code inspect and design review until the cows come home andC >>> still ship crap. >>B >>Certainly people test.  In this case, they could have tested for. >>54 minutes without encountering the problem. > J > Well, that's the surprising thing - it should have shown up on a call of one  > minute and 1 second.   Whoops, bad math on my part.  C As a previous poster said, you have to choose test cases carefully,s  and that means reading the code.   ------------------------------    Date: 07 Jun 2001 01:51:15 +0800, From: Paul Repacholi <prep@prep.synonet.com>2 Subject: Re: [OT] Current Microsoft v DOJ status ?- Message-ID: <87puch1oq4.fsf@prep.synonet.com>t  3 "David J. Dachtera" <djesys.nospam@fsi.net> writes:    > JF Mezei wrote:k  F > > Very few people *need* an SUV.  How many kilometres of road in the > > USA require a 4wd vehicle ?e  F > This is more of Dachtera's "idiotic" ranting (Hi, Bill!); so, now is+ > a good time to tune out (fair warning)...l   Oh well, in for a penny...  E > Like the M$ debate, the issue is not who *NEEDS* an SUV. The issue, B > IMHO, is that people DON'T WANT little rice-burning wimp-mobilesB > that can't even get out of their own way, especially at a corner6 > where you're trying to turn and can't see around the, > {SUV,VAN,pick-up truck, etc.}  beside you.  ? I have driven in the US, UK, Europe and here in Oz. Yes, I havea$ survived Boston trafic, both ways :)  E The vehicles in the US are, frankly, atrocious. Talk about a disaster4F looking for somewhere to happen! Some of the imports where OK, if theyD had not been 'got at' too much, and there was the odd good US model.  F But nowhere will you find anything to match a medium Merc, or 3xx BMW,G or most any other European car. Honda may come close, and in the sportyO0 end the Mitubitshi sold as I think, the Spectra.  ; In .AU, it is distance that matters. I could probably leave0> here, and head off to where I grew up, 320Km, and have an even+ chance of seeing NO other cars on the road.c  A The vehicle I put the most miles on where a Toyota Landcruser and-@ a DS21. The modern SUVs are the pits, over weight cars with near no off road ability.   --  < Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda.i@                                              West Australia 6076. Raw, Cooked or Well-done, it's all half baked.H Spam-To: uce@ftc.gov,enforcement@sec.gov,sness@fcc.gov,hfurchtg@fcc.gov,#   mpowell@fcc.gov,gtristan@fcc.gov B   ------------------------------   Date: 6 Jun 2001 19:03:18 GMTt1 From: bill@triangle.cs.uofs.edu (Bill Gunshannon)u2 Subject: Re: [OT] Current Microsoft v DOJ status ?, Message-ID: <9flupm$2k3c$1@info.cs.uofs.edu>  , In article <3B1D3D84.21D28283@videotron.ca>,0  JF Mezei <jfmezei.spamnot@videotron.ca> writes: |> tN |> What is the point of having a president if his decisions are meaningless toK |> international bodies because there is little chance that the legislative H |> assemblies will ratify what the president said his country would do ?  E When the president does his job, which is reperesenting the people of=D this country, and not acting like a tin plated dictator who is aboveG the law, he has a reasonable expectation of having his actions ratifiedsI by the legislature.  It is a relatively safe bet that Clinton knew before8H he signed the agreement and probably before he even left the country forE the trip that Congress would never ratify it.  It's not like it was ar close vote or something!!    bill   -- eJ Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesD bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. University of Scranton   |A Scranton, Pennsylvania   |         #include <std.disclaimer.h>   g   ------------------------------  $ Date: Wed, 6 Jun 2001 15:09:14 -0400' From: "Bill Todd" <billtodd@foo.mv.com>g2 Subject: Re: [OT] Current Microsoft v DOJ status ?( Message-ID: <9flv04$d5v$1@pyrite.mv.net>  9 "Paul Repacholi" <prep@prep.synonet.com> wrote in messagee' news:87puch1oq4.fsf@prep.synonet.com...c5 > "David J. Dachtera" <djesys.nospam@fsi.net> writes:  >p > > JF Mezei wrote:o >-H > > > Very few people *need* an SUV.  How many kilometres of road in the! > > > USA require a 4wd vehicle ?e  L In the snow belt, even less than 0.1 km can be significant, if it happens toJ be your driveway.  But that certainly doesn't explain their popularity in, e.g., Southern California.  L There are also people with legitimate needs for a fairly capacious vehicle -H and an SUV arguably uses its bulk more effectively in that respect than,L e.g., a Cadillac or Lincoln.  And since I believe at least to some degree inI a free market (with its excesses reined in by things like CAFE), the facttJ that some people just *like* something large doesn't bother me, as long asC the environmental costs (including the long-term cost of additionalKH consumption of non-renewable resources) are fairly borne by them if they make that choice.X  K So I'd have relatively little problem with SUVs (and small trucks, for that H matter) if they fell under the same CAFE regulations that cars do:  thatA would encourage manufacturers to make them more efficient withouthK sacrificing the things they're good at.  But I'd like to see gas prices for C *all* vehicles reflect the true value of the resources used and the  environmental impact caused.   >lH > > This is more of Dachtera's "idiotic" ranting (Hi, Bill!); so, now is- > > a good time to tune out (fair warning)...s >t > Oh well, in for a penny... >tG > > Like the M$ debate, the issue is not who *NEEDS* an SUV. The issue,:D > > IMHO, is that people DON'T WANT little rice-burning wimp-mobiles- > > that can't even get out of their own way,.  I Then they should buy a Neon:  mine has a bit over 40,000 miles on it now,hI and has consumed a bit less than 1,000 gallons of fuel during those milesiL (many of which were at 70+ mph on interstates).  It also is rated 0 - 60 mphH in 8.4 seconds (hardly supercar territory - my dad had an Olds 442 in myI youth, and I used to rally in a factory-prepared Porsche 911 - but faster8L than most cars on the road).  Handles well too, though the springs are a bitF too stiff and the shocks a bit too soft for anything other than smoothL tarmac (BMW got that combination right 30+ years ago, but others have yet to> catch up - possibly because good, stiff shocks are expensive).  H Some people, myself included, *don't* like large cars unless they reallyJ need them.  My first vehicle was a 1250-lb. Sprite with a 'glass hood, andG I've always tried to keep at least under twice that weight (some of thecL early Civics were under 2000 lbs., but finding anything under 2500 is pretty tough these days).  K As Theodore Sturgeon was fond of saying, 90% of everything is crap.  But ifoI you rummage carefully through the other 10%, you can often find something  acceptable to your tastes.    especially at a corner 8 > > where you're trying to turn and can't see around the. > > {SUV,VAN,pick-up truck, etc.}  beside you.  L So you're advocating having everyone drive a van so *no one* can see much ofD anything?  I'd rather see no one drive a van (or similar vehicle) soK *everyone* can see - but since that's not really feasible, I just make sure  that I can see what I need to.   >oA > I have driven in the US, UK, Europe and here in Oz. Yes, I have'& > survived Boston trafic, both ways :) >mG > The vehicles in the US are, frankly, atrocious. Talk about a disaster H > looking for somewhere to happen! Some of the imports where OK, if theyF > had not been 'got at' too much, and there was the odd good US model.  L I wonder how old your experience with U.S. vehicles is.  The later '70s wereK truly the pits:  the early-era emissions controls had crippled performance,FE handling was comparable to soft-suspension '50s behemoths, and safetye' engineering was a joke, as was quality.S  H The transformation since that time has been nothing short of remarkable.E U.S. manufacturers have learned how to produce relatively high-output>J engines with good efficiency and low emissions, improved suspension designJ (tire enhancements have likely contributed as well) has given even 'familyF cars' a level of handling once reserved for two-seat sportsters (whileE retaining a comfortable but far better-controlled ride - really major G advances occurred around the world in suspension design sometime in the8G early '80s, and I'd love to know exactly what they were), quality is atu7 least acceptable, and crash-safety is greatly improved.D  J 90% may still be (comparitively) crap, but the average quality of the crap is 'way up.e   - bill   ------------------------------  % Date: Wed, 06 Jun 2001 15:16:11 -0400i- From: JF Mezei <jfmezei.spamnot@videotron.ca>t2 Subject: Re: [OT] Current Microsoft v DOJ status ?, Message-ID: <3B1E8177.EF6BC73B@videotron.ca>   "David J. Dachtera" wrote:G > In a collision with a smaller, lighter vehicle, the other car is moreoE > likely to sustain severe, potentially fatal damage than is the SUV.t  K Damage to the car, perhaps. But damage to occupants is more a factor of theJM design of the interior and that of the safety systems. And it also depends on ? whether the collision of frontal, back, sideways, diagonal etc.e  E Furthermore, think about avoidinance of accidents. At high speed, the N tank/boat has much higher inertia and this cannot manoeuver as easily to avoidH a dangerous situation. (And due to designs, if you do decide to manouver@ radically to avoid the problem, you risk the roll-over scenario.  N You might feel safer because you are higher, but a higher centre of gravity is actually much more dangerous.o   ------------------------------  % Date: Wed, 06 Jun 2001 14:18:43 -0500o1 From: "David J. Dachtera" <djesys.nospam@fsi.net>i2 Subject: Re: [OT] Current Microsoft v DOJ status ?' Message-ID: <3B1E8213.3DFCED69@fsi.net>o   Jan Vorbrueggen wrote: > 5 > "David J. Dachtera" <djesys.nospam@fsi.net> writes:B > J > > F.Y.I: Chevy Cavalier is a small car, 2 doors, smallish back seat. TheJ > > only European equivalent I can think of is what was once called a Ford > > Sierra.o > L > A Sierra is the fourth size car Ford sells around here. Smalles is the Ka,O > then the Fiesta (what I drive), then the Escort, then the Sierra and company. = > Your description fits more closely to the Escort or Fiesta.   E No, it's really closer to the American version of the Escort, which I F believe is now replaced by the Focus. See http://www.ford.com/ for theC complete list (may be sensitive to your country of origin - dunno).@   -- C David J. Dachtera  dba DJE Systems  http://www.djesys.com/  : Unofficial Affordable OpenVMS Home Page and Message Board: http://www.djesys.com/vms/soho/   F This *IS* an OpenVMS-related newsgroup. So, a certain bias in postings is to be expected.  @ Feel free to exercise your rights of free speech and expression.  F However, attacks against individual posters, or groups of posters, are strongly discouraged.c   ------------------------------   Date: 6 Jun 2001 19:46:33 GMTe1 From: bill@triangle.cs.uofs.edu (Bill Gunshannon)i2 Subject: Re: [OT] Current Microsoft v DOJ status ?, Message-ID: <9fm1ap$2k3c$2@info.cs.uofs.edu>  H In article <y4puchy8xk.fsf@mailhost.neuroinformatik.ruhr-uni-bochum.de>,J  Jan Vorbrueggen <jan@mailhost.neuroinformatik.ruhr-uni-bochum.de> writes:B |> wayne@tachysoft.xxx.021386.killspam.015b (Wayne Sewell) writes: |>  Q |> > > Funnyt how , outside the USA, families have no problems with regular sized L |> > > cars, and it is only in the USA that families require oversized cars.N |> > Are parents in all these other countries required to have the child/baby = |> > seats mentioned above?  Those can take up a lot of room.  |> V5 |> Dunno about "all", but it certainly is in Germany.p  C It seems to me that I remember from when I lived there that Germany C was the first for requiring a lot of this child safety stuff.  LikesC not letting children even ride in the front seat if they were under 4 12 years old.  And that was more than 2 decades ago.   bill   -- pJ 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>   g   ------------------------------  % Date: Wed, 06 Jun 2001 16:13:54 -0400e- From: JF Mezei <jfmezei.spamnot@videotron.ca>e2 Subject: Re: [OT] Current Microsoft v DOJ status ?, Message-ID: <3B1E8EFB.115040B8@videotron.ca>   TO bring this back onto topic:  1 http://www.theregister.co.uk/content/2/19484.htmln  G Seems that Compaq's ass licking of Bill Gates is paying off. Seems thatLM Microsoft is preventing too many WINCE (Pocket PC) manufacturers from getting J licenses in order to give Compaq and HP signififant market share as "thankI you" for adopting WinCE and succeeding in making significant market sharebE inroads for an OS platform that had been stalled since its inception.>   ------------------------------   Date: 6 Jun 2001 16:30:17 -0500t- From: koehler@encompasserve.org (Bob Koehler)l2 Subject: Re: [OT] Current Microsoft v DOJ status ?3 Message-ID: <z2wIWnDcTi7t@eisner.encompasserve.org>M  [ In article <3B1E46D5.85DB7046@fsi.net>, "David J. Dachtera" <djesys.nospam@fsi.net> writes:e > Jan Vorbrueggen wrote: > E > F.Y.I: Chevy Cavalier is a small car, 2 doors, smallish back seat. n  H Chevy Cavalier is a particularly bad example.  I had an 84 (old design),D which wasn't too bad, but when they redisigned it in the late 80s orD early 90s it's crashworthiness went down.  Would probably loose in aD direct hit with a big SUV or a little import.  It's roughly the sameA size as a Toyota Corolla, but I'd feel much safer in the Corolla.a  F ----------------------------------------------------------------------? Bob Koehler                     | Computer Sciences Corporationa= NASA GSFC Flight Software       | Federal Sector, Civil GroupnE                                 | please remove ".aspm" when replyingt   ------------------------------   Date: 6 Jun 2001 16:41:38 -0500n- From: koehler@encompasserve.org (Bob Koehler)a2 Subject: Re: [OT] Current Microsoft v DOJ status ?3 Message-ID: <Pt6cK8aouwL$@eisner.encompasserve.org>   % At 12:45 PM 6/5/2001, JF Mezei wrote:e >tO >That is the main problem with the United States. In a parliementary system, if M >a government is elected with a majority, the prime minister has not only theiO >mandate but also the power to represent his people and sign treaties since hisiM >party has a majority in the house and thus has a very good chance of passingw4 >the law that implements the treaty in that country. >   @ Hm, seems to me several countries are and have experimented withF systems where one guy is the leader and everyone just does whatever he says.l  
 No thanks.  F ----------------------------------------------------------------------? Bob Koehler                     | Computer Sciences Corporationf= NASA GSFC Flight Software       | Federal Sector, Civil Group1E                                 | please remove ".aspm" when replyingr   ------------------------------   Date: 6 Jun 2001 13:33:15 -0700g1 From: nothome@spammers.are.scum (Malcolm Dunnett)l2 Subject: Re: [OT] Current Microsoft v DOJ status ?, Message-ID: <ymmj0tW1vREE@malvm1.mala.bc.ca>  - In article <9fm1ap$2k3c$2@info.cs.uofs.edu>, y7     bill@triangle.cs.uofs.edu (Bill Gunshannon) writes:  > E > It seems to me that I remember from when I lived there that GermanyeE > was the first for requiring a lot of this child safety stuff.  LikerE > not letting children even ride in the front seat if they were undere6 > 12 years old.  And that was more than 2 decades ago.  B    Funny how we hardly heard anything about "children in the back"< here in North America until airbags started killing them :-}   ------------------------------  % Date: Wed, 06 Jun 2001 12:28:59 -0700v! From: Shane.F.Smith@Healthnet.com32 Subject: Re: [OT] Current Microsoft v DOJ status ?D Message-ID: <OF9CAD8C07.16328FA4-ON88256A63.006AA11B@foundation.com>  G This puts me in mind of a short story called "Deadlier Weapon", IIRC bypF Larry Niven. The hero is carjacked by a guy with a gun. He spins a sobE story about his wife leaving him, then threatens to commit suicide byeK driving into a bridge support, taking the hijacker with him. Larry has saidwD he's received letters since about readers beating carackers by using variations of the story.  ? SUV vs Bridge, bridge wins. SUV+bridge vs gun, SUV driver wins.o   Shane.          E "David J. Dachtera" <djesys.nospam@fsi.net> on 06/06/2001 08:10:59 AMt  = Please respond to "David J. Dachtera" <djesys.nospam@fsi.net>e   To:   Info-VAX@Mvb.Saic.Comr cc:a  3 Subject:  Re: [OT] Current Microsoft v DOJ status ?      Jan Vorbrueggen wrote: >?) > Dan O'Reilly <dano@process.com> writes:c >tJ > > But if it seems germane, I drive SUV's because they give me and mine a MUCH: > > better chance of surviving a crash with a rice burner. > H > Nope, almost any other factor will contribute more to your safety than vehicler7 > mass. The most dangerous thing to drive in is a tank.    Well, yes and no.   E In a collision with a smaller, lighter vehicle, the other car is more*C likely to sustain severe, potentially fatal damage than is the SUV.r  F Vs. a truck or a bridge abutment, your chances are no better in an SUV' and, as you say, may actually be worse.    -- David J. Dachterau dba DJE SystemsN http://www.djesys.com/  : Unofficial Affordable OpenVMS Home Page and Message Board: http://www.djesys.com/vms/soho/   F This *IS* an OpenVMS-related newsgroup. So, a certain bias in postings is to be expected.  @ Feel free to exercise your rights of free speech and expression.  F However, attacks against individual posters, or groups of posters, are strongly discouraged.u   ------------------------------   Date: 6 Jun 2001 22:10:44 GMTe) From: leslie@clio.rice.edu (Jerry Leslie)n2 Subject: Re: [OT] Current Microsoft v DOJ status ?' Message-ID: <9fm9p4$6j0$1@joe.rice.edu>   . JF Mezei (jfmezei.spamnot@videotron.ca) wrote:G : Almost all trains in north america use that principle: diesel engine  D : that drives a generator that drives an electric motor that drives 
 : the wheels.n  0 Some now use alternators instead of generators,   9   http://www.gmemd.com/locomotives/naf/sd70mac/index.htmli7   GM Electro-Motive: SD70MAC Locomotive Specifications 9  * Here's GE's locomotive specs and drawings:  M    http://www.getscwc.com/general/apps/freight_rail/reference/specs/specs.htmr   --Jerry Leslie   ------------------------------  % Date: Wed, 06 Jun 2001 22:48:11 -0500c1 From: "David J. Dachtera" <djesys.nospam@fsi.net> 2 Subject: Re: [OT] Current Microsoft v DOJ status ?' Message-ID: <3B1EF97B.B7335A5B@fsi.net>f   Jan Vorbrueggen wrote: > A > wayne@tachysoft.xxx.021386.killspam.015b (Wayne Sewell) writes:$ > P > > > Funnyt how , outside the USA, families have no problems with regular sizedK > > > cars, and it is only in the USA that families require oversized cars.pL > > Are parents in all these other countries required to have the child/baby< > > seats mentioned above?  Those can take up a lot of room. > 4 > Dunno about "all", but it certainly is in Germany.  2 One of my memories of Germany from summer of 1989:  - In the U.S., Mercedes Benz is a prestige car.l   In Dusseldorf, it's taxicab.   --   David J. Dachterae dba DJE Systemsi http://www.djesys.com/  : Unofficial Affordable OpenVMS Home Page and Message Board: http://www.djesys.com/vms/soho/e  F This *IS* an OpenVMS-related newsgroup. So, a certain bias in postings is to be expected.  @ Feel free to exercise your rights of free speech and expression.  F However, attacks against individual posters, or groups of posters, are strongly discouraged.i   ------------------------------  % Date: Wed, 06 Jun 2001 22:51:03 -0500i1 From: "David J. Dachtera" <djesys.nospam@fsi.net> 2 Subject: Re: [OT] Current Microsoft v DOJ status ?' Message-ID: <3B1EFA27.4F0147A6@fsi.net>    Malcolm Dunnett wrote: > . > In article <9fm1ap$2k3c$2@info.cs.uofs.edu>,9 >     bill@triangle.cs.uofs.edu (Bill Gunshannon) writes:y > >mG > > It seems to me that I remember from when I lived there that GermanysG > > was the first for requiring a lot of this child safety stuff.  LikemG > > not letting children even ride in the front seat if they were undern8 > > 12 years old.  And that was more than 2 decades ago. > D >    Funny how we hardly heard anything about "children in the back"> > here in North America until airbags started killing them :-}  F When I was little, I rode in the front seat between Mom and Dad in ourE 1959 Ford "tank" - when cars were made of more than foil and plastic,h and still had steel dashboards.i   --   David J. Dachtera  dba DJE Systems, http://www.djesys.com/  : Unofficial Affordable OpenVMS Home Page and Message Board: http://www.djesys.com/vms/soho/   F This *IS* an OpenVMS-related newsgroup. So, a certain bias in postings is to be expected.  @ Feel free to exercise your rights of free speech and expression.  F However, attacks against individual posters, or groups of posters, are strongly discouraged.e   ------------------------------   Date: 7 Jun 2001 05:19:09 GMTa- From: Joe Heimann <heimann@nog.ecs.umass.edu>l2 Subject: Re: [OT] Current Microsoft v DOJ status ?, Message-ID: <9fn2sd$nnb$5@odo.ecs.umass.edu>  3 Robert Deininger <rdeininger@mindspring.com> wrote:gJ > In article <OF4118F03E.D586F273-ON80256A62.0043C01B@qedi.quintiles.com>,# > steven.reece@quintiles.com wrote:   F >> Something I'm not clear on with this thread (which isn't really VMS* >> related, unless - let's not go there) :K >> In the UK, hydro-electric power plants use water to drive turbines.  Th=t >> isiK >> is the principle used in one of the Welsh plants (whose name escapes me=  >>  atI >> the moment) in which water drives the turbines at high demand and then K >> during off-peak periods when electricity is cheap the water is pumped b=0 >> ack >> to the top of the plant.mH >> The power during the off-peak periods would, of course, come from theK >> base-load stations which are primarily nuclear or coal burning stations=n  K > I don't think we do that over here.  Generally, the hydro plants make the K > cheapest power, so pumping the water back uphill would be a money-loser. hI > We let the sun take care of that, but sometimes we have to wait quite a  > while.  H There are a number of pumped storage generation plants around in the US,E but not too many.  One is not too far from here, Northfield Mountain s, Pumped Storage Facility in Northfield, Mass.  G It uses water pumped up from the Connecticut River to a storage pond oneE the top of the hill, and releases the water back to generate power ate peak demand periods.     Joe Heimanni   heimann@ecs.umass.eduh   ------------------------------   End of INFO-VAX 2001.313 ************************