louie@sayshell.umd.edu (Louis A. Mamakos) (10/08/90)
All this flaming about how CMU doesn't make it easy to run Mach on <my favorite platform> for free is starting to annoy me. Why do people have an expectation that research organization have the resources to do this sort of thing? In article <8883@helios.TAMU.EDU> cnh5730@calvin.tamu.edu (Chuck Herrick) writes: >It is a fact that a lion's share of the funding for Mach has come >directly from the coffers of the U.S. government in the form of tax >dollars for funding of CMU (Carnegie Mellon University in >Pittsburg, PA). Do you know the particulars of the funding that CMU receives? I hardly think that they are funded by the feds to produce a release of software; I'd expect that the funding is for research in operating systems for multi-processor platforms and architectures. Why would you expect that producing a software "product" for release would further those goals? The folks are Berkeley are in somewhat the same boat. >While CMU has said that they will eventually rewrite >all of UNIX, incorporating the result into Mach, currently only the >kernel has been completed. I don't think that I've heard them say this. I suspect that they said they would produce a Mach kernel which is free from AT&T licensing restrictions. The Mach kernel is not UNIX. >Until they make good on their word, the >only people who can really use Mach are those who are willing to >pay the licensing fees to AT&T for the non-Mach UNIX residual in >the operating system. And these are the folks that can help contribute the the work that CMU is doing. Most educations sites don't have expensive licensing problems. >And it costs a bundle, folks. That it "costs a bundle" is an artifact of AT&T, not CMU. If you disagree with the pricing that AT&T sets for their code, go elsewhere or take it up with AT&T. >Of course, until >the Mach project is completed, really the only outfits who can profit >from our tax-dollar supported Mach are those who are can afford to >pay the pipers at AT&T. Also note that this is probably true of the >situation with BSD UNIX at Berkeley. > Those who wish to be a recipient of the benefit generated from >their own tax dollars should speak out to CMU through their own >voices (and perhaps those of their elected representatives) in order >to avoid another Berkeley-Sun debacle. In other words, you all >paid for BSD UNIX to be done at Berkeley with your tax dollars... and >now when you buy a Sun you're paying for it again, since Sun was >started by some of the same people who did BSD at the Berk and who >then took it to sell it for profit. What drivel. You clearly have no idea of the effort required to do a "release" of software as opposed to writing software to futher your research. Just the packaging of software onto media is difficult enough (how do you make 1000 copies of a 9 track tape distribution?). Who's going to write the documentation? Is CMU supposed to hire folks to man the phones and ask stupid questions? I hope not! I want the funding those guys receive to go toward doing productive research, and not a task that private industry seems willing to handle. I applaude folks like NeXT and Mt. Xinu for making recent technology available to groups outside the research community as quickly as they have. These are the guys who have taken the risk to buy a source license and binary redistribution licenses from AT&T, and to the integration and testing to release the software. Don't underestimate the effort involved in such a task.
ast@cs.vu.nl (Andy Tanenbaum) (10/23/90)
In article <1990Oct8.032554.23303@ni.umd.edu> louie@sayshell.umd.edu (Louis A. Mamakos) writes: > >All this flaming about how CMU doesn't make it easy to run Mach on ><my favorite platform> for free is starting to annoy me. Why do people >have an expectation that research organization have the resources to >do this sort of thing? I certainly agree with the premise that neither CMU nor any other university has any obligation to do all the dirty work of making a distribution. And believe me, I can testify from actual experience that it is a lot of work. Nevertheless, I believe CMU has indeed had a LOT of (tax) money from the government (e.g., DARPA), for BOTH research and development, thus enabling Carnegie-Mellon to give it away for free, provided that somebody else does the packaging. Since I'm in Europe, the situation is slightly different, but I can certainly imagine an American company that wants to compete in the UNIX-successor marketplace being pretty annoyed that Uncle Sam is funding their competitor to give them a big price advantage. If the funding agency were MITI instead of DARPA, and the recipient were the University of Tokyo instead of CMU, Uncle Sam would be screaming Unfair! Andy Tanenbaum (ast@cs.vu.nl)
bader+@andrew.cmu.edu (Miles Bader) (10/28/90)
ast@cs.vu.nl (Andy Tanenbaum) writes: > but I can certainly imagine an American company that wants to compete > in the UNIX-successor marketplace being pretty annoyed that Uncle Sam > is funding their competitor to give them a big price advantage. Annoyed companies can get a mach distribution, buy all the redistribution licences from AT&T, etc., and sell mach themselves... -Miles