1 INFO-VAX	Fri, 12 May 2006	Volume 2006 : Issue 262       Contents: Re: Compressing backup file  Hobbyist on Itanium  Re: Hobbyist on Itanium 9 linspire newsletter - interesting read, and maybe topical  Results of my straw poll.  Re: Results of my straw poll.  Re: Results of my straw poll.  Re: Results of my straw poll.  Re: SGI files for chapter 11 Re: SGI files for chapter 11 Re: SGI files for chapter 11 Re: SGI files for chapter 11 Re: SGI files for chapter 11 Re: SGI files for chapter 11 Re: SGI files for chapter 11 Re: SGI files for chapter 11P Re: This post will self-destruct in 10secs (Was Re: X windows  vulnerability)  v# Re: Updated VMS information May 7th # Re: Updated VMS information May 7th # Re: Updated VMS information May 7th & VMS indexed files - how did they work?* Re: VMS indexed files - how did they work?* Re: VMS indexed files - how did they work?* Re: VMS indexed files - how did they work?* Re: VMS indexed files - how did they work?* Re: VMS indexed files - how did they work?* Re: VMS indexed files - how did they work?* Re: VMS indexed files - how did they work?* Re: VMS indexed files - how did they work?* Re: VMS indexed files - how did they work?* Re: VMS indexed files - how did they work?* Re: VMS indexed files - how did they work?F Re: We need a recommendation of good VMS backup (non-library) softwareF Re: We need a recommendation of good VMS backup (non-library) softwareB Re: XtRemoveTimeOut() with value of 0 accvios under 7.3-2, but notB Re: XtRemoveTimeOut() with value of 0 accvios under 7.3-2, but notB Re: XtRemoveTimeOut() with value of 0 accvios under 7.3-2, but notB Re: XtRemoveTimeOut() with value of 0 accvios under 7.3-2, but notB Re: XtRemoveTimeOut() with value of 0 accvios under 7.3-2, but notB Re: XtRemoveTimeOut() with value of 0 accvios under 7.3-2, but notN Re: XtRemoveTimeOut() with value of 0 accvios under 7.3-2, but not under 7.1-2  F ----------------------------------------------------------------------  # Date: Thu, 11 May 2006 19:36:45 GMT , From: Hoff Hoffman <hoff-remove-this@hp.com>$ Subject: Re: Compressing backup file0 Message-ID: <hnM8g.386$uM4.257@news.cpqcorp.net>   Steven M. Schweda wrote:% > From: "Tom Linden" <tom@kednos.com>   7 >> Is DVD supported as output device for backup on 8.3?  >  >    Directly?  I doubt it.   D    The CD-R/RW and DVD+R/RW recording capabilities that are present E within V8.3 require the generation of a master, or the use of a file  H containing a disk image.  The recording capabilities then replicate the E master onto the target medium, akin to what CDRECORD and a gazillion  , other recording tools traditionally provide.  I    You can use BACKUP or COPY or other tools to populate the master, and  < you can then generate the optical medium from it, of course.  B    Folks that are involved with the V8.3 EFT or that acquired the C associated V8.3 SDK do have access to these recording capabilities.   G    Given what is involved with establishing and managing the recording  H process, I'd generally not tend to expect to see recording capabilities  layered into various utilities.   F    It is certainly reasonably possible to provide driver-level direct E write access to DVD+RW (media lifetimes are arguably too short to be  F useful) or for DVD-RAM media (far higher read-write cycles), but that F driver capability is not available within V8.3.   That approach would ? allow BACKUP or other tools to write directly to optical media.   F    If you are interested in direct-recording, do let the folks at the F customer support center or your formal HP contact know.  (I certainly C know how to implement the direct-recording capabilities within the  * current SYS$DQDRIVER device driver stack.)   ------------------------------    Date: 11 May 2006 11:40:56 -0700  From: banditstan2000@yahoo.co.uk Subject: Hobbyist on ItaniumC Message-ID: <1147372856.157532.316340@u72g2000cwu.googlegroups.com>    Hi, A I've seen this covered in a few places, but I can't seem to get a C definitive answer, and no reply so far from Montagar. Basically, is G OpenVMS available under the Hobbyist license on Itanium? If it is, what E organisation do I need to join before I can get a license and a media  kit?  F The information here seems a bit out of date and only mentions VAX and Alpha:  + http://www.montagar.com/hobbyist/mount.html    Any info appreciated.    Garry    ------------------------------  % Date: Thu, 11 May 2006 21:33:51 -0500 6 From: "David J. Dachtera" <djesys.no@spam.comcast.net>  Subject: Re: Hobbyist on Itanium0 Message-ID: <4463F40F.E92BF3BF@spam.comcast.net>  ! banditstan2000@yahoo.co.uk wrote:  >  > Hi, C > I've seen this covered in a few places, but I can't seem to get a E > definitive answer, and no reply so far from Montagar. Basically, is I > OpenVMS available under the Hobbyist license on Itanium? If it is, what G > organisation do I need to join before I can get a license and a media  > kit? > H > The information here seems a bit out of date and only mentions VAX and > Alpha: > - > http://www.montagar.com/hobbyist/mount.html  >  > Any info appreciated.   B I would direct that query to Sue Skonetski. She posts here, so youF should be able to find her address without me reposting it (increasing her likelihood of spam).   --   David J Dachtera dba DJE Systems  http://www.djesys.com/  & Unofficial OpenVMS Marketing Home Page! http://www.djesys.com/vms/market/   ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/   " Unofficial OpenVMS-IA32 Home Page: http://www.djesys.com/vms/ia32/   ) Unofficial OpenVMS Hobbyist Support Page: " http://www.djesys.com/vms/support/   ------------------------------  % Date: Thu, 11 May 2006 20:13:53 -0400 * From: "d b turner" <dbturner@islandco.com>B Subject: linspire newsletter - interesting read, and maybe topical9 Message-ID: <7rQ8g.40224$Kn4.6424@bignews2.bellsouth.net>   / Thought this might be of interest to someone...    DT     Freespire: The "Hybrid" OS   by Kevin Carmony   May 11, 2006      M In the last Linspire Letter, I shared with you our recent announcement about  J Freespire, a no-cost, community version of the Linspire operating system. L Freespire is designed more for developers and the traditional Linux devotee F than for consumers, who are Linspire's sweet spot. (If you missed the M Desktop Linux Summit, you can watch the video of my presentation here, along  L with all keynote addresses from the show.) Since this announcement, we have I had thousands join the Freespire community, participate on the Freespire  J forums, join the Freespire mailing lists, share in refining the Freespire J Vision, and influence the development of the Freespire leadership boards. M We're well on our way to the first and most important part for any community  C project, and that's organizing that community. We should be nicely  G functioning by August when the first beta version of Freespire will be   available for download. I As we've been organizing the Freespire project and community, detractors  D have also come forth, expressing their skepticism or disapproval of M Freespire. I fully expected this, for if there is one thing you can count on  M when anything happens in the open source community, it is that there will be  L disagreement, dissension and debate. I certainly didn't expect Freespire to K be an exception to this. Even though the vast majority of the response has  G been positive and supportive, I wanted to talk about why Freespire has  I created discomfort for some, and explain in more depth why the Freespire  I project is doing things different than most other community-driven Linux   distributions.  L The disagreement doesn't come from the fact that Freespire will be free, or M that it will focus on ease of use and providing a solid development platform  M for desktop Linux and Linspire. There seems to be pretty much agreement that  B these things are all good for open source. For the most part, the K disagreement stems from one area - the fact that Freespire will be offered  E in two versions. One version will be a traditional open source Linux  F distribution, void of any proprietary code and only using open source M software. The other version, however, is what makes Freespire different from  K other Linux community distributions - it will include proprietary software   in the core distribution.   M As I explained last week, we offer this second version so that Freespire can  L "out of the box" support things such as MP3, DVD, Windows Media, QuickTime, J Java, Flash, Real, enhanced fonts, and better hardware support for things K like 3-D ATI and nVidia graphic cards, modems, Wi-Fi, and so on. Freespire  J includes proprietary software when viable open source alternatives aren't 
 available.  K Some see the mixing of proprietary software with open source software as a  G bad thing, and something that could hamper the advancement of the open  F source movement. I can understand and respect that view, but there is J meaning to our madness. Here are some of the reasons I think the time has  come for this approach:             M 1. Freedom of choice. Even if one day there are open source replacements for  K everything, determining which software is better for your particular needs  M will still be a subjective decision. For example, today there is some pretty  K good open source software for viewing PDF files. Adobe also offers Acrobat  M Reader, their proprietary PDF viewer for Linux. Even if 99% of people decide  J they prefer the open source readers over Adobe's, shouldn't that 1% still 2 have the freedom to choose the reader they prefer?  K 2. Most people buy computers, not operating systems. Around 95% of all the  C computers running today are running the operating system that came  I pre-installed on that machine, which in most cases is Microsoft Windows.  J This means that if an open source operating system like Linux is going to M find broad adoption, it needs to be pre-installed on computers. I hear a lot  M of people in the open source community shouting that Dell should be shipping  K Linux on their desktop and laptop computers. Why would they? There is very  I little demand for it, and it doesn't work out of the box with everything  D (MP3, iPods, DVD, Windows Media, Java, Flash, QuickTime, etc.) like I Microsoft Windows does. Until there is demand and it can work as well as  K Windows, why bother? (This is why Linspire has always included proprietary  F software when necessary to support for as many formats, filetypes and  hardware devices as possible.)  G 3. To influence people, you must first become influential. Speaking of  K demand, today Linux is run on only around 1% of all desktop computers. Not  H exactly a very influential group ... yet. Linux on servers is extremely L influential. If you're IBM, Intel, Oracle, HP, etc., you have no choice but I to pay attention to Linux with your server products. But on the desktop,  H it's not worth their bother yet because there just aren't enough users. M Linux needs to work the way people need it to work, or they won't use it. If  M including the option for proprietary drivers, codecs and software means more  M people can use Linux, that will expand its circle of influence. If Linux can  G get even 5-10% more market share, there will be a much larger group of  I people who are experiencing the value of open source. Linspire only uses  J proprietary drivers when there are no viable alternatives. I long for the I day when a hardware company can take on nVidia and ATI by offering a new  I option that does come with open source drivers. If Linux had millions of  M users on the desktop, that would be a realistic opportunity, but with such a  I small market we pretty much have to play by the rules in existence today.   G 4. Competition is a good thing. I think it would be a mistake for open  K source solutions to only "compete" against other open source solutions. In  H other words, if open source GIMP only needs to be as good as other open L source image editors, then it can stop all work right now. However, if GIMP K has to also compete with the likes of Adobe Photoshop, it will continue to  I advance. If you have a software marketplace where GIMP and Photoshop are  J side by side, right next to each other, each just one click away from the L consumer, what do you think would happen? I believe GIMP would be compelled J even more to get better, and Photoshop would become a lot less expensive. H It's good for computer users when both the open source software and the H proprietary vendors are having to compete with each other. Ask yourself G this: Would it be a good or bad thing if Adobe announced tomorrow that  K Photoshop is now available for Linux users? Choice is always a good thing,   as is competition.  G 5. Most people don't want religion when buying a computer. Many of you  J reading this love your computers. You love technology. You have installed M many different operating systems in your lifetime. But, for ever person like  K you, there are 100 that view their computer as nothing more than a pile of  G metal, plastic and silicon. It's simply an appliance to help them play  I games, surf the web, type a letter, do some work, read email, and so on.  I They care about as much about the philosophy of open source as they care  I about what language the operating system in their refrigerator is using.  I They don't. You can't tell the average computer user, "Use Linux because  J it's a lot less expensive, more secure and open. Sure, it won't work with M your iPod, play MP3s, or play DVDs, but let me explain why open source is so  L great." Most will just look at you funny, thinking to themselves "Hey, it's L only a computer." Sure, they'll be interested in the benefits of Linux, but M not at the sacrifice of getting done what they need to get done. Most people  F are not going to give up usability just to support the values of open  source.                    Prius - The old and new  meet in Toyota's hybrid   F 6. Ideals vs Ideas. "Ideals" are a dime a dozen and easy to agree on. E "Ideas" on how to achieve an ideal, on the other hand, are much more  F difficult and they are rarely accompanied by universal agreement. For L example, most people believe peace, clean air, healthy food, etc., are good L things, and most agree that hunger, poverty, pollution, crime, etc. are bad G things. It's an "ideal" to have more of the good, and end the bad. Who  H doesn't think it would be good if the world had clean air and no crime? I However, the "ideas" to achieving those ideals will vary widely. Look at  K something like the debate over fuel. Everyone agrees that it would be good  L to reduce pollution, but do you do that by changing fuel economy standards, H taxing gas, requiring ethanol or other additives in gas supplies, or by J developing new non-gas dependent cars? Each decision has its own pros and M cons. It's tough to get to the ideal without varying ideas. Freespire shares  K the same ideal with most everyone in the open source community, in that we  L believe closed standards, software patents and proprietary software for the I most part can be harmful and hamper innovation. We just have a different  # idea than some on how to get there.   G 7. The "Hybrid" Open Source OS. I have friends who for years have been  I talking to me about alternative fuels for automobiles. I never paid much  I attention. I agreed with their ideals, but their ideas were just too far  L removed from my habits. Then along comes the hybrid automobile. This car is I part "the old way" (gasoline) and part "the new way" (electricity).  But  K because this car looks the same as a regular gar, fills up at the same gas  J stations, and drives similarly, it's not a stretch for anyone to consider M this type of car. The popularity of these new hybrid cars comes from letting  M people enjoy the ideal of alternative fuels (higher MPG, cleaner air, etc.),  J but without them having to learn something new or sacrifice what they are J used to. Granted, a hybrid car owner is still using oil and gasoline, but J he's certainly one step closer to the ideal, and that person will be much J more open minded to the next jump in the migration away from oil. I think I the hybrid automobile has perhaps done more to get people to become open  K minded and learn about alternative fuels than about anything else to date.  J Freespire is basically a "hybrid" open source operating system, combining K the new (open source) with the old, comfortable and traditional (mp3, DVD,  I etc.). The more Freespire Linux users there are, the more people who are  K being exposed to the wonders of open source software. People don't want to  G be "educated," they just want to use something that they love, and the  ! education is a byproduct of that.   I On the Freespire forums, we asked which version of Freespire people were  L planning on using most, and currently over 80% have said they would be more G than likely using the "hybrid" version of Freespire, not the 100% open  K source version.  When I gave my address at DLS, I asked by a show of hands  H how many people in the audience didn't have any proprietary software on > their desktop and laptop computers. Not even one hand went up.  J I share in the ideal of open source software, but I also believe the time H has come for some new ideas and approaches to exposing Linux to a wider J audience. In the end, it will get more people using open source software, ( which I believe to be a very good thing.   - Kevin               J The views expressed herein are those of the author and do not necessarily # reflect the views of Linspire, Inc.           - Want to become a Featured Partner? Learn more   H Please visit support.linspire.com for answers to questions you may have  about Linspire.      About LinspireK Linspire, Inc. (www.linspire.com) was founded in 2001 to bring choice into  J the operating system market. The company's flagship product, the Linspire M operating system, is an affordable, easy-to-use Linux-based operating system  M for home, school, and business users. Linspire pioneered CNR (click and run)  M Technology, which allows Linspire users to download and install thousands of  J software programs with just one mouse click. The more than 2,000 software L titles available in the CNR Warehouse (www.linspire.com/cnrservice) include F full office and productivity suites, games, multimedia players, photo 0 management software, accounting tools, and more.   ------------------------------   Date: 12 May 2006 01:20:04 GMT( From: bill@cs.uofs.edu (Bill Gunshannon)" Subject: Results of my straw poll., Message-ID: <4ci664F15ouoiU1@individual.net>  D You may remember I asked a few days ago about what questions I couldE ask to try and find out why people choose Unix over VMS.  I tried it. C Granted, my sampling size was very small and not statistically sig- E nificant.  But, I think even these results are interesting.  So, here @ we go.  Comments/discussion of the results would be appreciated.  H First, the sample size.  Just my department faculty, 8 people.  Of thoseH 8, only 4 responded.  I think this says something in its own right.  But we can cover that later.  C Next, the questions.  Given no specific questions, I kept it pretty  general.  . 1.  Between VMS and Unix, which do you prefer?   2.  Why?  " 3.  Which one did you learn first?  K 4.  What other OSes (ignoring PC OSes for the sake of this discussion) have 4     you used enough to consider yourself proficient?    And  now we go onto the answers.   ---------------- Professor #1   1.  VMS   = 2. In some ways, I view VMS as "Unix done right";  that is, I >    understand it as having being inspired by and adopting most?    of the good things about Unix, but done more professionally,     regularly, and securely.   B 3.  I studied and gained experience with Unix before VMS, but madeE     more substantial use of VMS before I made equivalent use of Unix.   H 4.  Various IBM mainframe OSs, such as VM, etc., Prime, MVS, even Xerox.     -------------------- Professor #2  > 1.  Unix.  Of course, here I'm referring strictly to the "user=     interface" and not to the "under the hood" type of stuff, !     about which I know even less.   > 2.  Mostly because I am more familiar with Unix and hence find=     it easier to do certain tasks.  These may be equally easy :     in VMS but I simply don't know how to do it there.  An>     example is to determine the current/working directory.  In=     Unix, the command 'pwd' provides that info.  I know of no @     analogous command in VMS.  Another directory-related issue I?     have with VMS is that if you change directories (using 'set @     default') into a non-existent directory, no error message isC     given.  And you end up in some kind of netherworld in the sense C     that it is not easy to get back to the directory where you were D     before issuing the erroneous command.  (Or at least I don't knowC     how to do so.)  I find the 'HELP' feature in VMS to be not very E     helpful and, in fact, even worse than Unix's dreaded "man" pages.   > 3.  VMS.  On a VAX 11/780 (750?) at the U of S in 1983.  I was?     introduced to Unix in 1985 at graduate school, where I also 
     used MTS.   + 4.  None.  (I've completely forgotten MTS.)    ------------------------ Professor #3   1. Unix     A 2.  Because VMS is not available for any computer that I OWN.  So A     I can play around with unix on my computers in a stronger way A     than I could  with VMS.  However, in general, I preferred the      command structure of VMS.    3.  VMS   C 4.  In the past I've used RSTS/E, RT-11, a couple of IBM O/S es and D     some from other hardware vendors but quite frankly right now I'dE     have trouble doing anything with them that wasn't at least pretty      similar to VMS or unix.    ---------------- Professor #4  = 1.  VAX, but unfortunately, it does not seem to be in demand.    2.  Protection/access detail.    3.  VAX    4.  none   ------------------A The order is not significant and reflects the order in which they A responded to me.  Needless to say, the first one surprised me and A I discussed it further with him in the hall (that's where we hold ? some of our best academic discussions! :-)  He then informed me ? that he actually would prefer to do more of his copurse work on ? VMS if the tools were available, citing specifically Java and a A good IDE.  While my using VAXen makes this impossible, Alphas can @ definitely run Java and I wouldthink that an IDE written in Java> (like Eclipse) would work on VMS.  Interesting that no one has< ever approached me before to ask if we could do more on VMS,A apparently because, as has often been states here, the perception = that VMS is legacy and these modern computer concepts are not 
 available.  @ Looking at the next two, choice of Unix appears to be based on a@ lack of knowledge or a misconception.  And 4th had little to sayA except that they would prefer VMS (Note the constant reference to A "VAX" which shows how long since he actually used it on a regular  basis.  A Looks to me like if HP actually wanted it they could have the EDU 
 market again.   E Oh yeah, I said I would comment on the 4/8 respondents.  Again, based C on informal discussions in the hall, the 4 who did not respond just E thought it was humorous as they consider VMS to be irrelevant at this  time.   	 Comments?    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: 11 May 2006 20:45:29 -0500- From: Kilgallen@SpamCop.net (Larry Kilgallen) & Subject: Re: Results of my straw poll.3 Message-ID: <sEFbXALxBFcE@eisner.encompasserve.org>   W In article <4ci664F15ouoiU1@individual.net>, bill@cs.uofs.edu (Bill Gunshannon) writes: F > You may remember I asked a few days ago about what questions I couldG > ask to try and find out why people choose Unix over VMS.  I tried it. E > Granted, my sampling size was very small and not statistically sig- G > nificant.  But, I think even these results are interesting.  So, here B > we go.  Comments/discussion of the results would be appreciated.   > Comments?   ! Thanks for putting in the effort.    ------------------------------    Date: 11 May 2006 19:43:10 -0700; From: "johnhreinhardt@yahoo.com" <johnhreinhardt@yahoo.com> & Subject: Re: Results of my straw poll.C Message-ID: <1147401790.332866.103600@i40g2000cwc.googlegroups.com>    Bill Gunshannon wrote:F > You may remember I asked a few days ago about what questions I couldG > ask to try and find out why people choose Unix over VMS.  I tried it. E > Granted, my sampling size was very small and not statistically sig- G > nificant.  But, I think even these results are interesting.  So, here B > we go.  Comments/discussion of the results would be appreciated. > J > First, the sample size.  Just my department faculty, 8 people.  Of thoseJ > 8, only 4 responded.  I think this says something in its own right.  But > we can cover that later. > E > Next, the questions.  Given no specific questions, I kept it pretty 
 > general. > 0 > 1.  Between VMS and Unix, which do you prefer? > 
 > 2.  Why? > $ > 3.  Which one did you learn first? > M > 4.  What other OSes (ignoring PC OSes for the sake of this discussion) have 6 >     you used enough to consider yourself proficient? > " > And  now we go onto the answers. >  > ---------------- > Professor #1 > 	 > 1.  VMS  > ? > 2. In some ways, I view VMS as "Unix done right";  that is, I @ >    understand it as having being inspired by and adopting mostA >    of the good things about Unix, but done more professionally,  >    regularly, and securely.  > D > 3.  I studied and gained experience with Unix before VMS, but madeG >     more substantial use of VMS before I made equivalent use of Unix.  > J > 4.  Various IBM mainframe OSs, such as VM, etc., Prime, MVS, even Xerox. >  >  > -------------------- > Professor #2 > @ > 1.  Unix.  Of course, here I'm referring strictly to the "user? >     interface" and not to the "under the hood" type of stuff, # >     about which I know even less.  > @ > 2.  Mostly because I am more familiar with Unix and hence find? >     it easier to do certain tasks.  These may be equally easy < >     in VMS but I simply don't know how to do it there.  An@ >     example is to determine the current/working directory.  In? >     Unix, the command 'pwd' provides that info.  I know of no B >     analogous command in VMS.  Another directory-related issue IA >     have with VMS is that if you change directories (using 'set B >     default') into a non-existent directory, no error message isE >     given.  And you end up in some kind of netherworld in the sense E >     that it is not easy to get back to the directory where you were F >     before issuing the erroneous command.  (Or at least I don't knowE >     how to do so.)  I find the 'HELP' feature in VMS to be not very G >     helpful and, in fact, even worse than Unix's dreaded "man" pages.  > @ > 3.  VMS.  On a VAX 11/780 (750?) at the U of S in 1983.  I wasA >     introduced to Unix in 1985 at graduate school, where I also  >     used MTS.  > - > 4.  None.  (I've completely forgotten MTS.)  >  > ------------------------ > Professor #3 > 	 > 1. Unix  >  > C > 2.  Because VMS is not available for any computer that I OWN.  So C >     I can play around with unix on my computers in a stronger way C >     than I could  with VMS.  However, in general, I preferred the  >     command structure of VMS.  > 	 > 3.  VMS  > E > 4.  In the past I've used RSTS/E, RT-11, a couple of IBM O/S es and F >     some from other hardware vendors but quite frankly right now I'dG >     have trouble doing anything with them that wasn't at least pretty  >     similar to VMS or unix.  >  > ---------------- > Professor #4 > ? > 1.  VAX, but unfortunately, it does not seem to be in demand.  >  > 2.  Protection/access detail.  > 	 > 3.  VAX  > 
 > 4.  none >  > ------------------C > The order is not significant and reflects the order in which they C > responded to me.  Needless to say, the first one surprised me and C > I discussed it further with him in the hall (that's where we hold A > some of our best academic discussions! :-)  He then informed me A > that he actually would prefer to do more of his copurse work on A > VMS if the tools were available, citing specifically Java and a C > good IDE.  While my using VAXen makes this impossible, Alphas can B > definitely run Java and I wouldthink that an IDE written in Java@ > (like Eclipse) would work on VMS.  Interesting that no one has> > ever approached me before to ask if we could do more on VMS,C > apparently because, as has often been states here, the perception ? > that VMS is legacy and these modern computer concepts are not  > available. > B > Looking at the next two, choice of Unix appears to be based on aB > lack of knowledge or a misconception.  And 4th had little to sayC > except that they would prefer VMS (Note the constant reference to C > "VAX" which shows how long since he actually used it on a regular  > basis. > C > Looks to me like if HP actually wanted it they could have the EDU  > market again.  > G > Oh yeah, I said I would comment on the 4/8 respondents.  Again, based E > on informal discussions in the hall, the 4 who did not respond just G > thought it was humorous as they consider VMS to be irrelevant at this  > time.  >  > Comments?  >  > bill >  >  > --L > Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesF > bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. > University of Scranton   |@ > Scranton, Pennsylvania   |         #include <std.disclaimer.h>  > I find it interesting that a professor in the Computer ScienceF department (I assume - could be Math) can make his/her way through theD tangled web of "man", yet cannot deduce that if you "set default" toG change directories then a perhaps the "show default" command might tell ? you where you are.  They started out on VMS in 1983.  The "Show E default" command has been there since I can remember (VMS V2.1).  And E "Help" being worse than 'man"?  I guess it's just what you're used to ) but I still find that hard to comprehend.   E Thanks for the info on your poll.  It is indeed interesting, although C without some hefty discounts on hardware (even the new IA64 systems B which are much cheaper then their Alpha counterparts) I doubt mostA colleges would be willing to bring back VMS. It' s a nice thought @ though.  I know I certainly had a grand time in college with our 11/780.    ------------------------------  % Date: Thu, 11 May 2006 21:43:04 -0500 6 From: "David J. Dachtera" <djesys.no@spam.comcast.net>& Subject: Re: Results of my straw poll.0 Message-ID: <4463F638.6FAF92C4@spam.comcast.net>   Bill Gunshannon wrote: >  > [snip] > ------------------------ > Professor #3 > 0 > 1.  Between VMS and Unix, which do you prefer?	 > 1. Unix  > 
 > 2.  Why?C > 2.  Because VMS is not available for any computer that I OWN.  So C >     I can play around with unix on my computers in a stronger way C >     than I could  with VMS.  However, in general, I preferred the  >     command structure of VMS.   F ...and of course, as we all "know", there's no market for OpenVMS-IA32 or -X86/64.   G Perhaps someday, someone will be able to give me a lucid explanation of H how out of hundreds of millions - if not billions - of IA32 machines outE there, not a single one of them would be a good candidate to run VMS.   7 Perhaps, but not likely - at least, not in my lifetime.   E I guess Charon-VAX is just an expensive toy for us die-hards, eh? ;-)    --   David J Dachtera dba DJE Systems  http://www.djesys.com/  & Unofficial OpenVMS Marketing Home Page! http://www.djesys.com/vms/market/   ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/   " Unofficial OpenVMS-IA32 Home Page: http://www.djesys.com/vms/ia32/   ) Unofficial OpenVMS Hobbyist Support Page: " http://www.djesys.com/vms/support/   ------------------------------  % Date: Thu, 11 May 2006 12:07:25 -0600 6 From: "Michael D. Ober" <obermd.@.alum.mit.edu.nospam>% Subject: Re: SGI files for chapter 11 / Message-ID: <Q1L8g.23$7J5.5440@news.uswest.net>      K The 8088 was a 16-bit processor with an 8-bit data interface.  The 8086 was H a 16-bit processor with a 16-bit data interface.  Both processors had 20L address lines and an identical instruction set.  IBM chose the 8088 over the8 8086 because 8 bit interfaces were the norm at the time.  
 Mike Ober.  > "Richard B. Gilbert" <rgilbert88@comcast.net> wrote in message* news:zfydnVcgne9Ezf7ZRVn-tg@comcast.com...  I > The 8088 was 8 bit and the 8086 was 16 bit IIRC.  The 8086 was not well G > known; the IBM PC, and PC/XT used the 8088.  The PC/AT used the 80286 J > and, from that point on, the 8088 and 8086 were pretty much "history" or% > "nostalgia" or something like that.  >    ------------------------------  + Date: Thu, 11 May 2006 18:21:05 +0000 (UTC) 1 From: legalize+jeeves@mail.xmission.com (Richard) % Subject: Re: SGI files for chapter 11 , Message-ID: <e3vvah$lfd$2@news.xmission.com>  / [Please do not mail me a copy of your followup]   ; koehler@eisner.aspm.encompasserve.org spake the secret code / <N5kYoewJ7G1D@eisner.encompasserve.org> thusly:    >   IIRC SGI owns MIPS.   ! They used to; they don't anymore.  --  E "The Direct3D Graphics Pipeline"-- code samples, sample chapter, FAQ: 3           <http://www.xmission.com/~legalize/book/> ( 	    Pilgrimage: Utah's annual demoparty,                <http://pilgrimage.scene.org>   ------------------------------  % Date: Thu, 11 May 2006 14:47:02 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> % Subject: Re: SGI files for chapter 11 , Message-ID: <446386A5.C9D32F45@teksavvy.com>  
 Andrew wrote: G > >From what you have said as far as your company is concerned HP would H > only lose their existing OpenVMS revenues and nothing else. Given thatF > HP's decision to axe OpenVMS if it ever happens would have to factorH > that loss in anyway its hard to see how your example helps make a case  C It isn't quite like that. When the Stallard/Winkler type manages to E convince Hurd to kill VMS, it will be because he will have some study D with him showing that X % of VMS customers would migrate to HP-UX orD Windows and that this percentage woudl generate revenus, which, whenF combined with the cost savings by firing all the VMS people, would end up to be a + for HP.  H HP would present that study to the Wall Street Casino Analysts who wouldH approve of that decision since it would rid HP of a little known productE without any market penetration and which costs a lot to maintain, and  voila. Done deal.   E It doesn't matter that afterwards that the X % will never materialise : and HP will find itself with much greater losses than that# Stallard/Winkler type had promised.     F Consider that Alpha was killed under Compaq. They would have generatedG whatever study was necessary to support that decision. And now that VMS G is under HP, nobody will go back and blame whoevere was responsible for H that because the negative impact of the murder of Alpha was much greater$ than had been anticipated/promised.   D So when the time comes for a Stallard/Winkler type to present such aA study to justify killing VMS, there probably won't be anyone in a E position to challenge based on the real numbers of the effects of the 
 Alpha murder.     C When Palmer told SWIFT to drop VMS as its key platform, he told his H folks that the expectation was that ST400 customers would all migrate toN Digital built Windows machines and hire Digital for support. It didn't happen.   ------------------------------  % Date: Thu, 11 May 2006 15:19:11 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> % Subject: Re: SGI files for chapter 11 + Message-ID: <44638E2A.AC35AE0@teksavvy.com>    Tom Linden wrote: K > all that was needed was a sane ISV policy.  Introducing new architectures L > is a risky, reckless endeavour.  VAX should have been extended to 64 bits.K > alpha should never have happened, Itanium should never have happened, not ' > without bringing all the ISV's along.     G Had VMS engineers been given the resources to fully integrate VEST into H Alpha VMS and allow VAX customers to safely run their VAX apps on Alpha,L including 3rd party apps, then the migration would have been more succesful.  G Apple had done a wonderful job going from 86k to PowerPC namely because H they provided a nearly seamless binary translator (and this one was more> sophisticated than VEST because it was "live" and even ran 68k extensions on powerpc machines.    ------------------------------    Date: 11 May 2006 12:40:06 -07001 From: nothome@spammers.are.scum (Malcolm Dunnett) % Subject: Re: SGI files for chapter 11 , Message-ID: <rJEcX0Or9Nlf@malvm9.mala.bc.ca>  4 In article <tUioAZzJ41zK@eisner.encompasserve.org>, @    koehler@eisner.nospam.encompasserve.org (Bob Koehler) writes:   > G >    VAX should not have been extended to 64 bits.  VAX was already too H >    damn slow.  VAX couldn't compete against 32 bit RISC.  The decisionJ >    to make Alpha 64 bit was along the line of a good idea as long as DEC- >    knew they had to replace the VAX anyhow.  >   D    The "conventional wisdom" of the time was that CISC architecturesC were too complex to be speeded up enough and that RISC was the only 
 way to go.  I    However that same thinking suggested that the 80x86 architecture could G never keep up either and was doomed. I'm not a chip designer, but Intel J certainly seems to have found ways to get those speed enhancements withoutE sacrificing binary compatibility. I'm certainly not suggesting anyone D would ever do it, but I wonder if it wouldn't be possible to build aI VAX chip today ( using all the latest and greatest tricks and processes ) 6 that would be competetive with the x86 chips of today.  H >    Jumping from Alpha to IA64 is a much more serious question, as wellF >    as deciding to stay with VAX instead of porting to 386 before DEC0 >    lost all its sales to UNIX RISC and Wintel. > D      Yes, and how would the world look today if DEC hadn't cancelled PRISM?   ------------------------------  % Date: Thu, 11 May 2006 15:50:46 -0400 ( From: Bill Todd <billtodd@metrocast.net>% Subject: Re: SGI files for chapter 11 = Message-ID: <zsmdnQx2Otf3CP7Z4p2dnA@metrocastcablevision.com>    Malcolm Dunnett wrote:6 > In article <tUioAZzJ41zK@eisner.encompasserve.org>, B >    koehler@eisner.nospam.encompasserve.org (Bob Koehler) writes: > H >>    VAX should not have been extended to 64 bits.  VAX was already tooI >>    damn slow.  VAX couldn't compete against 32 bit RISC.  The decision K >>    to make Alpha 64 bit was along the line of a good idea as long as DEC . >>    knew they had to replace the VAX anyhow. >> > F >    The "conventional wisdom" of the time was that CISC architecturesE > were too complex to be speeded up enough and that RISC was the only  > way to go.   True enough, so far.   > K >    However that same thinking suggested that the 80x86 architecture could & > never keep up either and was doomed.  I Indeed - that was supposedly one of the rationales behind Itanic, though  A the prospect of becoming a monopoly may have been a stronger one.    > I'm not a chip designer,  E People who *are* chip designers (such as John Mashey of MIPS and SGI  B fame) do not share the opinion which you voice below, as has been  discussed here previously.     but Intel L > certainly seems to have found ways to get those speed enhancements without# > sacrificing binary compatibility.   F Yes, but it took them until the late '90s to do so - well over half a D decade after Alpha solved the problem in a different manner for VMS.  %   I'm certainly not suggesting anyone F > would ever do it, but I wonder if it wouldn't be possible to build aK > VAX chip today ( using all the latest and greatest tricks and processes ) 8 > that would be competetive with the x86 chips of today.  G People like Mashey think not.  x86 is a very simple CISC compared with  D VAX, and the pipelining optimizations applicable to x86 and used in I modern x86 implementations would not work nearly as well for the VAX ISA  C due to its complexity (for starters, being a 3-address rather than  ! mostly a 1-address architecture).    - bill   ------------------------------  % Date: Thu, 11 May 2006 15:52:37 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> % Subject: Re: SGI files for chapter 11 + Message-ID: <446395FE.61195AC@teksavvy.com>    Malcolm Dunnett wrote:K >    However that same thinking suggested that the 80x86 architecture could I > never keep up either and was doomed. I'm not a chip designer, but Intel L > certainly seems to have found ways to get those speed enhancements without$ > sacrificing binary compatibility.   H That is because the 8086's instruction set was much simpler than VAX's. F Intel didn't steal Vax tricks for its 8086, they stole Alpha tricks to embed into the 8086.      F > would ever do it, but I wonder if it wouldn't be possible to build aK > VAX chip today ( using all the latest and greatest tricks and processes ) 8 > that would be competetive with the x86 chips of today.  G Yes, it would probably be possible. Put a risc engine are the very core ? with vax instruction decoders layers above it. Would it be cost G effective ? Would it allow engineers to boost the speed of VAX chips as H fast as Intel/AMD boost their 8086s and keep up with the industry ? That is the question.  F But you'd also need to bring the VAX compilers up to speed with modern times as well.   ------------------------------  % Date: Thu, 11 May 2006 20:44:19 -0700 # From: "Tom Linden" <tom@kednos.com> % Subject: Re: SGI files for chapter 11 ) Message-ID: <op.s9e5z5wezgicya@hyrrokkin>   H On Thu, 11 May 2006 12:50:46 -0700, Bill Todd <billtodd@metrocast.net>   wrote:   > Malcolm Dunnett wrote:8 >> In article <tUioAZzJ41zK@eisner.encompasserve.org>,  C >>    koehler@eisner.nospam.encompasserve.org (Bob Koehler) writes:  >>I >>>    VAX should not have been extended to 64 bits.  VAX was already too J >>>    damn slow.  VAX couldn't compete against 32 bit RISC.  The decisionJ >>>    to make Alpha 64 bit was along the line of a good idea as long as   >>> DEC / >>>    knew they had to replace the VAX anyhow.  >>> H >>     The "conventional wisdom" of the time was that CISC architecturesF >> were too complex to be speeded up enough and that RISC was the only
 >> way to go.  >  > True enough, so far. > I >>     However that same thinking suggested that the 80x86 architecture    >> could' >> never keep up either and was doomed.  > L > Indeed - that was supposedly one of the rationales behind Itanic, though  C > the prospect of becoming a monopoly may have been a stronger one.  >  >> I'm not a chip designer,  > H > People who *are* chip designers (such as John Mashey of MIPS and SGI  E > fame) do not share the opinion which you voice below, as has been    > discussed here previously. > 
 >   but Intel G >> certainly seems to have found ways to get those speed enhancements   
 >> without$ >> sacrificing binary compatibility. > I > Yes, but it took them until the late '90s to do so - well over half a   F > decade after Alpha solved the problem in a different manner for VMS. > ' >   I'm certainly not suggesting anyone G >> would ever do it, but I wonder if it wouldn't be possible to build a L >> VAX chip today ( using all the latest and greatest tricks and processes   >> )9 >> that would be competetive with the x86 chips of today.  > J > People like Mashey think not.  x86 is a very simple CISC compared with  G > VAX, and the pipelining optimizations applicable to x86 and used in   L > modern x86 implementations would not work nearly as well for the VAX ISA  F > due to its complexity (for starters, being a 3-address rather than  # > mostly a 1-address architecture).  Bill,   K When Mashey wrote the article you couldn't put half a billion transistors    on aK chip.  Also remember because VAX had a more advanced instruction set than    riscG with objects about 1/3 the size it didn't need multi-level caches and    correspondingly L broad (uncontended) channels to primary memory  sure there would have been   someL thorny issues to sort, Mashey mentions the problem of instruction decoding   stalls, L but I wrote such a decoder for the national 32032 which was pretty much a    vax H copy and I think there are ways to solve that issue.  Yes, I am taking   drugs,H vicodin, had surgery on right hand today, so don't expect a long reply    from ne   >  > - bill   ------------------------------  % Date: Thu, 11 May 2006 15:15:59 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> Y Subject: Re: This post will self-destruct in 10secs (Was Re: X windows  vulnerability)  v , Message-ID: <44638D6A.E6BC119A@teksavvy.com>   Bob Koehler wrote:I >    I know of no other product from HP and no other software vendor that % >    gets fixes for systems that old.     H Turns out that over 100,000 VAX systems remain on support today, roughlyC 1/3 of the installed base if you believe in the more recent 300,000 C number that was recentlty leaked, or 25% if you believe in the 400k & systems that was used a few years ago.    H If 5.5-2 is still such a popular version, it is most likely a tribute toF the damage done by Palmer (which in fairness was started under Olsen).A The dwindling software portfolio on VMS is probably a big reason, G especially since the Alpha portfolio was a subset of VAX and DEC didn't E want to provide VEST to everyone. So many would have decided to leave 2 their VAX systems "as is" until they are replaced.  A And by the time Alpha-VMS was finally on par with VAX, Palmer had E already begun his slash/burn and customers would have decided to stop $ spending money on their VMS systems.  > Also, consider that Digital decided at about that time to stopD developping user applications. So X/Motif didn't get much work sinceB then. So no big new appliaction came to VMS due to lack of visibleI future for VMS as workstation. So VMS was stuck with legacy applications.   H Lets be blunt here. Between 5.5-2 and 7.3, legacy applications don't seeF much difference.  Consider how quickly Linux got multiple desktop/ GUI2 stacks. VMS is still stuck with an ancient Motif.   G The big irony: the one big functionality that Alpha had that VAX didn't 3 have was sound support. And that is gone with IA64.       E The fact that HP isn't doing anything to try hard to reactivate those G old VAX customers and move them to more modern software and/or hardware F is an indication that they have given up on them and are just miliking; them for their money until they do move to Sun or whatever.   H Getting those 100k systems "back" wouldn't be easy. But it would be much@ easier than trying to acquire 100,000 customers from Sun or IBM.   ------------------------------  + Date: Thu, 11 May 2006 20:07:34 +0000 (UTC) P From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply), Subject: Re: Updated VMS information May 7th$ Message-ID: <e405i6$8h2$1@online.de>  3 In article <uiaFDQqh2EMS@eisner.encompasserve.org>, > koehler@eisner.nospam.encompasserve.org (Bob Koehler) writes:   8 >       2) SCS over IP protocols (a plan, not a promise)  E With 50 Mb/s DSL available already in some areas, does this mean one  E could implement a multi-site cluster over such a connection (perhaps  H with something similar to a VPN)?  Of course, the advertised 50 Mb/s is A downstream, not upstream, but 10 Mb/s upstream can't be far away.    ------------------------------  # Date: Thu, 11 May 2006 20:54:39 GMT , From: Hoff Hoffman <hoff-remove-this@hp.com>, Subject: Re: Updated VMS information May 7th/ Message-ID: <jwN8g.389$tM4.47@news.cpqcorp.net>   / Phillip Helbig---remove CLOTHES to reply wrote: 5 > In article <uiaFDQqh2EMS@eisner.encompasserve.org>, @ > koehler@eisner.nospam.encompasserve.org (Bob Koehler) writes:  > 9 >>       2) SCS over IP protocols (a plan, not a promise)  > G > With 50 Mb/s DSL available already in some areas, does this mean one  G > could implement a multi-site cluster over such a connection (perhaps  J > with something similar to a VPN)?  Of course, the advertised 50 Mb/s is C > downstream, not upstream, but 10 Mb/s upstream can't be far away.   A    Having sufficient bandwidth is certainly necessary though not  C necessarily sufficient for clustering; it's the round-trip latency  ! that's usually the classic limit.   H    Clusters are already operating over WAN bridges at a number of sites.  F    The classic rule of thumb for cluster operations over an arbitrary I datalink involves having the bandwidth and the latency and the multicast  I capabilities of a ten megabit Ethernet or IEEE 802.3 network, or better.  J   If you have all that, then you can probably cluster over the connection.   ------------------------------    Date: 11 May 2006 20:41:37 -0500- From: Kilgallen@SpamCop.net (Larry Kilgallen) , Subject: Re: Updated VMS information May 7th3 Message-ID: <iNgsYHy083N4@eisner.encompasserve.org>   w In article <e405i6$8h2$1@online.de>, helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply) writes: 5 > In article <uiaFDQqh2EMS@eisner.encompasserve.org>, @ > koehler@eisner.nospam.encompasserve.org (Bob Koehler) writes:  > 9 >>       2) SCS over IP protocols (a plan, not a promise)  > G > With 50 Mb/s DSL available already in some areas, does this mean one  G > could implement a multi-site cluster over such a connection (perhaps  J > with something similar to a VPN)?  Of course, the advertised 50 Mb/s is C > downstream, not upstream, but 10 Mb/s upstream can't be far away.   @ For many clusters the issue is not so much bandwidth as latency.   ------------------------------    Date: 11 May 2006 11:48:12 -0700 From: transnet@vrx.net/ Subject: VMS indexed files - how did they work? C Message-ID: <1147373292.684406.109440@v46g2000cwv.googlegroups.com>   8 My favorite has always been the VMS "indexed" file type.  @ I really enjoyed working with them. You could do anything (quite literally).   B There are still things you could (can) do with them so simply that can't be done, easily - if at all on modern RDBMS or other databases.   C I liked the fact that everything was contained in one data file, no 5 external indexes or other external files as overhead.   D And no need to keep track of the indexes in your programming, it was all taken care of for you.  F And you could literally "seek" any record by any string, even parts of descriptive fields, etc.  D Like today, if you wanted to use a flat text file for a fifo buffer,D you really can't. because you can't delete individual entries from aC text file, formatted (records) or not. You can blank out the record $ data, but the record is still there.  E I know the OS was hiding a lot of the overhead, but I always wondered B how indexed files on VMS actually worked. or is this still "secret
 sauce" today?    ------------------------------  % Date: Thu, 11 May 2006 21:38:01 +0200 3 From: Wilm Boerhout <w4OLD.boerhout@PAINTplanet.nl> 3 Subject: Re: VMS indexed files - how did they work? 5 Message-ID: <446392b2$0$5009$ba620dc5@nova.planet.nl>   , transnet@vrx.net wrote on 11-5-2006 20:48...: > My favorite has always been the VMS "indexed" file type. > B > I really enjoyed working with them. You could do anything (quite
 > literally).   > Please, please do not use the past tense here. There are many K application environments today using RMS indexed files. Long may they last.   G > I know the OS was hiding a lot of the overhead, but I always wondered D > how indexed files on VMS actually worked. or is this still "secret > sauce" today?   H It is a matter of philosophy whether RMS is part of VMS. The RMS record @ structure sure is older than VMS. OTOH, RMS is indeed very well  integrated into VMS.  G How things "work" for the user is pretty well described in the various   RMS manuals that come with VMS.   I How things "work" internally (what code is in place to fill the buckets,  G so to speak) well, there have been "advanced RMS" courses in the past,  J but I don't know if there's an "RMS Internals" manual around (/ping/ Hein)   /Wilm    ------------------------------    Date: 11 May 2006 12:53:51 -0700- From: "Steve Lionel" <steve.lionel@intel.com> 3 Subject: Re: VMS indexed files - how did they work? C Message-ID: <1147377231.741128.154580@i40g2000cwc.googlegroups.com>   F VMS indexed files initially were a copy of IBM ISAM files from OS/370.D They first appeared in VMS 2.0.  In the mid-70s, ISAM files were theD most popular "database" format and VMS needed them to break into theF commercial applications market (didn't hurt that VAX COBOL needed themE too.)  I designed and implemented the VAX FORTRAN support for them in  1979.    Steve    ------------------------------  % Date: Thu, 11 May 2006 15:41:48 -0400 ( From: Bill Todd <billtodd@metrocast.net>3 Subject: Re: VMS indexed files - how did they work? = Message-ID: <DPmdncY99IXNDv7ZRVn-rw@metrocastcablevision.com>    transnet@vrx.net wrote:    ...   G > I know the OS was hiding a lot of the overhead, but I always wondered D > how indexed files on VMS actually worked. or is this still "secret > sauce" today?   G The 'sauce' was never secret:  DEC did indexed file internal structure  B presentations starting in the late '70s.  There's probably a .pdf F presentation (ISTR one done in the late '80s or early '90s) available . somewhere still today, but I don't know where.  I The fundamentals of how an indexed file was managed can be understood by  D studying the mechanisms used in standard B+ trees (I justed checked E wikipedia.com's description and it seems reasonably accurate, though  I RMS, like many commercial implementations, did not - at least early-on -  H reshuffle data in lightly-filled leaf nodes but rather reclaimed a node E only if it became empty).  RMS extended the mechanisms by supporting  I leaf entries with duplicate key values, an unvarying identifier for each  F user data record valid for the life of the file (later and optionally I possibly only for the life of the record on VMS), alternate indexes with  H leaf entries that pointed to the relevant user data records using those E stable IDs, locking mechanisms at the record level (only at the page  I level in its 16-bit incarnations) allowing consistent access by multiple  G concurrent readers and writers), optimizations (as the product matured  I on VMS) like key- and data-compression, and (also later and only on VMS)  D features like journaling.  Extensions to support multiple user data G record types within a single file were AFAIK discussed from nearly the   start but never implemented.   - bill   ------------------------------  % Date: Thu, 11 May 2006 15:46:52 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> 3 Subject: Re: VMS indexed files - how did they work? , Message-ID: <446394A6.276B9D87@teksavvy.com>   transnet@vrx.net wrote: & > I really enjoyed working with them. G > I liked the fact that everything was contained in one data fileit was  > all taken care of for you. > : > And you could literally "seek" any record by any string,  G > I know the OS was hiding a lot of the overhead, but I always wondered + > how indexed files on VMS actually worked.       E Whis have you written everything in the past tense ? VMS still exists * today, and RMS files are still used today.   ------------------------------  # Date: Thu, 11 May 2006 20:00:01 GMT , From: Hoff Hoffman <hoff-remove-this@hp.com>3 Subject: Re: VMS indexed files - how did they work? 0 Message-ID: <5JM8g.388$1V4.311@news.cpqcorp.net>   transnet@vrx.net wrote:   G > I know the OS was hiding a lot of the overhead, but I always wondered D > how indexed files on VMS actually worked. or is this still "secret > sauce" today?   F    It's a standard non-relational database, with mechanisms for keyed G searches through the indexes and for sequential access -- I'm aware of  9 nothing here that is or was ever considered to be secret.   F    There is a book on file system internals available, and there is a I general intro and guide to the file system in the OpenVMS manuals posted  ' at <http://www.hp.com/go/openvms/doc/>.    ------------------------------    Date: 11 May 2006 15:02:40 -0500- From: Kilgallen@SpamCop.net (Larry Kilgallen) 3 Subject: Re: VMS indexed files - how did they work? 3 Message-ID: <nj2fYQIIODCz@eisner.encompasserve.org>   \ In article <1147373292.684406.109440@v46g2000cwv.googlegroups.com>, transnet@vrx.net writes:  G > I know the OS was hiding a lot of the overhead, but I always wondered D > how indexed files on VMS actually worked. or is this still "secret > sauce" today?   D To the level of approximation you seem to want, they work about likeA ISAM (Indexed Sequential Access Method) files on IBM's MVS (Z/OS) D operating system.  That is different from the builtin RDBMS on IBM's OS400.  C Lesser operating systems skip this capability, with the implication B that standardization of such on an operating system is not useful.   ------------------------------  % Date: Thu, 11 May 2006 16:17:50 -0400 ( From: Bill Todd <billtodd@metrocast.net>3 Subject: Re: VMS indexed files - how did they work? G Message-ID: <E-WdnQ5o_stfBv7ZnZ2dnUVZ_v2dnZ2d@metrocastcablevision.com>    Steve Lionel wrote: H > VMS indexed files initially were a copy of IBM ISAM files from OS/370.  E Not really - ISAM files were *strongly* associated with the physical  E layout of the disks they resided on, and not conventional b-trees at  ; all.  RMS resembled IBM's VSAM in several respects, though.   ! > They first appeared in VMS 2.0.   E Actually, they appeared in VMS 1.0, but only as a compatibility-mode   RMS-11 implementation.  &    In the mid-70s, ISAM files were the  > most popular "database" format  I I would have said that IBM's IMS was the most popular database format at  A that time - and by some measures (such as the relative amount of  I large-corporation centralized data under its management) it may still be  E close to that today.  Runners-up might perhaps have been Cullinane's  I IDMS and Cincom's TOTAL, both of which were also 'network' databases (in  G terms of internal and navigational structure, not in the 'distributed'  A sense) and neither of which was tied to a single computer vendor.   E Indexed files were file-system extensions that back then (before the  C wide-spread adoption of Unix dumbed file systems down) were fairly  E common across both minicomputer and mainframe vendors (there were no  E serious PCs at the time), but no one that I can remember presumed to  I characterize them as anything like 'databases' (though with a layer like  E DATATRIEVE on top of them they could begin to look somewhat similar).   '   and VMS needed them to break into the H > commercial applications market (didn't hurt that VAX COBOL needed them > too.)   F It might be closer to the mark to say that *DEC* needed them to break H into the commercial applications market.  RMS on both the PDP-11 and on B VAX was initially developed in the same group (Languages and Data @ Management, managed by Ron Ham) responsible for the 'commercial H languages' like COBOL, and VAX naturally inherited the products of this G group from their implementations on the 11 (though of course they were  2 mostly re-written to support the new environment).   - bill   ------------------------------    Date: 11 May 2006 14:09:33 -0700 From: bob@instantwhip.com 3 Subject: Re: VMS indexed files - how did they work? C Message-ID: <1147381773.311230.104770@u72g2000cwu.googlegroups.com>   D "Lesser operating systems skip this capability, with the implicationB that standardization of such on an operating system is not useful"   don't you mean "profitable"?   ------------------------------  % Date: Thu, 11 May 2006 17:43:42 -0500 / From: Chris Scheers <chris@applied-synergy.com> 3 Subject: Re: VMS indexed files - how did they work? 2 Message-ID: <4463BE1E.9010106@applied-synergy.com>   Steve Lionel wrote: H > VMS indexed files initially were a copy of IBM ISAM files from OS/370.F > They first appeared in VMS 2.0.  In the mid-70s, ISAM files were theF > most popular "database" format and VMS needed them to break into theH > commercial applications market (didn't hurt that VAX COBOL needed themG > too.)  I designed and implemented the VAX FORTRAN support for them in  > 1979.   B I was under the impression that Xerox's CP-V keyed files were the " inspiration for VMS indexed files.  + Many VMS ideas seem to have come from CP-V.    --  G ----------------------------------------------------------------------- $ Chris Scheers, Applied Synergy, Inc.  B Voice: 817-237-3360            Internet: chris@applied-synergy.com    Fax: 817-237-3074   ------------------------------  % Date: Thu, 11 May 2006 17:16:19 -0600 % From: Dan O'Reilly <dano@process.com> 3 Subject: Re: VMS indexed files - how did they work? A Message-ID: <6.1.2.0.2.20060511171514.0266f968@raptor.psccos.com>   + At 04:43 PM 5/11/2006, Chris Scheers wrote:  >Steve Lionel wrote:H >>VMS indexed files initially were a copy of IBM ISAM files from OS/370.F >>They first appeared in VMS 2.0.  In the mid-70s, ISAM files were theF >>most popular "database" format and VMS needed them to break into theH >>commercial applications market (didn't hurt that VAX COBOL needed themG >>too.)  I designed and implemented the VAX FORTRAN support for them in  >>1979.  > C >I was under the impression that Xerox's CP-V keyed files were the  # >inspiration for VMS indexed files.  > , >Many VMS ideas seem to have come from CP-V.  J Man, it's been a LONG time since I've heard talk of CP-V.  I used it on a H Xerox Sigma-7 a whole lotta (more than I care to think about) years ago!   ------J +-------------------------------+----------------------------------------+J | Dan O'Reilly                  |  "There are 10 types of people in this |J | Principal Engineer            |   world: those who understand binary   |J | Process Software              |   and those who don't."                |J | http://www.process.com        |                                        |J +-------------------------------+----------------------------------------+   ------------------------------  % Date: Thu, 11 May 2006 20:49:58 -0400 ( From: Bill Todd <billtodd@metrocast.net>3 Subject: Re: VMS indexed files - how did they work? G Message-ID: <2-2dnTMwAuIXRv7ZnZ2dnUVZ_tqdnZ2d@metrocastcablevision.com>    Chris Scheers wrote: > Steve Lionel wrote: I >> VMS indexed files initially were a copy of IBM ISAM files from OS/370. G >> They first appeared in VMS 2.0.  In the mid-70s, ISAM files were the G >> most popular "database" format and VMS needed them to break into the I >> commercial applications market (didn't hurt that VAX COBOL needed them H >> too.)  I designed and implemented the VAX FORTRAN support for them in >> 1979. > D > I was under the impression that Xerox's CP-V keyed files were the $ > inspiration for VMS indexed files. > - > Many VMS ideas seem to have come from CP-V.   H While I don't recall ever directly discussing with Ed Marison where his E ideas for RMS's indexed-file implementation came from, I do remember  F that his previous work was on MUMPS-11, which I think itself had some H kind of indexed file facility that may have provided at least *some* of  the inspiration for RMS's.   - bill   ------------------------------  % Date: Thu, 11 May 2006 17:03:45 -0400 3 From: "Richard B. Gilbert" <rgilbert88@comcast.net> O Subject: Re: We need a recommendation of good VMS backup (non-library) software 0 Message-ID: <Xt-dnY78Y_EvO_7ZRVn-uA@comcast.com>   John Vottero wrote:   . > <johnhreinhardt@yahoo.com> wrote in message ? > news:1147363888.554190.300470@y43g2000cwc.googlegroups.com...  > > >>Are any of these packages available with a Hobbyist license? >> >  > - > A JAMS hobbyist license is available.  See:  > ! > http://www.mvpsi.com/hobby.html  >  >  >    Hi John,  G JAMS is a very nice batch scheduler.  You can do things like running a  H job at 2130 every Wednesday except holidays.  And you can say it almost I exactly that way!  I had it set up to send e-mail to my pager whenever a  F System job failed.  It paged the applications people when their stuff * failed. I thought it was money well spent!   ------------------------------  % Date: Thu, 11 May 2006 20:59:51 -0400 / From: "William Webb" <william.w.webb@gmail.com> O Subject: Re: We need a recommendation of good VMS backup (non-library) software H Message-ID: <8660a3a10605111759s77de3ab3scfb1ac8b7d107c2@mail.gmail.com>  > On 5/11/06, Richard B. Gilbert <rgilbert88@comcast.net> wrote: > John Vottero wrote:  > / > > <johnhreinhardt@yahoo.com> wrote in message A > > news:1147363888.554190.300470@y43g2000cwc.googlegroups.com...  > > @ > >>Are any of these packages available with a Hobbyist license? > >> > >  > > / > > A JAMS hobbyist license is available.  See:  > > # > > http://www.mvpsi.com/hobby.html  > >  > >  > >  > 
 > Hi John, > H > JAMS is a very nice batch scheduler.  You can do things like running aI > job at 2130 every Wednesday except holidays.  And you can say it almost J > exactly that way!  I had it set up to send e-mail to my pager whenever aG > System job failed.  It paged the applications people when their stuff , > failed. I thought it was money well spent! >  >   , I second Richard's positive opinion of JAMS.  L We ran it on a cluster of systems running national stuff for a federal agen= cy.    Thousands of users.   8 MANY thousands of jobs spread out on hundreds of queues.  C Stone-Cold-Solid application; configurable as all get out, imports, " exports and user-friendly to boot.  A Oh, yeah.  Friendly and knowledgable product support on those odd 8 occasions when we had to talk with them about something.  C Their marketing folk asked me for a blurb which I would have gladly D given back then, but it was absolutely forbidden for me (both by theF federal agency and the contractor for whom I worked) to do so while on	 that gig.    WWWebb --=20 C NOTE: This email address is only used for noncommerical VMS-related  correspondence. C All unsolicited commercial email will be deemed to be a request for 8 services pursuant to the terms and conditions located at# http://bellsouthpwp.net/w/e/webbww/    ------------------------------   Date: 11 May 06 13:56:36 EDT) From: cook@wvnvms.wvnet.edu (George Cook) K Subject: Re: XtRemoveTimeOut() with value of 0 accvios under 7.3-2, but not ! Message-ID: <8VqLHVRebNpL@wvnvms>   : In article <lwH8g.1$8v4.0@fe05.lga>, Z <Z@ids.net> writes: > George Cook wrote: >> All the code I G >> have ever seen which uses XtAppAddTimeOut either validates the timer " >> or has a safety check for zero. >  > "Validate the timer?"  > ; > Is there an Xt...() function that can validate the timer?   B Sorry, "validate" was not the best word to use.  I know of no such Xt function.  = Is is up to the application to track the state of a timer and A only attempt to remove it if it is still set to go off.  Not much B can go wrong if you always set the timer id to zero when the timerD fires or when a "non-zero id timer" is manually removed.  Unless theE timer code is extremely complex and/or extremely broken, it shouldn't H take more than a few minutes to fix.  Even if there was an Xt validationH routine, it would be more work to implement than simply fixing the code.     George Cook  WVNET    ------------------------------  % Date: Thu, 11 May 2006 15:28:54 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> K Subject: Re: XtRemoveTimeOut() with value of 0 accvios under 7.3-2, but not , Message-ID: <44639071.139D59CA@teksavvy.com>   Z wrote: >  > George Cook wrote: > > All the code IH > > have ever seen which uses XtAppAddTimeOut either validates the timer# > > or has a safety check for zero.  >  > "Validate the timer?"     3 http://www.xfree86.org/4.3.0/XtAppAddTimeOut.3.html   G Also note that XtRemoveTimeout has been replaced by XtAppRemoveTimeOut. C This may explain why the old routine worked one way and then worked G differently when you got a new release of X if you continued to use the 	 old name.    ------------------------------  % Date: Thu, 11 May 2006 15:32:23 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> K Subject: Re: XtRemoveTimeOut() with value of 0 accvios under 7.3-2, but not , Message-ID: <44639141.75189050@teksavvy.com>   Hoff Hoffman wrote: H >    Is the timer null?  If it is, it's not valid, and it's a bug in the > calling code.   F Is there documentation that clearly states that getting a XtIntervalId? of 0 after calling XtAddTimeOut means that there was an error ?   H The doc I have seen doesn't say anything about it, and the routines hand' you an opaque value (the XtIntervalID).    ------------------------------  # Date: Thu, 11 May 2006 19:51:57 GMT , From: Hoff Hoffman <hoff-remove-this@hp.com>K Subject: Re: XtRemoveTimeOut() with value of 0 accvios under 7.3-2, but not 0 Message-ID: <xBM8g.387$mS4.260@news.cpqcorp.net>   Z wrote: > Hoff Hoffman wrote:   K > Hoff, checking for a non-zero value will not guarantee that the timer is  J > valid; it might be uninitialized data or the code might not have set it , > back to 0 after calling XtRemoveTimeOut().  I    Yes.  (Without intending any offense, that much is obvious; that there D are (probably) 2^32-1 wrong answers (or however big XtIntervalId is G declared), and one answer that's almost certainly always wrong (0), if  G there's just one timer outstanding?   :-)  Pointers and context blocks  H are like that -- bad I/O channel numbers are an equivalent construct on I OpenVMS.  Short of carnal knowledge of what's in the target structure or  @ the particular meaning of the context value -- something that's F certainly an inviting approach, but generally also not a good idea to G know about for reasons of upward-compatibility -- there's often no way  $ to verify these sorts of structures.  7 > Is there a function I can call to validate the timer?   "    If I knew of one, I'd tell you.  I > If I'm going to change this code, I want to change it *once* and do it  J > *correctly*. If there's a function available to validate, then I should  > be calling *that*.  G    When I can establish the correctness of all possible inputs, then I  G will have reached spiritual perfection.  (Sorry, having a bit of a Zen  I moment there. :-)  Perfection and correctness are not something that can  H be easily reached in a non-trivial environment -- if at all.  Much like H any good calculus function, you have to decide how much you are willing H to pay for how close you get to perfection, and the closer you get, the D more it all costs.  X Windows has generally chose to be rather more I cavalier about this sort of thing.  In this case, I'd certainly look for  I the zero value (and punt that), and I'd look at how I stored and managed  * the context values within the application.   > Jeez.   )    I'm not trying to make this difficult.   E    I know of no way to provide what you want within the context of X  I Windows, short of rebuilding the whole X Windows platform.  If Fred or I  F knew of a call that immediately solved this, we'd have certainly told 
 you about it.    ------------------------------  % Date: Thu, 11 May 2006 15:43:14 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> K Subject: Re: XtRemoveTimeOut() with value of 0 accvios under 7.3-2, but not + Message-ID: <446393CC.CDA6A30@teksavvy.com>   = > > Is there an Xt...() function that can validate the timer?   E The XtAppPending function returns a nonzero value if there are events @ pending from the X server, timer pending, or other input sourcesG pending. The value returned is a bit mask that is the OR of XtIMXEvent, F XtIMTimer, XtIMAlternateInput, and XtIMSignal (see XtAppProcessEvent).F If there are no events pending, XtAppPending flushes the output buffer and returns zero.       G Not quite what you want. But if, prior to XtAppAddTimeout the XtIMTimer D bit is not set, and after the call , it is set, then you'd know your call succeeded.     A It seems that having an timer event ID of 0 causes a problem.  So H checking for a value of 0 seems a fair practice since you know it causesF a problem later on. Looks to me like your XtAppAddTimeOut may not have the right arguments and fails.     Note:   G the old one :   XtIntervalId XtAddTimeout(interval, proc, client_data)  E the current one:  XtIntervalId XtAppAddTimeOut(app_context, interval,  proc, client_data)   ------------------------------  % Date: Thu, 11 May 2006 16:26:09 -0700  From: Z <Z@ids.net> K Subject: Re: XtRemoveTimeOut() with value of 0 accvios under 7.3-2, but not ( Message-ID: <jKP8g.675$Dv2.402@fe06.lga>   George Cook wrote: >>> All the code IH >>> have ever seen which uses XtAppAddTimeOut either validates the timer# >>> or has a safety check for zero.  >> "Validate the timer?"  < >> Is there an Xt...() function that can validate the timer?  D > Sorry, "validate" was not the best word to use.  I know of no such > Xt function.    Ok, thanks, that clears that up.   ------------------------------  % Date: Thu, 11 May 2006 07:44:01 -0400 * From: "FredK" <fred.nospam@nospam.dec.com>W Subject: Re: XtRemoveTimeOut() with value of 0 accvios under 7.3-2, but not under 7.1-2 * Message-ID: <4463237b@usenet01.boi.hp.com>  C "Z" <Z@ids.net> wrote in message news:FXA8g.130$2I1.125@fe05.lga...  > Hoff Hoffman wrote: F > > Assuming you are current with the installed DECwindows version andI > > with the DECwindows ECO kit (and if not, then I'd start with that), a : > > small reproducer would be useful in chasing this down. > . > Are the available ECO kits listed somewhere?  K I looked at the source.  There was a sanity check in the DEC code (in fact, K conditional #ifdef) to sanity check for a null ID.  In the upgrade to X11R6 J the routine changed in a way that the sanity check happens after the point8 that the ID is used to initialize an automatic variable.  G I'll ask the DECwindows group to fix the problem, but it probably isn't I something that we would do a special ECO for - since it is your code that I has the bug (and this would crash your program on a UNIX system as well). I If you have a support contract, feel free to file a bug report - in which - case they can probably get you a fixed image.    ------------------------------   End of INFO-VAX 2006.262 ************************