Return to BSD News archive
Path: euryale.cc.adfa.oz.au!newshost.anu.edu.au!harbinger.cc.monash.edu.au!news.rmit.EDU.AU!news.unimelb.EDU.AU!munnari.OZ.AU!spool.mu.edu!howland.reston.ans.net!newsfeed.internetmci.com!uwm.edu!newsspool.doit.wisc.edu!koala.uwec.edu!daffy.anetsrvcs.uwrf.edu!cakerwood!not-for-mail From: bl03@uwrf.edu (BENJAMIN A LINDSTROM) Newsgroups: comp.os.linux.advocacy,comp.unix.bsd.freebsd.misc,comp.unix.advocacy,comp.unix.misc Subject: Re: Linux vs FreeBSD Followup-To: comp.os.linux.advocacy,comp.unix.bsd.freebsd.misc,comp.unix.advocacy,comp.unix.misc Date: 8 Dec 1995 06:06:15 GMT Lines: 94 Message-ID: <4a8kkn$71n@daffy.anetsrvcs.uwrf.edu> References: <489kuu$rbo@pelican.cs.ucla.edu> <49rm0g$o8o@daffy.anetsrvcs.uwrf.edu> <DJ2IBL.71t@nntpa.cb.att.com> <4a2nvk$m7h@daffy.anetsrvcs.uwrf.edu> <DJ6L7r.8BG@nntpa.cb.att.com> NNTP-Posting-Host: cakerwood.acc.uwrf.edu X-Newsreader: TIN [UNIX 1.3 950726BETA PL0] Xref: euryale.cc.adfa.oz.au comp.os.linux.advocacy:29576 comp.unix.bsd.freebsd.misc:9968 comp.unix.advocacy:11856 comp.unix.misc:19908 John S. Dyson (dyson@inuxs.inh.att.com) wrote: : In article <4a2nvk$m7h@daffy.anetsrvcs.uwrf.edu>, : BENJAMIN A LINDSTROM <bl03@uwrf.edu> wrote: : > : >I admit that Linux is NOT CVS open to the public...But it's open to anyone : >who wants to do it...Yes, we might have have source code release every hour : >(well..A few times.=), but even if Linus openned up his CVS to everyone and : >let everyone read it...I doubt he recompiles his kernel more then 2 a week. : ><shrug> : > : Again, it is Linus' decision, or is there a group of people making the : decision? : From what it sounds like it.. It's like FreeBSD..Once you gain control over a driver or sub-set of the system. YOU make the choices on how to optimized and implement...and YOU deal with it when it's broken.. However, if you write something and it gets into the kernel and it never really get updated nor fixed it's removed.(Case in point Modem Load-Balancing has MAJORLY broke and was taken out in the 1.1.x group and later was added back in to 1.3.x when it was fully fixed by another group of programmers.) Does it matter if ONE or MANY people are deciding?? As long as the ONE person is reasonable (as Linus has been) and explains WHY he refuses to add it into the mainstream kernel...Does it really matter? : >: The approach used in FreeBSD has allowed for very quick development even : >: in the face of the Net/2 thing and the slow start from 386BSD. I think that : >: FreeBSD is really doing pretty well... : > : >Is FreeBSD any faster in development then Linux? (Honest question) From : >that I've been seeing on the Linux side is there is a major release every : >year, but it almost doubles the drivers that are contained in the : >year befores version. The only differents from what it sounds like for : >'speed' of development is FreeBSD freezes more offen and tries to get more : >stable kernels out...Where Linux currently gets a kernel that is 100% (or : >near..We all know about the 1.0.9 problems.=) perfect before moving on to : >the next phase. : : Well, even though FreeBSD was slowed down by about 1 year (basically catching : up the 2.x series to what the 1.x series could do.) -- it appears that it : is doing pretty well. We had your 1.0.9 problems embodied in our 2.0 problems. : Those problems set us back quite a ways -- but that was due to the Net/2 problem : and we are recovering mightily. FreeBSD has not concentrated on drivers : as much as it perhaps should, but the current stuff (equiv to the 1.3.xx : series) has R/W EXT2FS, a faster yet VM system, ibcs2, linux binary support : (BSDI Netscape doesn't have Java yet -- but Linux's does, and it works nicely : under *BSD :-).) Also, FreeBSD has been quicker in some cases to get stable : drivers that Linux has had trouble with. It is just that FreeBSD has been : slow to adopt the less common or lower performance device support -- but it : is coming. There are many individuals working to resolve those issues. : The FreeBSD group feels that there is a lot of opportunity for the Linux user : to graduate to FreeBSD when their systems and needs grow, so we will be : supplying better and better compatibility with Linux. : I'm sure if you did not have the legal issues then you would be a little more head of Linux in a lot of areas, but my point is (and many others) that it's a different method of management. Soon we have the *BSD* file systems for use under Linux...I don't know much about the VM system since I'm not a kernel hacker..IBCS2 works..Talk that was had.... BSDI I think compatability? (Please correct me..I have IBCS2 under RedHat, but have not played with it yet) : The biggest thing right now is that FreeBSD is great at providing performance : under high load with stability. And that the FreeBSD development is proceeding : at an increasing pace. The most important thing that Linux has ahead of : FreeBSD is user base -- and that (in a capitalistic society) can be the : whole story. :-(. But of course Microsoft has an even bigger user base, so : it has to be the best :-). : Where you fokes are focusing on High Load stability; we are working on pushing Networking..... "0.99" was pushing for a fully working systems "1.1.x" was working on over all stability and drivers for more support, and starting with Multi-platform support. "1.3.x" is for faster networking support and finishing off all the 'rogue' Linux ports to other hardware platforms. I don't know what 1.5/2.1 has in store for us, but I'm sure better highload support is one of them. I'm runing a 1.3.x kernel..I'm IMPRESSED! I've watched, Alpha, Sparc, and PowerPC mature. I get rather bugged when my old Mitsumi CDrom driver breaks under the beta kernels, but I learn to live with it. : Note that some of the above could be considered flame bait, but remember : if there was no *BSD -- Linux might have been the only real game in town. : So, it just so happens that users have a choice and the various groups : fortunately compete. If *BSD* had not been around..I'm scared to think of how UNIX machines would be running now.=) For as much as I see good points in sysV... I've seen some REALLY bad UNIXes bases on it.=) (Interactive? =)