Return to BSD News archive
Path: euryale.cc.adfa.oz.au!newshost.anu.edu.au!harbinger.cc.monash.edu.au!news.cs.su.oz.au!metro!metro!inferno.mpx.com.au!news.unimelb.EDU.AU!munnari.OZ.AU!news.ecn.uoknor.edu!solace!nntp.uio.no!news.cais.net!newsfeed.internetmci.com!newsxfer2.itd.umich.edu!agate!reason.cdrom.com!usenet From: "Jordan K. Hubbard" <jkh@FreeBSD.org> Newsgroups: comp.unix.bsd.freebsd.misc Subject: Re: Linux vs. FreeBSD ... Date: Thu, 09 May 1996 01:26:30 -0700 Organization: Walnut Creek CDROM Lines: 67 Message-ID: <3191AC36.41C67EA6@FreeBSD.org> References: <3188C1E2.45AE@onramp.net> <4mnsc5$6qo@sundial.sundial.net> <xcd3f5csajb.fsf@woodlawn.uchicago.edu> NNTP-Posting-Host: time.cdrom.com Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Mailer: Mozilla 3.0b2 (X11; I; FreeBSD 2.2-CURRENT i386) CC: b.j.smith@ieee.org Soren Dayton wrote: > > Our friend, b.j.smith@ieee.org (Bryan J. Smith, E.I.), wrote: > > > Linux, since it is written from the ground-up, is a much more efficient OS > > than FreeBSD (which has been written somewhat from the ground up, to prevent > > a lawsuit from BSDI, is still a lot of legacy OS code). > > You are going to have to explain this causal relationship to my little > mind Indeed, this is quite simply completely and utterly wrong (and shows that Mr Smith honestly doesn't know what the heck he's talking about). First off, being written from "the ground up" is hardly a guarantee of *anything* for those who actually understand software engineering. Sometimes this results in better code and sometimes not. In the case of Linux's networking code, I think _not_ is a fair assessment. There are literally thousands of weird interoperability issues with TCP/IP and these *take time* to find and fix. The BSD networking code has had some 10 years of very careful testing and tuning whereas the Linux folks have a fair ways to go. Will they get there? Sure, I've little doubt of that. Are they there now? No. Even the most die-hard Linux devotee will generally admit that one. To give Linux its due, in other areas, like dynamic loading of kernel objects, re-engineering things certainly didn't hurt and they were able to escape certain historical warts that BSD inheirited. Those are being worked on now, of course, and BOTH operating systems are making progress on their respective shortcomings. I'm not out to slander Linux here, simply to point out that pat summaries like Bryan's are simply laughably naive. It's also more than fair to point out that 4.4 BSD, upon which FreeBSD is now based, was hardly a mouldy pile of antiquated code. A *lot* of new stuff (like stackable filesystems, portals, LFS, etc) has been implemented "from the ground up" in BSD 4.4, and it is, in fact, much of that new code which is going to require the most polishing work for us. Yet another argument against "ground up superiority", I think. :-) This is not to say that the new 4.4 code is bad, far from it, but the oldest code from the CSRG is actually some of the most refined. Take Kirk McKusick's operating systems class someday (which uses FreeBSD) and you'll see that it's pretty damn difficult to catch him out on anything during the code review. Every time I've thought I spotted a flaw and raised my hand, there has turned out to be a very good reason for it. :-) The CSRG people were no dummies, that's for sure, and very experienced indeed in the art of creating operating systems. Second, there was NO lawsuit or threat thereof from BSDI. Bryan's got it almost exactly backwards. BSDI was the one who got sued by USL over what they claimed to be system V proprietary code in BSD, and it was their long-running refusal to knuckle under that finally led to a resolution of the matter when Novell (who'd aquired USL in the meantime) decided to "bless" 4.4 Lite as unencumbered on the condition that folks, from BSDI on down, stopped using the Net/2 release code. Finally, FreeBSD 2.1 was released in January of 1996 (check the CD!) so I don't know how Bryan can possibly assert that "it's a little dated" or was released in "late '94." One is left wondering whether Bryan actually spells his name "Brian", a final error which would give him a 100% failure rate from the top of the header to the bottom of the signature and thus ensuring him a place in the USENET hall of fame for "most inaccurate posting of 1996." :-) -- - Jordan Hubbard President, FreeBSD Project