gnu@hoptoad.uucp (John Gilmore) (05/23/88)
Nathaniel Mishkin wrote: > Unifying Unix (especially > the BSD and System V derivative) is clearly a desirable thing. Then why are the companies that are so set upon derailing the unified Unix, the ones with a heavy investment in locking their customers into proprietary software? The only one missing is Apple, and I expect them to join any day now. > The problem is the way in which the unification was happening. > [att] requires companies to sign a..."heavyweight" contract... It's hard to believe that if half the Unix licensees refused to sign SVR3 licenses [due to provisions specifying that you had to ship it all, not just the parts you liked], that AT&T would stick with them, and continue to make them more heinous in future releases. I know AT&T is stupid about software, but they *are* trying to retain control of a major fast-growing market, and alienating most of your customers is not the way. > no one other than Sun or AT&T appeared to have any opportunity > to have input into the process. Are you giving AT&T the opportunity to tell you how to build Apollo's Unix? It sounds like that was exactly what you were complaining about in the "heavyweight" contract. You won't take it, but you'd like to dish it out. > Do end users really want to see the set of computer vendors > reduced because all but two are put at the disadvantage of getting the > latest software at a 6-12 month lag? This seems to be a bogeyman. Motorola has already announced that the Unix for its 88000 will be SVR4 and that they will get early versions of it so that they can make a release 'in the same timeframe' as AT&T and Sun. -- John Gilmore {sun,pacbell,uunet,pyramid,ihnp4}!hoptoad!gnu gnu@toad.com "Use the Source, Luke...."
gnu@hoptoad.uucp (John Gilmore) (05/23/88)
Nat Mishkin wrote: > contrary to comments otherwise, at least some software > that OSF distributes will be subject to licenses created by organizations > other than OSF. (Although presumably OSF will not distribute software > subject to the kind of "heavyweight" licenses currently associated with > System V Release 3.) I don't see the difference between depending upon OSF to "never make its licenses objectionable in the future" and depending upon AT&T to do the same. Is there a contractural committment from OSF to provide new releases on the same, or less onerous, terms as its first releases? If not, you have just switched from the devil you know to a new devil. Do you think IBM and DEC will each throw dozens of millions at this and then sit meekly if the little companies and universities who join this "nonprofit" vote to do something that hurts the big guys? Or is there voting at all? How are such decisions to be made (I presume this is settled now, since it is the key to the viability of the OSF)? Is there an "OSF Security Council" that gets veto power over things? I can't see OSF competing on price with AT&T, since AT&T's prices drop precipitously with volume; Xenix, for example, costs them $10/copy since they have shipped 300,000 copies. As the Unix market explodes, more and more companies will move to this price point. The only way to supply a cheaper Unix is to make it free -- which, of course, the OSF is deliberately NOT doing. It's the same old "licensed software" BS, sources to the big companies, binaries to the mortals, legalese before you get a line of it. Happily, due to the GNU license, they will not be able to steal any GNU software and ship it in binary unless they distribute its full sources at cost, thus aiding and abetting the GNU effort. > I'd rather place the state of my future software > into the hands of an organization structured along the lines of OSF than > into the hands of for-profit companies that have already at least partially > closed the door on outside input. Has any of the companies in OSF publicly stated that they will *not* buy any more software from AT&T? Or that they will exclusively use OS software from OSF? This business of putting your future "into the hands of xxx" is a strawman; none of these companies is required to use the OSF software, nor to eschew Unix, SunOS, VMS, MVS, Domain, or whatever. Do these companies really expect to stop shipping real Unix anytime soon, and/or to convert their proprietary stuff to quote "open" unquote software hastily assembled by a nonprofit committee? I don't believe it. If you can build a better Unix than what AT&T ships, you can keep running on your old AT&T license (and indeed many Unix companies are doing this now, e.g. they had SVR2 licenses and they toss the tape and ship Berkeley Unix, or a mix). Your company's future software is in your own hands. -- John Gilmore {sun,pacbell,uunet,pyramid,ihnp4}!hoptoad!gnu gnu@toad.com "Use the Source, Luke...."
gnu@hoptoad.uucp (John Gilmore) (05/23/88)
I see too much resemblance between "Open Software Foundation"
and "Free Software Foundation". Given its constituency, the only
thing I expect to be "open" about it is its mouth. While Apollo
might have learned how to survive in an open systems market (I'm
keeping an open mind about that :-), IBM and DEC hate it like poison.
The name is just a marketing gimmick, like the "Citizens for Decency,
Justice, and the American Way" type political committees.
If the Free Software Foundation feels that its name has been unfairly
infringed upon, I would be glad to back it in a lawsuit, and I suspect
that other GNU users would rally to its support.
OSF could have chosen to take an approach like GNU, but deliberately
spurned it. I personally put some of the Hamilton Group people in
touch with the GNU leadership, as did Rich $alz. While I could believe
that a bunch of lawyer-bound companies might not want Richard Stallman
in charge of their new Unix-clone project (though his track record so
far is amazingly good), they could have chosen to write their code
under the same terms (anyone can distribute sources for any price, but
with no restrictions on redistribution; distributing binaries at any
price requires you to distribute matching sources at copying cost for 3
years). GNU has written some large parts of Unix and more are under
way; OSF could have contributed much of the remaining work and come up
with a complete, modern, working, non-AT&T, public source code clone of
Unix. The fact that they didn't speaks volumes to me about their
motives.
They want to keep this software under corporate control. They will be
"open" with each other, not with their customers. The whole brouhaha
is a standard "FUD" (fear, uncertainty, doubt)-generating marketing
operation. AT&T and Sun have made an effort to make it possible to
run the same applications software on hundreds of manufacturer's machines.
If the OSF companies ship systems that, while compatible with Posix,
have lots of extraneous differences from Unix, portable applications will
be hard to find, and there will remain a market for applications that
run on VMS, MVS, Domain, and other proprietary systems. If Sun and
AT&T succeed, an applications company will be able to cover the whole
market by writing an application once, and the resulting depth and breadth
of applications will obsolete applications that run only on the
proprietary systems, thereby obsoleting the proprietary OS's. IBM,
DEC, HP, and Apollo have a lot to gain by making Sun and AT&T fail at this.
If/when OSF ships a product, their next move is to start claiming that
Sun and AT&T, who pushed the whole midrange computer market market wide
open(*), are pushing "proprietary" software. You read it here first, folks...
John Gilmore
(*) AT&T did it unwittingly, by licensing Unix out cheaply before
they could sell software, thereby making it possible for all kinds of
new hardware to come with compatible operating systems. Sun did it
deliberately, because it was a foot in the door, an advantage for
buyers that a small company could supply better than, say, DEC. It is
a struggle for Sun to stay open as it grows, but so far it seems equal
to the challenge.
--
John Gilmore {sun,pacbell,uunet,pyramid,ihnp4}!hoptoad!gnu gnu@toad.com
"Use the Source, Luke...."rogerk@mips.COM (Roger B.A. Klorese) (05/24/88)
In article <4628@hoptoad.uucp> gnu@hoptoad.uucp (John Gilmore) writes: >This seems to be a bogeyman. Motorola has already announced that the >Unix for its 88000 will be SVR4 and that they will get early versions >of it so that they can make a release 'in the same timeframe' as AT&T >and Sun. Everybody who's negotiated an ABI agreement has been told this. We still harbor doubts that those early releases will be sufficiently frozen that we CAN release something in the same timeframe as AT&T and Sun and have it be a genuine SVR4. -- Roger B.A. Klorese MIPS Computer Systems, Inc. {ames,decwrl,prls,pyramid}!mips!rogerk 25 Burlington Mall Rd, Suite 300 rogerk@mips.COM Burlington, MA 01803 I don't think we're in toto any more, Kansas... +1 617 270-0613
mishkin@apollo.uucp (Nathaniel Mishkin) (05/24/88)
In article <4628@hoptoad.uucp> gnu@hoptoad.uucp (John Gilmore) writes: >Nathaniel Mishkin wrote: >> Unifying Unix (especially >> the BSD and System V derivative) is clearly a desirable thing. > >Then why are the companies that are so set upon derailing the unified >Unix, the ones with a heavy investment in locking their customers into >proprietary software? On the face of it at least, the OSF is *not* trying to derail the creation of a unified Unix. They're just trying to unify Unix by an alternate method that they feel is superior, clearly for themselves, but also for the market at large. Your statement is not so much a response as it is a claim that AT&T and Sun have the sole authority to unify Unix and to do so under the terms they specify. >It's hard to believe that if half the Unix licensees refused to sign >SVR3 licenses [due to provisions specifying that you had to ship it >all, not just the parts you liked], that AT&T would stick with them, >and continue to make them more heinous in future releases. Apparently, a number of organizations and companies *have* refused to sign. Is "half" a magic number? Maybe other, smaller companies have been afraid to refuse to sign for fear they'd be screwed (either intentionally or not). Does lack of political dissent in the Soviet Union mean that everyone is happy? (OK, that's ad hominem [ad corporatum?], but you get the idea.) Anyway, it's not so hard to believe that they'd make the terms more heinous over time. All they have to do is wait for a couple of more releases and get everyone hooked and then they can do whatever they damn well please, can't they? Are we basing corporations on wishful thinking about competitors these days? >> no one other than Sun or AT&T appeared to have any opportunity >> to have input into the process. > >Are you giving AT&T the opportunity to tell you how to build Apollo's Unix? >It sounds like that was exactly what you were complaining about in the >"heavyweight" contract. You won't take it, but you'd like to dish it out. Is Apollo trying to push its version of Unix on the rest of the world? Or more to the point, could it? If not, this is a non-issue. Come back and complain about Apollo's policies when Apollo is trying similar moves. >> Do end users really want to see the set of computer vendors >> reduced because all but two are put at the disadvantage of getting the >> latest software at a 6-12 month lag? > >This seems to be a bogeyman. Motorola has already announced that the >Unix for its 88000 will be SVR4 and that they will get early versions >of it so that they can make a release 'in the same timeframe' as AT&T >and Sun. And I'm sure all the Sun & AT&T salesman will add to their pitch "And by the way, before you make your purchasing decision, you really should wait 6 months for our competitors to come out with exactly what we offer right now". -- -- Nat Mishkin Apollo Computer Inc. Chelmsford, MA {decvax,mit-eddie,umix}!apollo!mishkin
mishkin@apollo.uucp (Nathaniel Mishkin) (05/24/88)
In article <4629@hoptoad.uucp> gnu@hoptoad.uucp (John Gilmore) writes: >I don't see the difference between depending upon OSF to "never make its >licenses objectionable in the future" and depending upon AT&T to do the same. >Is there a contractural committment from OSF to provide new releases on the >same, or less onerous, terms as its first releases? If not, you have just >switched from the devil you know to a new devil. Perhaps. At least the new devil is not a competitor. At least it's starting off with a different model of operation about software licensing. Sure, the OSF could turn out to be Godzilla of the software world, but I don't expect it to be. The fur is already showing on AT&T though... >Do you think IBM and DEC will each throw dozens of millions at this and >then sit meekly if the little companies and universities who join this >"nonprofit" vote to do something that hurts the big guys? Or is there >voting at all? How are such decisions to be made (I presume this is >settled now, since it is the key to the viability of the OSF)? Is there an >"OSF Security Council" that gets veto power over things? The model is that the OSF is a *company* and it is run by the staff of the company, just like every other company. Sure, the board of directors can fire the president and everyone under him if they think they're doing the wrong thing, but is that going to happen? Is it going to happen because they picked semantics A for "signal(2)" versus semantics B? Not very likely. If that happens, the OSF will have failed, and I think all the sponsors understand this. >Has any of the companies in OSF publicly stated that they will *not* >buy any more software from AT&T? Or that they will exclusively use OS >software from OSF? This business of putting your future "into the >hands of xxx" is a strawman; none of these companies is required to use >the OSF software, nor to eschew Unix, SunOS, VMS, MVS, Domain, or >whatever. Do these companies really expect to stop shipping real Unix >anytime soon, and/or to convert their proprietary stuff to quote "open" >unquote software hastily assembled by a nonprofit committee? I don't >believe it. It's a direction, not an instantaneous event. I believe the sponsors have said "We can't keep going on this way, let's find a new way". >If you can build a better Unix than what AT&T ships, you can keep >running on your old AT&T license (and indeed many Unix companies are >doing this now, e.g. they had SVR2 licenses and they toss the tape and >ship Berkeley Unix, or a mix). Your company's future software is in >your own hands. It could well *not* be in your own hands if you make a technical and market-oriented decision that you *have* to support a particular piece of software, base your company on it, and then have your chain yanked 2-3 years down the line. The only "choice" one has is to find a different direction. BTW, I object to the various speculation (on the part of several people) on the topic of the "ulterior motives" of DEC and IBM. (No one cares about Apollo's ulterior motives, I guess :-) "They really just want to wreak havoc", it is said. Aren't AT&T and Sun subject to ulterior motives? "Get them hooked on a standard *we* define and that *they're* contractually obligated to support and then add some whacko feature that kills some piece of added value our major competitor has managed to eek out". Geez. Talk about white hats and black hats! -- -- Nat Mishkin Apollo Computer Inc. Chelmsford, MA {decvax,mit-eddie,umix}!apollo!mishkin
dave@sdeggo.UUCP (David L. Smith) (05/24/88)
Where are Mt. Xinu, Interactive Systems, and the rest of the companies that
only do Unix ports in this battle? These should be the companies forming
consortiums to produce a "standard" Unix. I think that the only way to have
a "fair" and open operating system is to have it developed by a company that
does not make hardware. If I were going to be making computers, it would
be pretty stupid to rely on a major competitor for one of the most crucial
parts of my system.
AT&T should get out of the computer business and go back to just producing
software, or quit making noises about creating an "open" system.
--
David L. Smith
{sdcsvax!jack,ihnp4!jack, hp-sdd!crash, pyramid, uport}!sdeggo!dave
sdeggo!dave@amos.ling.edu
Sinners can repent but stupid is forever.guy@gorodish.Sun.COM (Guy Harris) (05/24/88)
> BTW, I object to the various speculation (on the part of several people) > on the topic of the "ulterior motives" of DEC and IBM. (No one cares > about Apollo's ulterior motives, I guess :-) "They really just want > to wreak havoc", it is said. Aren't AT&T and Sun subject to ulterior > motives? "Get them hooked on a standard *we* define and that *they're* > contractually obligated to support and then add some whacko feature that > kills some piece of added value our major competitor has managed to eek > out". Geez. Talk about white hats and black hats! Naah, let's not talk about white hats and black hats. If you choose your set of beliefs carefully, you can put the aforementioned hats on any heads you want to. Claiming that AT&T and Sun are really trying to come up with a standard that they can control just so that they can screw their competitors is no less objectionable than is claiming that DEC and IBM are really trying to muddy the waters so that *they* can either keep UNIX from eating into other OSes that they support. I agree that that a lot of the speculation about the motives of the OSF members is random noise. The same can, however, be said about a lot of the speculation about the motives of AT&T and Sun. Again, I ask: can we please keep the speculation and random DEC/Sun/IBM/AT&T/... bashing down to a dull roar, please? I realize this is USENET, where idle uninformed speculation and vendor-bashing is a fine art, but could we try to make an exception, just this once?
sas@pyrps5 (Scott Schoenthal) (05/24/88)
After some thought, I've decided to add my (rather long-winded) two cents. I think that the recent statements by <mishkin@apollo.UUCP> display a naive loyalty to his company's viewpoint. >Perhaps. At least the new devil is not a competitor. At least it's starting >off with a different model of operation about software licensing. Sure, >the OSF could turn out to be Godzilla of the software world, but I don't >expect it to be. The fur is already showing on AT&T though... >It's a direction, not an instantaneous event. I believe the sponsors >have said "We can't keep going on this way, let's find a new way". I think that if IBM, DEC, et al., were truly interested in delivering an open implementation of Mumblix (or whatever their derivative of Un*x is to be called), the money would have been best spent in funding an independent effort (e.g., FSF or a university). I have no doubt that OSF will do what is best for the OSF members. That's fine. It's their money. Just as, has been recently implied, AT&T and Sun will do what is best for themselves. Please don't represent OSF as a white knight riding in from the East (or wherever). Although I mostly agree with comments made by <gnu@hoptoad.UUCP>, I have a problem with his views on AT&T: > AT&T and Sun have made an effort to make it possible to > run the same applications software on hundreds of manufacturer's machines. > If the OSF companies ship systems that, while compatible with Posix, > have lots of extraneous differences from Unix, portable applications will > be hard to find, and there will remain a market for applications that > run on VMS, MVS, Domain, and other proprietary systems. My current impression (admittedly somewhat uninformed) is that the ABI standard (I assume this is what John is referring to) is a crock. Given the recent explosion of micro-processor designs which have come out in the last year, the notion of an ABI seems to me to be useless. The current System V release is based upon the 3b2 architecture. My understanding is that future releases will be based upon the SPARC architecture. In this sense, no effort has been made by AT&T/Sun to "make it possible to run the same applications software on hundreds of manufacturer's machines." It will be interesting to see what machine the OSF people choose to use as their architecture base. PC/RT, perhaps? :-) > If/when OSF ships a product, their next move is to start claiming that > Sun and AT&T, who pushed the whole midrange computer market market wide > open(*), are pushing "proprietary" software. You read it here first, folks. But, it is proprietary. Customers have to sign licenses. Customers have to pay money. Customers have to perform certain obligations (e.g., passing SVVS). Is this not 'proprietary'? Finally, my gut feeling is that 5 years from now we'll all have a good laugh over this while professing that each of us knew that "X" was the way to go in 1988 where "X" corresponds to whatever the Un*x world looks like in 1993. sas "I speak for myself only." ---- Scott Schoenthal sas@pyrps5.pyramid.com Pyramid Technology Corp. {sun,hplabs,decwrl}!pyramid!sas
che@pbhyf.PacBell.COM (Mitch Che) (05/24/88)
In article <3c3a336e.13422@apollo.uucp> mishkin@apollo.UUCP (Nathaniel Mishkin) writes: >BTW, I object to the various speculation (on the part of several people) >on the topic of the "ulterior motives" of DEC and IBM. Even the experts can only speculate, but only DEC is confused enough to try and convince the world that they are all for UNIX (whatever its form) and/or "open systems" on the one hand, while its president talks up VMS and complains about UNIX on the other. (Not too long ago, we requested a DEC microVAX II with Ultrix; DEC shipped it to us with VMS. That says something, but I don't know what.) IBM and DEC would prefer to have to deal with only proprietary operating systems, period. If UNIX disappeared off the face of the earth, we'd probably hear a (quiet) round of applause from DEC and IBM. But that's OK, that's business. No one expects them to be altruistic. I don't expect that from AT&T, I certainly don't expect it from Big Blue. I do expect a multi-vendor choice and if OSF is the way, that's fine. It might even work. After all, DEC and IBM have SUCH a fine track record of supporting a competitive hardware and operating system environment! (Personally I choose from the wide selection at "VMS 'r US" or was that "MVS 'r US"? :-) -- Mitch Che Pacific Bell 415-823-2454 "What, Me Worry?" --------------------------------------- Ed "10-20" Meese disclaimer, disclaimer, too uucp:{ames,bellcore,sun}!pacbell!pbhyf!che
guy@gorodish.Sun.COM (Guy Harris) (05/24/88)
> My current impression (admittedly somewhat uninformed) is that the ABI > standard (I assume this is what John is referring to) is a crock. Given > the recent explosion of micro-processor designs which have come out in the > last year, the notion of an ABI seems to me to be useless. Maybe, maybe not. If you expect every box out there to have its own architecture, yes. If you expect some number of architectures to be dominant, so that there are a large number of boxes with a particular architecture (68K, 80*86, MIPS, SPARC, whatever), probably not - if the ability to package one version of a particular piece of software (or, at least, one per distribution medium) and have it run on *all* boxes using members of a particular processor family causes lots more of this software to be available at reasonable prices, an ABI for that processor family is far from useless. Note that the phrase "*the* ABI standard" has no unique referent; the SPARC ABI is just one of many ABIs - AT&T presumably intends to support both the '386 and SPARC ABIs, as they currently have a '386 machine and will have SPARC machines, and Sun will probably end up supporting those and quite possibly a 68K ABI as well. (Note: I obviously don't speak for AT&T here, and can't speak for Sun either; I don't make those decisions.) > The current System V release is based upon the 3b2 architecture. My > understanding is that future releases will be based upon the SPARC > architecture. In this sense, no effort has been made by AT&T/Sun to > "make it possible to run the same applications software on hundreds of > manufacturer's machines." AT&T has *never* made such an effort, in the sense that they have never shipped UNIXes for every architecture in existence. They have promoted the "microports" of S5 to various chip families; I think the ports themselves were done outside AT&T. I would *personally* like to see the source release of UNIX provide support for some non-singleton set of processors - Sun builds its Sun-2/Sun-3 and Sun-4 versions from common source, and builds most of its Sun386i version from that source as well - but the problem here is "which processors get on the tape?" '386? SPARC? AT&T currently makes a '386 machine (or, at least, sells one; I don't know if they, Olivetti, or somebody else makes it), and eventually plans to make a SPARC machine. 68K? While Sun may make 68K machines, AT&T currently doesn't. MIPS? AT&T doesn't make a MIPS machine. VAX? Clipper? Cray-2? etc., etc., etc..
csg@pyramid.pyramid.com (Carl S. Gutekunst) (05/24/88)
Soapbox time, I'm getting irritated. [Don asbestos suit.] Nearly all the postings seem to make two assumptions that I have trouble with: - The presense of DEC, Apollo, and IBM in the Open Software Foundation has been widely accepted as proof that OSF is really only a plot to destroy UNIX and muddy the marketplace. This completely ignores the presense of Nixdorf, Siemens, and HP, all of which have made major commitments to UNIX. Nixdorf in particular has spent a tremendous amount of labor over the past few years converting their entire systems product line to UNIX. Both Nixdorf and Siemens struck strategic technical relationships with American pure-UNIX companies several years ago already (with Pyramid Technology and Sequent, respectively). These are hardly companies that want to see UNIX fail. - WHO SAYS unifying UNIX is *that* important? I've argued from the start that the Sun/AT&T deal was a waste of precious programmer resources. OSF is just piled higher and deeper. UNIX is an appalling mediocre commercial operating system. There is a *lot* of really important work that needs to be done be- fore most commercial customers are going to trust *any* UNIX box replacing their IBM 3090 running VM. And it's *not* getting done, because the lion's share of programmer resources are being gobbled up by Sun, AT&T, and now OSF to diddle with pedagogical puzzles. Anyone who thinks unified UNIX is *that* important should consider two things: - How difficult is it *really* to port applications to the different UNIX variants? There are definitely a few pieces I wish they all had in common -- for example, the V-Node filesystem, shared memory. But it doesn't take any major effort to get these to work. And even for really big commercial appli- cations engines (e.g., Sybase, Oracle, Unify, Informix) you'll have a much easier time going between System V and BSD than you will between UNIX and VMS. - There are lots of different sized platforms running System V. There are lots of different sized platforms running Berkeley UNIX. Pick your application, then pick a system to run it. What's the problem? People upset because the application the runs on your 3B2 won't run on your ISI? For pity's sake, it won't run on your Amiga or your 3090 either; why are you so irritated that the UNIX systems differ a little, when you've tolerated big differences in your other systems for so long? I'm not saying a unified UNIX isn't important. Some day, it will be a major priority. But we have far too many other important things to do first. (I'll be happy to offer a laundry list, if anyone's listening.) These are my personal opinions, of course. I'm quite sure Pyramid has no use for them. :-) <csg>
schwartz@swatsun.uucp (Scott Schwartz) (05/24/88)
In article <3c3a336e.13422@apollo.uucp> mishkin@apollo.UUCP (Nathaniel Mishkin) writes: >Perhaps. At least the new devil is not a competitor. At least it's starting >off with a different model of operation about software licensing. Sure, >the OSF could turn out to be Godzilla of the software world, but I don't >expect it to be. The fur is already showing on AT&T though... Digital, IBM, and Apollo not competitors? Wait, I have to sit down... Anyway, wasn't Godzilla a lizard? (no hair) :-) :-) But seriously, if licensing is the issue, why not just finance the FSF? That way everyone gets a top notch free unix. If input into the standardization process is the issue, why not either support the FSF, or like Sun, Motorola, and others, join with AT&T? I think the real issue is that unix is now (or is going to be, depending on how you feel) an important force in the computer world, and AT&T OWNS it. Competing corporations just can't tolerate something important that the other guy controls. No matter what the outcome of the OSF effort, it just can't be worse that the status quo for the coalition members. If they succeed in promulgating a new standard, they win in the market. If they simply slow down unix and promote {VMS,MVS,Aegis...} then they win in the market. If nothing comes of it except that they get a unix clone sans AT&T license restrictions, well at least they are free of AT&T, and so they win in the market. >BTW, I object to the various speculation (on the part of several people) >on the topic of the "ulterior motives" of DEC and IBM. (No one cares >about Apollo's ulterior motives, I guess :-) Yup. Says a lot about how people feel about DEC and IBM. But why do you object to such speculation? Truth of accusition is an acceptable defense against charges of slander. Unix politics aside, the OSF does have all sorts of ulterior motives. You and I are concerned with technical issues, and with promoting things that benefit the unix community. The ceo of any given computer company may or may not share those concerns. > Geez. Talk about white hats and black hats! Not to mention expensive business suits. Can we all go back to being computer scientists now? Or should we petition for the creation of comp.unix.mba? -- Scott Schwartz, schwartz@swarthmore.edu, psuvax1!vu-vlsi!swatsun!schwartz
hutch@net1.ucsd.edu (Jim Hutchison) (05/24/88)
<3c3a336e.13422@apollo.uucp> mishkin@apollo.UUCP (Nathaniel Mishkin) writes: >BTW, I object to the various speculation (on the part of several people) >on the topic of the "ulterior motives" of DEC and IBM. (No one cares >about Apollo's ulterior motives, I guess :-) "They really just want >to wreak havoc", it is said. Aren't AT&T and Sun subject to ulterior >motives? "Get them hooked on a standard *we* define and that *they're* >contractually obligated to support and then add some whacko feature that >kills some piece of added value our major competitor has managed to eek >out". Geez. Talk about white hats and black hats! Speculation? Who just made a funny shaped pc box so that they could cut out other vendor's add-in boards? IB-something? What 3 letter company is relatively famous for sitting on the meaning of error codes generated by its personal computers self-test? IB-somthing-else? Please! DEC, large volumes of documentation which tell you how to break into the system. :-) After having VMS/Bliss locked up in $$$ for so many years, how are we supposed to believe that they have a need to change their tune? Why OFS? What is wrong with an ANSI/IEEE/ACM/OSI/... standards commitee? Not that I am overly fond of commitees, but a little iterative specification couldn't hurt. Start with POSIX, and revise it as appropriate, as we learn what it most appropriate. I have enjoyed Apollo computers, the network root concept is very very nice. I have not worked with your Unix-oid which ran under Domain in several years. How POSIX acceptable is your OS? >... Geez. Talk about white hats and black hats! Gosh, I'm naive. Good will has gone the way of the passenger pigeon. Why not set up an interaction. Yes, make it mutually profitable for you and them to remain compatible. Admittedly this goes contrary to some notions of commerce, so I'll just learn Japanese. If we get stuck in such a destructive loop, we shall in fact be their "farmland" soon enough. sigh. This discussion is really depressing. Jim Hutchison UUCP: {dcdwest,ucbvax}!cs!net1!hutch ARPA: Hutch@net1.ucsd.edu Disclaimer: The cat agreed that it would be o.k. to say these things.
sullivan@vsi.UUCP (Michael T Sullivan) (05/24/88)
In article <3c3a336e.13422@apollo.uucp>, mishkin@apollo.uucp (Nathaniel Mishkin) writes: > > BTW, I object to the various speculation (on the part of several people) > on the topic of the "ulterior motives" of DEC and IBM. (No one cares > about Apollo's ulterior motives, I guess :-) "They really just want > to wreak havoc", it is said. Aren't AT&T and Sun subject to ulterior > motives? "Get them hooked on a standard *we* define and that *they're* I can't see why you have objections. I myself am suspicious of a company that is supposed to write a unified Unix being headed by DEC and IBM, who would like nothing more than to see Unix go away. Despite what DEC and IBM execs say, I believe that is still the case. People's objections are understandable. -- Michael Sullivan {uunet|attmail}!vsi!sullivan sullivan@vsi.com HE V MTL Anybody out there remember Max Webster?
runyan@hpirs.HP.COM (Mark Runyan) (05/24/88)
While I tend to agree with Scott, I wanted to make a minor point... >/ hpirs:comp.unix.wizards / sas@pyrps5 (Scott Schoenthal) >I think that if IBM, DEC, et al., were truly interested in delivering an open >implementation of Mumblix (or whatever their derivative of Un*x is to be >called), the money would have been best spent in funding an independent effort >(e.g., FSF or a university). It seems that OSF will be funding some university research. Does this count for something? >I have no doubt that OSF will do what is best for the OSF members. Yes, but can't anyone become an OSF member (including AT&T)? This is supposed to be an organization for standards, and I don't remember them saying that it was going to be an elite organization. >That's fine. It's their money. Just as, has >been recently implied, AT&T and Sun will do what is best for themselves. >Please don't represent OSF as a white knight riding in from the East (or >wherever). I won't represent them as White Knights, but I don't see them as Black Villians, either (yet?). Of course, I have a biased point of view, I suppose. Mark Runyan Hewlett-Packard, Cupertino, CA. , USA {ucbvax|hplabs}!hpda!runyan
davidsen@steinmetz.ge.com (William E. Davidsen Jr) (05/25/88)
In mishkin@apollo.UUCP (Nathaniel Mishkin) writes: | And I'm sure all the Sun & AT&T salesman will add to their pitch "And | by the way, before you make your purchasing decision, you really should | wait 6 months for our competitors to come out with exactly what we offer | right now". Honestly, are there that many people who insist on having a new machine the moment a new version of the o/s is available? If a group is running Suns, or Apollo, or PC/RT, will they change just to get something a few months earlier? It looks to me as though currently AT&T has had the source code earlier than anyone else, and that obviously has not let them dominate the hardware market. In fact they barely survive in the range above PCs, and the unix-pc, which would have been a killer at the right price, just plain didn't sell. I think it would be justified to (a) have AT&T release a "work in progress" source when something reasonably works, (b) give the final version as a set of deltas, so that customized code could be upgraded, and (c) hold release of the SRVR4 until 3 months after the code was frozen to give other vendors a chance to be in the right timeframe. Of course they would have to feed back any enhancements and bug fixes to AT&T, as Sun says they have agreed (as in make it work better, not giving up totally new features) rather than call bug fixes "proprietary enhancements." Then AT&T perhaps could agree to allow systems to be labeled "SRVR4" if they met SVVS, "POSIX conforming" if they satisfied posix (if and when), and something like "UNIX-based" if there was AT&T code. As a user I don't want to have something labeled SRVR4 if it won't run my programs, while not forbidding any sub/supersets as long as I am warned that they may not run my programs. There is a need for a validation suite for BSD as well, as anyone who runs programs in a multi-vendor BSD environment knows. Of course NONE of the big players has the user in mind, except as a source of revenue. If they can do something for the user while not hurting sales, or helping a competitor, they will for good will, but don't expect them to say "it's only money." As a user I see SysV and BSD running on parallel courses. The idea of a merged UNIX is a great one, and in the long run will be better for the vendors, too. Would Apollo like to stop supporting two flavors of UNIX on their systems? And how do they feel about supporting a third version if OSF doesn't knock UNIX out of the market. AT&T said somewhere (or was quoted as saying) that at some time they might turn UNIX over to a marketing group, vendor independent. Perhaps some of the best people could be stationed on such an organization, mush as Sun and AT&T are sharing people and code now. Finally I think that OSF is a clear effort to infringe on the repution of FSF, and that it should be promptly renamed or sued. -- bill davidsen (wedu@ge-crd.arpa) {uunet | philabs | seismo}!steinmetz!crdos1!davidsen "Stupidity, like virtue, is its own reward" -me
davidsen@steinmetz.ge.com (William E. Davidsen Jr) (05/25/88)
In article <24369@pyramid.pyramid.com> sas@pyrps5.pyramid.com (Scott Schoenthal) writes: | But, it is proprietary. Customers have to sign licenses. Customers have to | pay money. Customers have to perform certain obligations (e.g., passing | SVVS). Is this not 'proprietary'? The only reason SRV has been as portable as it has is that there is a reasonable validation suite, and you must meet it. It is not perfect, but to the user it gives a fighting chance that system will support a product. Ada used to be that way, but now that there is no longer a requirement to pass the validation suite, I see some real crap compilers called Ada. Does it hurt the other vendors to have to deliver a working product? Obviously they think so, they put up $90 mil to be able to do what they want with they version of UNIX, making it probably that there will be 100 flavors of OSFix, AIX, or whatever. This will provide a nice set of proprietary o/s to keep the vendor happy. Will we see a "better" filesystem from one? A "better" shell from another? A "replacement" for curses which isn't call for call compatible? If the OSF version really goes anywhere, I'd bet on it. -- bill davidsen (wedu@ge-crd.arpa) {uunet | philabs | seismo}!steinmetz!crdos1!davidsen "Stupidity, like virtue, is its own reward" -me
gallen@apollo.uucp (Gary Allen) (05/25/88)
In article <4630@hoptoad.uucp> gnu@hoptoad.uucp (John Gilmore) writes: >I see too much resemblance between "Open Software Foundation" >and "Free Software Foundation". Given its constituency, the only >thing I expect to be "open" about it is its mouth. While Apollo >might have learned how to survive in an open systems market (I'm >keeping an open mind about that :-), IBM and DEC hate it like poison. >The name is just a marketing gimmick, like the "Citizens for Decency, >Justice, and the American Way" type political committees. > >If the Free Software Foundation feels that its name has been unfairly >infringed upon, I would be glad to back it in a lawsuit, and I suspect >that other GNU users would rally to its support. >..... >They want to keep this software under corporate control. They will be >"open" with each other, not with their customers. The whole brouhaha >is a standard "FUD" (fear, uncertainty, doubt)-generating marketing >operation. AT&T and Sun have made an effort to make it possible to >run the same applications software on hundreds of manufacturer's machines. >..... >run on VMS, MVS, Domain, and other proprietary systems. If Sun and >AT&T succeed, an applications company will be able to cover the whole >market by writing an application once, and the resulting depth and breadth >of applications will obsolete applications that run only on the >proprietary systems, thereby obsoleting the proprietary OS's. IBM, >DEC, HP, and Apollo have a lot to gain by making Sun and AT&T fail at this. > >If/when OSF ships a product, their next move is to start claiming that >Sun and AT&T, who pushed the whole midrange computer market market wide >open(*), are pushing "proprietary" software. You read it here first, folks... > > John Gilmore >...... First of all, its unlikely that 7 (ok, 6 1/2) of the worlds giant corporations are sufficiently freaked out by GNU that they are out to steal FSF's thunder. C'mon, how about a reality break? If reality is not your bag (it isn't always mine either), feel free to strike fear into the heart of 60 billion/year IBM, 12 billion/year DEC, ...... Lawsuits? You're talking about the people who invented the word. The HUGE majority of application software DOES run on proprietary OS's, or are you unaware that VMS outsells UNIX on VAXen 15 or so to 1? Are you unaware that MS-DOS (PC-DOS if you prefer) is proprietary, the most common OS on earth, the most portable, and probably also the worst? While I like FSF and the idea of GNU (probably from all the drugs I did in the'60s), I again suggest a reality pill. The whole market? REALITY dude, REALITY. These 7 companies ARE the vast majority of the market. Obsolete IBM, DEC, HP, etc. On second thought, DRUGS dude DRUGS! Just in case you haven't heard Apollo's position before, I'll try to summarize it, but you get the standard disclaimer, I don't represent Apollo. The deal is this. UNIX is not and never was an "Open System". You bought it (usually indirectly) from AT&T. AT&T put what they wanted in it, left out what they wanted, just like DEC, IBM, Apollo, etc. Open? When was the last time you had anything to say about what UNIX is/isn't? And, if you think that the Sun/AT&T UNIX will not favor particular machine architectures such as SPARC and 3Bx, then I'd like to talk to you about some bridges that I have for sale. The idea that Sun and AT&T are in it for karma and the rest of us are just in it for the money is pretty bizzare. Each of the companies involved in OSF have particular needs/focus that must be addressed by any PORTABLE OS, which is really what you're after isn't it? For instance, whose notion of ISAM files will be used in UNIX? I don't know, do you? Perhaps AT&T will get around to it as THEY need it. Otherwise, we have to provide interim solutions that our customers will use until there is some sort of "standard". Then, our customers will have to rewrite their applications to match. Hardly portable. Finally, Sun is our major competitor (or is it lost on you that we have an installed base of >60,000 workstations). It is simply unfair of our Licensor [is that really a word?] to put our major competitor in such a position. The day that Sys5.Next hits the streets, it will be available on Sun hardware. We at Apollo will have to go to the lab with it for a few months (you know, porting, QA, etc). That's hardly "open". We were told point blank by AT&T that that was going to be the way it is. They refused to make any concessions (even though we pay the same license fees) vis-a-vis the time-to-market problem. Our options were: A) Sit in the back of the bus B) Get off the UNIX bus C) Start a new bus company. Put yourself in our place, what would you have done? And by the way, FSF was never considered, thought about, infringed upon, or even mentioned. The name came from "the suits" who can barely pronounce the word "software" (you know the type, they use words like "focus", "thrust", and "calendar intensive"). Sorry about the flame, Gary Allen Apollo Computer Chelmsford Ma Even I am not responsible for my opinions. I learned this from Reagan.
lkw@csun.UUCP (Larry Wake) (05/25/88)
In article <24369@pyramid.pyramid.com>, Pnews swears sas@pyrps5.pyramid.com (Scott Schoenthal) wrote: >It will be interesting to see what machine the OSF people choose to use as >their architecture base. PC/RT, perhaps? :-) "IBM's AIX-based RT PC may be used as the porting base for the software environment." "Anti-AT&T UNIX Group Forms," Jim Duffy; MIS Week, May 23, 1988 -- Larry Wake lkw@csun.edu CSUN Computer Center uucp: {hplabs,rdlvax}!csun!lkw Mail Drop CCAD sun!tsunami!valley!csun!lkw Northridge, CA 91330 BITnet: LKW@CALSTATE
ok@quintus.UUCP (Richard A. O'Keefe) (05/25/88)
In article <3c3fdf1b.4bee@apollo.uucp>, gallen@apollo.uucp (Gary Allen) writes: > if you think that the Sun/AT&T > UNIX will not favor particular machine architectures such as SPARC and 3Bx, then > I'd like to talk to you about some bridges that I have for sale. > Gary Allen > Apollo Computer Well, I think that SUN just _might_ want to favour a couple of antique no-hoper architectures called, oh what were they, 80386 and 68020 (you know 68020s, the things a lot of Apollos come with...). With SUN wanting merged UNIX to run on 68020s, 80836s, and SPARCs, and AT&T wanting it to run on 3Bs, sounds like porting S5R4 ought to be a lot more portable than any previous version (to start with, the 68020 and 80386 have opposite byte sex). What puzzles me is what took the OSF so long. The SUN/AT&T UNIX merge has been publicised for a couple of years.
reggie@pdn.UUCP (George W. Leach) (05/25/88)
In article <3c3fdf1b.4bee@apollo.uucp> gallen@gallen.UUCP (Gary Allen) writes: >The HUGE majority of application software DOES run on proprietary OS's, or >are you unaware that VMS outsells UNIX on VAXen 15 or so to 1? Ah, but how many VAX sales have been lost to DEC due to other competition in the marketplace who only depend upon UNIX? Are you aware that UNIX outsells VMS on just about everything else :-) >Are you unaware that MS-DOS (PC-DOS if you prefer) is proprietary, the most >common OS on earth, the most portable, and probably also the worst? ^^^^^^^^^^^^^^^^^ Excuse me? Want to talk about reality pal. Just take a good look at what you said and *think* about it. >The whole market? REALITY dude, REALITY. These 7 companies ARE the vast >majority of the market. Obsolete IBM, DEC, HP, etc. Well then if they are the "vast majority of the market" and the "HUGE majority of application software DOES run on proprietary OS's", then why should they worry themselves about UNIX???????? -- George W. Leach Paradyne Corporation ..!uunet!pdn!reggie Mail stop LF-207 Phone: (813) 530-2376 P.O. Box 2826 Largo, FL 34649-2826
karl@triceratops.cis.ohio-state.edu (Karl Kleinpaste) (05/25/88)
Climbing momentarily onto a soapbox... I'm really, really tired of hearing all the OSF sponsors moaning that "AT&T & Sun's relationship will put all the other companies at a disadvantage for N months after release of SysVRel(M+1) while we port it to our machines." Question: So what else is new? When was the last time that AT&T involved its competitors heavily in a new release of SysV? How many other companies had a significant part in the actual development of SysVRel4? Who helped AT&T with SysVRel3? What co-sponsors were there of SysVRel2? All other companies, including Sun, have been playing catch-up to AT&T's releases since Day One. This is nothing new. Claims to the contrary by OSF sponsors are red herrings, useless in logical discussion due to the evidence of the past weighing against them. Nothing in the AT&T/Sun plan changes that. The A/S plan provides for the development of the merged UNIX system - little more than that - and therefore for the day when I won't have to say "#ifdef BSD" and mutter the same functional phrases in my programs twice. OSF, if it goes anywhere at all, will force me right back into this mode with "#ifdef OSFMumblix" and a different pair of code fragments to do the same things 2 ways. A short digression concerning the need for a merged UNIX: I disagree with claims (from folks like csg@pyramid and others) that the merge is not needed, not required, not being requested by customers. Sorry, csg, I *can't* agree - Pyramid in particular is in a funny position with its dual-universe port. Pyramid's customers aren't asking for it because they've already got it - and I say that as a Pyramid customer. But Pyramid's solution (dual port) is different from Sun's solution to date (partial merge) which is different from HP's solution (SysVified 4.2BSD with retrofitted enhancements) which is different from... everybody else's solution to the problem of getting both BSD and SysV capabilities into a single box. How much time does each company which advertises some variant of "both universes" or "SysV with BSD enhancements" or "pick-your-buzzphrase" spend in the creation of their local incarnation of a merge? I think it must unavoidably be rather a lot. Pyramid supports 4.3BSDish UUCP in the UCB universe plus HDB in the ATT universe... It just occurred to me that HP, while joining in the chorus of "we're at a disadvantage due to delay," has voluntarily put itself in an even worse position: HP wants to be SysV, thus forcing it to wait on AT&T releases, and supports NFS and YP, thus forcing it to wait on Sun releases. HP has no business claiming that they're suddenly in a worse position than they used to be in light of the A/S plan; if anything, their timing will be *improved* by getting an entire release from a single source at one time. Now I have to go work on implementing an idea or two I've got for motivating OSF sponsors to abandon OSF... Stepping off the soapbox, --Karl
barnett@vdsvax.steinmetz.ge.com (Bruce G. Barnett) (05/25/88)
In article <3c3fdf1b.4bee@apollo.uucp> gallen@gallen.UUCP (Gary Allen) writes: > Put yourself in our place, what would you have done? The OSF is a gamble. If it works, we will have two or more different Unicii. Different specifications for different features. If it doesn't, the people developing OSF will have to do twice as much work when adding extensions. Because they will have to implement two sets of extensions (networking, window systems, lightweight processes, dynamic linking, network file system, security, toolkits, real-time, redundancy, sys-admin, mapped files, etc.) If the gamble works, we lose. If the gamble doesn't work, they lose and we lose. It might be one thing if they were extending Unix into an areas where AT&T wasn't going. But they seem to be extending OSF Un*x into the same areas but going into a different direction. And I believe in most cases they will have their extensions done AFTER AT&T releases the similar extension/feature. I predict that the members of OSF, noted for their stubbornness, will ignore the incompatible AT&T extensions until their customers demand them. Then they will have to provide compatibility for the 'Pure' Unix. Net result: A lot of wasted effort to provide two ways to accomplish the same feature. If I were a member of OSF, I would try to find out about the future extensions and make my extensions as similar to SysV.Next as possible. And if *MY* requirements weren't being met, I would start screaming at AT&T NOW. (And some companies *ARE* doing this *NOW*). If I were a user of an OSF unix, I would start asking my vendor for AT&T compatibility. But I wouldn't hold my breath. -- Bruce G. Barnett <barnett@ge-crd.ARPA> <barnett@steinmetz.UUCP> uunet!steinmetz!barnett
vixie@palo-alto.DEC.COM (Paul Vixie) (05/25/88)
Karl Kleinpaste climbs momentarily onto a soapbox... > >I'm really, really tired of hearing all the OSF sponsors moaning that >"AT&T & Sun's relationship will put all the other companies at a >disadvantage for N months after release of SysVRel(M+1) while we port >it to our machines." > >Question: So what else is new? When was the last time that AT&T >involved its competitors heavily in a new release of SysV? > >All other companies, including Sun, have been playing catch-up to >AT&T's releases since Day One. This is nothing new. >[...] Nothing in the AT&T/Sun plan changes that. AT&T was never likely to sell enough iron to be a competitor when it was the only entity working on new releases. Teamed up with Sun, the picture changes a lot. With their combined name recognition and technical talents, with Sun's marketing talents and AT&T's dollars, they make a fearsome pair. Maybe more fearsome than they realized :-/. I'm just glad there's going to be competition. Bright as he is, I don't want to be stuck with Joy et al's vision of what needs to exist... I'd rather see both sides sweat a bit, try to out-do eachother. That was the great strength of AT&T's non-product for a long time: everybody did their own thing. I'm very glad that AT&T couldn't sell the thing back when BSD was being born -- we'd all still be using V7's IPC. Which is not to say at all that I think every strain should survive and that we should all #ifdef our code 'til we die. New strains ought to have a chance, though, and if one company owns the design, that doesn't happen. We needed a balance, a struggle, and now we've got one. I hope neither side capitulates or screws up. I shouldn't need to say this, but: I'm not speaking for DEC. -- Paul Vixie Digital Equipment Corporation Work: vixie@dec.com Play: paul@vixie.UUCP Western Research Laboratory uunet!decwrl!vixie uunet!vixie!paul Palo Alto, California, USA +1 415 853 6600 +1 415 864 7013
mlandau@bbn.com (Matt Landau) (05/26/88)
In comp.unix.wizards (<14181@tut.cis.ohio-state.edu>), Karl Kleinpaste writes: >I'm really, really tired of hearing all the OSF sponsors moaning that >"AT&T & Sun's relationship will put all the other companies at a >disadvantage for N months after release of SysVRel(M+1) while we port >it to our machines." > >Question: So what else is new? When was the last time that AT&T >involved its competitors heavily in a new release of SysV? There is a difference in the new situation, though. The fact is that AT&T has been notoriously unsuccessful at selling Unix machines, even though they controlled one of the two major branches of the Unix operating system. This has meant that DEC, IBM, Apollo, etc. didn't have to worry about AT&T as a competitor. Sun, on the other hand, is one of the most successful companies going, and the OSF cabal *has* had to worry quite a bit about them (to put it mildly). Historically, though, Sun has had to play catch-up on the OS at the same rate as everyone else. (That's not really true, of course, since Sun has tended to innovate more than most of its competition, but it IS true from the point of view of SysV compatibility features and passing SVVS.) The OSF contingent now fears that they will have to worry not only about Sun's well-regarded machines and generally superior price/performance ratio, but also about its ability to deliver a verified "standard" System V implementation long before any of them can do so. This is a reasonable argument. If it had come from a group with more of an historical committment to open systems (even as much as Sun's committment, which involves things like publishing the NFS and ONC specs, licensing SPARC to anyone who wants it, etc.), I might be more inclined to believe that it's more than just petty political posturing. -- Matt Landau The happiest cold and lonely guy mlandau@bbn.com stuck in the Yukon without a dog.
glennw@nsc.nsc.com (Glenn Weinberg) (05/26/88)
I never, ever, thought I'd see the day when the great majority of people on Usenet would come to the opinion that AT&T had their best interests at heart! I continue to be amazed at how so many people are just gobbling up the AT&T corporate line on the effects of OSF on the Unix* market. I realize that I will get severely flamed for these opinions, but for what it's worth here are my ideas on several of the points that have been discussed by a variety of people: 1) Timing of releases Contrary to what some people have claimed, the circumstances surrounding the timing of releases HAVE changed recently. Prior to the OFFICIAL announcement of the AT&T/Sun merger, SVR4 and ABI's, (which was quite recent, not a couple of years old as at least one person has implied) AT&T did get a head start on the rest of the industry, but basically no other vendors were favored. As of SVR4, Sun in particular, and other vendors with ABI's to a lesser degree, would be favored. 2) ABI's Now, you say, "well, why doesn't everyone just sign up for an ABI, then?" The answer is simple: because AT&T wouldn't let them. AT&T alone decided which vendors it would sign ABI agreements with. I can't provide much more detail than that, but I think you can guess what the criteria were if you look at the list of founding members of OSF. (And by the way, AT&T isn't just giving away ABI agreements for free, either.) Also, keep in mind why the Hamilton group got together in the first place. Initially, THERE WAS TO BE ONLY ONE ABI--THE SPARC ABI. AT&T had to retreat from that position when they saw the reaction, and opened up the ABI idea to other vendors. Of course, in the process they eliminated ABI's greatest attraction, that of being able to provide shrink-wrapped software for Unix. (Now, AT&T and probably Sun will deny that a SPARC-only ABI was ever their intention, but if you believe that, I too have a bridge to sell you.) 3) Vendor neutrality Many people have commented on the fact that OSF will be dominated by IBM and DEC. This shows a clear lack of understanding of how OSF will be structured. OSF is an independent foundation, with its own board, president and technical director. In many ways, it's set up like MCC, for those of you who are familiar with that organization. IBM and DEC will be board members just like everyone else. They will still have only one vote each. Furthermore, the actual technical decisions will be made by the president and/or technical director, not by the board or by a committee. If the board decides the president and/or technical director show a bias, they can fire them. 4) Porting base Yes, it has been stated that AIX will be the base software for the OSF version of Unix. But all of the members of the OSF will be contributing software to the foundation, and all will have the resulting whole ported to their systems. So there will be an advantage to being a member of the OSF, but anyone can join the OSF, unlike the ABI club, where AT&T has exclusive control over the membership. 5) Licensing Anyone who has seen the changes in the licensing agreements for System V from SVR1 through SVR3 must understand one of the greatest motiviations for creating the OSF. AT&T has arbitrarily changed the definition of Unix, how it can be distributed, to whom it can be distributed, what it costs, etc., etc. Major portions of the system have been unbundled or dropped altogether (e.g., DWB, man pages--did you know that the SVR3 license does not allow you to ship man pages?) What will they change next? Of course it's possible that the OSF will play the same games, but I don't think they will, and besides, the MEMBERSHIP has control over the OSF's licensing terms. Only AT&T has control over AT&T's licensing terms. 6) Standardization Sorry, the AT&T/Sun effort doesn't result in standardization either. First, even if you grant portability over all the ABI's (which I don't), as discussed above, not all vendors will be allowed to have an ABI. Second, there is already the question of POSIX. Third, since AT&T has shown no propensity to accept input regarding the contents of future releases, each vendor has been forced to extend Unix in its own way. This is standardization? At least with OSF I see the POSSIBILITY of setting up working groups to define and implement standard versions of market-specific extensions to Unix. I wanted to present my opinions to the net, especially since I do not (nor have I ever) worked for any of the OSF members and therefore consider myself to be at least somewhat neutral. It has pained me to see all the negative publicity surrounding OSF, and maybe even more, the positive light in which AT&T has been portrayed recently. It is my humble opinion that AT&T has, in fact, precipitated the new "divisiveness" in the Unix community with its ABI creation, and that OSF is a positive reaction to this problem. Obviously, there are many others who disagree with me, but hey, life wouldn't be any fun otherwise! So, I've said my piece, and I'm ready. Fire away! Glenn Weinberg Manager, Operating Systems Development, Series 32000 National Semiconductor Corporation STRONG DISCLAIMER: I speak strictly for myself. I do not mean to imply in any way a commitment by National Semiconductor to OSF or, in fact, to any of the opinions expressed above. (Unix is a registered trademark of AT&T.) -- Glenn Weinberg Email: glennw@nsc.nsc.com National Semiconductor Corporation Phone: (408) 721-8102 (My opinions are strictly my own, but you can borrow them if you want.)
guy@gorodish.Sun.COM (Guy Harris) (05/26/88)
> Of course, in the process they eliminated ABI's greatest > attraction, that of being able to provide shrink-wrapped software > for Unix. While a single ABI would mean you could provide shrink-wrapped software "for UNIX", as opposed to providing it for "UNIX on the '386" or "UNIX on the 68K" or "UNIX on the MIPS" or..., that's not its "greatest attraction" all by itself. It might be *nicer* for application vendors if there were only one processor they had to port to - you'd have to ask them that - but it's not at all clear that the difference between "one procesor" and "more than one processor" is the difference between shrink-wrapped software and no shrink-wrapped software. (The people working on the 68K binary standard, for example, do not appear to be dropping their efforts simply because a '386 binary standard may emerge; in fact, their efforts may have been *spurred* by the relative uniformity of the '386 UNIX world.) > (Now, AT&T and probably Sun will deny that a SPARC-only > ABI was ever their intention, but if you believe that, I too have > a bridge to sell you.) Well, I believe it, but then I have the disadvantage of working at Sun and being too close to the facts; if you're coming from the outside, you can believe or disbelieve all sorts of things without having the facts get in the way. Or, maybe, there was some Secret Illuminati Plot that they didn't tell us grunts about, but I tend to doubt that. Both AT&T and Sun have '386 machines, for example. There has certainly been motion towards establishing an ABI for the '386; if ABIs for a particular processor truly cause more versions of software to appear for machines using that processor, an ABI for the '386 would be a Good Thing both for AT&T and Sun, as well as for other '386 machine vendors. The same can be said for a 68K binary standard (except that AT&T no longer sells 68K machines, unless they still have some 7300's they're flogging). > 5) Licensing > > (e.g., DWB, man pages--did you know that the SVR3 license does > not allow you to ship man pages?) Since you don't *get* machine-readable man pages with the S5R3 tape - the machine-readable documentation is a separate product - this is not surprising. I personally have some qualms with that particular bit of unbundling, but that's a different matter.
ok@quintus.UUCP (Richard A. O'Keefe) (05/26/88)
In article <5085@nsc.nsc.com>, glennw@nsc.nsc.com (Glenn Weinberg) writes: > 2) ABI's > Now, you say, "well, why doesn't everyone just sign up for an ABI, > then?" The answer is simple: because AT&T wouldn't let them. > AT&T alone decided which vendors it would sign ABI agreements > with. It is important to be clear about "vendors of WHAT?" SUN are doing a SPARC ABI. Motorola (not SUN) are doing a 68020 ABI and an 88000 ABI. If Intel aren't doing an 80386 ABI I'll be much surprised (there already is a de facto definition). The relevant vendors here are the companies that make the chips. These companies can listen to anybody _they_ want to. You can easily get a copy of the Motorola ABI drafts and tell them what you think. HP and Apollo are ok as far as their 68020-based machines are concerned; as far as their proprietary machines are concerned ABIs *would* be an issue except that no-one else is making UNIX boxes out of thosse machines, so they have de facto total control over "ABI" definitions for their machines anyway. IBM don't need an ABI for the RT PC. If anyone were to say what a 370 ABI ought to look like, I think we'd be better off with Amdahl deciding that. > Many people have commented on the fact that OSF will be dominated > by IBM and DEC. This shows a clear lack of understanding of > how OSF will be structured. Never mind the structure, who pays the bills? Somehow I don't see IBM letting Apollo tell them what to do. The giveaway as far as I am concerned is that > Yes, it has been stated that AIX will be the base software for the > OSF version of Unix. I've used AIX on an RT. Rhymes with "aches", and my word, that's appropriate. > So there will be an > advantage to being a member of the OSF, but anyone can join the OSF, > unlike the ABI club, where AT&T has exclusive control over the > membership. Serious question: just _how_ is it guaranteed that anyone can join the OSF?
wnp@dcs.UUCP (Wolf N. Paul) (05/26/88)
In article <3c3fdf1b.4bee@apollo.uucp> gallen@gallen.UUCP (Gary Allen) writes: >Are you unaware that MS-DOS (PC-DOS if you prefer) is proprietary, the most >common OS on earth, the most portable, and probably also the worst? Well, it's proprietary in the sense that someone owns it, and that's true of UNIX also; that will no doubt be true of any OSF product as well. In fact, I can't see anyone investing the amount of time and effort needed to produce a viable OS and not wanting to maintain some control over it. Even GNU is copyrighted, and in that sense proprietary. It is not proprietary in that it's not owned by the manufacturer of the hardware required to run it, unlike VMS, MVS, etc. In fact it's owned by a company which does not make hardware at all. -- Wolf N. Paul * 3387 Sam Rayburn Run * Carrollton TX 75007 * (214) 306-9101 UUCP: ihnp4!killer!dcs!wnp ESL: 62832882 INTERNET: wnp@DESEES.DAS.NET or wnp@dcs.UUCP TLX: 910-280-0585 EES PLANO UD
mishkin@apollo.uucp (Nathaniel Mishkin) (05/26/88)
In article <10978@steinmetz.ge.com> davidsen@crdos1.UUCP (bill davidsen) writes: >In article <24369@pyramid.pyramid.com> sas@pyrps5.pyramid.com (Scott Schoenthal) writes: > >| But, it is proprietary. Customers have to sign licenses. Customers have to >| pay money. Customers have to perform certain obligations (e.g., passing >| SVVS). Is this not 'proprietary'? > > The only reason SRV has been as portable as it has is that there is a >reasonable validation suite, and you must meet it. It is not perfect, >but to the user it gives a fighting chance that system will support a >product. Ada used to be that way, but now that there is no longer a >requirement to pass the validation suite, I see some real crap compilers >called Ada. > > Does it hurt the other vendors to have to deliver a working product? >Obviously they think so, they put up $90 mil to be able to do what they >want with they version of UNIX, making it probably that there will be >100 flavors of OSFix, AIX, or whatever. This will provide a nice set of >proprietary o/s to keep the vendor happy. I don't know how many people have had to deal with SVVS, but Apollo has. (Apollo's OS version that's in beta test passes SVVS.) It's not fun. It's not fun unless you started with AT&T's source code. It's not fun if your view of the world extends beyond 1970s timesharing systems. It's not fun if you want to add functionality because SVVS has a way of thinking certain added functionality is "bugs". SVVS is a little worse than "not perfect". The Ada analogy doesn't hold up. No one is changing and/or adding new features to Ada every 12-18 months. If AT&T adds some feature to System V and tests for that feature SVVS, if I want to ship the next version of my system, and my system contains System V stuff in it, I have to make that feature work, but fast. And from what I've heard, the problems with the "obligations" go beyond SVVS. I'm afraid that since the details of all this (presumably) appears in contracts which (a) I've never actually seen (only hear about), and (b) for all I know could contain restrictions about what parts of it can be made public, I can't say any more. I'm sure there are some bright people out there who could supply more information though. -- -- Nat Mishkin Apollo Computer Inc. Chelmsford, MA {decvax,mit-eddie,umix}!apollo!mishkin
glennw@nsc.nsc.com (Glenn Weinberg) (05/27/88)
>In article <5085@nsc.nsc.com>, I wrote: >> 2) ABI's >> Now, you say, "well, why doesn't everyone just sign up for an ABI, >> then?" The answer is simple: because AT&T wouldn't let them. >> AT&T alone decided which vendors it would sign ABI agreements >> with. > In article <1022@cresswell.quintus.UUCP> ok@quintus.UUCP (Richard A. O'Keefe) writes: >It is important to be clear about "vendors of WHAT?" [Stuff about who has or might soon have ABIs deleted] For generality, let's say "vendors of CPUs." As I mentioned in my original article, AT&T has excluded an entire class of vendors from ABI consideration. What you have to understand is that having an official ABI involves more than just defining a binary standard for a particular CPU architecture. Unfortunately, you'll just have to take my word for the above because I don't feel that I can say any more due to non-disclosure rules. (And you can take that for what it's worth.) >> Many people have commented on the fact that OSF will be dominated >> by IBM and DEC. This shows a clear lack of understanding of >> how OSF will be structured. >Never mind the structure, who pays the bills? Somehow I don't see IBM >letting Apollo tell them what to do. The giveaway as far as I am concerned >is that As I understand it, the bills are paid in equal shares by all the founding members of the OSF. The long-term goal is for the OSF to become self-funding. >> Yes, it has been stated that AIX will be the base software for the >> OSF version of Unix. >I've used AIX on an RT. Rhymes with "aches", and my word, that's appropriate. I should restrain myself, but can't. You could run The World's Greatest Operating System on an RT and it wouldn't help... > >> So there will be an >> advantage to being a member of the OSF, but anyone can join the OSF, >> unlike the ABI club, where AT&T has exclusive control over the >> membership. > >Serious question: just _how_ is it guaranteed that anyone can join the OSF? As far as I can tell, it's in the organization's charter. I can't make an absolute statement to that effect, since I haven't actually gotten a physical copy of the charter and read it. But that's my understanding. -- Glenn Weinberg Email: glennw@nsc.nsc.com National Semiconductor Corporation Phone: (408) 721-8102 (My opinions are strictly my own, but you can borrow them if you want.)
davidsen@steinmetz.ge.com (William E. Davidsen Jr) (05/27/88)
In article <5085@nsc.nsc.com> glennw@nsc.UUCP (Glenn Weinberg) writes: | 2) ABI's | | Now, you say, "well, why doesn't everyone just sign up for an ABI, | then?" The answer is simple: because AT&T wouldn't let them. An ABI is not for a vendor, it's for a CPU type. There will be one for SPARC, 386, 68020, and hopefully for VAXen. These are all markets in which there are either multiple hardware or software vendors. ABI describes how things call the O/S, and you don't need to "get one," you just buy the standard and write to it. | Also, keep in mind why the Hamilton group got together in the | first place. Initially, THERE WAS TO BE ONLY ONE ABI--THE | SPARC ABI. AT&T had to retreat from that position when they | saw the reaction, and opened up the ABI idea to other vendors. I hate to say this, but is there some reason to believe that anyone ASKED for more than one ABI? | Of course, in the process they eliminated ABI's greatest | attraction, that of being able to provide shrink-wrapped software | for Unix. (Now, AT&T and probably Sun will deny that a SPARC-only | ABI was ever their intention, but if you believe that, I too have | a bridge to sell you.) Beats hell out of me how this eliminates shrink wrap compatibility... on the 386 market I'll be able to sell software for Xenix, V/386, IN/ix, and probably some others I've forgotten. Right now there are (at least) two standards, and I believe three. | 3) Vendor neutrality | | Many people have commented on the fact that OSF will be dominated | by IBM and DEC. This shows a clear lack of understanding of | how OSF will be structured. OSF is an independent foundation, | with its own board, president and technical director. In many ways, I'm sorry, if you believe that any foundation is "independent" of the people who pay the bills, I disagree with you. If they want to be truly independent, the board should be TOTALLY made up of users and software vendors who don't sell hardware (or at least CPU's, I'd hate to count the Microsoft Mouse and stuff like peripherals). | 5) Licensing | | Anyone who has seen the changes in the licensing agreements for | System V from SVR1 through SVR3 must understand one of the | greatest motiviations for creating the OSF. AT&T has arbitrarily | changed the definition of Unix, how it can be distributed, to | whom it can be distributed, what it costs, etc., etc. Major | portions of the system have been unbundled or dropped altogether | (e.g., DWB, man pages--did you know that the SVR3 license does | not allow you to ship man pages?) What will they change next? And here I thought that the reason was so that I didn't pay for something I don't need... BellTech is shipping a UNIX runtime, C compiler, and manuals (unlimited license) for about $400. If they make money at that rate the license fees can't be too onerous. | Of course it's possible that the OSF will play the same games, | but I don't think they will, and besides, the MEMBERSHIP has | control over the OSF's licensing terms. Only AT&T has control | over AT&T's licensing terms. True. In either case the user's have nothing to say about it. If OSF gets really obnoxious you can always get the nice cheap proprietary o/s from the vendor. I don't disagree with you on this point, but as long as the user's have no say, we're at the mercy of whoever, anyway. | | 6) Standardization | | Sorry, the AT&T/Sun effort doesn't result in standardization either. | First, even if you grant portability over all the ABI's (which I | don't), as discussed above, not all vendors will be allowed to | have an ABI. Second, there is already the question of POSIX. AT&T has said they will conform with POSIX. Just getting away from having versions of stuff for SysV, Xenix, Ultrix, SunOS, BSD, Apollo, etc^3, would make me happy. Anything which gives me another set of ifdefs rots. | Third, since AT&T has shown no propensity to accept input regarding | the contents of future releases, each vendor has been forced to | extend Unix in its own way. This is standardization? At least Isn't having AT&T get Sun and Microsoft together to make a common product for all of them an indication that they're trying to integrate the major flavors? Sure, it's driven by the market, but it's happening, and the software vendors (and developers) and users will be better off for it. A common user interface which runs initially on both X and NeWS? Can you give up all the proprietary unterfaces? I sure can. * * * I'm not ridiculing your opinions, I just feel that you are looking at this from a vendor standpoint, and I'm looking at is as a developer and purchaser of software. We have Suns, VAXen, Xenix, V/AT, PC/ix, and about seven other flavors of UNIX here. I would love to think that we could all share the same versions of the same programs, *and I will give up a little of the benefits of either SysV or BSD to do it!* -- bill davidsen (wedu@ge-crd.arpa) {uunet | philabs | seismo}!steinmetz!crdos1!davidsen "Stupidity, like virtue, is its own reward" -me
shankar@hpclscu.HP.COM (Shankar Unni) (05/27/88)
There has been an incredible amount of heat and smoke in the various
discussions I have seen on this issue, and practically no light.
As far as I can see, one *major* motivation for this reaction on the part of
the big 3 (or 2 or 1) has been AT&T's adamant refusal to let *any* of their
potential customers participate in the Unix standardization process. Bill Joy
has been dropping inflammatory statements like comparing the other vendors'
demands to "getting 9 pregnant women together to produce a baby in one month".
And so on.
Don't forget that Unix is no longer the exclusive preserve of small,
workstation-based engineers in the US. European and Asian customers have long
been screaming for a decent standardized OS (which is why the big 3 are
suddenly showing such concern in the issue). They have been traditionally
ignored by almost all the large US companies, and have had little say in
correcting this situation. Well, now they command a significant amount of
clout, and they want a good solution NOW, that addresses all their needs
(national language support, commercial application interface, user interface,
etc etc).
Also, many commercial and defense customers and vendors are showing increasing
interest in a standardized OS, something that can accomplish for the mini and
mainframe market what MSDOS achieved for the PC market.
AT&T and Sun's parochial and short-sighted attitude does nothing to address
these problems. Their attitude seems to be reflected in all the diatribes
emanating from the old-fashioned, berzerkeley jocks on the net ("it's small
and it's mine and I like it and keep your !$^% hands off it and ...")
WAKE UP, GUYS! UNIX HAS HIT THE BIG TIME!!!! Unix is the future of both
commercial and engineering computing, believe it or not. There is no reason
why Unix users cannot have it both ways (lean & efficient, and robust(slow?)
& commercially viable). It just takes a broad viewpoint on the part of the
designers/standardizers. AT&T and Sun have *NOT* inspired any confidence on
the part of the above-mentioned user bases as far as these issues go...
And yes, OSF really wouldn't care to have two UNIX standards around. They'd
much rather that other vendors would see the light (including AT&T and Sun)
and come around to join them in the effort. Remember, in spite of all the
dripping sarcasm and cynicism that has greeted the proposal, OSF *is* really
an independent organization (not a quasi-subsidiary of any of the participants,
not even of *BM)
---
Shankar Unni.
#include <std_disclaimer.h>
These opinions are mine and mine alone and do not necessarily reflect those of
my employer and all that..jima@hplsla.HP.COM ( Jim Adcock) (05/27/88)
| I see too much resemblance between "Open Software Foundation" | and "Free Software Foundation". Well, as long as we're complaining about naming conventions, how about complaining about "Free Software Foundation" ??? Free means Free. Free does not mean you have an obligation to send someone bucks. Free does not mean that you are forced to buy into someone else's political/economic philosophies before one is allowed to use his/her software. Free does not mean you're going to threaten to sue the ass off anyone who says or does something you don't like. Free does not mean you take someone else's software that was given to you without restriction, and add your own licensing restrictions. Free means free. Free means being able to speak your honest mind without having to consult with a lawyer first. Free means being able to speak your honest mind without having the one you're talking about sending his/her lawyer to come talk to you. Free means being able to charge a buck for one's efforts, if one feels the marketplace is willing to pay you a buck for those efforts. Free means being able to put one ideas in the public domain, if that is what one chooses to do, for the betterment of all man-kind, without restriction, to do with as they might. Free means free. I for one, am sick and tired of having to run to a lawyer everytime I want to write some software, or to speak my mind, or having someone else sick his/her lawyers on me when I do speak my mind. I don't see where "Free Software Foundation's" software is any more "free" than the "Open Software Foundation's" software is "open" "A rose by any other name ..... " Mine own opinion only.
vixie@palo-alto.DEC.COM (Paul Vixie) (05/27/88)
In article <4457@vdsvax.steinmetz.ge.com> barnett@steinmetz.ge.com writes: >If I were a member of OSF, I would try to find out about the future >extensions and make my extensions as similar to SysV.Next as possible. >And if *MY* requirements weren't being met, I would start screaming at >AT&T NOW. (And some companies *ARE* doing this *NOW*). > >If I were a user of an OSF unix, I would start asking my vendor >for AT&T compatibility. But I wouldn't hold my breath. Let's imagine, for the sake of argument, that there are two UNIX-making entities in the world. One is AT&T+Sun and their licensees; one is OSF and its licensees. Just imagine this so we can see what kind of world it would be. Imagine (I know this is asking a lot) that as of 1995 or so, the world of UNIX computing is about evenly split between the two major UNIX products. Half the machines in the world run SUNix, the other half run OSFix. If I were a vendor in this imaginary world, I'd want to do as much as possible to make sure that my OS was competitive (which means, among other things: as compatible as possible) with both of the major UNIX variants. If I were an OSF member, I'd want to be competitive against SUNix licensees. If I were a SUNix licensee, I'd want to be competitive against OSF members. If I were a user in this imaginary world, I would want all my machines to have at least the same core of functionality (and method), and if one of my vendors started falling behind (which means: not keeping up with BOTH UNIX camps on major features), I'd switch vendors. The two UNIX camps need competition to spur them into new developments. Bright young programmers need to be able to get their ideas into the mainstream, too, and that's much more likely if you can approach either of the two camps with an idea the other one doesn't have yet. Anyway, if you can imagine this world I describe, then maybe you can tell me why it isn't possible. Or tell me why AT&T/Sun has to be the only one making standards, the one other people have to keep up with. I note with amusement that AT&T doesn't think that way anymore, and that they have for these last years has let Sun and everyone else beat up on them horribly in the marketplace. 14-character file names indeed. -- Paul Vixie Digital Equipment Corporation Work: vixie@dec.com Play: paul@vixie.UUCP Western Research Laboratory uunet!decwrl!vixie uunet!vixie!paul Palo Alto, California, USA +1 415 853 6600 +1 415 864 7013
gwyn@brl-smoke.ARPA (Doug Gwyn ) (05/27/88)
In article <14181@tut.cis.ohio-state.edu> karl@triceratops.cis.ohio-state.edu (Karl Kleinpaste) writes: >Nothing in the AT&T/Sun plan changes that. The A/S plan provides for >the development of the merged UNIX system - little more than that - >and therefore for the day when I won't have to say "#ifdef BSD" and >mutter the same functional phrases in my programs twice. OSF, if it >goes anywhere at all, will force me right back into this mode with >"#ifdef OSFMumblix" and a different pair of code fragments to do the >same things 2 ways. This is right on the mark. Application development is hindered by the existence of slightly different variants of the UNIX programming environment. Even the current hybrids of BSD and SysV suffer from this. The systems came close enough together with 4.3BSD and SVR3 to make it feasible to attempt to really merge them for once and for all and to provide a single good naive-user interface. If we can get the UNIX vendors to all start shipping systems with the common merged application support environment, then that would help UNIX considerably in the marketplace, due to improved quantity and availability of useful applications. To compete with systems such as OS/2, such a unified front seems to be essential. The last thing we needed was for yet another UNIX variant to arise. Whoever is pushing that either does not have the interests of the UNIX community at heart, or they haven't been paying close attention to the factors that have really been hindering the spread of UNIX.
fangli@ihlpl.ATT.COM (Chang) (05/27/88)
In article <4457@vdsvax.steinmetz.ge.com>, barnett@vdsvax.steinmetz.ge.com (Bruce G. Barnett) writes: > > If it works, we will have two or more different Unicii. > Different specifications for different features. > > If it doesn't, the people developing OSF will have to do twice > as much work when adding extensions. Because they will have to > implement two sets of extensions (networking, window systems, > lightweight processes, dynamic linking, network file system, > security, toolkits, real-time, redundancy, sys-admin, mapped > files, etc.) > > > If the gamble works, we lose. > If the gamble doesn't work, they lose and we lose. > (stuff deleted) > -- > Bruce G. Barnett <barnett@ge-crd.ARPA> <barnett@steinmetz.UUCP> > uunet!steinmetz!barnett I don't think it is a gamble. It is a carefully planned business move. Especially for IBM, they don't depend on saling UNIX box at all. Even if this move are going to failed, they've already muddy the water and create a lot of confusion. Whenever people have doubt or confused they tend to hold back and wait. You've already seem what happened in PC market when they announced PS/2. They have nothing to lose and probably something to gain. If the move works, this move will prevent UNIX ever merging into a single standard and challenge their proprietary OS market and at the same time they can get an easy entry into the "AIX" market. If the gamble works, they win. If the gamble doesn't work, they win too. Fangli Chang (312)979-2502 Disclaimer: above are my opinion only. My employer do not share my opinion, despite that I hope they do.
rsalz@bbn.com (Rich Salz) (05/27/88)
-- Please send comp.sources.unix-related mail to rsalz@uunet.uu.net.
gore@eecs.nwu.edu (Jacob Gore) (05/27/88)
/ comp.unix.wizards / gallen@apollo.uucp (Gary Allen) / May 24, 1988 / >Are you unaware that MS-DOS (PC-DOS if you prefer) is proprietary, the >most common OS on earth, the most portable, and probably also the worst? ___________________________________^^^^^^^^ Yep! You can put it on a laptop computer, and carry it anywhere! Jacob Gore Gore@EECS.NWU.Edu Northwestern Univ., EECS Dept. {oddjob,gargoyle,ihnp4}!nucsrl!gore
lid@cernvax.UUCP (lid) (05/27/88)
In article <685@vsi.UUCP> sullivan@vsi.UUCP (Michael T Sullivan) writes: >In article <3c3a336e.13422@apollo.uucp>, mishkin@apollo.uucp (Nathaniel Mishkin) writes: >> >> BTW, I object to the various speculation (on the part of several people) >> on the topic of the "ulterior motives" of DEC and IBM. (No one cares > >I can't see why you have objections. I myself am suspicious >of a company that is supposed to write a unified Unix being >headed by DEC and IBM, who would like nothing more than to >see Unix go away. Despite what DEC and IBM execs say, I believe >that is still the case. People's objections are understandable. I really don't see the whole point of the discussion: a) they (OSF) have got 90M$ to spend, they are free to spend it as they wish, b) they could come out with a real Unix system, compatible with ATT Unix and we will get 2 vendors for Unix, OR their system is different and if it better than Unix we (users) gain, if it is worse we are stuck with ATT and they've lost 90M$. In any case the user community will gain, Sun and ATT will for sure change their attitude to better counter OSF effort. So why people complain about OSF ? There is no reason to fear IBM or DEC, the market is asking for Unix and the market will decide what is going to win, just wait and see. Achille Petrilli, Cray Operations
karl@triceratops.cis.ohio-state.edu (Karl Kleinpaste) (05/27/88)
gwyn@brl-smoke.ARPA writes:
To compete with systems such
as OS/2,...a unified front seems to be essential.
Exactly. Everyone should remember the porting base from which OSF has
said it will work.
IBM's AIX. On the IBM PC/RT.
Guess who wants to do major market things with OS/2?
Yup, I thought so.
I wouldn't be surprised to find that IBM (in particular, and DEC to a
lesser degree) *wants* OSF to fail...right when OS/2 is about to
become `real.' IBM's ability to park a new product into an existing
market at exactly the right time is one of its long-standing
strengths.
$90M? It seems like a lot, but not in the context of IBM's billions,
and the prospect for potential sales of PS/2's with OS/2.
Whoever is
pushing that either does not have the interests of the UNIX community
at heart, or they haven't been paying close attention to the factors
that have really been hindering the spread of UNIX.
Considering IBM's interest in OS/2, the lack of concern for the UNIX
community's interests is kinda vacuously obvious.
--Karlmlight@hpiacla.HP.COM (Mike Light ) (05/27/88)
> It just occurred to me that HP, while joining in the chorus of "we're > at a disadvantage due to delay," has voluntarily put itself in an even > worse position: HP wants to be SysV, thus forcing it to wait on AT&T > releases, and supports NFS and YP, thus forcing it to wait on Sun > releases. HP has no business claiming that they're suddenly in a > worse position than they used to be in light of the A/S plan; if > anything, their timing will be *improved* by getting an entire release > from a single source at one time. You're quite right - HP wishes to be SysV and support NFS - and we don't really mind being a few months "behind" on releases from AT&T/Sun. The apparent fear was that future releases might be "slanted" toward SPARC machines which might not run well on HP architectures thus leaving HP at a price/performance disadvantage. And all this talk about "standard binaries"... huh. right. SPARC standard binaries. I believe HP got a wee bit nervous that its $400 million investment in HP Precision Architecture might be in jeopardy. -- Mike Light. P.S. I speak only for myself. I am considered a heretic by some.
rogerk@mips.COM (Roger B.A. Klorese) (05/28/88)
In article <4629@hoptoad.uucp> gnu@hoptoad.uucp (John Gilmore) writes: >I don't see the difference between depending upon OSF to "never make its >licenses objectionable in the future" and depending upon AT&T to do the same. >Is there a contractural committment from OSF to provide new releases on the >same, or less onerous, terms as its first releases? If not, you have just >switched from the devil you know to a new devil. The difference is that the companies porting the software are not licensees, but co-owners of the source. -- Roger B.A. Klorese MIPS Computer Systems, Inc. {ames,decwrl,prls,pyramid}!mips!rogerk 25 Burlington Mall Rd, Suite 300 rogerk@mips.COM Burlington, MA 01803 I don't think we're in toto any more, Kansas... +1 617 270-0613
rogerk@mips.COM (Roger B.A. Klorese) (05/28/88)
In article <4630@hoptoad.uucp> gnu@hoptoad.uucp (John Gilmore) writes: >They want to keep this software under corporate control. They will be >"open" with each other, not with their customers. If you honestly believe that platform vendors ought to be freely distributing their software, source and all, to the user community, then you have no understanding of (a) what it takes to support the average customer (and it is the average customer, not the wizard, for whom support plans should be oriented, despite the history of the "here-it-is-do-something-with-it-if-you-can" wonder from AT&T) and (b) exactly how large a portion of systems vendors' income is derived from software licensing. >AT&T and Sun have made an effort to make it possible to >run the same applications software on hundreds of manufacturer's machines. How have they done this? I would be happy to see MIPS make its operating system available to others as a porting base, but I wouldn't delude myself into yhinking of that as selflessly offering a standard, but as just another means of growing my revenue base. >It is a struggle for Sun to stay open as it grows, but so far it seems equal >to the challenge. Sun's original attempt at foisting the SPARC ABI on the public as the only ABI to exist is an illustration of how two-faced this whole issue is. You see it as offering an open hardware architecture to the world; I see it as Sun permitting people to hop on a standard as long as they pay Sun to do it, and then enter the market with a later, more expensive product. -- Roger B.A. Klorese MIPS Computer Systems, Inc. {ames,decwrl,prls,pyramid}!mips!rogerk 25 Burlington Mall Rd, Suite 300 rogerk@mips.COM Burlington, MA 01803 I don't think we're in toto any more, Kansas... +1 617 270-0613
rogerk@mips.COM (Roger B.A. Klorese) (05/28/88)
In article <24369@pyramid.pyramid.com> sas@pyrps5.pyramid.com (Scott Schoenthal) writes: >I think that if IBM, DEC, et al., were truly interested in delivering an open >implementation of Mumblix (or whatever their derivative of Un*x is to be >called), the money would have been best spent in funding an independent effort >(e.g., FSF or a university). This is tremendously naive. "Open" here means equally available for input and for delivery to all vendors who participate in the group. This does not imply "free". Free software is fine, but *supported* software is more important, and outside the hacker community, sales and service cannot be uncoupled in the majority of situations. The user community is better served by a full-cost operating system which is compatible among platforms and offers support than by a hunk of wizard code that can be run on any box, but offers no single point of control and support for the hardware-software combination. -- Roger B.A. Klorese MIPS Computer Systems, Inc. {ames,decwrl,prls,pyramid}!mips!rogerk 25 Burlington Mall Rd, Suite 300 rogerk@mips.COM Burlington, MA 01803 I don't think we're in toto any more, Kansas... +1 617 270-0613
rogerk@mips.COM (Roger B.A. Klorese) (05/28/88)
In article <1818@thebes.UUCP> schwartz@swatsun.UUCP (Scott Schwartz) writes: >In article <3c3a336e.13422@apollo.uucp> mishkin@apollo.UUCP (Nathaniel Mishkin) writes: >>Perhaps. At least the new devil is not a competitor. >Digital, IBM, and Apollo not competitors? Wait, I have to sit down... No, no, you don't understand. *OSF* is not a competitor to its members. AT&T/Sun *is* a competitor to its licensees. -- Roger B.A. Klorese MIPS Computer Systems, Inc. {ames,decwrl,prls,pyramid}!mips!rogerk 25 Burlington Mall Rd, Suite 300 rogerk@mips.COM Burlington, MA 01803 I don't think we're in toto any more, Kansas... +1 617 270-0613
rogerk@mips.COM (Roger B.A. Klorese) (05/28/88)
In article <10976@steinmetz.ge.com> davidsen@crdos1.UUCP (bill davidsen) writes: > Honestly, are there that many people who insist on having a new >machine the moment a new version of the o/s is available? If a group is >running Suns, or Apollo, or PC/RT, will they change just to get >something a few months earlier? Some groups are *always* buying systems, and they want to be as sure as possible that each system they buy has the latest features they can get, no matter the vendor. This is especially true in universities and research labs. We are selling in some situations where people are deciding which machines to add to their networks next winter based on our plans for follow-on products during the two years following. > It looks to me as though currently AT&T has had the source code >earlier than anyone else, and that obviously has not let them dominate >the hardware market. That's because AT&T can't build computers, or to be fair, has given no evidence of being able to build or price them based on the 3B series. The real fear is that Sun will have a source code advantage with somewhat competitive hardware. > I think it would be justified to (a) have AT&T release a "work in >progress" source when something reasonably works, (b) give the final >version as a set of deltas, so that customized code could be upgraded, >and (c) hold release of the SRVR4 until 3 months after the code was >frozen to give other vendors a chance to be in the right timeframe. This is a reasonable idea, but only if it is practiced for *every* release. Also, three months is nowhere near sufficient lead time for some companies to beta-test and package a release. > Of course NONE of the big players has the user in mind, except as a >source of revenue. If they can do something for the user while not >hurting sales, or helping a competitor, they will for good will, but >don't expect them to say "it's only money." Finally, a user with a rational understanding of a vendor's position. > AT&T said somewhere (or was quoted as saying) that at some time they >might turn UNIX over to a marketing group, vendor independent. Perhaps >some of the best people could be stationed on such an organization, mush >as Sun and AT&T are sharing people and code now. Too bad they didn't come up with this *before* OSF... > Finally I think that OSF is a clear effort to infringe on the repution >of FSF, and that it should be promptly renamed or sued. Interesting... most of the people watching OSF in the user community have never even heard of FSF. -- Roger B.A. Klorese MIPS Computer Systems, Inc. {ames,decwrl,prls,pyramid}!mips!rogerk 25 Burlington Mall Rd, Suite 300 rogerk@mips.COM Burlington, MA 01803 I don't think we're in toto any more, Kansas... +1 617 270-0613
rogerk@mips.COM (Roger B.A. Klorese) (05/28/88)
In article <1018@cresswell.quintus.UUCP> ok@quintus.UUCP (Richard A. O'Keefe) writes: >What puzzles me is what took the OSF so long. The SUN/AT&T UNIX merge has >been publicised for a couple of years. Yes, but when it was started, AT&T wasn't an owner of Sun... -- Roger B.A. Klorese MIPS Computer Systems, Inc. {ames,decwrl,prls,pyramid}!mips!rogerk 25 Burlington Mall Rd, Suite 300 rogerk@mips.COM Burlington, MA 01803 I don't think we're in toto any more, Kansas... +1 617 270-0613
rogerk@mips.COM (Roger B.A. Klorese) (05/28/88)
In article <14181@tut.cis.ohio-state.edu> karl@triceratops.cis.ohio-state.edu (Karl Kleinpaste) writes: >When was the last time that AT&T >involved its competitors heavily in a new release of SysV? AT&T was no threat because it was never a viable competitor in the systems business. >All other companies, including Sun, have been playing catch-up to >AT&T's releases since Day One. Yes, but "including Sun" is the key phrase. Now they are holding an inappropriate advantage. -- Roger B.A. Klorese MIPS Computer Systems, Inc. {ames,decwrl,prls,pyramid}!mips!rogerk 25 Burlington Mall Rd, Suite 300 rogerk@mips.COM Burlington, MA 01803 I don't think we're in toto any more, Kansas... +1 617 270-0613
rogerk@mips.COM (Roger B.A. Klorese) (05/28/88)
In article <54544@sun.uucp> guy@gorodish.Sun.COM (Guy Harris) writes: >In response to Glenn Weinberg: >> (Now, AT&T and probably Sun will deny that a SPARC-only >> ABI was ever their intention, but if you believe that, I too have >> a bridge to sell you.) >Well, I believe it, but then I have the disadvantage of working at Sun and >being too close to the facts; if you're coming from the outside, you can >believe or disbelieve all sorts of things without having the facts get in the >way. ...Or maybe there was some marketing decision they felt was too poisonous for you techies to hear. At last year's UNIX Expo in New York, a representative of Sun Home Office Marketing presented the notion, in a seminar on "Big Iron on Wall Street", that if the vendors of minisupers would only drop their proprietary hardware and use SPARC, they'd be able to jump on "the ABI, the only source of shrink-wrapped UNIX software". Seems pretty explicit to me. -- Roger B.A. Klorese MIPS Computer Systems, Inc. {ames,decwrl,prls,pyramid}!mips!rogerk 25 Burlington Mall Rd, Suite 300 rogerk@mips.COM Burlington, MA 01803 I don't think we're in toto any more, Kansas... +1 617 270-0613
price@decwrl.dec.com (Chuck Price) (05/28/88)
In article <11006@steinmetz.ge.com> davidsen@crdos1.UUCP (bill davidsen) writes: > An ABI is not for a vendor, it's for a CPU type. There will be one for >SPARC, 386, 68020, and hopefully for VAXen. These are all markets in >which there are either multiple hardware or software vendors. ABI >describes how things call the O/S, and you don't need to "get one," you >just buy the standard and write to it. This is yet another example of the blind speculation that seems to be rampant throughout this topic. Getting an ABI is not as simple as you state. DEC cannot get an ABI for the VAX. AT&T wouldn't allow it. Do we need any more reason to rebel and form OSF? How tight does AT&T have to hold the leash before you realize they are in control of you future? Did you like it when they wouldn't let you plug your own phone into your own telephone outlet? Their new license is *that* binding on the other vendors, and *worse*. > >| 3) Vendor neutrality >| >| Many people have commented on the fact that OSF will be dominated >| by IBM and DEC. This shows a clear lack of understanding of >| how OSF will be structured. OSF is an independent foundation, >| with its own board, president and technical director. In many ways, > > I'm sorry, if you believe that any foundation is "independent" of the >people who pay the bills, I disagree with you. If they want to be truly >independent, the board should be TOTALLY made up of users and software >vendors who don't sell hardware (or at least CPU's, I'd hate to count >the Microsoft Mouse and stuff like peripherals). Again, positions based upon pure speculation. If you *read* the press release, or an accurate description of the organization, you would see that all the sponsors pay an *equal* amount into the OSF, to eliminate the possibility of one vendor being "more equal" than another. There *is no* Security Council. Now, if you just plain don't believe the statements made by the OSF members, what makes you any more confident in the ulterior motives of AT&T? And why should software vendors or users be any more independent than hardware vendors? I can't imagine that Claris, for example, would be any more independent than HP. And I doubt that Microsoft has DEC's best interest at heart. The hardware vendors' business depends upon shipping operating systems to run on their platforms. If these vendors commit to ship an OSF based system, they are pretty stongly motivated to make the OSF produce that system. Remember, their customers are holding them to the statements that were made at the OSF announcement. > >| 5) Licensing >| >| Anyone who has seen the changes in the licensing agreements for >| System V from SVR1 through SVR3 must understand one of the >| greatest motiviations for creating the OSF. AT&T has arbitrarily >| changed the definition of Unix, how it can be distributed, to >| whom it can be distributed, what it costs, etc., etc. Major >| portions of the system have been unbundled or dropped altogether >| (e.g., DWB, man pages--did you know that the SVR3 license does >| not allow you to ship man pages?) What will they change next? > > And here I thought that the reason was so that I didn't pay for >something I don't need... BellTech is shipping a UNIX runtime, C >compiler, and manuals (unlimited license) for about $400. If they make >money at that rate the license fees can't be too onerous. Read the operative words here: SVR3 license *DOES NOT ALLOW* you to ship man pages. AT&T controls the content of you distribution. Anything you add becomes the property of AT&T, and/or AT&T can arbitrarily refuse to renew your license. None of the vendors involved in OSF have voiced *any* objection to paying license fees. It is the nature and binding rules of the license that the OSF members object to. > ... as long as >the user's have no say, we're at the mercy of whoever, anyway. Users can participate by paying the membership fee. If you are too poor to pay, or are a cheapskate :-> like me, you can make your feelings known through the standards organizations, like the POSIX committee. Remember, OSF abides by the POSIX standard, a public, OPEN standard. > Isn't having AT&T get Sun and Microsoft together to make a common >product for all of them an indication that they're trying to integrate >the major flavors? Sure, it's driven by the market, but it's happening, >and the software vendors (and developers) and users will be better off >for it. A common user interface which runs initially on both X and NeWS? >Can you give up all the proprietary unterfaces? I sure can. The OSF members have no argument with trying to integrate the major flavors of Unix. But the needs of the players in the industry *must* be taken into account. AT&T/Sun's plan did not take the needs of the rest of us into account at all. They explicitly ruled out our participation. The software vendors and developers will *not* be better off, because the major computer vendors will be *forced* to abandon Unix in favor of a business in which we can fairly compete. So it would be much worse without OSF. Our only other option is to go our independent ways. We *could not* accept AT&T's "deal". Question: If AT&T is so committed to open systems, why doesn't it join OSF? Remember that they are a big company too, and are not likely to be dominated by IBM, even if the OSF rules allowed it (which they *don't*). They were invited to be a sponsor, and their refusal is clear indication that they have no intention of opening Unix. Question: If the OSF is so bad, why did a Sun VP say that he thought it was a pretty good idea, and hinted that Sun might be interested in joining (from quote in this week's CSN). My opinion: Before taking positions which you cannot substantiate, why not do a little homework on the motivations and mechanisms behind the AT&T position and the OSF group? In fact, I would love to see someone independent of the vendors perform a study of the AT&T license, and compare it to the mechanisms which make up the OSF, and report their findings to this conference. It would make much more interesting reading than the baseless flames recorded thus far. -chuck * I speak only for myself. DEC and my boss typically don't agree with me. * * But I know they are usually wrong. *
shan@mcf.UUCP (Sharan Kalwani) (05/28/88)
In article <1818@thebes.UUCP> schwartz@swatsun.UUCP (Scott Schwartz) writes: >I think the real issue is that unix is now (or is going to be, depending >on how you feel) an important force in the computer world, and AT&T >OWNS it. Competing corporations just can't tolerate something important >that the other guy controls. .... I think you've hit the nail on the head here. It does seem to explain practically most of the moves made so far. >> Geez. Talk about white hats and black hats! > >Not to mention expensive business suits. I like that! >Can we all go back to being computer scientists now? Or should >we petition for the creation of comp.unix.mba? Nope, it is high time that we petition for the creation of comp.unix.politics ;-) -- sharan kalwani ...!{uunet!umix, pur-ee!iuvax, ucbvax!mtxinu}!mcf!shan internet: shan%mcf.uucp@umix.cc.umich.edu shan@mcf.uucp UNIX:They say the learning curve is steep, but you have to climb it only once!
ron@topaz.rutgers.edu (Ron Natalie) (05/28/88)
The Apollo systems I've been forced to use do such a poor imitation of UNIX that I have no wonder that you have problems with SVVS. -Ron
ekrell@hector.UUCP (Eduardo Krell) (05/28/88)
In article <503@bacchus.DEC.COM> price@decwrl.UUCP (Chuck Price) writes: (about SVR3) >Anything >you add becomes the property of AT&T, ... I've read a lot of nonsense in this discussion, but this is clearly untrue. I have a Counterpoint Workstation on my desk running their port of SVR3 to their multiprocessor architecture, and their additions are clearly Counterpoint proprietary source code. In fact, we had to sign a non-disclosure agreement with them before they let us have source. Now, if their code became the property of AT&T by some magic act, why did we have to sign such an agreement (after all, it's our code now, isn't it). >Question: If AT&T is so committed to open systems, why doesn't >it join OSF? Because AT&T owns Unix and doesn't want to give it away (especially to those who have always benefited from closed/proprietary systems) and because committee don't design good operating systems/programming languages/whatever. Eduardo Krell AT&T Bell Laboratories, Murray Hill, NJ UUCP: {ihnp4,ucbvax}!ulysses!ekrell ARPA: ekrell@ulysses.att.com
gwyn@brl-smoke.ARPA (Doug Gwyn ) (05/29/88)
In article <503@bacchus.DEC.COM> price@decwrl.UUCP (Chuck Price) writes: >DEC cannot get an ABI for the VAX. AT&T wouldn't allow it. Is this a fact? Has DEC tried? >Read the operative words here: SVR3 license *DOES NOT ALLOW* you to >ship man pages. AT&T controls the content of you distribution. Anything >you add becomes the property of AT&T, and/or AT&T can arbitrarily >refuse to renew your license. This is completely wrong. I just reviewed our UNIX System V, Release 3.0 license, sublicensing agreement, and schedule. There is no mention of on-line documentation, presumably because that is not supplied as part of the base UNIX System V Release 3.0 (it's available as an add-on, and presumably sublicensable as such). Printed documentation can be copied (no more than 2 copies per designated CPU). You can of course distribute SVR2 on-line manual entries under terms of SVR2 sublicensing. The only constraints on what software is distributed are that there are several categories, such as the networking utilities, that require (at least after June 30, 1988) that if ANY part of that category is supplied, ALL parts of that category will be supplied. There is no restriction against vendors adding their own extensions, and nowhere is it stated that such extensions become AT&T property. There is no expiration date on the license, so refusal to renew is a non-issue. >Remember, OSF abides by the POSIX standard, a public, OPEN standard. Remember, the SVID abides by the POSIX standard, a public, OPEN standard. >The software vendors and developers will *not* be >better off, because the major computer vendors will be *forced* to >abandon Unix in favor of a business in which we can fairly compete. Excuse me, but that would make very little difference. Those vendors are the ones whose UNIX offerings have not been competitive all along. >If AT&T is so committed to open systems, why doesn't it join OSF? Assuming AT&T has made a decision not to, which I don't think we know at this time, it could well be that they are reluctant to abandon the operating system that they've been gradually improving with certain long-range goals in mind just to pick up a version that starts way behind where AT&T's system currently is. >In fact, I would love to see someone independent of the vendors >perform a study of the AT&T license, and compare it to the mechanisms >which make up the OSF, and report their findings to this conference. I did this, and as previously noted there is nothing particularly obnoxious about AT&T's UNIX licensing terms. In fact the per-CPU binary sublicensing fees under SVR3 are much less expensive than before.
guy@gorodish.Sun.COM (Guy Harris) (05/29/88)
> Getting an ABI is not as simple as you state. DEC cannot get an ABI for > the VAX. AT&T wouldn't allow it. What does "get an ABI" mean here? In some sense, there already exists an ABI for the VAX, defined by Ultrix; it's not as if other people are out there building VAX clones with their own OSes, etc. AT&T may not be willing to put their imprimatur on such an ABI, but that's a different matter. However, if it is the case that, say, you get early access to code from AT&T iff you have an ABI agreement with AT&T, I agree that this unfairly favors vendors with whom they have ABI agreements. > Read the operative words here: SVR3 license *DOES NOT ALLOW* you to > ship man pages. SVR3 tape *DOES NOT CONTAIN* man pages, so I'm not surprised the SVR3 licence doesn't cover them. We (Sun) signed a separate license with AT&T so that we could 1) get the S5R3 machine-readable documentation and 2) ship documents - including machine-readable man pages - derived from those documents. (And no, before you ask, we signed this stuff before we signed the agreement announced on October 19.) Whether UNIX should be unbundled to that degree is a separate argument. > AT&T controls the content of you distribution. Anything you add becomes > the property of AT&T, I second Eduardo Krell's comments on this. SunView is not AT&T's property, for example; if you claim that *anything* you add to your UNIX distribution becomes AT&T's property, you'd better cite chapter and verse of the license if you expect to convince me of this.
henry@utzoo.uucp (Henry Spencer) (05/29/88)
> I think that if IBM, DEC, et al., were truly interested in delivering an open > implementation of Mumblix... > the money would have been best spent in funding an independent effort > (e.g., FSF or a university)... Not necessarily. One reason for funding a not-quite-independent effort like OSF is to have some small control over the result. They want something that complies with reasonable standards and is enough like current Unixes that their programs will run without too much trouble. FSF's ultimate output (in my opinion) will bear only a passing resemblance to Unix. And the major experience we've had so far with university control of Unix, to wit BSD, has not been exactly encouraging to those who prefer software to remain compatible unless there is a good reason to change it. (Not that AT&T hasn't contributed its share of stupid, gratuitous, incompatible changes, but Berkeley has them beat by several light-years.) -- "For perfect safety... sit on a fence| Henry Spencer @ U of Toronto Zoology and watch the birds." --Wilbur Wright| {ihnp4,decvax,uunet!mnetor}!utzoo!henry
henry@utzoo.uucp (Henry Spencer) (05/29/88)
> Why OFS? What is wrong with an ANSI/IEEE/ACM/OSI/... standards commitee?
Standards committees do not write code. What is wanted is runnable code,
not specifications for it.
--
"For perfect safety... sit on a fence| Henry Spencer @ U of Toronto Zoology
and watch the birds." --Wilbur Wright| {ihnp4,decvax,uunet!mnetor}!utzoo!henrycsg@pyramid.pyramid.com (Carl S. Gutekunst) (05/29/88)
>> Read the operative words here: SVR3 license *DOES NOT ALLOW* you to >> ship man pages. > >SVR3 tape *DOES NOT CONTAIN* man pages... We (Sun) signed a separate license >with AT&T so that we could 1) get the S5R3 machine-readable documentation and >2) ship documents - including machine-readable man pages - derived from those >documents. And the flexability is still there for the licensees. Note that although AT&T has chosen not to bundle machine-readable man pages with SVR3, Pyramid and Sun both do. I personally think it was idiotic the way AT&T handled the SVR3 docs, but that was certainly *not* because of a desire to withhold information; just AT&T's internal confusion about how to bundle their releases. >> AT&T controls the content of you distribution. Anything you add becomes >> the property of AT&T, > >I second Eduardo Krell's comments on this. I third it. Pyramid has added many extensions to UNIX, as has any serious UNIX vendor. Value-added is essential for product differentiation, and our lawyers would never sign a license that required us to give that away. <csg>
gwyn@brl-smoke.ARPA (Doug Gwyn ) (05/30/88)
In article <700@cernvax.UUCP> lid@cernvax.UUCP () writes: > their system is different and if it better than Unix > we (users) gain, if it is worse we are stuck with ATT > and they've lost 90M$. The counter-argument is that no matter whether it is marginally better or marginally worse, so long as it is different, application developers will have to deal with it (unless it were not much used, but with DEC and IBM pushing it, it will be used regardless of its merit). The users lose because the developers lose.
david@dhw68k.cts.com (David H. Wolfskill) (05/31/88)
In article <700@cernvax.UUCP> lid@cernvax.UUCP (Achille Petrilli) writes: >.... >There is no reason to fear IBM or DEC, the market is asking for Unix >and the market will decide what is going to win, just wait and see. Well, "fear" may or may not be appropriate, but a large dose of concern would seem reasonable (from my perspective). I work on an IBM mainframe (running MVS/XA, which I am paid to support); I retreat to UNIX here at home.... (Make of that what you will.... :-) I perceive a few things about IBM's actions in the past that lead me to believe that IBM is very good at marketing things to the people who make the $$ decisions. Unfortunately, I also perceive that some of their more successful marketing efforts have been dismal from a technical standpoint; as an illustration of this, I point out the s/36. IBM has a considerable capability to do marvelous things; if they live up to that potential, I hope that I will be one of the first to congratulate them -- but I'm not expecting it. david -- David H. Wolfskill uucp: ...{trwrb,hplabs}!felix!dhw68k!david InterNet: david@dhw68k.cts.com
stan@sdba.UUCP (Stan Brown) (05/31/88)
> > In article <11006@steinmetz.ge.com> davidsen@crdos1.UUCP (bill davidsen) writes: > > > An ABI is not for a vendor, it's for a CPU type. There will be one for > >SPARC, 386, 68020, and hopefully for VAXen. These are all markets in > >which there are either multiple hardware or software vendors. ABI > >describes how things call the O/S, and you don't need to "get one," you > >just buy the standard and write to it. > > This is yet another example of the blind speculation that seems to > be rampant throughout this topic. Getting an ABI is not as simple > as you state. DEC cannot get an ABI for the VAX. AT&T wouldn't allow > it. Do we need any more reason to rebel and form OSF? How tight Exactly what do you base this statement on ? -- Stan Brown S. D. Brown & Associates 404-292-9497 (uunet gatech)!sdba!stan "vi forever"
clewis@spectrix.UUCP (Chris Lewis) (05/31/88)
In article <5085@nsc.nsc.com> glennw@nsc.UUCP (Glenn Weinberg) writes: >I never, ever, thought I'd see the day when the great majority of people >on Usenet would come to the opinion that AT&T had their best interests >at heart! I continue to be amazed at how so many people are just gobbling >up the AT&T corporate line on the effects of OSF on the Unix* market. I find it even harder to accept that (by inference) that anything IBM is involved in has our best interests at heart either. Speaking as one who has worked for IBM in C language definition, I have a few points: - All appearances to the contrary, IBM has been firmly against UNIX since the mid seventies. You see, they have these major mid and mainframes to flog, and UNIX just cuts into that market. Even though they've built, or have had built for them, at least a dozen variants of UNIX, only two or three have ever seen the light of day, and customers are discouraged as much as possible from using them (eg: IX/370, VM/IX etc.). They'd much rather UNIX just simply quietly fade away. Many times in the past IBM has announced that they have taken steps to change this. Nothing much ever happened (with the minor exception of A/IX). Why should this time be any different? I'll believe it when I see it. - IBM believes only in standards that they invent. I mean, who else uses EBCDIC? And they can't even agree on what version of EBCDIC for crying out loud. - As part of the C definition group we were under a great deal of pressure to customize C divergent from ANSI: Enum cardinality operators, language tie-ins to graphics subsystems, language tie-ins so that you could call BASIC subroutines etc. Not functions, but language modifications! Not nearly as simple as "fortran", "basic", or "rpg" keywords, but a complete revision of function/variable implementation and definition syntax. Yes, IBM had (this was in the 83/84 timeframe) a representative on the ANSI C Standards Committee. Some rep. Had about 3 weeks exposure with C. No programming. What a jerk. As far as I could tell, he was fresh out of high school - he would never once give me a straight answer what his background was. He just loved inventing new features to make C look more like some sort of mutant hybrid of Pascal, Ada and Assembler. Out of twelve people in my department (language design centre), there were only two people (myself and a colleague from a previous employer) who had *any* background whatsover in computer languages/compilers/interpreters. The two of us with 6-7 years experience, both from University and compiler building with previous employers were the juniors, the rest were fresh out of high school or what Americans would call Junior College. The oldest was 22. Some of these people became world-wide "IBM Prime" for the languages we were working with (Pascal, COBOL, BASIC, RPG, C, FORTRAN) simply because they had been with IBM a couple of weeks longer. I resisted these changes to C as hard as I could, largely by trying to teach our C rep something about C. Didn't do any good. However, the presence of experienced people in our department evoked paranoia on the part of many of the others. Stories started to flow. Eg: my resume was a lie and I didn't have any University degrees. T'was a really wonderful place to work.... So I left. (Gave some of my friends a chance to stop laughing.) As it turned out, my old department never produced anything - several other colleagues from this same previous employer were in the "back-end" group and gave up on all of the language teams in our dept. The back-end group built the language specs and frontends themelves and it's shipping quite nicely thank you. Since the C rep was funded by IBM HQ, they couldn't get rid of him. So, they eventually made him a manager of his own department (with no underlings) and put him in a closet where he wouldn't bother anybody else. I think he's still there.... You want these guys to define UNIX? [BTW: I have absolutely no complaint with what you might call the "old hands" at IBM. Whatever their background, they were a joy to work with, and we would come up with combined efforts. It's the 18 year-olds you have to worry about... :-(] OSF: a wonderful opportunity for IBM to divide and submerge. Just like OS/2-PS/2. -- Chris Lewis, Spectrix Microsystems Inc, UUCP: {uunet!mnetor, utcsri!utzoo, lsuc, yunexus}!spectrix!clewis Phone: (416)-474-1955
mishkin@apollo.uucp (Nathaniel Mishkin) (06/01/88)
In article <May.28.09.17.30.1988.25758@topaz.rutgers.edu> ron@topaz.rutgers.edu (Ron Natalie) writes: >The Apollo systems I've been forced to use do such a poor imitation >of UNIX that I have no wonder that you have problems with SVVS. I'm crushed that you have such a low opinion of our systems, but other than to say that I think we're doing far better now than we used to, I'll ignore the comment and press the generic issue: I think the lesson to be learned from the Apollo experience with SVVS is that at least in some ways it's hard to be innovative (or even just to stray from the AT&T implementation, of which SVVS is supposed to be independent) yet pass SVVS. In at least some ways, SVVS as it is currently constituted thus stifles innovation. I think stifling innovation is something none of us want. -- -- Nat Mishkin Apollo Computer Inc. Chelmsford, MA {decvax,mit-eddie,umix}!apollo!mishkin
sullivan@vsi.UUCP (Michael T Sullivan) (06/02/88)
In article <700@cernvax.UUCP>, lid@cernvax.UUCP (lid) writes: > In article <685@vsi.UUCP> sullivan@vsi.UUCP (Michael T Sullivan) writes: > > >that is still the case. People's objections are understandable. > > There is no reason to fear IBM or DEC, the market is asking for Unix > and the market will decide what is going to win, just wait and see. Okay, maybe I should have said suspicions. I myself am curious as to how this whole thing is going to go, so I'm not about to write it off as garbage. I was merely trying to point out that people's suspicions were understandable. I'm suspicious. I'm also curious. -- Michael Sullivan {uunet|attmail}!vsi!sullivan sullivan@vsi.com HE V MTL <- Can anybody guess the significance of that?
karl@triceratops.cis.ohio-state.edu (Karl Kleinpaste) (06/02/88)
rogerk@mips.COM writes: AT&T was no threat because it was never a viable competitor in the systems business. That seems inconsistent. An awful lot of vendors have been bending over backwards for several years in order to be able to advertise their conformity with SysVRelN. They must have viewed AT&T as a viable competitor, or else they would not be looking for a way to be able to claim to provide at least as much capability as AT&T. It would be profitless to fight for compatibility with something which did not matter. Ergo, AT&T has always mattered; it is and always has been viewed as a viable competitor. At the very least, AT&T's lead in SysV development has always been taken very, very seriously. Witness IBM's AIX' compatibility with SysVRel[?1?2?] (exactly which is unclear to me), HP's SVVS-compliant HP-UX, DEC's adoption of similar standards in the face of losing the USAF contract a year ago, ad nauseum. AT&T has been very significant in all these companies' plans for quite some time. >All other companies, including Sun, have been playing catch-up to >AT&T's releases since Day One. Yes, but "including Sun" is the key phrase. Now they are holding an inappropriate advantage. If Sun is so dangerous to everyone else, why have so many companies been so careful to provide compatibility with so many of Sun's enhancements, notably including NFS and RPC? These companies have been holding their heads in the "lion's mouth" for a rather long time to be so suddenly disturbed by that position. @begin[speculation] I honestly detect a positively horrendous case of "sour grapes." The OSF sponsors seem merely upset because they're not the ones that approached AT&T about such an agreement, preferring (until now) to work in their own private, proprietary corners until faced with Sun. If they were so intensely interested in standards, they should not have waited so long - they should have beaten Sun to the competitive punch by offering such a deal with AT&T quite some time before. Does anyone else notice that the OSF press release reads a lot like early Sun press releases from, say, 1983 or so? Does this say anything about Sun's ability to stay ahead of market demands as well as its competitors? By, say, 5 years or so? @end[speculation] --Karl
nelson@sun.soe.clarkson.edu (Russ Nelson) (06/02/88)
I'm surprised that no one has come to the following conclusion:
AT&T+Sun AND the OSF are gathering the wagons into circles because they're
afraid of the GNUs. Think about what's going to happen when rms finishes
the kernel? There is a tremendous market opportunity for a firm that
simply offers support for GNUnix. They need no investment other than
that needed to become a GNUnix wizard. And, because of the restrictions
on derivatives of GNU, they will benefit from everyone else's work.
Who do you think will provide better customer support? A company that
sells customer support, i.e. GNU Wizards, Inc.? Or a company that
sells Unix boxes i.e. AT&T+SUN+OSF?
Parting shot: If IBM is so impressed by Unix, why aren't they on Usenet?
Can you say FUD?
--
char *reply-to-russ(int network) {
if(network == BITNET) return "NELSON@CLUTX";
else return "nelson@clutx.clarkson.edu"; }perry@apollo.uucp (Jim Perry) (06/02/88)
In article <7986@brl-smoke.ARPA> gwyn@brl.arpa (Doug Gwyn (VLD/VMB) <gwyn>) writes: >In article <503@bacchus.DEC.COM> price@decwrl.UUCP (Chuck Price) writes: > >>If AT&T is so committed to open systems, why doesn't it join OSF? > >Assuming AT&T has made a decision not to, which I don't think we know >at this time, it could well be that they are reluctant to abandon the >operating system that they've been gradually improving with certain >long-range goals in mind just to pick up a version that starts way >behind where AT&T's system currently is. Ah, but that's exactly the point. This is the mindset of the owner of a proprietary operating system. I was not around when the issue of supporting Unix was debated at Apollo, but I would imagine the anti-Unix argument could have been pretty well summed up as "reluctance to abandon the operating system that they've been gradually improving with certain long-range goals in mind just to pick up an OS that starts way behind where Apollo's system currently is". As it happens, people who buy computers want them to run Unix because it is a standard. Whether it's state-of-the-art or antiquated is not relevant, it's a standard. Sort of like FORTRAN IV in the '70's, people who know better sneer at it but everybody sells it (and devotes a lot of resources to it). But I digress. A standard is successful because it is a common base on which everyone can build, with each player adding extensions suited to their strengths, be it parallel processing, real-time, whatever. Unix is a standard because Bell Labs was not in the commercial computer business and therefore made it widely available, and many people implemented it on a variety of hardware. Now AT&T (along with Sun) is in the position of developing it in the character of a product rather than a research item. The argument goes that if AT&T+Sun not only get to make their proprietary extensions but also get to fold those extensions back into the base at intervals, this puts them at a considerable competitive advantage. Having a 'standard' defined by one (or two) companies' proprietary system may not be a good idea. Does everybody scramble to update their Fortran compilers because IBM releases a new version? Disclaimer: I don't speak for anyone but myself, and I'm not so sure I'm always in complete agreement with what I say. As it happens, I'm an iconoclastic upstart who doesn't much care for premature standardization... but that's not the way the wind blows these days. Jim Perry perry@apollo.UUCP
dora@erc3ba.UUCP (Dora_Group) (06/02/88)
In article <637@spectrix.UUCP>, clewis@spectrix.UUCP (Chris Lewis) writes: > I find it even harder to accept that (by inference) that anything IBM > is involved in has our best interests at heart either. > > Speaking as one who has worked for IBM in C language definition, I have > a few points: > > - All appearances to the contrary, IBM has been firmly against > UNIX since the mid seventies. You see, they have these major > mid and mainframes to flog, and UNIX just cuts into that market. > Even though they've built, or have had built for them, at least > a dozen variants of UNIX, only two or three have ever seen the > light of day, and customers are discouraged as much as possible > from using them (eg: IX/370, VM/IX etc.). They'd much rather > UNIX just simply quietly fade away. Many times in the past IBM > has announced that they have taken steps to change this. Nothing > much ever happened (with the minor exception of A/IX). > > Why should this time be any different? I'll believe it when I > see it. > > - IBM believes only in standards that they invent. I mean, who > else uses EBCDIC? And they can't even agree on what version of > EBCDIC for crying out loud. AND In article <7988@brl-smoke.ARPA>, gwyn@brl-smoke.ARPA (Doug Gwyn ) writes: > The counter-argument is that no matter whether it is marginally better > or marginally worse, so long as it is different, application developers > will have to deal with it (unless it were not much used, but with DEC > and IBM pushing it, it will be used regardless of its merit). > > The users lose because the developers lose. People, Doesn't this get to the central motivation for the OSF arrangement -- splintering and confusion. We all know OSF has nothing to do with standards and open systems. It is a blatant attempt by the market captors to retain control of their captive audience. The idea of a customer shopping around for the best/cheapest/most reliable hardware makes them wretch. Let's face it, UNIX is the best thing since sliced bread. With it, a system designer/programmer is bound only by his/her imagination. We can make it do whatever we please and if done carefully, in a way that is hardware independent. I think developers, users, and key OSF members have always realized this, but had also known that it would take a lot of clout to give UNIX momentum and credibility. A perturbation in captivity has come and it is alarming the captors. It has finally come to the point where the developers -- who have been designing successful applications on UNIX boxes and selling them everywhere else but on UNIX boxes -- have seen $$ in a single UNIX that has mass appeal. I think the masses are finally seeing the light also -- we don't have to be locked in, we do have choices. This is a very alarming thought to the captors -- "We can't turn the users' world upside down every three or four years anymore with new architectures and new operating systems and new bus interfaces and total system re-writes etc...." If the users want the new technology they can buy it from who is offering the best price/performance and not worry about their applications. Can you imagine the fear that instills in the captors when threatened by users who would no longer be captive -- enough fear to get in bed with an arch rival to kill the infant in its cradle. Something else troubles me, I've read the phrase "we'll be stuck with AT&T" -- what does that mean? What am I missing? Is there something inherently wrong with AT&T that I don't know about? Can someone elaborate please? From the info I currently have I'd have to say we're all lucky to be STUCK with the transistor and lasers and fiber optics and etc....... What has AT&T done to so many of you that you would see them fail? Hell -- all of this noise about OSF was and is inspired by UNIX. Where did it come from? Is AT&T that bad? When was the last time any of you picked up the phone and it didn't work? When couldn't you complete a call? How many of you realize the technology behind the network that enables the reliability? Ask you parents and grandparents the same question and see what they say. It seems to me that AT&T has a history of quality goods and services. So someone tell me what the problem is. I for one have been very happy with the quality of the service and products I've received from them. If it's that they were a monopoly and that bothers you, then what about the account control that captors use to lock users in. Isn't that very similar? What about captors who would promise a disk drive manufacturer unlimited demand and then telling him he isn't needed due to the captors decision to manufacture its own new standard drive leaving the little guy out of business with unwanted inventory. Isn't that very similar? What about captors that drop support for certain pieces of hardware and software forcing a user to upgrade? Isn't that very similar? I could go on. We have the opportunity to stop being captive and to start using some really robust, EXTENSIBLE, tools and software -- let's not screw it up. Thanx, Dora Allen 6/1/88
ok@quintus.UUCP (Richard A. O'Keefe) (06/02/88)
In article <358@mipseast.mips.COM>, rogerk@mips.COM (Roger B.A. Klorese) writes: > In article <1018@cresswell.quintus.UUCP> ok@quintus.UUCP (Richard A. O'Keefe) writes: > >What puzzles me is what took the OSF so long. The SUN/AT&T UNIX merge has > >been publicised for a couple of years. > > Yes, but when it was started, AT&T wasn't an owner of Sun... AT&T owning 7.5% of the shares of Sun is a big threat, but AT&T and Sun co-operating on the actual product wasn't? C'mon, when you have a joint project that important, you *expect* some such stock arrangement. If IBM's market analysts couldn't see that coming the day the UNIX merge was announced, they weren't doing their job. It has been said that the final thing will be a three-way AT&T/Sun/Xenix merge. Anyone know what's happening on the Xenix side?
ok@quintus.UUCP (Richard A. O'Keefe) (06/02/88)
In article <14807@tut.cis.ohio-state.edu>, karl@triceratops.cis.ohio-state.edu (Karl Kleinpaste) writes: > Witness IBM's AIX' compatibility with SysVRel[?1?2?] (exactly which is > unclear to me), ... This has been mentioned before, but remains a puzzle to me. We have BSD boxes, SysV boxes, and a couple of RTs, and porting our stuff to AIX was not a matter of saying make -DSYS5 nor even of mixing and matching BSD bits and SysV bits, but in several cases of finding third solutions. I only read about half of the RT manuals, but for what it's worth I don't remember seeing any claim of SysV conformance.
friedl@vsi.UUCP (Stephen J. Friedl) (06/02/88)
Ron Natalie @ Rutgers notes: > The Apollo systems I've been forced to use do such a poor imitation > of UNIX that I have no wonder that you have problems with SVVS. A friend of mine who uses Apollo says the same thing as well. In Apollo's defense, Nathanial Mishkin (mishkin@apollo.uucp) says: > In at least some ways, SVVS as it is currently constituted > thus stifles innovation. I think stifling innovation is something > none of us want. Isn't it *wonderful* how people can make sunshine out of sh*t? :-) -- Steve Friedl V-Systems, Inc. (714) 545-6442 3B2-kind-of-guy friedl@vsi.com {backbones}!vsi.com!friedl attmail!vsi!friedl Nancy Reagan on ptr args with a prototype in scope: "Just say NULL"
ekrell@hector.UUCP (Eduardo Krell) (06/02/88)
In article <358@mipseast.mips.COM> rogerk@mipseast.mips.COM (Roger B.A. Klorese) writes: > >Yes, but when it was started, AT&T wasn't an owner of Sun... And it still isn't. AT&T owns a little over 7% of Sun's stocks. That can hardly be called ownership. Recall that AT&T owns around 22% of Olivetti and noone ever said that Olivetti was owned by AT&T. Just look at the AT&T-Olivetti relationship over the last few years, and especially the recent events. Then tell me whether AT&T owns Olivetti. Then tell me how AT&T can own Sun (with only 7% of the stocks). Eduardo Krell AT&T Bell Laboratories, Murray Hill, NJ UUCP: {ihnp4,ucbvax}!ulysses!ekrell ARPA: ekrell@ulysses.att.com
guy@gorodish.Sun.COM (Guy Harris) (06/03/88)
> I'm surprised that no one has come to the following conclusion: > > AT&T+Sun AND the OSF are gathering the wagons into circles because they're > afraid of the GNUs. I'm not surprised in the least. None of the management here has indicated that they're afraid of GNU; I doubt any of the management at AT&T or in the management of the OSF are, either. Inventing motivations in the absence of evidence may be barrels of fun, but it's not particularly productive.
guy@gorodish.Sun.COM (Guy Harris) (06/03/88)
> Parting shot: If IBM is so impressed by Unix, why aren't they on Usenet? > Can you say FUD? Fascinating theory; the only trouble with it is that IBM *IS* on USENET. Contributions from IBMers are't frequent, but they do occur. In addition, I fail to see what membership in USENET, or the lack thereof, has to do with FUD.
allbery@ncoast.UUCP (Brandon S. Allbery) (06/03/88)
As quoted from <3c3fdf1b.4bee@apollo.uucp> by gallen@apollo.uucp (Gary Allen): +--------------- | First of all, its unlikely that 7 (ok, 6 1/2) of the worlds giant corporations | are sufficiently freaked out by GNU that they are out to steal FSF's thunder. | C'mon, how about a reality break? +--------------- One reality break, coming up. It is, of course, good technique to obscure any competitors... and the name "Open Software Foundation" does so by (1) implying that AT&T/Sun isn't, and (2) by obscuring the existence of the FSF, which is potentially their BIGGEST problem. (Let us hope that the creation of the OSF causes many companies to fund the FSF! I wouldn't trust IBM as far as I could throw it.) +--------------- | If reality is not your bag (it isn't always mine either), feel free to strike | fear into the heart of 60 billion/year IBM, 12 billion/year DEC, ...... | Lawsuits? You're talking about the people who invented the word. +--------------- Awww, poor wittwe IBM! Maybe it's time IBM got a rude awakening. I rather suspect that RMS would be quite willing to do the honors... and the fact that IBM can afford lawyers doesn't necessarily mean the FSF would lose. +--------------- | are you unaware that VMS outsells UNIX on VAXen 15 or so to 1? Are you +--------------- I am given to understand that that is no longer true. +--------------- | unaware that MS-DOS (PC-DOS if you prefer) is proprietary, the most common OS | on earth, the most portable, and probably also the worst? While I like FSF +--------------- "Most portable"?! And you think *John* needs a reality check?! The only reason that MS-DOS runs on so many different kinds of PCs is that they're all *hardware* clones of each other. You don't believe me? Try booting generic MS-DOS on an old Sanyo MBC-550. Or on a Tandy 2000. Or a PCjr. Or an Altos 586. Get the idea? (You can get modified MS-DOS for three of them... because the hardware is *sufficiently* close to a PC. The Altos box, on the other hand, is too smart for MS-DOS because of its memory management unit and other features. MS-DOS in particular requires that certain things be done in ROM... if they aren't, fat chance!) +--------------- | The whole market? REALITY dude, REALITY. These 7 companies ARE the vast majority | of the market. Obsolete IBM, DEC, HP, etc. On second thought, DRUGS dude DRUGS! +--------------- Oh, I see. Now let us all kneel and worship the Gods of the Computer. To put it mildly: up theirs! Of *course* they're the vast majority of the market! And they d*mned well want to make sure that they have all their customers locked up -- but Unix has begun to make that impossible. Of *course* DEC and IBM in particular will want to derail Unix as soon as possible! "Why pick on DEC and IBM?" Because they're the ones with the track records for shafting their customers -- *especially* IBM, who (not incidentally) is providing the development base for OSF's FUDnix. (my name, not theirs) If IBM has an overwhelming history of locking its customers into a death grip, is it not reasonable to assume that they are doing so yet again? And DEC is the company which tried to ignore the existence of Unix for as long as possible, then finally was forced to produce Ultrix -- which is known to be broken. And it seems (see other messages in this group, and if there are any archives of this group you might find them instructive as well) that DEC has been known to either (1) misrepresent Ultrix in order to force customers into VMS or (2) simply ignore a customer order for Ultrix and ship them VMS instead. Why pick on DEC and IBM? Because we have only their past actions to go on -- and based on those past actions, the OSF spells major trouble for Unix. +--------------- | they wanted, just like DEC, IBM, Apollo, etc. Open? When was the last time you | had anything to say about what UNIX is/isn't? +--------------- 4.2BSD, actually. You will note that BSD features have been popping up in System V right and left -- vi here, "reliable signals" there, NFS in the future. +--------------- | UNIX will not favor particular machine architectures such as SPARC and 3Bx, then | I'd like to talk to you about some bridges that I have for sale. The idea that +--------------- And, of course, the fact that UCB won't sell me a copy of 4.3BSD for a 65C816 CPU makes UCB an evil corporate giant, perhaps? Let's leave out the arguments which have no relation to the point at hand, please. +--------------- | Each of the companies involved in OSF have particular needs/focus that must be | addressed by any PORTABLE OS, which is really what you're after isn't it? For | instance, whose notion of ISAM files will be used in UNIX? I don't know, do you? | Perhaps AT&T will get around to it as THEY need it. Otherwise, we have to provide | interim solutions that our customers will use until there is some sort of | "standard". Then, our customers will have to rewrite their applications to match. | Hardly portable. +--------------- Gee, I could have sworn that AT&T had announced they would support POSIX... guess the OSF knows better, right? C'mon, AT&T isn't *stupid*. 4.xBSD vs. System V proved the need for a standard, that's why there is POSIX. SVID is an evident failure, else AT&T wouldn't be working with Sun Microsystems from the other side of the fence. Let's give the people at AT&T credit for some intelligence. They would NOT be integrating with BSD Unix if they wanted to shaft everyone else: they could do *that* much more easily the same way they countered termcap with terminfo. +--------------- | A) Sit in the back of the bus | B) Get off the UNIX bus | C) Start a new bus company. +--------------- And, to use one of your earlier examples, which of these has been done with MS-DOS? Think about it. (Oh, but of course: IBM is (was) involved with MS-DOS, so it must be perfectly all right! Since they're *not* involved with Unix, *that* must be horribly evil!) <-- Can you say "double standard"? How about "hypocrisy"? And show me the difference between AT&T-Sun collusion with DEC, IBM, etc. on the sidelines and Microsoft-IBM collusion with Compaq, Tandy, etc. on the sidelines. Think about it. And then tell me the salient difference between them. (Hint: IBM) +--------------- | Put yourself in our place, what would you have done? And by the way, FSF was never | considered, thought about, infringed upon, or even mentioned. +--------------- Which is the most conclusive proof I've seen yet that the "Open" Software Foundation is a sham. If the reasons for the OSF were *truly* as stated, then all those companies would have supported the FSF. That they didn't says quite a bit about th eintentions of the member companies.... -- Brandon S. Allbery, moderator of comp.sources.misc {well!hoptoad,uunet!marque,cbosgd,sun!mandrill}!ncoast!allbery Delphi: ALLBERY MCI Mail: BALLBERY
allbery@ncoast.UUCP (Brandon S. Allbery) (06/03/88)
As quoted from <5085@nsc.nsc.com> by glennw@nsc.nsc.com (Glenn Weinberg): +--------------- | I never, ever, thought I'd see the day when the great majority of people | on Usenet would come to the opinion that AT&T had their best interests | at heart! I continue to be amazed at how so many people are just gobbling | up the AT&T corporate line on the effects of OSF on the Unix* market. +--------------- ??? Many of the messages I've seen have NOT been pushing AT&T/Sun: they're pushing FSF's GNU. Wake up out there! If the Hamilton Group and/or the OSF really meant what they said, they'd be funding Gnu development. The OSF is yet another devil, masquerading as Unix's guardian angel. Hah! Certainly, AT&T/Sun isn't the answer -- but if you believe DEC/IBM is, you're one heck of a gullible person. GNU FOREVER!!! -- Brandon S. Allbery | "Given its constituency, the only uunet!marque,sun!mandrill}!ncoast!allbery | thing I expect to be "open" about Delphi: ALLBERY MCI Mail: BALLBERY | [the Open Software Foundation] is comp.sources.misc: ncoast!sources-misc | its mouth." --John Gilmore
guy@gorodish.UUCP (06/03/88)
> Somehow this strikes me as naive...OSF strikes me as the trojan horse > that is waiting to be brought into the city walls...every single sponser > is selling a rival proprietary system... Well, of the major names associated with OSF: IBM sells a variety of IBM-specific OSes as well as two Microsoft OSes; however, on at least one of their machines (the RT PC), UNIX is (as far as I know) the only OS. I don't think MVS sells as a UNIX competitor; I don't see all that mainframe software being converted to UNIX tomorrow, so I don't think MVS has anything to fear from UNIX in the near term. DEC does sell VMS as well as selling ULTRIX. Apollo does sell DOMAIN/OS, which includes an Aegis environment. HP sells MPE on the Spectra (before anybody complains, note that "spectra" is the plural of "spectrum"); *however*, I don't think they sell it at all in *competition* with UNIX. HP seems pretty solidly committed to UNIX, even if you don't consider the others solidly committed to UNIX. The various other names (I'm not sure which of them qualify as "sponsors") include several that have been pushing UNIX: Honeywell-Bull, NCR, etc.. Even for the majors that do sell non-UNIX OSes in competition with UNIX, I see no indication that any of them are, as corporations, solidly opposed to UNIX. I suspect there are camps within IBM and DEC, at least, that would like UNIX to go away (I think Apollo, whose own OS may not be as able to stand on its own as VMS or some of IBM's can, is less likely to have such camps); however, I suspect that there are also camps within IBM and DEC that would like their companies to get a significant share of the UNIX market.
bak@csd-v.UUCP (Bruce) (06/05/88)
In article <55234@sun.uucp> guy@gorodish.Sun.COM (Guy Harris) writes: >> Parting shot: If IBM is so impressed by Unix, why aren't they on Usenet? >> Can you say FUD? > >Fascinating theory; the only trouble with it is that IBM *IS* on USENET. >Contributions from IBMers are't frequent, but they do occur. In addition, I >fail to see what membership in USENET, or the lack thereof, has to do with FUD. Funny I've never seen a contribution with "Organization: IBM..." in any posting to comp.unix.... So IBM employee contributions must be *very* infrequent. FUD can be engendered amongst employees even more easily than customers. -- Bruce Kern | uunet!swlabs!csd-v!bak Computer Systems Design | 1-203-270-0399 29 High Rock Rd., Sandy Hook, Ct. 06482 | This space for rent.
rogerk@mips.COM (Roger B.A. Klorese) (06/05/88)
In article <10344@ulysses.homer.nj.att.com> ekrell@hector (Eduardo Krell) writes: >In article <358@mipseast.mips.COM> rogerk@mipseast.mips.COM (Roger B.A. Klorese) writes: >>Yes, but when it was started, AT&T wasn't an owner of Sun... >And it still isn't. AT&T owns a little over 7% of Sun's stocks. That can >hardly be called ownership. If it owns *one* share, then it is AN owner, which is what I said. 7% (plus, as I recal, a board seat) can have quite a bit of influence, especially when AT&T is contractually committed to being able to own 20%. -- Roger B.A. Klorese MIPS Computer Systems, Inc. {ames,decwrl,prls,pyramid}!mips!rogerk 25 Burlington Mall Rd, Suite 300 rogerk@mips.COM Burlington, MA 01803 I don't think we're in toto any more, Kansas... +1 617 270-0613
henry@utzoo.uucp (Henry Spencer) (06/06/88)
> AT&T was no threat because it was never a viable competitor in the > systems business. > > That seems inconsistent. An awful lot of vendors have been bending > over backwards for several years in order to be able to advertise > their conformity with SysVRelN... There is no inconsistency at all; SVRn was a de-facto software standard, not a computer, runnable software for it, or a complete system. The vendors kept up with it for the same reason that they keep up with other standards. It had nothing to do with AT&T's feeble attempts to sell computers. -- "For perfect safety... sit on a fence| Henry Spencer @ U of Toronto Zoology and watch the birds." --Wilbur Wright| {ihnp4,decvax,uunet!mnetor}!utzoo!henry
ekrell@hector.UUCP (Eduardo Krell) (06/06/88)
In article <371@mipseast.mips.COM> rogerk@mipseast.mips.COM (Roger B.A. Klorese) writes: >7% (plus, as I recal, a board seat) can have quite a bit of influence, >especially when AT&T is contractually committed to being able to own >20%. You still haven't answered my question: Show us how AT&T, owning over 20% of Olivetti (and a board seat) has been able to control or influence it (the answer is they haven't been able to). Then show us, why AT&T would be able to influence Sun with only 7% of the stock when it was unable to influence Olivetti with over 20%. Eduardo Krell AT&T Bell Laboratories, Murray Hill, NJ UUCP: {ihnp4,ucbvax}!ulysses!ekrell ARPA: ekrell@ulysses.att.com
henry@utzoo.uucp (Henry Spencer) (06/06/88)
> ... Having a 'standard' defined by one (or two) companies' > proprietary system may not be a good idea... An amusing sidelight on this one is that in a talk a while ago here, some Sun folks criticized MSDOS et al for being controlled by two companies! They didn't have a satisfactory answer when I challenged them about it. -- "For perfect safety... sit on a fence| Henry Spencer @ U of Toronto Zoology and watch the birds." --Wilbur Wright| {ihnp4,decvax,uunet!mnetor}!utzoo!henry
henry@utzoo.uucp (Henry Spencer) (06/06/88)
> AT&T+Sun AND the OSF are gathering the wagons into circles because they're > afraid of the GNUs. It is to laugh. Most of the people involved in setting up OSF wouldn't know a GNU if it bit them. > Think about what's going to happen when rms finishes the kernel? Yeah, all the potential users will look at, see all the unnecessary incompatible changes, say "ugh", and go back to Unix/OSFnix/whatever. > ... because of the restrictions > on derivatives of GNU, they will benefit from everyone else's work. Such of it as gets done, of course. Because of those same restrictions, many companies have decided that GNU is not worth working on. > Who do you think will provide better customer support? A company that > sells customer support, i.e. GNU Wizards, Inc.? Or a company that > sells Unix boxes i.e. AT&T+SUN+OSF? Considering the disparity of resources (and the increasing tendency of hardware manufacturers to lock customers into their own software by not documenting the hardware well enough for somebody else to port to it), clearly the latter. -- "For perfect safety... sit on a fence| Henry Spencer @ U of Toronto Zoology and watch the birds." --Wilbur Wright| {ihnp4,decvax,uunet!mnetor}!utzoo!henry
henry@utzoo.uucp (Henry Spencer) (06/06/88)
> ... Wake up out there! If the Hamilton Group and/or the OSF > really meant what they said, they'd be funding Gnu development. Uh, Brandon, were you half asleep when you wrote that? One of the things HG/OSF explicitly said was "Unix compatible", which GNU won't be. (This is a personal prediction, not FSF policy, but it's a prediction I feel fairly confident about.) There are also some small philosophical differences between, say, IBM and RMS that just might get in the way. The idea of supporting GNU probably was never even raised, but if it had been, I'm sure it would have been laughed out of consideration even if the HG/OSF people had perfectly pure motives. (Note that I don't claim that they do, just that their non-support of GNU says nothing about the issue.) > Certainly, AT&T/Sun isn't the answer -- but if you believe DEC/IBM is, > you're one heck of a gullible person. > GNU FOREVER!!! Certainly AT&Sun and OSF aren't the answer, but if you believe GNU is, you're one heck of a gullible person. It's an interesting project that will at least have useful byproducts (notably the compiler), but seeing it as the salvation of Unix strikes me as a triumph of enthusiasm and ideology over rational thought. -- "For perfect safety... sit on a fence| Henry Spencer @ U of Toronto Zoology and watch the birds." --Wilbur Wright| {ihnp4,decvax,uunet!mnetor}!utzoo!henry
nelson@sun.soe.clarkson.edu (Russ Nelson) (06/06/88)
In article <196@csd-v.UUCP> bak@csd-v.UUCP (Bruce) writes: >In article <55234@sun.uucp> guy@gorodish.Sun.COM (Guy Harris) writes: >>And I wrote: >>> Parting shot: If IBM is so impressed by Unix, why aren't they on Usenet? [I'm wrong -rn] >>Fascinating theory; the only trouble with it is that IBM *IS* on USENET. [you're right -rn] >Funny I've never seen a contribution with "Organization: IBM..." in >any posting to comp.unix.... [you didn't look -rn] This is the output of `find /usr/spool/news -exec grep Organization:.\*IBM {} \;` Organization: IBM AES, Austin, TX Organization: IBM T. J. Watson Research Center, Yorktown Heights, NY That makes two out of about 11,000 articles. I think that I can be forgiven for not having run across them, but I should have looked before I leaped... -- signed char *reply-to-russ(int network) { /* Why can't BITNET go */ if(network == BITNET) return "NELSON@CLUTX"; /* domainish? */ else return "nelson@clutx.clarkson.edu"; }
bzs@bu-cs.BU.EDU (Barry Shein) (06/06/88)
>Then show us, why AT&T would be able to influence Sun with only 7% of >the stock when it was unable to influence Olivetti with over 20%. > > Eduardo Krell AT&T Bell Laboratories, Murray Hill, NJ I for one am very concerned at the prospect of someone writing drivers under the influence...
gallen@apollo.uucp (Gary Allen) (06/07/88)
In article <11410001@eecs.nwu.edu> naim@eecs.nwu.edu (Naim Abdullah) writes: >Nat Mishkin writes: >>I think the lesson >>to be learned from the Apollo experience with SVVS is that at least in >>some ways it's hard to be innovative (or even just to stray from the >>AT&T implementation, of which SVVS is supposed to be independent) yet >>pass SVVS. >> In at least some ways, SVVS as it is currently constituted >>thus stifles innovation. I think stifling innovation is something none >>of us want. > >Could you be more specific about why Apollo finds it difficult to >pass SVVS ? I read in Unix Review sometime ago that there is an OS >which is NOT based on proprietary AT&T source code and yet it comes very >close to passing the SVVS (I *think* only a couple of tests amongst 7000 >or so tests in the SVVS, failed). Also, Sun has been very innovative >in the last couple of years and yet they either already pass SVVS >(I am not sure if they do or not) or can come very close to it. > >Why do you think SVVS depends upon the AT&T implementation of Unix ? >Can you give us any specifics ? This is not rhetoric. I am genuinely >curious. To start with, Apollo's Domain/OS does indeed pass SVVS; it is not a native kernel. To site one instance of something that might be considered "implementation specific" is that SVVS assumes that system calls (those things from chapter 2) really are system calls. Some of the tests invoke "system calls" and then check that the "system time" increased as a result. This implies that not only must a particular subroutine work as specified (totally reasonable), but must also be a system call (arguable), and must have the side effect of incrementing a global variable (questionable). For instance, our write(2) is not a system call; it is a library call that makes use of our file/virtual memory mapping (very similar to the unimplemented 4.2bsd 'mmap' call). Our write simply copies data to a piece of library- controlled virtual memory to which the file had been previously mapped during the 'open' call (which in our case is not a system call either, but does in fact make a system call to do the mapping). I believe that Sun and HP have implemented similar facilities in their most recent VM implementations. Since our system does lots of stuff like this via our shared libraries, we have never had the dichotomy of user/system time. Therefore, our 'system' time (pre-SVVS) was always 0 with the *actual* (not profiled) time accredited to user time. When AT&T "flunked" us on this one, a waiver was requested and denied (we did get a couple of others). In order to pass, half the time was credited to sys, the other half to user. We passed SVVS, but had lots of complaints and problems with benchmarks (which also assume such things). Since then, we put in profiled times which are more accurate but still don't exactly match a native UNIX due to the library/system call mismatch, and suck up cycles that could be used to execute user processes. Caveats: I didn't actually do any of this myself; 'write' was just an example, I don't know which system calls were being measured; I don't claim that our way is better; I don't claim AT&T was wrong, unfair or arbitrary; I don't represent anybody (I do have a strange little brother who claims to represent God). Flames to /dev/upyours. Gary Allen Apollo Computer Chelmsford, Ma {decvax,yale,umix}!apollo!gallen Roses are red Violets are blue Most poems rhyme But I seem to have trouble making mine do that
gallen@apollo.uucp (Gary Allen) (06/07/88)
In article <55368@sun.uucp> guy@gorodish.Sun.COM (Guy Harris) writes: >Even for the majors that do sell non-UNIX OSes in competition with UNIX, I see >no indication that any of them are, as corporations, solidly opposed to UNIX. >I suspect there are camps within IBM and DEC, at least, that would like UNIX to >go away (I think Apollo, whose own OS may not be as able to stand on its own as >VMS or some of IBM's can, is less likely to have such camps); however, I >suspect that there are also camps within IBM and DEC that would like their >companies to get a significant share of the UNIX market. Refreshing! Finally, someone with an IQ larger than a medium-sized grapefruit! Apollo wants (as I'm sure does Sun, AT&T, IBM, etc) a share of everything. We sell what customers demand, not what some little geek CS major from Some-U finds ideologically appealing. Gary Allen Apollo Computer Chelmsford, Ma {decvax,yale,umix}!apollo!gallen Behind every great computer Sits a skinny little geek. -- Charles LaQuidera WBCN radio, Boston
speyer@apollo.uucp (Bruce Speyer) (06/07/88)
In article <55368@sun.uucp> guy@gorodish.Sun.COM (Guy Harris) writes: >... >Even for the majors that do sell non-UNIX OSes in competition with UNIX, I see >no indication that any of them are, as corporations, solidly opposed to UNIX. >I suspect there are camps within IBM and DEC, at least, that would like UNIX to >go away (I think Apollo, whose own OS may not be as able to stand on its own as >VMS or some of IBM's can, is less likely to have such camps); however, ... If what you mean by "... I think Apollo, whose own OS may not be as able to stand on its own as >VMS or some of IBM's can ..." is that Apollo obviously must solidly support UNIX as their primary OS platform since they don't have the clout to push an Apollo-only AEGIS solution onto a customer - then you are correct. Unix is the key to Apollo surviving and thriving and that simple fact is very well understood. Marketing ploy subsitutes won't work. The latest release of the software puts AEGIS into a maintainance mode. The three supported environments BSD4.3, SYS5.3, and Aegis all use the same kernel, shared global libraries, file system, naming system, and network. What this translates to is that many features that are normally associated with one of these environments is actually available in all three. So the AEGIS user may see future enhancements but it will only be through UNIX improvements. Furthermore, where AEGIS and UNIX conflict it is now the AEGIS OS and user that loses rather than the other way around. For example: case sensitive, allocate free space upon malloc, pathname syntax discrepencies, etc. I hope these arguments help to demonstrate how marketing pressure has given Apollo no option but to make UNIX number one. We have no choice. Whether or not Apollo internally prefers UNIX or AEGIS is a dead issue. Its UNIX or no Apollo. We (at least r&d) are throwing ourselves into UNIX with wild abandon. .. Disclamer: I'm an engineer, not in PR, not management, not privy to OSF, take it or leave it - its my opinion alone. Bruce Speyer CAE engineer Apollo Computer Chelmsford, MA. speyer@apollo.uucp {decvax,yale,umix}!apollo!speyer
jhh@ihlpl.ATT.COM (Haller) (06/08/88)
In article <371@mipseast.mips.COM> rogerk@mipseast.mips.COM (Roger B.A. Klorese) writes: >7% (plus, as I recal, a board seat) can have quite a bit of influence, >especially when AT&T is contractually committed to being able to own >20%. Sun probably has done its best to avoid any hostile takeover attempts with its agreement with AT&T. With 20% ownership, although AT&T may have some influence, it does not have control. And, because AT&T can buy 2x the amount of anyone else that buys more than 20%, other potential takeover artists are likely to look for other, easier targets.
gallen@apollo.uucp (Gary Allen) (06/09/88)
In article <55540@sun.uucp> guy@gorodish.Sun.COM (Guy Harris) writes: >> To site one instance of something that might be considered "implementation >> specific" is that SVVS assumes that system calls (those things from chapter >> 2) really are system calls. Some of the tests invoke "system calls" and then >> check that the "system time" increased as a result. This implies that not >> only must a particular subroutine work as specified (totally reasonable), but >> must also be a system call (arguable), and must have the side effect of >> incrementing a global variable (questionable). > >Umm, to which version of SVVS are you referring? I found no such tests of >system time in SVVS3.0. This is as it should be; neither of the two later >assumptions are "arguable" or "questionable", they are simply *wrong* - the >SVVS should make *no* assumptions about what a "system call" is. [.....some deletions.....] Ok, so are you saying that I should know what I'm talking about before posting? If we all did that, most of these newsgroups would go away! Unfortunately, I posted the anecdotal tale of this instead of obtaining the facts; my apologies. I checked with the person in the know and here are the facts: In the file time.c, there are messages to be printed out to the effect that time (both user and system) are not increasing. On closer inspection of this, those messages only appear if the time DECREASES. In the file exec1.c, there are messages indicating that the child process did not inherit the system and user times of its parent. However, the test actually only checks to see that the times are NON-ZERO. This appears to have been the sticking point for Apollo since our system time was always 0. Note that it could have been any non-zero number and passed this section, I didn't look to see if there were any other peeks at it. Thanks for pointing out my error. Apologies for erroneousness[sp?]. I must have been contaminated by the current administration in Washington. Gary Allen Apollo Computer Chelmsford, MA
guy@gorodish.Sun.COM (Guy Harris) (06/09/88)
> Unfortunately, I posted the anecdotal tale of this instead of obtaining the > facts; my apologies. Apology accepted, but unnecessary; the problem is that it wasn't clear to what you were referring. I thought you were saying that they explicitly made sure that a variety of procedures increased the system CPU time; in fact, they were just *assuming* that certain procedures increased the system CPU time for the purposes of testing. In either case, the assumption being made by the SVVS is bogus; all they should assume is that the *sum* of user and system time increases - and even there, they shouldn't assume that it increases by a measureable amount after a certain amount of real time, but should run in a loop until at least certain amount of real *and* (user+system) CPU time have passed. > I checked with the person in the know and here are the facts: > > In the file time.c, Presumably you mean "times.c" here. > there are messages to be printed out to the effect that time (both user and > system) are not increasing. On closer inspection of this, those messages > only appear if the time DECREASES. In which case, it appears to print a message claiming that there was "No change in {user,system} time", which is a misleading message. "times.c" is buggy; it should be fixed as specified above. > In the file exec1.c, there are messages indicating that the child process > did not inherit the system and user times of its parent. However, the test > actually only checks to see that the times are NON-ZERO. This appears to > have been the sticking point for Apollo since our system time was always 0. In this case, they should test whether the *sum* of user+system time is non-zero. I don't think any of this is the result of a deliberate decision to require that certain procedures be "system calls" whose CPU time is counted as "system time", it's just the result of naive assumptions about implementations and of careless programming. Of course, the reason *why* they botched the SVVS isn't the high-order bit, the fact that it was botched is. Perhaps you should try to present this as a test failure due to a bug in SVVS, rather than presenting it as a test failure for which a waiver should be approved? After all, your system would be *in*compliant with the SVID if it charged time spent in user mode as "system time", unless certain user-mode library routines are to be counted as part of "the system". Requiring you to make your system incompliant with the SVID in order to get it to pass SVVS makes no sense whatsoever....
allbery@ncoast.UUCP (Brandon S. Allbery) (06/10/88)
As quoted from <360@mipseast.mips.COM> by rogerk@mips.COM (Roger B.A. Klorese): +--------------- | In article <54544@sun.uucp> guy@gorodish.Sun.COM (Guy Harris) writes: | >In response to Glenn Weinberg: | >> (Now, AT&T and probably Sun will deny that a SPARC-only | >> ABI was ever their intention, but if you believe that, I too have | >> a bridge to sell you.) | >Well, I believe it, but then I have the disadvantage of working at Sun and | >being too close to the facts; if you're coming from the outside, you can | >believe or disbelieve all sorts of things without having the facts get in the | >way. | | ...Or maybe there was some marketing decision they felt was too poisonous for | you techies to hear. At last year's UNIX Expo in New York, a representative | of Sun Home Office Marketing presented the notion, in a seminar on "Big Iron | on Wall Street", that if the vendors of minisupers would only drop their | proprietary hardware and use SPARC, they'd be able to jump on "the ABI, the | only source of shrink-wrapped UNIX software". Seems pretty explicit to me. +--------------- Sigh. People, I have *yet* to meet a salesman who got technical issues right. Before you go off flaming ABI proponents, remember the following: If you tell a marketing type that "we're coming up with an application binary interface in order to be able to sell shrink-wrap software -- of course, an ABI only works on a single CPU type", it's fairly certain that the marketing type will NOT recognize that the limitation is the fact that a 68020 ABI will *never* work on a SPARC CPU. Instead, the marketing type will conclude that you want the ABI for exactly one CPU type -- marketing- type thinking. (Keep in mind that to be at all workable, the concept of an ABI would *have* to have been developed by tech types -- the salesmen would insist on an ABI that spanned all processor types, since they don't know any better.) Don't assume ill will where common stupidity is sufficient. -- Brandon S. Allbery | "Given its constituency, the only uunet!marque,sun!mandrill}!ncoast!allbery | thing I expect to be "open" about Delphi: ALLBERY MCI Mail: BALLBERY | [the Open Software Foundation] is comp.sources.misc: ncoast!sources-misc | its mouth." --John Gilmore
allbery@ncoast.UUCP (Brandon S. Allbery) (06/10/88)
As quoted from <3c67f93d.ddc1@apollo.uucp> by perry@apollo.uucp (Jim Perry): +--------------- | Bell Labs was not in the commercial computer business and therefore made | it widely available, and many people implemented it on a variety of hardware. | Now AT&T (along with Sun) is in the position of developing it in the character | of a product rather than a research item. The argument goes that if AT&T+Sun +--------------- Bogus argument. AT&T has been developing Unix as a commercial product since 1984 at least, quite successfully, and doing all those evil things you mentioned with it. Sun has been doing so for at least that long as well. Why, then, did it take 4 years for the OSF to be founded? -- Brandon S. Allbery | "Given its constituency, the only uunet!marque,sun!mandrill}!ncoast!allbery | thing I expect to be "open" about Delphi: ALLBERY MCI Mail: BALLBERY | [the Open Software Foundation] is comp.sources.misc: ncoast!sources-misc | its mouth." --John Gilmore
allbery@ncoast.UUCP (Brandon S. Allbery) (06/11/88)
As quoted from <1017@sun.soe.clarkson.edu> by nelson@sun.soe.clarkson.edu (Russ Nelson): +--------------- | I'm surprised that no one has come to the following conclusion: | | AT&T+Sun AND the OSF are gathering the wagons into circles because they're | afraid of the GNUs. Think about what's going to happen when rms finishes +--------------- WHO hasn't come to that conclusion? Check my "desperation move" article again. GNU-style full openness is preferable to two lockouts. +--------------- | the kernel? There is a tremendous market opportunity for a firm that | simply offers support for GNUnix. They need no investment other than | that needed to become a GNUnix wizard. And, because of the restrictions | on derivatives of GNU, they will benefit from everyone else's work. +--------------- Before anyone says "can't happen": ask yourself "what is mtXinu?" That's right... it's been done before. It WILL be done if AT&T and/or the OSF don't keep their mercenary tendencies in check. -- Brandon S. Allbery | "Given its constituency, the only uunet!marque,sun!mandrill}!ncoast!allbery | thing I expect to be "open" about Delphi: ALLBERY MCI Mail: BALLBERY | [the Open Software Foundation] is comp.sources.misc: ncoast!sources-misc | its mouth." --John Gilmore
rogerk@mips.COM (Roger B.A. Klorese) (06/13/88)
In article <7942@ncoast.UUCP> allbery@ncoast.UUCP (Brandon S. Allbery) writes: >People, I have *yet* to meet a salesman who got technical issues right. The person in question was a corporate MARKETING manager, not a salesman. You know, product planning and marketing? The ones who TELL engineering which of their brainstorms will and will not be shipped, in what guise, and to whom? -- Roger B.A. Klorese MIPS Computer Systems, Inc. {ames,decwrl,prls,pyramid}!mips!rogerk 25 Burlington Mall Rd, Suite 300 rogerk@mips.COM Burlington, MA 01803 I don't think we're in toto any more, Kansas... +1 617 270-0613
friedl@vsi.UUCP (Stephen J. Friedl) (06/13/88)
In article <7942@ncoast.UUCP>, allbery@ncoast.UUCP (Brandon S. Allbery) writes: > People, I have *yet* to meet a salesman who got technical issues right. Q - What's the difference between a car salesman and a computer salesman? A - A car salesman *knows* he's lying. :-) :-) :-) :-) :-) Steve :-) :-) :-) :-) :-) -- Steve Friedl V-Systems, Inc. (714) 545-6442 3B2-kind-of-guy friedl@vsi.com {backbones}!vsi.com!friedl attmail!vsi!friedl Nancy Reagan on the Mac-II architecture: "Just say Nu"
grzesiak@a3bee2.UUCP (John Grzesiak ) (07/01/88)
In article <23148@bu-cs.BU.EDU>, bzs@bu-cs.BU.EDU (Barry Shein) writes: >Then show us, why AT&T would be able to influence Sun with only 7% of >the stock when it was unable to influence Olivetti with over 20%. > > Eduardo Krell AT&T Bell Laboratories, Murray Hill, NJ Why indeed, I belong to a company that would give it's eye teeth to be developing with AT&T. It seems to me to be a feather in the cap....not to mention (possibly) pocket! John Grzesiak GHM Inc , East Hartford ,Ct