Return to BSD News archive
Xref: sserve comp.os.386bsd.development:797 comp.os.386bsd.misc:396 Path: sserve!newshost.anu.edu.au!munnari.oz.au!news.Hawaii.Edu!ames!agate!soda.berkeley.edu!wjolitz From: wjolitz@soda.berkeley.edu (William F. Jolitz) Newsgroups: comp.os.386bsd.development,comp.os.386bsd.misc Subject: Re: SHARED LIBRARIES - THE END Date: 25 May 1993 18:24:16 GMT Organization: U.C. Berkeley, CS Undergraduate Association Lines: 99 Message-ID: <1tto8g$off@agate.berkeley.edu> References: <1993May23.003623.24102@fcom.cc.utah.edu> <1tr05o$qaa@terminator.rs.itd.umich.edu> <1993May24.225014.23425@fcom.cc.utah.edu> NNTP-Posting-Host: soda.berkeley.edu Terry, You've done an admirable job of explaining the details and background of the problem, and I'm quite please at the balance you provided with it while still showing your opinion. That's precisely the kind of discourse I enjoy when reading netnews for feedback. I wish there was more of it as well from others. Unsurprisingly, I don't agree with you about pragmatics, and since we're in a unique position here and it's hard to provide a sufficently detailed view of it, I will refrain from explaining it this time around. The underlying choices present here are made entirely on the basis of time, resources, and user base. We try utmost to set a good standard for quality, content, and software direction (both with work from others as well as ourselves) in the releases, modulo the pragmatics already mentioned. Sometimes we miss -- and we count on those with agile minds to tell us how we missed and why. If the position is defensible, you bet we'll listen and move the position. However, if it's not, we'll just continue to listen and consider. As to the supposed "politics" -- all we care about is the following: a) that the user base can continue to use the software without assuming additional obligations other than attribution, b) that we can continue to provide it in subsequent versions to our user base, and c) that we can verify that the work represented under our name or of that of 386BSD is appropriate to be labelled as such (since we *must* accept responsibility for the consequences). In addition, we're overtaxed even doing just one major release, so we can't yet participate in numerous others going in divergent directions. There are groups which have worked with us to meet the needs of users which we could not service directly, and that has been of great help to us and 386BSD, since providing access to 386BSD is important -- what good is teaching people to use 386BSD if they can't get a copy? Two commercial firms, Infomagic and Tuttle Designs (Germany) have done so for 0.1, and it helped users to get the real 386BSD that they read about. Not imitations or derivations, but the real thing. We've heard from these people, via phone, email, and written correspondance, and they're happy they have had an opportunity to participate in this effort. Some of our research associates in goverment labs and universities also have been "straight shooters". The upshot is that everyone wins, especially the 386BSD user. However, others have not been so helpful, all the while saying different things publicly and privately. I wish they would play fair, but the net effect is just more irrelevancy that we are straining to avoid. I'm sorry for the confusion and aggravation. The only people I'm really annoyed with are those who know all of this anyways and see fit to inject it back into the discourse ocassionally. Nothing really has changed in certain quarters since December of 1991. The goals remain the same. The "Road Not Taken" (March 1992) applies now, perhaps more than ever before. We did not have an opportunity to be as critical as we had liked with 0.1. This release we are attempting to be far more exacting, and I'm certain that results will be appreciated. Please continue to be as technically critical of details to us as you can in email, and the net, both before and after the release. But realize that 386BSD's audience is far larger than just the net, and tastes along are not uniform in this broad group. Ironically, we've been told that we cater to the net too much. I think it helps, however, to remind people of the forest, and stop niggling about a fungus on the trunk. There are larger issues than just "my religion, right or wrong", and we pay heed to it. In my opinion, narrow-minded uncritical (technically) just plain arrogance dooms the UNIX philosophy of systems to a fourth-rate position in the future, in comparison to the more methodical approach of the world's largest (and lousiest) operating system vendor Microsoft (I can't really slam success, and I won't). I've held this view since the late 70's, and seen no change whatsoever in the better part of 20 years. This is why when I talk to a major newspaper columnist and he tells me his editor won't allow stories longer then a couple column inches on UNIX, I can understand why. Flame me all you want, it still won't change this basic fundamental. In any case, those who wish to stagnate redoing the stale homilies of the past will only have a few years to continue the argument before it is rendered irrelevant anyways by the methodical, pragmatic march of the most successful software giant around. Even when you have something innovative as an alternative like NextStep, and a reputation as Steve Jobs has, people are so cynical that they'll wait TWO YEARS (or more) for Microsoft to dodder along and work it into their plans eventually. This *amazes* me. Before you flame yet again, no, I'm not comparing 386BSD to the "big guys" at all. I'm just trying to illustrate the significance of my remarks above in the wider perspective of the world. Also, to the deconstrutionists that latch upon the fragments of this posting to vomit forth, don't bother -- you'll have better luck using a wide-angle lens to see the point, and I *don't* make any claims that this posting will stand up to rigorous semantic analysis. All I ask is that you think a little beyond the present. Thanks for listening, and keep on pressing the issues of the moment, especially in a way by which we can see the scope as well as the opinion. Bill Jolitz.