andy@Theory.Stanford.EDU (Andy Freeman) (11/21/90)
In article <EGDORF.90Nov20102032@zaphod.lanl.gov> egdorf@zaphod.lanl.gov (Skip Egdorf) writes: >I think that the point is well made and that Sun (for all its faults) >is unique in the industry for such a long-term view. In that sense, it >is "an extraordinary action!" I also believe that this is the proper >way to ensure the domination of the SPARC architecture regardless of >the technical quibbles voiced so often (and often correctly!) in this >newsgroup. While they didn't have to be sued into doing it Sun, with SPARC, has done what IBM did decades ago with their 360/370 machines. Both allow others to produce hardware compatible systems. In what sense is Sun "first", let alone "unique"? What is its equiv of the IBM Technical Disclosure Bulletin? Hint: disclosure via source code is convenient, but it isn't a guarantee of future compatibility. Considering the bugs in the distributed version, it isn't even a guarantee of current compatibility. -andy -- UUCP: {arpa gateways, sun, decwrl, uunet, rutgers}!neon.stanford.edu!andy ARPA: andy@neon.stanford.edu BELLNET: (415) 723-3088
davidsen@crdos1.crd.ge.COM (Wm E Davidsen Jr) (11/21/90)
In article <1990Nov20.195623.28061@Neon.Stanford.EDU> andy@Theory.Stanford.EDU (Andy Freeman) writes: | While they didn't have to be sued into doing it Sun, with SPARC, has | done what IBM did decades ago with their 360/370 machines. Both allow | others to produce hardware compatible systems. Where did you get the info? I didn't realize that IBM gave Amdahl and the others the plans for their CPU and let them make it. Actually I thought they were reverse engineered. Does the license to manufacture include the right to use the software, like SPARC? | In what sense is Sun "first", let alone "unique"? What is its equiv | of the IBM Technical Disclosure Bulletin? Hint: disclosure via source | code is convenient, but it isn't a guarantee of future compatibility. | Considering the bugs in the distributed version, it isn't even a | guarantee of current compatibility. Bear in mind the the TDB is intended to protect IBM, not benefit the public. Not that it isn't useful, but it's not a gesture of goodwill. The basic idea is that once IBM decides they can't make money by using or licensing an idea, they tell the world, so no one else can patent it. A cheap way to assure that if they ever do decide to use it they won't have to pay for it. -- bill davidsen (davidsen@crdos1.crd.GE.COM -or- uunet!crdgw1!crdos1!davidsen) VMS is a text-only adventure game. If you win you can use unix.
andy@Theory.Stanford.EDU (Andy Freeman) (11/22/90)
In article <2940@crdos1.crd.ge.COM> davidsen@crdos1.crd.ge.com (bill davidsen) writes: >In article <1990Nov20.195623.28061@Neon.Stanford.EDU> andy@Theory.Stanford.EDU (Andy Freeman) writes: >| While they didn't have to be sued into doing it Sun, with SPARC, has >| done what IBM did decades ago with their 360/370 machines. Both allow >| others to produce hardware compatible systems. > > Where did you get the info? I didn't realize that IBM gave Amdahl and >the others the plans for their CPU and let them make it. Actually I >thought they were reverse engineered. Either IBM's lawyers are worse than almost every other manufacturers, or IBM allowed the 360/370 clone makers to exist. Sun helps more, but that's a difference of degree. >Does the license to manufacture include the right to use the software, >like SPARC? I don't know whether or not IBM will license its software to run on non-IBM hardware, but 3rd party suppliers will. The clone makers provide enough to make it reasonable for a customer to buy a 360/370 that isn't blue, so IBM proprietary software hasn't closed the market. To the extent that Sun licenses everything they write to run on non-Sun hardware, sun is merely discouraging competition on it. Note that there isn't a VMS clone for non-Dec vaxen. >| In what sense is Sun "first", let alone "unique"? What is its equiv >| of the IBM Technical Disclosure Bulletin? Hint: disclosure via source >| code is convenient, but it isn't a guarantee of future compatibility. >| Considering the bugs in the distributed version, it isn't even a >| guarantee of current compatibility. > > Bear in mind the the TDB is intended to protect IBM, not benefit the >public. Not that it isn't useful, but it's not a gesture of goodwill. >The basic idea is that once IBM decides they can't make money by using >or licensing an idea, they tell the world, so no one else can patent it. >A cheap way to assure that if they ever do decide to use it they won't >have to pay for it. It may be in IBM's best interests, but I wonder why other computer companies don't see things the same way. -andy -- UUCP: {arpa gateways, sun, decwrl, uunet, rutgers}!neon.stanford.edu!andy ARPA: andy@neon.stanford.edu BELLNET: (415) 723-3088
egdorf@zaphod.lanl.gov (Skip Egdorf) (11/22/90)
In article <2940@crdos1.crd.ge.COM> davidsen@crdos1.crd.ge.COM (Wm E Davidsen Jr) writes: > | While they didn't have to be sued into doing it Sun, with SPARC, has > | done what IBM did decades ago with their 360/370 machines. Both allow > | others to produce hardware compatible systems. > > Where did you get the info? I didn't realize that IBM gave Amdahl and > the others the plans for their CPU and let them make it. Actually I > thought they were reverse engineered. Gene Amdahl DESIGNED the 360. He then left to build a "better" version from his new company. > Does the license to manufacture > include the right to use the software, like SPARC? Fujitsu won a long-standing fight a few years back that forced IBM to sell OS sources rather than witholding them in an attempt to stave off the clone makers. (others with better memories of this may wish to correct/expand) > > bill davidsen (davidsen@crdos1.crd.GE.COM -or- uunet!crdgw1!crdos1!davidsen) It is interesting, in this day of stretching copyright and patent to cover anything related to computers, to speculate on the legal ramifications of producing a VAX or 386 or MIPS instruction-set clone. I don't mean copying design and microcode, but a pure reverse engineering from the instruction-set up. Were there any issues in the Intel AMD tiff that related to AMD's ability to produce a chip with the same instructions-set etc. as the 386 rather than details of who gets what layout schematics and microcode? Would Amdahl, NAS, etc. get sued by IBM in today's environment if they were to do their cloning afresh? What legal recourse does Sun (or the "free" spin-off that controlls SPARC) have if IEEE does make a SPARC instruction-set public domain? (gee, I don't really have to change that subject line after all...) Skip Egdorf hwe@lanl.gov
khb@chiba.Eng.Sun.COM (Keith Bierman fpgroup) (11/22/90)
In article <1990Nov21.194240.28355@Neon.Stanford.EDU> andy@Theory.Stanford.EDU (Andy Freeman) writes:
Either IBM's lawyers are worse than almost every other manufacturers,
Far better than most. They fought hard and lost. Don't forget the
large number of anti-trust cases (including the one with CDC ....
which was settled out of court by yeilding an entire IBM division to
CDC).
or IBM allowed the 360/370 clone makers to exist.....
I don't know whether or not IBM will license its software to run on
After suits, yes. It was the great unbundling war.
Note that there isn't a VMS clone for non-Dec vaxen.
Note that there WAS. At least one, Elxsi. I vaguely recall at least
one other, but they too are history.
It may be in IBM's best interests, but I wonder why other computer
companies don't see things the same way.
Other companies use other journals. There is one for DEC. There are
normal journals, like those of IEEE and ACM. IBM is large and
produtive enough to keep a mill going full time ;>
--
----------------------------------------------------------------
Keith H. Bierman kbierman@Eng.Sun.COM | khb@chiba.Eng.Sun.COM
SMI 2550 Garcia 12-33 | (415 336 2648)
Mountain View, CA 94043dhinds@elaine10.stanford.edu (David Hinds) (11/22/90)
In article <1990Nov21.194240.28355@Neon.Stanford.EDU> andy@Theory.Stanford.EDU (Andy Freeman) writes: >In article <2940@crdos1.crd.ge.COM> davidsen@crdos1.crd.ge.com (bill davidsen) writes: >> >> Where did you get the info? I didn't realize that IBM gave Amdahl and >>the others the plans for their CPU and let them make it. Actually I >>thought they were reverse engineered. > >Either IBM's lawyers are worse than almost every other manufacturers, >or IBM allowed the 360/370 clone makers to exist. Sun helps more, but >that's a difference of degree. I believe that IBM OS's are public domain - you can just pick up an MVS tape and pop it into an Amdahl clone. However, I think that some of this was forced on IBM as a result of anti-trust suits a long time ago - I don't remember the details. -David Hinds dhinds@cb-iris.stanford.edu
aglew@crhc.uiuc.edu (Andy Glew) (11/22/90)
>Either IBM's lawyers are worse than almost every other manufacturers, >or IBM allowed the 360/370 clone makers to exist. Sun helps more, but >that's a difference of degree. IBM being under the shadow of an anti-trust prosecution at the time certainly help motivate IBM not to pursue the clone makers. Also, in Europe an anti-trust decision required IBM to provide technical details of their I/O system interface to clone makers. -- Andy Glew, a-glew@uiuc.edu [get ph nameserver from uxc.cso.uiuc.edu:net/qi]
jmaynard@thesis1.hsch.utexas.edu (Jay Maynard) (11/22/90)
In article <1990Nov21.205753.13677@portia.Stanford.EDU> dhinds@elaine10.stanford.edu (David Hinds) writes: > I believe that IBM OS's are public domain - you can just pick up an MVS >tape and pop it into an Amdahl clone. However, I think that some of this >was forced on IBM as a result of anti-trust suits a long time ago - I don't >remember the details. This was true once upon a time, but hasn't been true since the introduction of MVS/SE (the predecessor to MVS/SP). I don't know if you can still get MVS 3.8 (the last PD version) all by itself; later versions still contain MVS 3.8 code, but it would take major surgery to separate it from the licensed code, especially in later systems such as MVS/ESA (otherwise known as MVS/SP version 3). I don't know houw an Amdahl 5990 site would get MVS/ESA - or any other IBM program product - licensed, but I know it's possible; I do know that Amdahl users get their support from Amdahl, not IBM. -- Jay Maynard, EMT-P, K5ZC, PP-ASEL | Never ascribe to malice that which can jmaynard@thesis1.hsch.utexas.edu | adequately be explained by stupidity. "With design like this, who needs bugs?" - Boyd Roberts
dricejb@drilex.UUCP (Craig Jackson drilex1) (11/25/90)
In article <4343@lib.tmc.edu> jmaynard@thesis1.hsch.utexas.edu (Jay Maynard) writes: > [ A response to an article asserting that IBM system software is public > domain, and discussing how Amdahl users get their O.S. software. ] > [ It was pointed out that everything after MVS/SE is not PD. ] > >I don't know houw an Amdahl 5990 site would get MVS/ESA - or any other IBM >program product - licensed, but I know it's possible; I do know that Amdahl >users get their support from Amdahl, not IBM. When we had our Amdahl, I believe our 'support' came from IBM for IBM code. However, there were Amdahl patches which were required to run many products, so we had to wait for each new release. Amdahl gets their IBM software the same way that any other IBM owner would-- they of course have to buy one of nearly every processor IBM ships. (At SHARE, the IBM user's group, Amdahl users technically cannot join. However, site 'AMD' has always one of the biggest representations--legally, because of their IBM boxes. It's always interesting seeing an Amdahl guy asking a 'user question' at an IBM presentation. Caveat: things may have changed, I haven't been to SHARE for about three years.) All this gets around to: if you aren't in anti-trust danger, you can do all you can to screw your competitors, especially cloners. If you *are* in anti-trust danger, the whole color of normal business decisions changes..... Whether a company is in danger depends on their size, their attitude, and who occupies the White House. -- Craig Jackson dricejb@drilex.dri.mgh.com {bbn,axiom,redsox,atexnet,ka3ovk}!drilex!{dricej,dricejb}
herrickd@iccgcc.decnet.ab.com (11/26/90)
In article <2940@crdos1.crd.ge.COM>, davidsen@crdos1.crd.ge.COM (Wm E Davidsen Jr) writes: > In article <1990Nov20.195623.28061@Neon.Stanford.EDU> andy@Theory.Stanford.EDU (Andy Freeman) writes: > > | While they didn't have to be sued into doing it Sun, with SPARC, has > | done what IBM did decades ago with their 360/370 machines. Both allow > | others to produce hardware compatible systems. > > Where did you get the info? I didn't realize that IBM gave Amdahl and > the others the plans for their CPU and let them make it. Actually I > thought they were reverse engineered. I think a better word might be forward engineered. IBM set up an internal competition to design the next generation of System 360. There were something like thirteen teams in the competition. Gene Amdahl headed one of those teams. The products of about four of the teams were combined to get the architecture of System 370. Gene's was one of the four. He was an important IBM manager in bringing 370 to market. I got the impression reading industry rags at the time that he left because he thought the decision making process in IBM had created a competitive opportunity. (In other words, no more compromises, let's do it right this time.) Of course, Amdahl is big enough now to have hardening of the categories. >Does the license to manufacture > include the right to use the software, like SPARC? > > | In what sense is Sun "first", let alone "unique"? What is its equiv > | of the IBM Technical Disclosure Bulletin? Hint: disclosure via source > | code is convenient, but it isn't a guarantee of future compatibility. > | Considering the bugs in the distributed version, it isn't even a > | guarantee of current compatibility. > > Bear in mind the the TDB is intended to protect IBM, not benefit the > public. Not that it isn't useful, but it's not a gesture of goodwill. > The basic idea is that once IBM decides they can't make money by using > or licensing an idea, they tell the world, so no one else can patent it. > A cheap way to assure that if they ever do decide to use it they won't > have to pay for it. > -- > bill davidsen (davidsen@crdos1.crd.GE.COM -or- uunet!crdgw1!crdos1!davidsen) > VMS is a text-only adventure game. If you win you can use unix. dan herrick herrickd@astro.pc.ab.com
xxremak@csduts1.lerc.nasa.gov (David A. Remaklus) (11/27/90)
In article <4343@lib.tmc.edu> jmaynard@thesis1.hsch.utexas.edu (Jay Maynard) writes: > (stuff deleted) > >I don't know houw an Amdahl 5990 site would get MVS/ESA - or any other IBM >program product - licensed, but I know it's possible; I do know that Amdahl >users get their support from Amdahl, not IBM. >-- An Amdahl user can order any IBM licensed software paying the same license fee that a user of a comperable IBM machine would pay. As a licensee, they are entitled to the same level of support from IBM as an IBM customer would receive. In addition they can also obtain support and assistance for problems with the IBM software from Amdahl. A later posting to this thread commented that numerous Amdahl changes and patches were required. This is no longer the case. Amdahl poineered a firmware feature named Macrocode which provides for all Recovery Management Services and remaining IBM compatibility (that not directly supported by the box). As part of Macrocode, they introduced the Multiple Domain Feature (MDF). MDF permits a customer to operate multiple OS'es on the same physical hardware as if each had total control of the machine. It is interesting that IBM has cloned this feature as PRSM as has HDS. It is now safe to say that in this case, IBM is the clone. -- David A. Remaklus Currently at: NASA Lewis Research Center Amdahl Corporation MS 142-4 (216) 642-1044 Cleveland, Ohio 44135 (216) 433-5119 xxremak@csduts1.lerc.nasa.gov