Return to BSD News archive
Xref: sserve comp.os.386bsd.questions:12255 comp.os.386bsd.misc:3142 Path: sserve!newshost.anu.edu.au!harbinger.cc.monash.edu.au!msuinfo!agate!howland.reston.ans.net!news.cac.psu.edu!news.pop.psu.edu!ra.nrl.navy.mil!sundance!cmetz From: cmetz@sundance.itd.nrl.navy.mil (Craig Metz) Newsgroups: comp.os.386bsd.questions,comp.os.386bsd.misc Subject: Re: Whats wrong with Linux networking ??? Date: 11 Aug 1994 13:35:58 GMT Organization: Information Technology Division, Naval Research Laboratory Lines: 102 Message-ID: <32d9fu$fbi@ra.nrl.navy.mil> References: <RSANDERS.94Aug9003813@hrothgar.mindspring.com> <CuA6w1.5tF@calcite.rhyolite.com> <32acqn$ghb@ra.nrl.navy.mil> <CuBzq9.BMx@calcite.rhyolite.com> NNTP-Posting-Host: sundance.itd.nrl.navy.mil In article <CuBzq9.BMx@calcite.rhyolite.com>, Vernon Schryver <vjs@calcite.rhyolite.com> wrote: >Have I looked at Linux? no. >Have I looked at the U.Guelph NFS code? yes. >Have I looked at other NFS code? yes. For example, I worked on the first por > of NFS to SVR3 as a Lachman employee in Bob Lyon's group at Sun. Oh, I could tell stories about how much sorrow LAI products have brought me... >Have I ever put 4.*BSD network code into a non-BSD system? yes. >More than once? yes. Into a non-UNIX system? Remember, you're talking about no ancestory here. It really shows that you've never looked at Linux. You just don't get it. >Do I think that my spare, recreational kernel hacking time would be well > spent putting the U.Guelph NFS code into Linux using the renowned > Linux networking code? no. It's easy to say that *other* people should do things. >Is it my professional judgement having looked at and ported NFS source > that porting the U.Guelph NFS source to any system with reasonable > UDP/IP networking would be a tiny job compared to writing a new NFS > clone for scratch for that system, even if the target system is not > like UNIX and/or not written in C? yes. Well, that's very nice. >Given that the last several articles in this thread have been confined > to comp.os.386bsd.questions and comp.os.386bsd.misc, do I think > you've made any converts? no. Only one of us is here pushing philosophy. I am the first to admit that the Linux network code and NFS code sucks. However, you immediately reject anything that is not BSD as inherently inferior without any consideration. Then you immediately respond to any further questioning with insults ``oh, that's just student code'' and trying to pull rank, rather than actually address things. >>>There was an excuse for inception of Linux. Big-bad-nasty-mean-USL/AT&T >>>is obviously a sufficent reason for most of the kernel. Who says >>>otherwise? >> Well, Linus. But he's not important, right? >What? He says that USL/AT&T nastiness an <<insufficient>> reason >for the core of Linux? I'm assuming that you meant to say ``is an insufficient''... the main initial reason for the inception of Linux was as an exercise by Linus himself to see if he could write a simple operating system. At the time, it was not designed to be UNIX-like, though it was somewhat MINIX-like. Contrary to your hardened beliefs, Linux did not come about because a bunch of arrogant people decided that the One True UNIX (pure BSD, of course) wasn't good enough for them and that it had to be rewritten from scratch because only they could do it. Initially, it was an exercise - to see if it could even be done. Face some facts - there are not exactly a lot of original, from-scratch UNIX systems out there. NIH, you say. NIH!, NIH! If the (in many places, but certainly NOT in all) from-scratch work of a lot of random people on the Net is so far inferior to your precious BSD, then why do you feel the need to continuously take pot-shots at it? If you're happy with your BSD being superior to everything else for all time, then why not just ignore Linux? Or do you just enjoy taking pot-shots at the work of people you consider to be inferior to you? >> Maybe (gasp!) Linux isn't a clone... > >That's nonsense. Of course Linux is a clone. "Clone" does not mean >"recompile", it means "write from scratch something with the same external >interfaces." Linux is supposed to have the same form, fit and function >of UNIX(tm). That makes it a UNIX(tm) clone. Linux is compatible, not a clone. A clone is made to work exactly the same as the original. A compatible is made to appear to work exactly the same as the original. There is a subtle, but extremely important, difference. It's not what it does, it's how it does it, that is different. >This inclination to label the student exercises or bad-NIH parts of >Linux (e.g. Linux networking and NFS) as other than they are is not >sane. Things are what they are. Calling one's student excercises Great >Works or declaring that one's clone of a mildly popular operating system >is something more original is not an indication of a good adjustment to >reality. Student work is very valuable, not because its direct utility >but because of what it has taught the student. Clones can be good, >breaking noxious license restrictions or improving on the original. >(e.g the IBM-PC BIOS clones did both.) Nevertheless, they are what they >are. Am I correct to summarize your opinion: 1. Anything that is not BSD is inferior to BSD 2. Linux is not BSD 3. Linux is inferior to BSD 4. Linux is not valuable as a work If this is really your opinion, then why do you care? Why do you insist on taking every possible opportunity to say something bad (and frequently unfounded) about Linux? -Craig