1 INFO-VAX	Sun, 22 Sep 2002	Volume 2002 : Issue 523       Contents:1 Re: Announcing the first annual OpenVMS Symposium 1 Re: Announcing the first annual OpenVMS Symposium  COPY/FTP passive mode * Re: delete/erase & init/erase w/dod_erapatP Re: ETS is only a couple of weeks away - Leper colony - how many of you are atte& Re: How to run a C program on VAX/VMS?D Kudos to HP on JavaScript (was: Oracle RDB 8.0 = Oracle RDB for IAF)H Re: Kudos to HP on JavaScript (was: Oracle RDB 8.0 = Oracle RDB for IAF)A Re: Legacy Win: NASA Kills 'Wounded' Launch System Upgrade at KSC A Re: Legacy Win: NASA Kills 'Wounded' Launch System Upgrade at KSC A Re: Legacy Win: NASA Kills 'Wounded' Launch System Upgrade at KSC A Re: Legacy Win: NASA Kills 'Wounded' Launch System Upgrade at KSC  Re: Marketing suggestion Re: Marketing suggestion Re: Marketing suggestion Re: Marketing suggestion Re: Marketing suggestion Re: Marketing suggestion Re: Marketing suggestion Re: Marketing suggestion Re: Marketing suggestion Re: Marketing suggestion# New Miss America - Miss Ill-a-Noise  Re: OpenVMS SIG 0 Re: Oracle RDB 8.0 = Oracle RDB for Itanium ????3 Re: Storing username/password for external services 3 Re: Storing username/password for external services 0 Re: VMS future (oh not not another one of these)0 Re: VMS future (oh not not another one of these)0 Re: VMS future (oh not not another one of these) Re: VMS, C++ and flush Re: VMS, C++ and flush Re: VMS, C++ and flush Re: VMS, C++ and flush Re: VMS, C++ and flush Re: VMS, C++ and flush Re: VMS, C++ and flush Re: VMS, C++ and flush( Will they offer WebLogic on VMS as well?, Re: Will they offer WebLogic on VMS as well? XP1000's going cheap!  Re: XP1000's going cheap! 2 Re: [OT] Could Philip Zimmermann work for the CIA?  F ----------------------------------------------------------------------  # Date: Sun, 22 Sep 2002 02:06:15 GMT # From: "John Smith" <a@nonymous.com> : Subject: Re: Announcing the first annual OpenVMS SymposiumG Message-ID: <rA9j9.87679$U_.67369@news04.bloor.is.net.cable.rogers.com>   @ "Fred Kleinsorge" <kleinsorge@star.zko.dec.com> wrote in message+ news:ipJh9.34$AM3.96351@news.cpqcorp.net...  > ? > JF Mezei wrote in message <3D861CE8.9BCA8E66@videotron.ca>...  > >Fred Kleinsorge wrote: G > >> If you want to do anything aside from hanging around at the hotel,  you'llI > >> need a car.  Nashua is only modestly mass-transit friendly, and most  > places) > >> are a bit more than waling distance.  > > G > >But surely Sue or any VMS engineer would offer to drive participants  where  > >they want to go ? > >  > >:-) :-) :-) :-) :-) > ' > I drive a Z3.  So only one at a time.     K You could probably strap a couple more to the trunk-mounted ski rack. A few 0 bungee cords or some duct tape as seat belts....   ------------------------------  % Date: Sun, 22 Sep 2002 01:09:02 -0400 - From: JF Mezei <jfmezei.spamnot@videotron.ca> : Subject: Re: Announcing the first annual OpenVMS Symposium, Message-ID: <3D8D505C.CB2587E9@videotron.ca>   John Smith wrote: ) > > I drive a Z3.  So only one at a time.  > M > You could probably strap a couple more to the trunk-mounted ski rack. A few 2 > bungee cords or some duct tape as seat belts....  M I have a feeling that if I were the lucky one, they would conveniently forget N to unstrap me once they got to the restaurant/pub and leave me strapped to the ski rack... :-(    ------------------------------  % Date: Sun, 22 Sep 2002 00:03:56 -0400 * From: "Stanley F. Quayle" <stan@stanq.com> Subject: COPY/FTP passive mode. Message-ID: <3D8D08EC.13672.6BEBD78@localhost>  C I have an application which uses COPY/FTP to send files to another   system  C on the Internet.  Due to a firewall change, I must now use passive   mode   for all transfers.  F Is it possible to force passive transfers on a user or system basis?    C I'm currently using VMS 7.3 and UCX 5.1 on both VAX and Alpha (and   could upgrade to  $ UCX 5.3 when the latest CD arrives).  
 --Stan Quayle  Quayle Consulting Inc.  
 ----------C Stanley F. Quayle, P.E. N8SQ  +1 614-868-1363  Fax: +1 614 868-1671 1 8572 North Spring Ct. NW, Pickerington, OH  43147 = Preferred address:  stan@stanq.com       http://www.stanq.com    ------------------------------  % Date: Sat, 21 Sep 2002 15:43:44 -0400 - From: JF Mezei <jfmezei.spamnot@videotron.ca> 3 Subject: Re: delete/erase & init/erase w/dod_erapat , Message-ID: <3D8CCBEC.E9F93426@videotron.ca>   Dec Vax Vms wrote:G > anyone know if when I use the delete/erase or initialize/erase if the 4 > eraparloa.exe ($erapat) is automatically called?    K I though I had an easy answer to this, but then I read the HELP DELETE file A /ERASE and it mentions that it uses a "system specified pattern".   I The dod_erapat.mar explains how this is to be setup as a kernel mode user ; written system service that creates the routine EXE$ERAPAT    < There is also a built in system service called SYS$ERAPAT  .  ? The file SYS$SYSTEM:SYSINIT.EXE contains the following message: 8 	"error loading installation -specific $ERAPAT service".4  And the SYS.STB contains a reference to EXE$ERAPAT.  L So it would be pretty fair to assume that the dod_erapat is meant to replace& the default SYS$ERAPAT system service.  M Note also the year of the dod_erapat.mar file... 1982 !!!!!!!   I recall some H discussions in the late 1980s early 1990s about someone requiring a moreK robust disk "destruction" software because such didn't exist yet on VMS and K some time later, the request had been satisfied. ( This was to dispose of a I disk drive, not during daily usage, so perhaps the requirements were much 4 stricter than just regular, daily deleting of files.  F DELETE.EXE isn't a very big file, most of the work seems to be done byN system/rtl services. Interestingly, SYS$ERASE  does not provide for an obviousJ /ERASE functionlality, unless perhaps one of the XABs contains a flag that3 specify that file should be overwritten on delete.      H What you could do to be *really* sure is to create your own routine thatI places a known pattern (always all "01010101") and install that as a user K written system $ERAPAT service, delete a file and then dump raw contents of L the drive where the file was located to see if your pattern was used. If so,Q then you can install the file in sys$examples and be confident it is really used.     J What utility does one use to dump the contents of a specific location on a disk drive ?   ------------------------------  # Date: Sat, 21 Sep 2002 19:02:46 GMT $ From: Jim Duff <jim@eight-cubed.com>Y Subject: Re: ETS is only a couple of weeks away - Leper colony - how many of you are atte . Message-ID: <3D8CC254.9020701@eight-cubed.com>   Sue Skonetski wrote:- > For the non leper's here is an explanation.  > = > At DECUS there are a few of us that get together for liquid H > refreshment.  We called ourselves the Leper colony because some of theG > folks smoke and they have to sit outside (at least in LA)and the rest E > of them would join them.  Mainly because they are great company and 
 > VMS bigots.  > H > Send me mail and we can arrange a time.  If I remember correctly there$ > are several brew pubs in St Louis. > / > This is going to be great week, see you soon.  >  > Sue     Did someone mention "brew pubs"?   ;-)    --   jim@eight-cubed.com    ------------------------------  # Date: Sat, 21 Sep 2002 19:42:18 GMT 1 From: "David J. Dachtera" <djesys.nospam@fsi.net> / Subject: Re: How to run a C program on VAX/VMS? ' Message-ID: <3D8CD133.F643054E@fsi.net>    Alex Ulis wrote: > I > I compiled C program using CC command. However, instead of .EXE file, I M > found .OBJ object file in my main directory. What should I do to be able to  > run C program on Vax?   G This is another problem that has arisen with the (ab)use of terminology  in the UN*X world.  G See, UN*X-land compilers typically produce a runnable binary, sometimes B by themselves, sometimes by invoking a linker transparently to the@ programmer/user. This tends to not be true outside of UN*X-land.  B In the word of IBM mainframe O.S.-es, you must compile/assemble to@ produce object code, then invoke the linkage editor to produce a runnable binary.  A In DOS-land, you compile/assemble then link to produce a runnable  binary.   F Likewise, in the realms of DEC-heritage o.s.-es, the compilers produceA object code and the LINKers (sometimes known as "Task Builder"'s)  produce runnable binaries.  B So, the more correct concept is to "compile and link", even though; UN*X-land frequently compresses that down to a single step.    --   David J. Dachtera  dba DJE Systems  http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/    ------------------------------    Date: 21 Sep 2002 14:21:53 -0600- From: Kilgallen@SpamCop.net (Larry Kilgallen) M Subject: Kudos to HP on JavaScript (was: Oracle RDB 8.0 = Oracle RDB for IAF) 3 Message-ID: <KR0TA1XzUIjj@eisner.encompasserve.org>   Z In article <3D8CC96B.644EC4D@fsi.net>, "David J. Dachtera" <djesys.nospam@fsi.net> writes: > Michael Unger wrote:  B >> The result is simply a plain white page, nothing else. CheckingF >> www.oracle.com there is a tiny remark at the top of the page saying5 >> "This page requires a JavaScript-enabled browser."  >>  B >> I suppose JavaScript isn't liked very much in the VMS communityE >> regardless of the system (VMS, Winwoes or something else) which is  >> used to view that page. > I > I rather expected Larry K. to note that javascript is avoided by anyone  > who is security conscious.  G I did not see the initial post, but since you know my feelings it would 3 probably be a waste of bandwidth for me to comment.   C I would like to say I think switching to the HP style web page from A the Compaq style web page is a great step forward, as HP seems to > indicate in the upper left corner "This page makes some use of> JavaScript and here is a link to a list of what aspects of theA page you are missing by having JavaScript disabled."  HP at least B recognizes that their customer base includes people who care about
 such things !    ------------------------------  % Date: Sat, 21 Sep 2002 17:41:41 -0400 - From: JF Mezei <jfmezei.spamnot@videotron.ca> Q Subject: Re: Kudos to HP on JavaScript (was: Oracle RDB 8.0 = Oracle RDB for IAF) , Message-ID: <3D8CE789.AD52513F@videotron.ca>   Larry Kilgallen wrote:E > I would like to say I think switching to the HP style web page from 4 > the Compaq style web page is a great step forward,  - Have you tried http://www.nasdaq.com lately ?   V I sent a rather nasty message to them.  (Try their page with Netscape 4.7 on a mac...)  I And don't ever try to validate that front page with the w3c validator, it 2 barfs right at the top and refuses to continue :-)  N I do not know what sort of deal NASDAQ has with microsoft, but it gives NASDAQ8 a very bad image to anyone know knows anything about IT.   ------------------------------    Date: 21 Sep 2002 16:41:01 -0700( From: bob@instantwhip.com (Bob Ceculski)J Subject: Re: Legacy Win: NASA Kills 'Wounded' Launch System Upgrade at KSC< Message-ID: <d7791aa1.0209211541.9e86659@posting.google.com>  d winston@SSRL.SLAC.STANFORD.EDU wrote in message news:<00A144AB.A567010B@SSRL04.SLAC.STANFORD.EDU>...u > In article <3D8AF738.1F43F473@mediasec.de>, Jan C. =?iso-8859-1?Q?Vorbr=FCggen?= <jvorbrueggen@mediasec.de> writes: H > >At some place in the article, the cost-to-completion was estimated atP > >around $560M. That's at least 3000 person-years of effort, for some 3 millionI > >lines of code (why so much, anyway?). So a person on this project does I > >1000 lines of code per year, or about 4 lines per average working day. 2 > >Man, what would I give to have a job like that. > P > That might be four _debugged_ lines per working day, or four formally-verified< > lines, etc.  And it might have included, oh, say, testing. > Q > I personally wrote an episodic-television budgeting package that was considered Q > the best in the industry at that time.  After I left Lorimar got a VAX 8600 and Q > converted all this working Fortran stuff to COBOL and DBMS, because it was what  > management understood.)  >  > 	 > -- Alan   C maybe if they would have used DIBOL, there wouldn't be any rewrites E necessary ... I started with and still work with code and subroutines D from '80's mcba dibol acct apps and everything ported thru the years< from RSTS/E to VAX to ALPHA OpenVMS without any rewrites ...   ------------------------------  % Date: Sat, 21 Sep 2002 21:21:22 -0400  From: <rob@netcarrier.net>J Subject: Re: Legacy Win: NASA Kills 'Wounded' Launch System Upgrade at KSC8 Message-ID: <cu8j9.322$I%5.87068766@news.netcarrier.net>  E " maybe if they would have used DIBOL, there wouldn't be any rewrites G > necessary ... I started with and still work with code and subroutines F > from '80's mcba dibol acct apps and everything ported thru the years> > from RSTS/E to VAX to ALPHA OpenVMS without any rewrites ...    3     Too bad there are only   4 programers remaining    ------------------------------  # Date: Sun, 22 Sep 2002 01:44:12 GMT 1 From: "David J. Dachtera" <djesys.nospam@fsi.net> J Subject: Re: Legacy Win: NASA Kills 'Wounded' Launch System Upgrade at KSC' Message-ID: <3D8D260B.EC8622CA@fsi.net>    rob@netcarrier.net wrote:  > G > " maybe if they would have used DIBOL, there wouldn't be any rewrites I > > necessary ... I started with and still work with code and subroutines H > > from '80's mcba dibol acct apps and everything ported thru the years@ > > from RSTS/E to VAX to ALPHA OpenVMS without any rewrites ... > 5 >     Too bad there are only   4 programers remaining   D Make it five(5). I could do DIBOL if I had to, but it wouldn't be my
 first choice.    --   David J. Dachtera  dba DJE Systems  http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/    ------------------------------  % Date: Sat, 21 Sep 2002 21:59:08 -0400 - From: JF Mezei <jfmezei.spamnot@videotron.ca> J Subject: Re: Legacy Win: NASA Kills 'Wounded' Launch System Upgrade at KSC, Message-ID: <3D8D23E6.5F7CD6F4@videotron.ca>   rob@netcarrier.net wrote:  > G > " maybe if they would have used DIBOL, there wouldn't be any rewrites I > > necessary ... I started with and still work with code and subroutines H > > from '80's mcba dibol acct apps and everything ported thru the years@ > > from RSTS/E to VAX to ALPHA OpenVMS without any rewrites ...  L When one decides to write a new system from scratch, one often tends to haveN "blue sky" goals of adding plenty of new stuff and changing everything. And asJ you start to get going, all those brand new requirements probably constalyL change as you realise some stuff is superfluous and realise you forgot otherQ stuff. In the end, you have a much more complex project that you had anticipated.   L Remember that those who push for such a project tend to lowball the costs inM order to get the project approved and later just request for additional money " to cover those unforeseen costs...   ------------------------------  % Date: Sat, 21 Sep 2002 16:02:49 -0400 - From: JF Mezei <jfmezei.spamnot@videotron.ca> ! Subject: Re: Marketing suggestion , Message-ID: <3D8CD063.79438643@videotron.ca>   John Smith wrote:  about the marketing of SUVs.  5 The "Subaru Outback" is not very common in Australia.   L But once, as I was nearing the end of a stretch on an outback road on a ROADK BIKE (not even mountain bike), I met a some folks at the roadhouse where we K stayed overnight. They had to turn back because their car wasn't suited for Q the road ahead (the part I had already done, they were going opposite direction).   L They were driving a, you guessed it, SUBARU OUTBACK. It may have had 4 wheelE drive, but its "urban marketing" which resulted in a low car, made it D unsuitable for the road they had though they could do on such a car.  D I also met some "real" folks with some real outback vehicles (toyotaN landrunners). These were experts and had just done the connie-sue highway, butI in doing so, damaged an axle in one of their 2 vehicles, so they used the M second one to drive to Alice Springs to get the part, came back to install it N themselves. They told me that any 4 wheel drive that does not have the abilityM to disengage the all  differentials was totally useless.  If you are stuck in K the mud with only one wheel having traction, you need to give power to that N wheel. A diferential results in the wheels that have least resistance spinning/ while the one with traction given little power.   J I also saw some "real" landrovers, the ones with beaten up aluminium skin,J oil-based filters for air intake (paper/fibre based air filters doN't lastJ long in the outback/dust/sand.) and of course the snorkel to take air muchL higher when you are crossing a stream. Now, that was a real outback vehicle.  I So when I see those TV ads for SUVs, I laugh.  The folks are given such a N false sense of security by the advertsing. Heck, BMW came out with on that hasQ a motorised step to help the ladies in their gowns get on and off the vehicle :-)    ------------------------------  # Date: Sat, 21 Sep 2002 20:42:11 GMT * From: "Bill Todd" <billtodd@metrocast.net>! Subject: Re: Marketing suggestion B Message-ID: <DQ4j9.191124$AR1.8120787@bin2.nnrp.aus1.giganews.com>  : "JF Mezei" <jfmezei.spamnot@videotron.ca> wrote in message& news:3D8CD063.79438643@videotron.ca...   ...   K > So when I see those TV ads for SUVs, I laugh.  The folks are given such a , > false sense of security by the advertsing.  @ Actually, no.  Virtually all the people who buy SUVs will not beI encountering anything but low-traction situations in which the 4WD may in K fact be useful and the other limitations you note won't matter:  they won't L be fording deep streams, or driving on non-roads, or through vast amounts ofK desert dust.  The closest they're likely to come to having ground-clearance J problems will be deep mud (some Northern New England dirt roads are closedK for this reason for a few weeks each year, but are bad enough that even the J highest vehicles can have problems) and deep snow (where you have a prettyL good idea of what you're getting into before you get into it, and should act
 accordingly).   J The false sense of security is much more likely to come from the incorrectJ belief that they'll be safer in a crash, as was just noted (I think here).   - bill   ------------------------------  % Date: Sat, 21 Sep 2002 17:53:30 -0400 - From: JF Mezei <jfmezei.spamnot@videotron.ca> ! Subject: Re: Marketing suggestion , Message-ID: <3D8CEA4D.4387075A@videotron.ca>   Bill Todd wrote:B > Actually, no.  Virtually all the people who buy SUVs will not beK > encountering anything but low-traction situations in which the 4WD may in A > fact be useful and the other limitations you note won't matter:   B The problem is that the urban SUVs have tractions systems that useK differentials. Unless you can disengage the differentials in an intelligent @ way, then you don't really have added traction when you need it.  M The idea of 4 wheel drive isn't to spread the traction on the nice paved road K to all 4 wheels (probably reduce tire wear), it is to be able to use any of M the 4 wheels to get you out of a mud, snow etc. The problem with a diffential L is that the one wheel that is able to get you out will be "disengaged" while all other 3 spin.   K Some models may be architectured to at least give you forward traction when J rear wheels are sliding, while others won't.  So just saying you have a "4B wheel drive" isn't enough to say that it gives you added traction.  N You need to be aware of hwo the differentials are configured for your car, and/ need the ability to disengage them when needed.    ------------------------------  # Date: Sat, 21 Sep 2002 22:25:08 GMT * From: "Bill Todd" <billtodd@metrocast.net>! Subject: Re: Marketing suggestion B Message-ID: <8l6j9.187426$z91.7889454@bin3.nnrp.aus1.giganews.com>  : "JF Mezei" <jfmezei.spamnot@videotron.ca> wrote in message& news:3D8CEA4D.4387075A@videotron.ca... > Bill Todd wrote:D > > Actually, no.  Virtually all the people who buy SUVs will not beJ > > encountering anything but low-traction situations in which the 4WD may inC > > fact be useful and the other limitations you note won't matter:  > D > The problem is that the urban SUVs have tractions systems that use > differentials.  L And the solution is limited-slip differentials:  do you know for a fact thatE they aren't used, because it's certainly my impression that they are.    - bill   ------------------------------   Date: 22 Sep 2002 00:05:14 GMT( From: bill@cs.uofs.edu (Bill Gunshannon)! Subject: Re: Marketing suggestion 5 Message-ID: <amj1fq$672pa$1@ID-135708.news.dfncis.de>   G In article <R50j9.10791$q41.6999@news02.bloor.is.net.cable.rogers.com>, & 	"John Smith" <a@nonymous.com> writes:M > As found on-line today while searching for reviews of a new book, "High And J > Mighty - SUVs: The World's Most Dangerous Vehicles And How They Got That > Way".   5 Not having read it, I am obviously not an expert but, $  it smacks of "Unsafe at any speed".   bill   --  J Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesD bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. University of Scranton   |A Scranton, Pennsylvania   |         #include <std.disclaimer.h>       ------------------------------  % Date: Sat, 21 Sep 2002 19:35:54 -0400  From: Everhart <ge@gce.com> ! Subject: Re: Marketing suggestion + Message-ID: <amivou$gka$1@bob.news.rcn.net>   C The 4WD in our SUV is of some value in snow (which is not that much E of an issue here) but the SUV is hellishly useful in towing our horse F trailer. Towing capacity of 12000 pounds is plenty for the trailer and a couple large animals...   D As my cousin puts it, "there is no replacement for displacement" ;-)  C We of course use smaller vehicles when possible for other transport E needs. Said vehicles would quickly lose their transmissions if one of ' THEM were to attempt to tow the horses.   H Who gets off trying to characterize all buyers of these things?  Go back0 and climb Mont Blanc in your 2CV and have fun...     Bill Todd wrote:< > "JF Mezei" <jfmezei.spamnot@videotron.ca> wrote in message( > news:3D8CD063.79438643@videotron.ca... >  > ...  >  > K >>So when I see those TV ads for SUVs, I laugh.  The folks are given such a , >>false sense of security by the advertsing. >  > B > Actually, no.  Virtually all the people who buy SUVs will not beK > encountering anything but low-traction situations in which the 4WD may in M > fact be useful and the other limitations you note won't matter:  they won't N > be fording deep streams, or driving on non-roads, or through vast amounts ofM > desert dust.  The closest they're likely to come to having ground-clearance L > problems will be deep mud (some Northern New England dirt roads are closedM > for this reason for a few weeks each year, but are bad enough that even the L > highest vehicles can have problems) and deep snow (where you have a prettyN > good idea of what you're getting into before you get into it, and should act > accordingly).  > L > The false sense of security is much more likely to come from the incorrectL > belief that they'll be safer in a crash, as was just noted (I think here). >  > - bill >  >  >    ------------------------------  # Date: Sun, 22 Sep 2002 01:55:27 GMT # From: "John Smith" <a@nonymous.com> ! Subject: Re: Marketing suggestion F Message-ID: <jq9j9.15860$q41.117@news02.bloor.is.net.cable.rogers.com>  5 "Bill Gunshannon" <bill@cs.uofs.edu> wrote in message / news:amj1fq$672pa$1@ID-135708.news.dfncis.de... I > In article <R50j9.10791$q41.6999@news02.bloor.is.net.cable.rogers.com>, ' > "John Smith" <a@nonymous.com> writes: K > > As found on-line today while searching for reviews of a new book, "High  And L > > Mighty - SUVs: The World's Most Dangerous Vehicles And How They Got That	 > > Way".  > 7 > Not having read it, I am obviously not an expert but, & >  it smacks of "Unsafe at any speed".    L Corvair was just a mid-engine car with a low polar moment of inertia. EngineE was ahead of the axle, so it qualifies for the mid-engine status. The H problem it suffered from is the same problem that Porsche 911's sufferedK from up until about 20 years ago - a desire to swap ends under certain road J and driving conditions. With the Porsche, the average driver is a bit moreJ knowledgeable and competent about anticipating and correcting this state -" the average Corvair driver wasn't.  L My personal dislike of SUV's (for a variety of reasons) aside, my real pointG in posting the article was the very last line and thinking of it in the L context of IT decision makers who can choose to spend approximately the sameD amount of dough and get either a fully robust operating system, or aK reasonably robust o/s, or an o/s with which they can play roulette with the " future of their business and data.  ; It all comes down to perception - and that means marketing.    ------------------------------  # Date: Sun, 22 Sep 2002 02:32:02 GMT % From: "-Andy-" <acs@fcgnet.works.net> ! Subject: Re: Marketing suggestion > Message-ID: <Xns9290E4E4F2480acsfcgnetworksnet@216.166.71.232>  8 "Bill Todd" <billtodd@metrocast.net> enlightened us with< news:8l6j9.187426$z91.7889454@bin3.nnrp.aus1.giganews.com on
 21 Sep 2002:     > < > "JF Mezei" <jfmezei.spamnot@videotron.ca> wrote in message( > news:3D8CEA4D.4387075A@videotron.ca...< >> The problem is that the urban SUVs have tractions systems >> that use differentials.  5 Hmmm.... must just be the cheap ones. Don't most have < some sort of "traction control" system that effectively does what you are asking ?   9 > And the solution is limited-slip differentials:  do you 5 > know for a fact that they aren't used, because it's ) > certainly my impression that they are.    8 Hmmmm..... some non-4WD vehicles come with that. My tiny< 91 BMW 318is (a rear-wheel drive vehicle) has a limited-slip7 differential and does quite well on bad roads (assuming < appropriate tires for the conditions) in New England in the 
 winter....   -Andy- ---    ------------------------------  # Date: Sun, 22 Sep 2002 03:52:19 GMT * From: "Bill Todd" <billtodd@metrocast.net>! Subject: Re: Marketing suggestion B Message-ID: <T7bj9.190738$z91.8141132@bin3.nnrp.aus1.giganews.com>  0 "-Andy-" <acs@fcgnet.works.net> wrote in message8 news:Xns9290E4E4F2480acsfcgnetworksnet@216.166.71.232...: > "Bill Todd" <billtodd@metrocast.net> enlightened us with> > news:8l6j9.187426$z91.7889454@bin3.nnrp.aus1.giganews.com on > 21 Sep 2002: >  > > > > > "JF Mezei" <jfmezei.spamnot@videotron.ca> wrote in message* > > news:3D8CEA4D.4387075A@videotron.ca...> > >> The problem is that the urban SUVs have tractions systems > >> that use differentials. > 7 > Hmmm.... must just be the cheap ones. Don't most have > > some sort of "traction control" system that effectively does > what you are asking ?  > ; > > And the solution is limited-slip differentials:  do you 7 > > know for a fact that they aren't used, because it's * > > certainly my impression that they are. > : > Hmmmm..... some non-4WD vehicles come with that. My tiny> > 91 BMW 318is (a rear-wheel drive vehicle) has a limited-slip9 > differential and does quite well on bad roads (assuming = > appropriate tires for the conditions) in New England in the  > winter....  @ Actually, technology may by now have supplanted the limited-slipI differential (a purely mechanical approach which has existed for at least J the last 40 years:  it was popular with muscle cars in the '60s as a meansK of avoiding asymmetrical tire-slippage while drag-racing), at least in some L installations:  if some kind of anti-lock braking system is already present,K it's probably possible to use its sensors and intelligence to apply braking : force to individual wheels that are sensed to be spinning.   - bill   ------------------------------  # Date: Sun, 22 Sep 2002 04:15:14 GMT * From: "Bill Todd" <billtodd@metrocast.net>! Subject: Re: Marketing suggestion B Message-ID: <mtbj9.195891$5r1.8269170@bin5.nnrp.aus1.giganews.com>  , Just to wander a bit farther from the topic:  . "John Smith" <a@nonymous.com> wrote in message@ news:jq9j9.15860$q41.117@news02.bloor.is.net.cable.rogers.com...   ...   G > Corvair was just a mid-engine car with a low polar moment of inertia.    Not really (I owned one).     Engine C > was ahead of the axle, so it qualifies for the mid-engine status.   I No, most (possibly all - I can't recall exactly) of the engine was behind % the rear axle, very much like the VW.     TheJ > problem it suffered from is the same problem that Porsche 911's suffered" > from up until about 20 years ago   Not exactly.  I The most significant problem the Corvair had was not its engine placement J but its 'swing axle' configuration, which was identical to similar-vintageK VWs (but not to the 911, though the 356 may have had them).  Coupled with a L relatively high pivot-point for the axles (again similar to the VW; MercedesE also used swing axles, but with a much lower pivot-point that greatly L reduced their influence on handling), this severely aggravated the oversteerH that it already tended to due to its rear weight bias:  the outside rearK tire would start to 'tuck under' during hard cornering (and the inside rear J tire somewhat as well), reducing its traction noticeably and significantlyC increasing the possibility of a spin (or under sufficiently adverse & conditions possibly even a roll-over).  K However, GM changed the swing-axle design to a double-U-joint configuration C in 1964 (and changed the suspension geometry as well to promote the I understeer characteristics Americans were used to) - long before (IIRC at K least a couple of years) Nader's book went to the printer.  So the problems G Nader described (though very similar to those affecting VWs back around L 1960 - I forget exactly when they changed) no longer existed well before the2 book came out, but it killed the car all the same.  J (Porsche did similar things - a good deal more than 20 years ago - to makeK the 911 a bit more 'civilized' than the earlier Porsches and Karmann-Ghias, F but had the additional problem that the engine was far more capable of8 breaking the rear wheels loose at inconvenient moments.)  < My Corvair was a 1967 model, and I adored it.  With a bit ofJ engine-massaging, over-sized Michelin tires, and Koni shocks all around itI really did handle quite a bit like a 911 (one of which I occasionally had & the pleasure to drive while rallying).   - bill   ------------------------------    Date: 21 Sep 2002 22:02:15 -0600+ From: young_r@encompasserve.org (Rob Young) , Subject: New Miss America - Miss Ill-a-Noise3 Message-ID: <IhK2kSKZvGQ4@eisner.encompasserve.org>    	Yep...   6 	With much shouting and fanfare... the announcer says:  ( 	"the new Miss America Miss Ill-a-noise"   	I rest my case.   				Rob    ------------------------------  # Date: Sat, 21 Sep 2002 19:11:58 GMT 1 From: "David J. Dachtera" <djesys.nospam@fsi.net>  Subject: Re: OpenVMS SIG' Message-ID: <3D8CCA1C.6C04F9AE@fsi.net>    Sue Skonetski wrote: >  > Brian, > E > If you go to the url in the orignial post you will see that you can G > register for the SIG and there is also something like a notes file as . > well (I forgot the name of what its called). > W > system@SendSpamHere.ORG wrote in message news:<00A14388.39157DEE@SendSpamHere.ORG>... k > > In article <amdef2$675$2@skates.gsfc.nasa.gov>, koehler@bessta.gsfc.nasa.aspm.gov (Bob Koehler) writes:  > > > U > > >In article <00A14370.D82AE600@SendSpamHere.ORG>, system@SendSpamHere.ORG writes: h > > >>In article <amcqri$hn1$1@web1.cup.hp.com>, "Sue Skonetski" <susan.skonetski@hp.nospam.com> writes: > > >>>Dear Newsgroup, > > >>> P > > >>>Just so you know Encompass has restarted the OpenVMS SIG.  Dave Cathey is > > >>>the SIG Chair.  > > >>> 3 > > >>> To join the VMS SIG you need to register on ; > > >>>http://www.encompassus.org/community/sigs/index.html  > > >>R > > >>Is joining the mailing list akin to registering to be an OpenVMS SIG member? > > > K > > >   I think I joined the SIG first and the mailing list came out later. K > > >   In this case the sign up claims "for members in good standing", but A > > >   in at least one case that includes non-paying associates.  > > H > > Well, the question is, where do I sign up for the SIG?  The web pageI > > mentions as a link to join the mailing list but there's nothing abouto! > > simply becoming a SIG member.p  . Seconding Brian's post - don't find it either.   Help (URL) please?   -- d David J. Dachterae dba DJE Systemso http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/o   ------------------------------  # Date: Sat, 21 Sep 2002 19:09:02 GMT81 From: "David J. Dachtera" <djesys.nospam@fsi.net>r9 Subject: Re: Oracle RDB 8.0 = Oracle RDB for Itanium ????W& Message-ID: <3D8CC96B.644EC4D@fsi.net>   Michael Unger wrote: > 7 > "norm lastovica" <norman.lastovica@oracle.com> wrote:e > > > > According to what you can read and learn on the Oracle Rdb; > > web set (www.oracle.com/rdb) and so on, Rdb 8.0 was theS9 > > name for the (now) unsupported Oracle Rdb NT release.  > >t	 > > [...]i >   > Re: "reading and learning" ... > A > The result is simply a plain white page, nothing else. CheckingeE > www.oracle.com there is a tiny remark at the top of the page sayingi4 > "This page requires a JavaScript-enabled browser." > A > I suppose JavaScript isn't liked very much in the VMS communityrD > regardless of the system (VMS, Winwoes or something else) which is > used to view that page.M  G I rather expected Larry K. to note that javascript is avoided by anyoneh who is security conscious.   -- h David J. Dachterae dba DJE SystemsM http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/+   ------------------------------  # Date: Sat, 21 Sep 2002 18:25:52 GMTa6 From: Jeffrey Coffield <jeffrey@digitalsynergyinc.com>< Subject: Re: Storing username/password for external services4 Message-ID: <3D8C4576.1020408@digitalsynergyinc.com>   JF Mezei wrote: L > I currently have a POP client that interfaces to Message router. It has anO > indexed file which contains information on my various POP accounts on various N > servers as well as the message router destination for mail picked up there.  >  > H > However, I know that in the long term, I will have to migrate this off/ > message-router/all-in-1 probably to VMS mail.J > M > In doing so, I would like to make something that can be used by others. AndAO > for this, I would want some sort of account information/password storage thatn > would be acceptable. > M > If I kept a single file to store account information for all users who wantfP > their pop mail transfered automativally from remote systems to the VMS system,P > would this be totally unacceptable ? If the file were owned by the UIC runningP > the POP software with no access by anyone else, except one utility that allowsO > a user to register/modify his profile (such a utility woudl ensure a user can 6 > only see his own record), would that be acceptable ? > O > In a central file concept, would some simple 2 way encryption of the passwordrN > be required ? Or is keeping the file secure sufficient without requiring its > contents to be encrypted ? > N > Or would the only solution to have the central file contain only the list ofM > users, and then have each user have some control file in his SYS$%LOGIN andn, > that file would contain the account info ? >  > Any other paradigm ?  C We use the SYSUAF file for passwords. You can read the appropriate nH fields from a user account using SYS$GETUAI then call SYS$HASH_PASSWORD C to encrypt the entered password using the method and seed from the -C account. If the encrypted password matches what is stored, they aree authenticated.   ------------------------------  % Date: Sat, 21 Sep 2002 17:37:37 -0400a- From: JF Mezei <jfmezei.spamnot@videotron.ca>/< Subject: Re: Storing username/password for external services, Message-ID: <3D8CE694.ECDCB80E@videotron.ca>   Jeffrey Coffield wrote:rD > We use the SYSUAF file for passwords. You can read the appropriateI > fields from a user account using SYS$GETUAI then call SYS$HASH_PASSWORDdD > to encrypt the entered password using the method and seed from theE > account. If the encrypted password matches what is stored, they are  > authenticated.  L The problem is that it is impossible to "fetch" the password from SYSUAF andN then transmit it to the remote POP server. The software really needs access toN a password "in the clear" in order to send it to the remote POP server (eitherL in the clear, or with minimal encryption based on a string provided when the& remorte POP server identifies itself).   ------------------------------  # Date: Sat, 21 Sep 2002 19:33:18 GMTa1 From: "David J. Dachtera" <djesys.nospam@fsi.net>r9 Subject: Re: VMS future (oh not not another one of these)a' Message-ID: <3D8CCF18.5B49FCA2@fsi.net>E   Hans Vlems wrote:d >  > [snip] > >tI > > Pro-VMS: Some active, but relatively impotent efforts, though the IPFsF > > port would be more significant if IPF were a currently marketable,L > > volume-shippable, ready-for-prime-time product - it remains vaporware atB > > this writing by that definition. IA32 is virtually universallyG > > ubiquitous and would make by far a better target, regardless of the K > > technological challenges that represents. (Again, understand that I am, I > > AFAIK, the only person currently active in this newsgroup who holds aP/ > > working understanding of why that is true.)o > I > Interesting remark. Would you care to elaborate on that last statement?0  D My posts in favor of OpenVMS-IA32 are routinely met with a chorus ofF follow-ups talking about since there are so many variants of chipsets,A third-party controllers, graphics adapters and other incompatibleeG hardware, not to mention the addressing scheme and other limitations ofrD that architecture, that the chances of any of those millions of IA32H boxes being able to run any OVMS-IA32 that might be developed are indeedF so small as to make it financially infeasible for the OpenVMS group toC justify to the corporate bean-counters. (I heartily disagree, but Ih digress...).  D Personally, I'd like to know how that differs from any scenario thatH might evolve as IA64 boxes start to replace IA32s (if, indeed, that ever does actually happen).    F ...but I already get enough heat from the OVMS Engineers and the Larry> Kilgallen types, and almost everyone who is anyone has alreadyC blacklisted and/or killfiled me since my reply-to address is highlya/ predictable (djesys.nospam@any-domain.any-tld).h   -- i David J. Dachterad dba DJE Systemso http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/1   ------------------------------  % Date: Sat, 21 Sep 2002 17:45:54 -0400y- From: JF Mezei <jfmezei.spamnot@videotron.ca>n9 Subject: Re: VMS future (oh not not another one of these)e, Message-ID: <3D8CE885.E9CB9485@videotron.ca>   "David J. Dachtera" wrote:F > My posts in favor of OpenVMS-IA32 are routinely met with a chorus ofH > follow-ups talking about since there are so many variants of chipsets,  L The other day, while cycling in traffic, I thought about you. PDP-11 used to/ be used a lot in manufacturing/process control.a  H I think that one reason to port VMS to the 8086  *today* would not be toK target to low end users, but perhaps to offer manufacturing/process control N solutions that are ultra reliable, yet affordable. One doesn't need a wildfireB to do the job that used to be done by a PDP-11 with 64k of memory.  H In other words, with Alpha now targetted at the very high end (and dyingM anyways), there needs to be some low end solution for those applications thatVP would benefit from VMS witout requiring large scalability. (eg: process control)   ------------------------------  # Date: Sun, 22 Sep 2002 01:48:24 GMT 1 From: "David J. Dachtera" <djesys.nospam@fsi.net>I9 Subject: Re: VMS future (oh not not another one of these)t' Message-ID: <3D8D2704.F27179DF@fsi.net>u   JF Mezei wrote:  >  > "David J. Dachtera" wrote:H > > My posts in favor of OpenVMS-IA32 are routinely met with a chorus ofJ > > follow-ups talking about since there are so many variants of chipsets, > N > The other day, while cycling in traffic, I thought about you. PDP-11 used to1 > be used a lot in manufacturing/process control., > J > I think that one reason to port VMS to the 8086  *today* would not be toM > target to low end users, but perhaps to offer manufacturing/process controltP > solutions that are ultra reliable, yet affordable. One doesn't need a wildfireD > to do the job that used to be done by a PDP-11 with 64k of memory.  E Agreed. My 386/387SX-16 running Red Hat is faster than a MicroVAX-II,N5 maybe even approaching the VAXstation-400/VLC, but...p  J > In other words, with Alpha now targetted at the very high end (and dyingO > anyways), there needs to be some low end solution for those applications that-R > would benefit from VMS witout requiring large scalability. (eg: process control)   *Heartfelt Sigh*   -- u David J. Dachterao dba DJE Systems: http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/t   ------------------------------  # Date: Sat, 21 Sep 2002 18:09:59 GMTe* From: "Bill Todd" <billtodd@metrocast.net> Subject: Re: VMS, C++ and flushaB Message-ID: <XB2j9.189060$AR1.7987762@bin2.nnrp.aus1.giganews.com>  3 "Jan-Erik Sderholm" <aaa@aaa.com> wrote in messagea! news:3D8C388D.322EE71D@aaa.com...  > Hi.s9 > The last couple of days, I'v looked into a problem withy< > a C++ program. I have now found a solution to the problem,- > but I still don't understand what happened.  >h > Scenario:a: > A C++ application works just OK on all systems, but one.; > The application is rather simple, reads on text file with = > 500 char blocks of data, moves the data around a little andn8 > writes a new textfil with 80 char blocks with a subset > of the data. > = > On this single system, the output file got currupted, there + > was a lot of <null>'s around in the file.e >e8 > After studying the output file, I found that the first6 > <null> character always was after 1024 characters in: > the output file. I thought about buffering and flushing,; > and tried to insert a flush() on the output stream in theo> > application after each 80 char records was written. And that > solved it. >:< > It looks like this system fills the current 80 char record; > with nulls up to a full record at each (automatic) flush. ? > Doing a flush when I know that the record is complete, solved> > it.r  F Are these RMS FIXED-format records, rather than a simple (undelimited)L byte-stream?  If so, it's possible that the interaction between the language@ run-time system's automatic flush activity (which may not know aF fixed-length record from a hole in the wall) and RMS (only in that oneK configuration, for some reason) is causing the behavior - e.g., possibly onfK the other systems automatic flush is occurring sufficiently less frequentlydB that you always flush the buffer because it's full before doing so automatically.   - bill   ------------------------------    Date: 21 Sep 2002 13:52:01 -0700: From: craig.berry@SignalTreeSolutions.com (Craig A. Berry) Subject: Re: VMS, C++ and flush = Message-ID: <7f15589f.0209211252.4bc2b497@posting.google.com>2   Jan-Erik Sderholm wrote:l > Scenario:s: > A C++ application works just OK on all systems, but one.; > The application is rather simple, reads on text file with1= > 500 char blocks of data, moves the data around a little andr8 > writes a new textfil with 80 char blocks with a subset > of the data. > = > On this single system, the output file got currupted, there0+ > was a lot of <null>'s around in the file.  > 8 > After studying the output file, I found that the first6 > <null> character always was after 1024 characters in; > the output file. I thought about buffering and flushing, o; > and tried to insert a flush() on the output stream in thei> > application after each 80 char records was written. And that > solved it. > < > It looks like this system fills the current 80 char record; > with nulls up to a full record at each (automatic) flush.e? > Doing a flush when I know that the record is complete, solved8 > it.  > 7 > Now, how come that this application works fine on allS- > other systems without the explicit flush()..  F You don't fully specify the record format or attributes of the file inC question but I infer that it is a record-oriented file with a "fullhH record" size of 1024 bytes. Using C RTL I/O on record-oriented files hasB various limitations, such as that fseek() can only position you onD record boundaries, and seeking beyond the end of the file causes theF intervening space to be zero-filled upon the next write. That *may* beF what is happening. IIRC, fflush() resets the current file position, soH that could be why it solves your problem. It's probably not relevant butB note that fflush() only flushes the C RTL buffers to RMS; you need% fsync() to flush all the way to disk.e  E Why it's different on different systems is still a mystery. Things tos check would include:  E Determine whether the program's file positioning actions (if any) arey8 performed differently according to differing input data.  F Compare the results of DIRECTORY/FULL on the output files. Even if theE identical code creates the file on each system, file attributes couldn4 default to those of prior versions of the same file.  A Compare the results of SHOW RMS_DEFAULT on the different systems.l  F Is the system in question by any chance the only one with XFC enabled?    ? These are all just guesses but may narrow things down somewhat.m   ------------------------------  % Date: Sun, 22 Sep 2002 00:01:03 +0200o9 From: Jan-Erik =?iso-8859-1?Q?S=F6derholm?= <aaa@aaa.com>o Subject: Re: VMS, C++ and flushr' Message-ID: <3D8CEC1F.D235A7C2@aaa.com>d  	 Hi again.h1 I'll try to answer the followup-questions posted.   % No, XFC is not enabled on this systemt  5 The output file is opened with fixed 80 char records.e  7 When I "flush" after each 80 char record is *complete*,-4 and exactly 80 chars are written by the application, all is fine.  8 When I tried to flush after each field in the record was6 written, I got a file with each field on an own record> with the rest of the record filled with <null> up to 80 chars.  8 Note that on another system where I compiled and linked,; the output file was OK (and, I expect, on all other systemsa0 also where the application worked since before).  6 It looks as something is going on between the C++ RTL, RMS and the file itself.  5 B.t.w, we have checked the C++ RTL versions, and theye are OK.t   Thanks so far !m   Jan-Erik Sderholm   ------------------------------  # Date: Sat, 21 Sep 2002 22:51:23 GMTg* From: "Bill Todd" <billtodd@metrocast.net> Subject: Re: VMS, C++ and flushnB Message-ID: <LJ6j9.192379$5r1.8005153@bin5.nnrp.aus1.giganews.com>  3 "Jan-Erik Sderholm" <aaa@aaa.com> wrote in messageh! news:3D8CEC1F.D235A7C2@aaa.com...t > Hi again.c3 > I'll try to answer the followup-questions posted.r >c' > No, XFC is not enabled on this systemg >t7 > The output file is opened with fixed 80 char records.  >e9 > When I "flush" after each 80 char record is *complete*,-6 > and exactly 80 chars are written by the application, > all is fine.  G From what Craig just said, this merely transfers the RTL buffers to RMS-L buffers (which if true sounds like a bug:  isn't flush supposed to flush allL the way to disk?).  But (true or not) as long as RMS never sees anything but* complete records that clearly should work.  I It had been my impression that the RTL bypassed RMS record I/O and either C used RMS block I/O or (gasp!) QIOs, but I don't know that for sure.e   > : > When I tried to flush after each field in the record was8 > written, I got a file with each field on an own record@ > with the rest of the record filled with <null> up to 80 chars.  I You don't mention whether this was on *all* systems or just on the systemi: giving you problems (which might be illuminating to know).  D In any event, it's pretty clearly tied up with the file's RMS recordH definition:  either the RTL is adding padding that it believes it has toK because of the FIXED record format (I don't know whether to call this a bugsG in the current context or not:  it's hard to tell what would constituteaL correct behavior in this mish-mash of C-style byte-oriented extension on theK one hand and RMS fixed-length data on the other, but my initial inclinationlK would be to have the RTL flush only up to the last complete RMS record), ornG it's using record-oriented access to pass the records to RMS and RMS isaB obliging you by null-padding the short records (RMS's only obviousG alternatives being to pad them with whatever garbage is in its buffer - J seldom a good idea - or to fail the operation due to invalid record size).   >r: > Note that on another system where I compiled and linked,= > the output file was OK (and, I expect, on all other systemsr2 > also where the application worked since before).  K If you're saying above that flushing after each field did *not* cause extrasC nulls to be added on other systems, as I suggested above that seems H significant.  Have you verified that the output file is RMS FIXED-format! 80-byte records on *all* systems?-   - bill   ------------------------------  % Date: Sun, 22 Sep 2002 01:27:40 +0200.9 From: Jan-Erik =?iso-8859-1?Q?S=F6derholm?= <aaa@aaa.com>9 Subject: Re: VMS, C++ and flush1& Message-ID: <3D8D006C.E2E9676@aaa.com>   Bill Todd wrote: > 5 > "Jan-Erik Sderholm" <aaa@aaa.com> wrote in message # > news:3D8CEC1F.D235A7C2@aaa.com...f > >V< > > When I tried to flush after each field in the record was: > > written, I got a file with each field on an own recordB > > with the rest of the record filled with <null> up to 80 chars. > K > You don't mention whether this was on *all* systems or just on the systems< > giving you problems (which might be illuminating to know). >    Sorry...D It was on the single trouble system. On my dev system (and maybe all others)tE the records was correct 80 char records in all three cases (no flush,t9 flush after each field, flush after each 80 char record).    > >e< > > Note that on another system where I compiled and linked,? > > the output file was OK (and, I expect, on all other systemso4 > > also where the application worked since before). > M > If you're saying above that flushing after each field did *not* cause extrakE > nulls to be added on other systems, as I suggested above that seems,J > significant.  Have you verified that the output file is RMS FIXED-format# > 80-byte records on *all* systems?    All output file are :f2 "Record format:      Fixed length 80 byte records" according to DIR/FULL.  @ Both the file without excplicit flush (with the nulls after 1024C characters), the one with one field in each record and nulls filledg= to 80 chars, and the OK one with a flush at axactly 80 chars. D This is on the "trouble" system, I don't have access to the other(s)? right now, but as far as I remember, this was the case on thosee
 systems also..   Jan-Erik Sderholm.g   >  > - bill   ------------------------------  # Date: Sun, 22 Sep 2002 00:05:08 GMTe* From: "Bill Todd" <billtodd@metrocast.net> Subject: Re: VMS, C++ and flushhA Message-ID: <UO7j9.95032$216.3726023@bin4.nnrp.aus1.giganews.com>   3 "Jan-Erik Sderholm" <aaa@aaa.com> wrote in messageu  news:3D8D006C.E2E9676@aaa.com... > Bill Todd wrote: > >a7 > > "Jan-Erik Sderholm" <aaa@aaa.com> wrote in messagen% > > news:3D8CEC1F.D235A7C2@aaa.com...e > > > > > > > When I tried to flush after each field in the record was< > > > written, I got a file with each field on an own recordD > > > with the rest of the record filled with <null> up to 80 chars. > >aF > > You don't mention whether this was on *all* systems or just on the system> > > giving you problems (which might be illuminating to know). > >- >-
 > Sorry...F > It was on the single trouble system. On my dev system (and maybe all	 > others)sG > the records was correct 80 char records in all three cases (no flush,.; > flush after each field, flush after each 80 char record).p  L Weird.  Very.  Unless the RMS record formats in the output files differ (see below).    >: > > >c> > > > Note that on another system where I compiled and linked,A > > > the output file was OK (and, I expect, on all other systems46 > > > also where the application worked since before). > >nI > > If you're saying above that flushing after each field did *not* cause  extra G > > nulls to be added on other systems, as I suggested above that seemsaL > > significant.  Have you verified that the output file is RMS FIXED-format% > > 80-byte records on *all* systems?  >- > All output file are :-4 > "Record format:      Fixed length 80 byte records" > according to DIR/FULL. >cB > Both the file without excplicit flush (with the nulls after 1024E > characters), the one with one field in each record and nulls filledt? > to 80 chars, and the OK one with a flush at axactly 80 chars.sF > This is on the "trouble" system, I don't have access to the other(s)A > right now, but as far as I remember, this was the case on thosee > systems also."  J At this point, verifying that would be desirable.  If your recollection is@ correct that the output files on the other systems had identicalL characteristics, then check the set-up variables for the run-time systems toB see if they're identical too (e.g., something like a difference inK 'write-through' behavior could conceivably lead to the behavior differences-@ you're seeing, though if so that would still be a possible bug).  L Sorry I can't be more insightful, but it's been close to two decades since IH worked on RMS internals and I never knew that much about exactly how theG language run-time systems worked.  Perhaps Hein will come to the rescueuI (putting RMS in the subject line might have helped attract his attention,nK unless those brilliant leaders at The New HP caused him to get down-sized).t   - bill   ------------------------------  % Date: Sat, 21 Sep 2002 20:39:42 -0400n- From: JF Mezei <jfmezei.spamnot@videotron.ca>i Subject: Re: VMS, C++ and flushn, Message-ID: <3D8D114D.D83C0F7C@videotron.ca>   Jan-Erik Sderholm wrote:o: > When I tried to flush after each field in the record was8 > written, I got a file with each field on an own record@ > with the rest of the record filled with <null> up to 80 chars.  G Which is not unexpected. With record access mode, a "flush" tells it to M complete outstanding writes, and if you have so far written 20 bytes, it will>M do a write of 20 of your bytes to a fixed length file, which means padding ofa 60 bytes to make the record.    8 The next field you write starts a new record and so on. E Is there a reason you are using flush midway while writing a record ?u  5 when you open the file, do you supply any options ?  m2 (eg : myfile = fopen("myfile.dat","w","cxt=rec") ?   do you use fopen, or "open" ?h  N More importantly, does your program create the file or use an existing file to which it appends ?  : > Note that on another system where I compiled and linked,= > the output file was OK (and, I expect, on all other systemsc  K Not sure about C++, but could you take a look at "expanded" listings (whereoJ #define directives are replaced with their equivalents in your code) ? You> might find different meanings of "flush" on different systems.  M Remember that C++ installs on top of C, and vice-versa. So while the compiler I itself may all be the same version, the support files might be different.    ------------------------------  % Date: Sun, 22 Sep 2002 02:57:46 +0200 9 From: Jan-Erik =?iso-8859-1?Q?S=F6derholm?= <aaa@aaa.com>0 Subject: Re: VMS, C++ and flushe' Message-ID: <3D8D158A.C4CA9153@aaa.com>s   JF Mezei wrote:  > 9 > The next field you write starts a new record and so on.nG > Is there a reason you are using flush midway while writing a record ?   C No, that was a "misstake", but anyway, on other systems the recordse% was not padded and written like that.H   > 5 > when you open the file, do you supply any options ?i4 > (eg : myfile = fopen("myfile.dat","w","cxt=rec") ? >  > do you use fopen, or "open" ?e  C Well, it's some kond of C++ syntax, can't remember now, and I don'ta# hav access to the source right now.n   > P > More importantly, does your program create the file or use an existing file to > which it appends ?   creates.   ------------------------------  # Date: Sun, 22 Sep 2002 01:44:52 GMTs# From: "John Smith" <a@nonymous.com>r1 Subject: Will they offer WebLogic on VMS as well?eH Message-ID: <og9j9.15724$q41.11533@news02.bloor.is.net.cable.rogers.com>  + BEA and Hewlett-Packard Team Up Against IBMg Fri Sep 20, 5:00 PM ET By Ilaina Jonasi  K NEW YORK (Reuters) - Hewlett-Packard Co. on Friday said it would bundle BEAaI Systems Inc. flagship software product WebLogic on its best-known line ofiL high-end computers, stepping up their alliance to do battle against a common	 foe, IBM.r  L Hewlett-Packard, which is tied with International Business Machines Corp. asL the No. 1 maker of hardware, will ship a copy of the most up-to-date version" of WebLogic with its UX11i server.  K "With this bundle, what we provide is a unique competitive offering that HP0G and BEA together can bring to the market relative to our natural commonrB enemies here," said Gamiel Gran, BEA's vice president of strategic
 alliances.  D "Clearly, we're eager to present our message relative to IBM and IBMI WebSphere, and similarly HP has an interest to erode IBM market share, as  well."  G The announcement is the most recent between the two companies that have7L periodically dangled hints of their relationship ever since Peter Blackmore,I HP's head of high-end computers, said that the company would "retire" its' middleware business.  H Subsequent to the merger of HP and Compaq, HP shuttered its money-losingH Bluestone middleware division, which had competed poorly against BEA and, IBM's WebSphere software application server.  B BEA has retained a slim lead over IBM in the market of applicationI servers -- software that programmers use as a base for their application. E Application servers perform important but routine tasks, such as loadm7 balancing, for networks of servers that run businesses.d  K However, the alliance is not exclusive and industry sources said that HP is / in talks with IBM to offer IBM's WebSphere too.   J In the new agreement BEA has named HP's OpenView, which monitors networks,E computers and software application performance to pinpoint sources ofnG bottlenecks. WebLogic has been adjusted for optimum performance on HP's6 servers.  L "BEA is rally just trying to find a way to regain momentum that they clearly3 lost to IBM," Steve Eisenstadt, IBM spokesman said.1  G About 60 percent of the programs running on BEA WebLogic operate on Sunn Microsystems Inc.3  L "This is a clear indicator that we are attempting to diversify and grow into> markets that we have not been as strong in before," Gran said.  F In June, the companies announced that HP would put its giant sales andJ services force behind BEA's software products. That alliance created teamsL of consultants from HP's services arm, one of the world's largest, specially trained in BEA's products.  D "Our intention is to bundle BEA's WebLogic server with all of our HPH platforms in the future," Don Jenkins, Hewlett-Packard vice president ofJ marketing for operating environment software said. Those platforms include? Windows 2000 ( news - web sites) and Linux ( news - web sites).a  F The new agreement will permit customers to use WebLogic during a free,+ six-month trial period, the companies said.c   ------------------------------  % Date: Sat, 21 Sep 2002 22:39:15 -0400s- From: JF Mezei <jfmezei.spamnot@videotron.ca>s5 Subject: Re: Will they offer WebLogic on VMS as well?i, Message-ID: <3D8D2D4A.E8AFB1A3@videotron.ca>   John Smith wrote:rN > periodically dangled hints of their relationship ever since Peter Blackmore,K > HP's head of high-end computers, said that the company would "retire" its  > middleware business.  M That is a scary statement coming from Blackmore, especially when you considerBM they have already retired so much VMS software such as ALL-IN-1.  Mailbus-400-F and the X.500 directory server is probably next on the chopping block.  I The morale of the story is not to buy any software from Digital/Compaq/HPm! because it might be killed soon. w  F > "Our intention is to bundle BEA's WebLogic server with all of our HPF > platforms in the future," Don Jenkins, <...> Those platforms includeA > Windows 2000 ( news - web sites) and Linux ( news - web sites).m  C That Jenkins guy needs a bit of education about HP's product lines.o   ------------------------------  % Date: Sat, 21 Sep 2002 17:27:46 -0400 , From: "David Turner" <dbturner@islandco.com> Subject: XP1000's going cheap!C Message-ID: <3g5j9.142049$AY5.58777582@e3500-atl1.usenetserver.com>t  
 XP1000 667Mhza 512MBw 9GB UW SCSIg 10/100 Ethernetu   Only $2695 for COV   ------------------------------  # Date: Sun, 22 Sep 2002 01:52:51 GMT11 From: "David J. Dachtera" <djesys.nospam@fsi.net> " Subject: Re: XP1000's going cheap!' Message-ID: <3D8D2812.D16A1710@fsi.net>d   David Turner wrote:e >  > XP1000 667Mhzw > 512MBp
 > 9GB UW SCSIl > 10/100 Ethernetn >  > Only $2695 for COV  E What would be a fair resale value for that box running Alpha Linux orrA FreeBSD-Alpha? Would the potential margin make it worth my while?'   --   David J. Dachtera  dba DJE Systems  http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/y   ------------------------------  % Date: Sat, 21 Sep 2002 19:20:40 -0400- From: Everhart <ge@gce.com> ; Subject: Re: [OT] Could Philip Zimmermann work for the CIA?1+ Message-ID: <amiusb$eji$1@bob.news.rcn.net>e   Leo Demers wrote:nG > OpenVMS engineering is investigating a port of GnuPG 1.0.7 to be made  > availablee9 > like Stunnel is under GPL as Open Source Tools project.m > I > Given recent news on GnuPG some folks may have second thoughts about it9* > http://online.securityfocus.com/news/576J > Right now we are proceeding but I'm not sure how many folks will want it > when it's done > C > It'll be available from http://www.openvms.compaq.com/opensource/1 > -- > Leo Demers" > OpenVMS Security Product Manager > Leo_dot_Demers_at_HP_dot_COM > @ > "Franz-Josef Fornefeld" <jo.fornefeld@gmx.de> wrote in message$ > news:amfbi5.318.1@jo.dyndns.org... >  >>Nic Clews wrote: >> >> >>>Franz-Josef Fornefeld wrote:  >>>n >>>mD >>>>Didn't You realize that many (and most serious) people still useH >>>>V2.6.1i? This is the last version without backdoors (as rumors say). >>>>? >>>>GnuPG will be the successor (whenever it will be bug free).- >>>- >>>We're sort of off topic,0 >>/ >>Sure. comp.security.pgp.* would be better ...g >> >>2 >>>however of interest because OpenVMS should have2 >>>representation as a platform for said software. >>> I >>>Where does 2.6.3i fit into the 'backdoor' theory? I'm interested in iteK >>>because it has 2048 bit support and is probably the latest available for- >>>OpenVMS.  >>> I >>>2.6.2 is the 1024 bit capable OpenVMS compilable version, should 2.6.1i >>>be considered?: >>I >>Well, maybe I'm wrong with this version and V2.6.3i would be the better I >>one. I took this from brain and may have missed one or two releases ;-)t >> >>+ >>>My (perhaps limited) understanding [...]o >>2 >>Yes, that's my problem too. But I found the FAQ:C >>http://www.uk.pgp.net/pgpnet/pgp-faq/. Maybe this is what you area >>interested in. >> >>4 >>>Is anyone working for OpenVMS interests in GnuPG? >>I >>I found no reference on VMS on http://www.gnupg.org/. So maybe there is 	 >>nobody.  >  >  > I GPG is a whole other kettle of fish from PGP in that regard. It has never > been the captive of a company, never had backdoors added, etc.K On the whole it is IMO a better tool to be using. I don't believe Zimmerman>F did any nefarious things with it, but PGP has been out of Phil's handsF for a while now. GPG on the other hand is pure open source; if someone= thinks there is a backdoor in it, he is free to exhibit same./     Good for you, Leo!   Glenn Everhart   ------------------------------   End of INFO-VAX 2002.523 ************************k I joined the SIG first and the mailing list came out later. K > > >   In this case the sign up claims "for members in good standing", but A > > >   in at least one case that includes non-paying associates.  > > H > > Well, the question is, where do I sign up for the SIG?  The web pa