/ INFO-VAX	Sun, 06 Jan 2002	Volume 2002 : Issue 9       Contents:P Re: BLISS pros and cons, was: Re: historical evidence of what went wrong  at DECO Re: BLISS pros and cons, was: Re: historical evidence of what went wrong at DEC > Re: Buffer Overflows - again Was: (Re: Congratulations for the2 Re: Compaq still tries to spin Alphacide both ways Re: DDCMP and RAS  Re: DDCMP and RAS % DIBOL better than Cobol, C anyday ... . Re: Hey Intel, VMS is your ticket to high end!. Re: Hey Intel, VMS is your ticket to high end!1 Re: historical evidence of what went wrong at DEC 1 Re: historical evidence of what went wrong at DEC 1 Re: historical evidence of what went wrong at DEC 1 Re: historical evidence of what went wrong at DEC 0 Re: HP admits it will kill VMS if merger suceeds0 Re: HP admits it will kill VMS if merger suceeds Re: locking display CDE + Re: NAT: publishing the remote IP of router  Re: Problem unpacking TSM  Re: VAX in a VT-103? RE: VAX in a VT-103?5 [Q] Installing the VXT2000+ software on OpenVMS/Alpha 9 Re: [Q] Installing the VXT2000+ software on OpenVMS/Alpha   F ----------------------------------------------------------------------   Date: 5 Jan 2002 17:18:50 -0800 ( From: bob@instantwhip.com (Bob Ceculski)Y Subject: Re: BLISS pros and cons, was: Re: historical evidence of what went wrong  at DEC = Message-ID: <d7791aa1.0201051718.560660d4@posting.google.com>   _ "David J. Dachtera" <djesys.nospam@fsi.net> wrote in message news:<3C3675B3.BBB92CB@fsi.net>...  > Bob Ceculski wrote:  > >  > > Mark Crispin <mrc@CAC.Washington.EDU> wrote in message news:<Pine.NXT.4.50.0201041017210.22025-100000@Tomobiki-Cho.CAC.Washington.EDU>... P > > > > I'll go along with COBOL. I did some COBOL porting, that was unpleasant. > > > P > > > I think that COBOL's problem was creeping featurism and too much leeway toL > > > add non-portable means of I/O to the language.  COBOL itself should beJ > > > quite portable, except that hardly anyone ever wrote a 100% standard > > > compliant COBOL program. > > >  > > > -- Mark -- > > > % > > > http://staff.washington.edu/mrc L > > > Science does not emerge from voting, party politics, or public debate. > > N > > should have used DIBOL ... way easier and more powerful to use than cobol! > / > Um, sorry, Bob - I gotta call ya on that one.  > I > I've done both COBOL (on VMS and MP/M (RM/COBOL)) and DIBOL (on VMS and G > UN*X (SIBOL)). DIBOL would be a miserable excuse for a language if it > > weren't to RT-11 what BASIC-Plus is (or once was) to RSTS/E. > H > I'd hardly want to work exclusively in COBOL (though I'd do it to saveA > my VMS career, if it came to that), but I'd hardly want to work  > exclusively in DIBOL, either.   J synergy dibol today has come a long way from the original rsts/e dibol ...F it is not only superior to cobol but to c as well ... and I still will, out program any cobol programmer w/dibol ...   ------------------------------  # Date: Sun, 06 Jan 2002 05:01:50 GMT $ From: Ric Werme <werme@mediaone.net>X Subject: Re: BLISS pros and cons, was: Re: historical evidence of what went wrong at DEC> Message-ID: <2TQZ7.15184$zX1.10960172@typhoon.ne.mediaone.net>  ? system@SendSpamHere.ORG (Brian Schenkenberger, VAXman-) writes:   d >In article <ys9OeZFRsATI@eisner.encompasserve.org>, Kilgallen@SpamCop.net (Larry Kilgallen) writes:h >>In article <qhitahj434.fsf@ruckus.brouhaha.com>, Eric Smith <eric-no-spam-for-me@brouhaha.com> writes:0 >>> John Reagan <john.reagan@compaq.com> writes:J >>>> While Compaq never sold BLISS for NT, we certainly continue to use itJ >>>> for the Fortran products (both on Alpha NT and IA32 NT) since the GEM: >>>> component of the compiler is written mostly in BLISS. >>> G >>> Would it be that hard to do an automatic translation of BLISS to C?  >>F >>C's macro capability is an incredibly tiny subset of the Bliss macroG >>capability.  To translate the expanded form would be like translating D >>Bliss to machine language, and there are already tools to do that.   >C has a macro capability?  K To those of us who used MACRO-10, the answer, as you know, "There's a macro  capability in ANSI C?"  :-)   C I like C, I'd like it even better if the preprocessor did something 
 useful.... --K "When we allow fundamental freedoms to be sacrificed in the name of real or D perceived emergency, we invariably regret it.   -- Thurgood MarshallC    Ric Werme                            | werme@nospam.mediaone.net >    http://people.ne.mediaone.net/werme  |       ^^^^^^^ delete   ------------------------------  # Date: Sun, 06 Jan 2002 06:56:26 GMT * From: "Bill Todd" <billtodd@metrocast.net>G Subject: Re: Buffer Overflows - again Was: (Re: Congratulations for the B Message-ID: <uySZ7.248050$Zd.21766153@bin1.nnrp.aus1.giganews.com>  5 "Peter da Silva" <peter@taronga.com> wrote in message , news:a14q71$202c$1@citadel.in.taronga.com.... > In article <3C34CC38.D9615F71@jetnet.ab.ca>,/ > Ben Franchuk  <bfranchuk@jetnet.ab.ca> wrote: > > >I think virtual memory causes more problems than it solves! > / > "Virtual memory leads to virtual performance"  > 2 > "Memory is like sex, it's better when it's real" > % > -- both attributed to Seymour Cray.  > I > VM solves a lot of problems, and if you're working in an area where you  needG > those problems solved a VM system sure beats the heck out of spending  months4 > tweaking TKB maps to fit everything into overlays.  J Then again, unless you carefully lay out your code such that related areasK will be paged in together by whatever clustering algorithm the VM uses, the F TKB approach can often offer better performance (though of course with) significantly higher development effort).    - bill   ------------------------------  $ Date: Sat, 5 Jan 2002 23:27:35 -0500+ From: Keith R. Williams <krw@attglobal.net> ; Subject: Re: Compaq still tries to spin Alphacide both ways : Message-ID: <MPG.16a19c3d9345f5169898c5@enews.newsguy.com>  5 In article <3C376882.A3FCCB9F@moene.indiv.nluug.nl>,  ! toon@moene.indiv.nluug.nl says...  > Wow wrote: >  > > Back in middle 90s... > > > Intel was beating the drums saying "RISCs will be crushed,( > > resistance is futile, blah, blah..."( > > How the tables turned.... Oh well... > = > Not at all.  Now we know *how* the "RISCs will be crushed".  > : > The difficult part is to predict which one will be next.  & Ok, which one is next, oh seer of all?  ? I find this amusing, since CISC turned to RISC to maintain its  B market.  Look into any CISC processor and it looks pretty RISCy.     ----   Keith    ------------------------------  # Date: Sun, 06 Jan 2002 03:57:56 GMT - From: "John E. Malmberg" <wb8tyw@qsl.network>  Subject: Re: DDCMP and RAS* Message-ID: <3C37D016.7060900@qsl.network>   Jerome DIDIER wrote:  J > Is it possible to access to a VMS server from a PC under Winnt/2000 via  > a modem ?      Yes.  H You can use a terminal emulation program to directly access the OpenVMS D system, or you can use the modem to dial up a RAS server and tunnel . through to it for a TCP/IP network connection.  H I have used Telnet, X-Windows, and Advanced Server LANMAN (file / print E sharing) protocols by tunnelling through RAS.  X-Windows is slow but  1 possibly usuable with the so called "56K" modems.     I Asynchronous DDCMP is possible in some configurations also.  It requires  I a Asynchronous DDCMP DECNET stack on both the Windows NT or Windows 2000  F system.  The Asynchronous DDCMP program must also be installed on the  OpenVMS Server.    What are you trying to do?     -John  wb8tyw@qsl.network Personal Opinion Only    ------------------------------   Date: 6 Jan 2002 04:22:53 GMT ) From: leslie@clio.rice.edu (Jerry Leslie)  Subject: Re: DDCMP and RAS' Message-ID: <a18jet$q82$1@joe.rice.edu>   , John E. Malmberg (wb8tyw@qsl.network) wrote: : > : he Asynchronous DDCMP program must also be installed on the  : OpenVMS Server.  : : IIRC, Async DDCMP is NOT an option for ALPHAs, just VAXes.  4 --Jerry Leslie     (my opinions are strictly my own)   ------------------------------   Date: 5 Jan 2002 19:47:30 -0800 ( From: bob@instantwhip.com (Bob Ceculski). Subject: DIBOL better than Cobol, C anyday ...= Message-ID: <d7791aa1.0201051947.774c0ec0@posting.google.com>   ? For those still sweating over wordy cobol code, there is a more  powerfulC version of cobol that has really evolved ... DIBOL!  It is an easy, 	 powerful, E c-like implementation of cobol that for years has made cobol sick ...  current C synergy version has many gui and web tools that cobol can't touch!   Our portE from vax dibol to alpha synergy dibol required only a compile!  It is  the bestD programming language out there, including c!  Completely portable to
 many otherE platforms also, unix, windoze, although vms is superior for it to run  on ...    3 Synergex Connects Dibol Users With New Technologies E GOLD RIVER, CA -- According to Digital partner Synergex, an estimated E 2.5 million users at 210,000 sites have such a significant investment C in legacy applications written in DIBOL, Digital's early, extremely A powerful development language, that they are reluctant to move to = newer hardware and software technologies in many areas of the  enterprise.   F Now, thanks to software development tools from Synergex, many of theseF customers are breathing a sigh of relief. Synergy Server, Synergy ODBCA Driver, and Synergy SQL Connection allow DIBOL developers to move D applications to OpenVMS, Digital UNIX, and Windows NT with little orD no reprogramming. From their existing applications, users can access@ remote systems across heterogeneous networks, in a client/serverF environment, using popular 32-bit software such as Microsoft Excel and? databases such as Oracle and SQL Server -- all from a graphical  Windows interface.  F "We want to get the word out to these organizations that they can haveE it all," said William J. Mooney, Vice President, Business Development > for Synergex. "The Synergy products give DIBOL-based customers< complete platform independence, the ability to introduce newF technologies such as Alpha-based client/server computing, and a bridge2 to popular 32-bit software and database products."  @ Formerly known as DISC (Digital Information Services Corp.), theE company has a 20-year track record in solving application portability C issues for Digital customers. In addition, as a member of Digital's A Affinity Program, Synergex provides the tools to allow their vast > customer base of vertical market software providers to develop& three-tier client/server applications.   ------------------------------  # Date: Sun, 06 Jan 2002 02:18:09 GMT   From: cjt <cheljuba@prodigy.net>7 Subject: Re: Hey Intel, VMS is your ticket to high end! + Message-ID: <3C37B3A6.442898D3@prodigy.net>    "Terry C. Shannon" wrote: 	 ><snip>>  N > Not only does Alpha have more software than IA64, it also has a proven trackK > record. From what I've seen in the Marvel lab, Compaq has nothing to fear ! > performance-wise from McKinley.   K That last sentence is Delphic.  If McKinley performance is low, and Compaq  J hitches itself to the McKinley wagon, do they THEN have something to fear?   ------------------------------  # Date: Sun, 06 Jan 2002 02:38:51 GMT 4 From: "Terry C. Shannon" <terryshannon@mediaone.net>7 Subject: Re: Hey Intel, VMS is your ticket to high end! < Message-ID: <%MOZ7.2978$864.50605@rwcrnsc51.ops.asp.att.net>  - "cjt" <cheljuba@prodigy.net> wrote in message % news:3C37B3A6.442898D3@prodigy.net...  > "Terry C. Shannon" wrote: 
 > ><snip>>J > > Not only does Alpha have more software than IA64, it also has a proven track H > > record. From what I've seen in the Marvel lab, Compaq has nothing to fear# > > performance-wise from McKinley.  > L > That last sentence is Delphic.  If McKinley performance is low, and CompaqL > hitches itself to the McKinley wagon, do they THEN have something to fear?  H Depends on which part of CPQ... the ISSG will be using McKinley in theirI 32-way WildFire-derived IA64box. Users of REAL CPQ OSes (NSK, VMS, Tru64) 6 need not worry about McKinley. EV78 will whip its ass!   ------------------------------  $ Date: Sun, 6 Jan 2002 14:23:10 +1300 From: "AG" <ang@xtra.co.nz> : Subject: Re: historical evidence of what went wrong at DEC4 Message-ID: <fDNZ7.9386$_02.1078293@news.xtra.co.nz>  5 "Peter da Silva" <peter@taronga.com> wrote in message , news:a12sqc$107o$1@citadel.in.taronga.com... > H > So how often have you needed to rename ALL your ".somethingelse" filesI > to ".c"? Ever? When was the last time that you needed to rename a whole J > directory of files and the translation was as simple as one extension to
 > another?  L Well, except for the ".c" bit, several times a day, the last time being this last Friday, see below ...    > G > This specific example is very popular with VMS users, but in practice G > when have you ever needed to do this on UNIX? I don't think I've ever - > got a reasonable response to this question.  >   , Here is a try, let's see if it's reasonable:  G Suppose you are developing a program that must read and process a whole H lot of files from a Unix directory which are called *.something_or_other where J the "*" is any valid file name. After processing a file, it must rename it to somethingF like *.archive (this is a cast-iron spec requirement, non-negotiable).   So far so good, but ...   L The program is still under development/debugging/tuning so you would like to beJ able to re-run it as often as needed with the same set of input files. But	 remember? G Some (or all) of them have been renamed by the previous run... The most  natural < and obvious option would seem to be to just do some sort of:J "rename *.archive *.something_or_other" and off you go (you could even put that& thing on a hot-key somewhere perhaps).D Of course, I know there are work-arounds (and I use some), but those work-aroundsK are simply trying to implement that semantics in a bit of a roundabout way. C For example, the need to write a (even small) script means that the  functionality *is* needed but *is* lacking ...   
 Best regards,  AG  H [BTW, I don't claim a lot of expertise with Unixes, so perhaps there are better solutionsL than writing a shell script (or using backup/restore;)) but the question was when/why/how< often you need this. Hope this is a "reasonable" example...]   ------------------------------   Date: 6 Jan 2002 02:00:07 GMT ( From: peter@taronga.com (Peter da Silva): Subject: Re: historical evidence of what went wrong at DEC1 Message-ID: <a18b37$ogl$1@citadel.in.taronga.com>   O In article <fDNZ7.9386$_02.1078293@news.xtra.co.nz>, AG <ang@xtra.co.nz> wrote: H >Suppose you are developing a program that must read and process a wholeI >lot of files from a Unix directory which are called *.something_or_other G >where the "*" is any valid file name. After processing a file, it must M >rename it to something like *.archive (this is a cast-iron spec requirement,  >non-negotiable).   K Is it writing the file during this processing step,or is it just processing . it and renaming it to indicate that it's done?  M >The program is still under development/debugging/tuning so you would like to J >be able to re-run it as often as needed with the same set of input files.G >But remember?  Some (or all) of them have been renamed by the previous G >run... The most natural and obvious option would seem to be to just do J >some sort of: "rename *.archive *.something_or_other" and off you go (you; >could even put that thing on a hot-key somewhere perhaps).   K Well, what I generally do when testing software that might modify the files M its working on is to have a directory full of "clean" unmunged data, and then / I copy it to a testbed when I need to do a run.   I If the operation is non-destructive, which I assume it is or you wouldn't H be renaming the files back, I'd use links or symlinks instead of copies.  L And in any case, I would never put part of the testbed in a hotkey. It wouldG always be driven by a script that could be shipped with the sources. In + most cases that script would be a makefile.   D >For example, the need to write a (even small) script means that the/ >functionality *is* needed but *is* lacking ...   L By that logic you should never need a scripting language in the first place.  K A better test for whether the functionality is needed is how often a scriptt< to provide precisely that functionality is created AND used.  I >[BTW, I don't claim a lot of expertise with Unixes, so perhaps there aree >better solutions M >than writing a shell script (or using backup/restore;)) but the question wass
 >when/why/how = >often you need this. Hope this is a "reasonable" example...]   H Not really. A good software development testbed is frequently a lot moreI complex than that. One project I'm working on has scripts to simulate theeL timing of messages recieved by an application... and it can reconstruct thatJ timing and the original messages by analysing a database and logfiles sent from the field.w   -- l@ Rev. Peter da Silva, ULC.	                                 WWFD?  F "Be conservative in what you generate, and liberal in what you accept" 	-- Matthew 10:16 (l.trans)S   ------------------------------  # Date: Sun, 06 Jan 2002 04:07:59 GMT0 From: robert@bonomi.invalid : Subject: Re: historical evidence of what went wrong at DEC8 Message-ID: <z4QZ7.679$Kf.12054@ord-read.news.verio.net>  O In article <fDNZ7.9386$_02.1078293@news.xtra.co.nz>, AG <ang@xtra.co.nz> wrote:C >l6 >"Peter da Silva" <peter@taronga.com> wrote in message- >news:a12sqc$107o$1@citadel.in.taronga.com...g >>I >> So how often have you needed to rename ALL your ".somethingelse" files J >> to ".c"? Ever? When was the last time that you needed to rename a wholeK >> directory of files and the translation was as simple as one extension tot >> another?n >:M >Well, except for the ".c" bit, several times a day, the last time being this  >last  >Friday, see below ... >  >>H >> This specific example is very popular with VMS users, but in practiceH >> when have you ever needed to do this on UNIX? I don't think I've ever. >> got a reasonable response to this question. >> >l- >Here is a try, let's see if it's reasonable:a >nH >Suppose you are developing a program that must read and process a wholeI >lot of files from a Unix directory which are called *.something_or_other  >whereK >the "*" is any valid file name. After processing a file, it must rename it 
 >to somethingwG >like *.archive (this is a cast-iron spec requirement, non-negotiable).g >a >So far so good, but ... > M >The program is still under development/debugging/tuning so you would like toX >beOK >able to re-run it as often as needed with the same set of input files. But-
 >remember?H >Some (or all) of them have been renamed by the previous run... The most >natural= >and obvious option would seem to be to just do some sort of:aK >"rename *.archive *.something_or_other" and off you go (you could even put  >thato' >thing on a hot-key somewhere perhaps). E >Of course, I know there are work-arounds (and I use some), but those 
 >work-aroundslL >are simply trying to implement that semantics in a bit of a roundabout way.D >For example, the need to write a (even small) script means that the >functionality *is*R >needed but *is* lacking ...    J Beg to differ.  If functionality is 'lacking', that means that you *can't*J do the desired thing.  That is a far cry from having the 'building blocks'F to construct the composite function.  In actuality, the fact that the I requisite functionality _can_ be accomplished within the 'limitations' oftM the semantic structure existant -- that *proves* that the 'desired' semanticsw are =not= necessary.  D If one has a -recurring- need to mass-rename by arbitrary extension,B then creating the following routine *once*  provides the requisite functionality:              #!/bin/cshe              foreach file ( *$1 )r,               mv $file `basename $file $1`$2            end  ; it gets invoked as "rename .archive .something_or_other".     J if you don't like that, then a shell 'alias' for the following single-lineI command:   'find . -name '*$1' -prune -exec mv {} \`basename {} $1\`$2 \;l    J One can debate the relative merits of having such a command as a 'default'K feature of the environemt, vs. having to add an 'extension' to the factory-  shipped environment.      D There is a difference in fundamental 'design philosophy" involved.    G In the UNIX world, by default, the 'command' never sees wildcards. All iJ wildcard 'expansion' is done by the command-line-interpreter, in a single G consistent manner.  One instance of the code, One pattern of behavior. mL Universally _consistent_ behavior.  smaller, simpler, code for the commands,D as they _do_not_ have to be aware of *anything* regarding wildcards.H The system, and command-interpreter, provide a set of 'building blocks',J allowing the user to assemble *whatever*he*desires* for behavior/semantics of the 'advanced functions'.  H In the TOPS world, each command has knowledge of the original tokens on H the command-line.  And can do "whatever kind of magic" with them that itL 'damn well pleases'.    Consistency in behavior occurs *only*if* the variousI program authors co-operate to that end.  Wildcard expansion logic, if any L is employed, must be installed in _each_ application.  makes the applicationH larger, more complex.  Hopefully, it uses a 'library' routine to provideH 'standard' expansion -- but that fails if 'non-standard' expansion rulesL are desired.  TOPS provides a 'richer' set of 'built-in' advanced functions.H but if you desire _different_ behavior, the implementation effort is far greater.    G Further, the semantics of TOPS wildcard expansion is ambiguous. to wit:l  M     "copy *.foo *.bar"  implies 'create a list of all names ending in ".foo",aN                                  create a second list from the first, changingK                                  the suffix to "bar". perform the indicatedrJ                                  action on the corresponding names in each'                                  list'.  *BUT*aM     "wc *.foo *.bar"  would seem to imply something completely different, no?n  L Whereupon, one is either into 'context-sensitive' wild-card expansion (whereI "*.foo *.bar" means one thing to 'wc', and something else to 'copy'), or rI silliness like  "(.foo|.bar)"  for 'anything that matches either suffix'..H and the additional hyper-silliness that accrues if one suffix (or both) 8 actually contains some of those magic 'meta-characters'.    J TThe TOPS user-interface does have a bunch of really *great* _convenience_K features -- command-completion, and the [ESC] prompting leading you throughc8 required arguments/parameters, come immediately to mind.  N The supposedly superior semantics of TOPS implementaton of "copy *.foo *.bar",K is not a strong argument.  what does "copy b*.foo c*.bar" do?  Or how about3G "copy b*xx  charlie*foo.baz"?  Or "copy *.* boo*-*.old" ??  suppose youtM want to rename *every* file, by appending '-old' to the suffix.  how's *that*.I done?  "copy *.* *.old" would replace the prior suffix, right?  how about  "copy *.* *.*-old"?  e  M I've been away from TOPS too long,  i'm not even sure which, _if_any_, of thekK above (semi-)flippant examples are valid in that environment.  If not, theylI illustrate deficiencies of the TOPS semantics. <grin>  Corollary, for thewJ ones that are -not- valid, how hard is it to *implement* an equivalent, orL work-around?  Doesn't have to be a 'generalized' solution -- specific to the/ format of the individual example is sufficient.d   ------------------------------  $ Date: Sun, 6 Jan 2002 18:38:50 +1300 From: "AG" <ang@xtra.co.nz> : Subject: Re: historical evidence of what went wrong at DEC4 Message-ID: <WmRZ7.9492$_02.1089146@news.xtra.co.nz>  ( <robert@bonomi.invalid> wrote in message2 news:z4QZ7.679$Kf.12054@ord-read.news.verio.net...J > In article <fDNZ7.9386$_02.1078293@news.xtra.co.nz>, AG <ang@xtra.co.nz> wrote: > >tG > >Of course, I know there are work-arounds (and I use some), but thoseo > >work-arounds I > >are simply trying to implement that semantics in a bit of a roundabout  way.F > >For example, the need to write a (even small) script means that the > >functionality *is*  > >needed but *is* lacking ... >t >wL > Beg to differ.  If functionality is 'lacking', that means that you *can't* > do the desired thing.7  A Beg to differ. When was the last time you ran the Turing machine? A You know, you can do everything on it (in principle) that you cane7 do on a Unix or VMS or Windows or Mac OS etc etc. WouldiA you really claim though that it is NOT lacking in functionallity?   6 >  That is a far cry from having the 'building blocks'G > to construct the composite function.  In actuality, the fact that theuK > requisite functionality _can_ be accomplished within the 'limitations' ofdE > the semantic structure existant -- that *proves* that the 'desired'z	 semantics  > are =not= necessary.  & Are we talking the Turing machine yet?  F > If one has a -recurring- need to mass-rename by arbitrary extension,D > then creating the following routine *once*  provides the requisite > functionality:  C Please note that in my original post above I said that I *know* and C do use the work-arounds like you suggest. That's no big deal either F way really. But - the original statement/question that sparked it was:   [Quote]e  6 >"Peter da Silva" <peter@taronga.com> wrote in message- >news:a12sqc$107o$1@citadel.in.taronga.com...i >>I >> So how often have you needed to rename ALL your ".somethingelse" filesaJ >> to ".c"? Ever? When was the last time that you needed to rename a wholeK >> directory of files and the translation was as simple as one extension to  >> another?e   <<snip>>   >>but in practiceeH >> when have you ever needed to do this on UNIX? I don't think I've ever. >> got a reasonable response to this question.  	 [Unquote]d   >SL > One can debate the relative merits of having such a command as a 'default'D > feature of the environemt, vs. having to add an 'extension' to the factory- > shipped environment. >e  K Fully agree with that. The challenge though was to show a real-life examplepJ where you do need (or would like) such functionality on a day-to-day basis' *regardless* of how it was implemented.n   >rD > There is a difference in fundamental 'design philosophy" involved. >nH > In the UNIX world, by default, the 'command' never sees wildcards. AllK > wildcard 'expansion' is done by the command-line-interpreter, in a singlesH > consistent manner.  One instance of the code, One pattern of behavior.D > Universally _consistent_ behavior.  smaller, simpler, code for the	 commands,vF > as they _do_not_ have to be aware of *anything* regarding wildcards.J > The system, and command-interpreter, provide a set of 'building blocks',L > allowing the user to assemble *whatever*he*desires* for behavior/semantics > of the 'advanced functions'.  B Once again, are we talking the Turing machine yet? Yes, of course,G you can do it either way - or mix both, it's just a trade-off question. B Do you need this thing often enough to warrant it's implementetionG as an in-built feature? Or is it so infrequent that if and when someoneMD needs it he is welcome to go ahead and write his own implementation?  A For example, I don't notice many objections from the users of any-E number of languages that have an in-built "sort" function saying thatn? they surely can roll one themselves (and, given a bit of a mathy; background, probably better than the default). So, let's gow0 minimalistic and abolish all in-built functions?   [Pass on the TOPS discussion]3   ------------------------------  # Date: Sun, 06 Jan 2002 06:25:18 GMT * From: "Bill Todd" <billtodd@metrocast.net>9 Subject: Re: HP admits it will kill VMS if merger suceedsaB Message-ID: <i5SZ7.547972$8q.44522473@bin2.nnrp.aus1.giganews.com>  > "John McLean" <mcleanj@swissonline.delete.ch> wrote in message/ news:3C3624D5.94396538@swissonline.delete.ch...i   ...t  , On the other hand, I suspect MC would not be> adverse to a well-reasoned argument using sound logic and with+ irrefutable evidence to clearly support it.-  9 < Could you describe in detail how that would differ frome6 < what we did 20 months ago?  And why you'd expect any < different result this time?  <a6 < You keep saying you *suspect* this.  What *possible*6 < evidence do you have to cause you to ignore the many: < existing precedents that so strongly indicate otherwise?  ? Constructive criticism might achieve a lot more than just plaine
 criticism.  ; < Both have been tried, often.  Neither has been effective,d5 < ever (at least not at the level you're discussing).? <e5 < It's either time to try other means or stop wastingh < our time.t <  < - bill   ------------------------------  # Date: Sun, 06 Jan 2002 06:32:35 GMT * From: "Bill Todd" <billtodd@metrocast.net>9 Subject: Re: HP admits it will kill VMS if merger suceeds C Message-ID: <7cSZ7.170130$m05.14578369@bin5.nnrp.aus1.giganews.com>a  ? "Terry C. Shannon" <terryshannon@mediaone.net> wrote in messagen& news:gFvZ7.4701$Sf2.39825@rwcrnsc52...   ...i  H > If CPQ makes money on VMS in 4FQ (a given) and loses money on consumerJ > peecees in 4FQ (almost surely a given), I would have to imagine that the guyo# > is smart enough to see the light..  I You mean like he did in all the previous quarters when the same situationa	 occurred?u   - bill   ------------------------------  % Date: Sun, 06 Jan 2002 04:03:43 +0100-2 From: martin@radiogaga.harz.de (Martin Vorlaender)  Subject: Re: locking display CDE; Message-ID: <3c37be8f.524144494f47414741@radiogaga.harz.de>o  " Tom Linden (tom@kednos.com) wrote:@ > How does one turn off the automatic locking due to inactivity?$ > I am running CDE under 7.3 on AXP.  H From the Front Panel, select the Style Manager (second icon in the rightD half). In the pop-up window, select Screen, and you can turn off the screen blanker.h   cu,s   Martin -- sG  Your mouse has moved.     | Martin Vorlaender  |  VMS & WNT programmer 4  Windows must be restarted | work: mv@pdv-systeme.deH  for the change to take    |    http://www.pdv-systeme.de/users/martinv/;  effect. Reboot now? [OK]  | home: martin@radiogaga.harz.deb   ------------------------------  % Date: Sat, 05 Jan 2002 20:05:20 -0500e, From: Michael Austin <maustin@spacelots.com>4 Subject: Re: NAT: publishing the remote IP of router- Message-ID: <3C37A2D0.B2870C9A@spacelots.com>y   JF Mezei wrote:d > O > OK, I have a lan connected to the internet with a NAT router. A vax can go toaP > the router and extract  the IP address that the router has negotiated with theO > ISP so it knows what the internet IP address is used to reach my lan from thee > rest of the world. > I > Various scripts on the cluster will want to know that address when they G > generate HTLM pages stored elsewhere to point to my hosts  (or do thee5 > negotiation of dynamic dns with a dyn dsn service).* > N > Any suggestion on how I should store that address ? If through a clusterwideO > logical name, any suggestions on its name ? (is there a standard for this ?).i   This is what I do:  G I have a program that logs into the router every 5 minutes and gets thepB current  IP address, I then update my dyn dns service with this IP2 address(using an old Pentium/166 running Linux).    F On the Alpha, I have Apache running using Virtual Hosts and IP addressF 192.168.1.200 (internal to my network only)  I have port forwarding onG the router for the appropriate ports and all point to the Alpha.  DoinglB it this way, I never have to really know what that IP address is.    http://www.spacelots.com http://www.firstdbasource.com   E are both "virutal hosts" on the same box using a dynamic dns provideriD www.zoneedit.com  -- very easy to configure "A" records, MX records,? etc...(free for up to 5 domain names and up to ~1M DNS queries)a  C I have a variation of the perl app that updates the dynamic dns forvC OpenVMS Perl, but have yet to implement it.  I am doing all of this C until I can afford to get a real T1 or T3. Until then, this is goodu enough.p -- r   Regards,   Michael Austin7 First DBA Source, Inc. -- http://www.firstdbasource.comp President/Sr. DBA Consultant 704-947-1089 (Office)d 704-236-4377 (Mobile)r   ------------------------------  # Date: Sun, 06 Jan 2002 04:45:19 GMTo- From: "John E. Malmberg" <wb8tyw@qsl.network>t" Subject: Re: Problem unpacking TSM( Message-ID: <3C37DB28.50409@qsl.network>  A There is a known problem with the TSM on the FREEWARE 5.0 CDROMS.r  J I do not know what the issue is with the download images on the DNPG site:  / http://digitalnetworks.net/dnpg/dr/npg/tsm.htmlh  C But the download images on for the FREEWARE have been fixed by the i0 freeware maintainer, and can be downloaded from:  6 http://www.openvms.compaq.com/freeware/freeware50/tsm/  H I just downloaded the ALPHA kit and decompressed it.  It installed just - fine as documented in the freeware_readme.txtI  I The required PAK is not on the CD-ROM or the FREEWARE download site, but  E can be downloaded from the Digital Network Products Group URL that I s posted earlier.   G Some tools that may be useful with TSM are also on the FREEWARE CD-ROM.,  < http://www.openvms.compaq.com/freeware/freeware50/tsm_tools/   -Johny wb8tyw@qsl.network Personal Opinion Only"   ------------------------------  % Date: Sat, 05 Jan 2002 20:12:17 -0500u, From: Michael Austin <maustin@spacelots.com> Subject: Re: VAX in a VT-103?o- Message-ID: <3C37A471.920635BF@spacelots.com>g   "C.W.Holeman II" wrote:r > E > Anyone ever try putting a VAX in a VT-103 and boot something on it?z >  > -- > C.W.Holeman II > cwhii@mail.com > http://also.as/cwhii  C The VT103 was actually a PDP11 in a VT box.  I have seen it running>F RSX11 on two floppies.  I saw it when it was a product offered by DataG Checker(??) Systems of MA. They used it for programming electronic cashaG registers using a "inter-register connection that looked a lot like a 3t/ or 4 wire serial connection.  That was in 1979.h -- "   Regards,   Michael Austin   ------------------------------  $ Date: Sat, 5 Jan 2002 18:01:13 -0800# From: "Tom Linden" <tom@kednos.com>M Subject: RE: VAX in a VT-103?e9 Message-ID: <CIEJLCMNHNNDLLOOGNJIIELHDNAA.tom@kednos.com>v  ? I took one apart around 1982.  As i recall it had 4 Q-bus slotsr (but I could be wrong)   > -----Original Message-----5 > From: Michael Austin [mailto:maustin@spacelots.com]s* > Sent: Saturday, January 05, 2002 5:12 PM > To: Info-VAX@Mvb.Saic.Comg > Subject: Re: VAX in a VT-103?n >  >  > "C.W.Holeman II" wrote:  > > G > > Anyone ever try putting a VAX in a VT-103 and boot something on it?n > >  > > -- > > C.W.Holeman II > > cwhii@mail.com > > http://also.as/cwhii > E > The VT103 was actually a PDP11 in a VT box.  I have seen it runningOH > RSX11 on two floppies.  I saw it when it was a product offered by DataI > Checker(??) Systems of MA. They used it for programming electronic cashrI > registers using a "inter-register connection that looked a lot like a 3o1 > or 4 wire serial connection.  That was in 1979.e > -- o > 
 > Regards, >  > Michael Austin >    ------------------------------    Date: 06 Jan 2002 00:23:44 -05001 From: "Matthew X. Economou" <xenophon@irtnog.org>u> Subject: [Q] Installing the VXT2000+ software on OpenVMS/Alpha0 Message-ID: <w4owuyw8333.fsf@eco-fs1.irtnog.org>  C I own a VXT2000+ X terminal and an AlphaStation 200 running OpenVMSuA 7.2.  I would like to boot the terminal from the AlphaStation.  Ig= downloaded the VXT software (a backup set and an installationaC script).  The script presumes the existance of an InfoServer on thee6 network, but, of course, I only have the AlphaStation.  F Would someone be so kind as to help me get this software installed?  IB can't seem to figure out how to start the MOP server.  Neither theE webbed OpenVMS documentation nor Google have turned up any useful (oreE understandable) information.  If hand-holding is out of the question,r@ I'd be grateful for just the right keywords to feed to Google orD Altavista.  Playing around with LATCP and LANCP and so forth haven't  gotten me anything but confused.  
 Kind regards,a	 #\Matthew    -- iF Matthew X. Economou <xenophon@irtnog.org> - Unsafe at any clock speed!J "Little grey men are coming our way (tastes just like chicken, they say)."	  - Clutch    ------------------------------  # Date: Sun, 06 Jan 2002 06:21:13 GMTn+ From: Ryan Moore <rmoore@rmoore.dyndns.org>vB Subject: Re: [Q] Installing the VXT2000+ software on OpenVMS/Alpha< Message-ID: <Pine.LNX.4.31.0201052214160.2688-100000@jaipur>  G Well, the 7.2 documention that talks about MOP loading is in the System I Manager's Manual.  For 7.2, it is chapter 22.  Pay attention to the LANCPtD commands like ADD NODE, etc.  The boot file should be something likeJ VXTLDR.SYS.  You'll need to know the hardware address of the VXT, but that$ shouldnt' be hard to get figure out.  H http://www.openvms.compaq.com:8000/72final/6017/6017pro_091.html#6017lan  G The bad news is that I believe an Infoserver is *required* to run a VXTaD terminal.  The InfoServer is what serves the I/O to the VXT after itK boots.  The Alpha can only send the VXT bootloader over MOP, it can't serve J the rest of the files to the VXT as far as I know.  Without the InfoSever, I believe you are out of luck.  ) On 6 Jan 2002, Matthew X. Economou wrote:e  E > I own a VXT2000+ X terminal and an AlphaStation 200 running OpenVMSiC > 7.2.  I would like to boot the terminal from the AlphaStation.  Ik? > downloaded the VXT software (a backup set and an installationlE > script).  The script presumes the existance of an InfoServer on then8 > network, but, of course, I only have the AlphaStation. >rH > Would someone be so kind as to help me get this software installed?  ID > can't seem to figure out how to start the MOP server.  Neither theG > webbed OpenVMS documentation nor Google have turned up any useful (orpG > understandable) information.  If hand-holding is out of the question, B > I'd be grateful for just the right keywords to feed to Google orF > Altavista.  Playing around with LATCP and LANCP and so forth haven't" > gotten me anything but confused.   ------------------------------   End of INFO-VAX 2002.009 ************************