Date: Wed, 9 Feb 2000 01:35:56 +0000 (GMT) From: Terry Lambert <tlambert@primenet.com> To: rcarter@pinyon.org (Russell L. Carter) Cc: grog@lemis.com (Greg Lehey), rcarter@pinyon.org (Russell L. Carter), wes@softweyr.com (Wes Peters), bpechter@shell.monmouth.com, freebsd-fs@FreeBSD.ORG Subject: Re: JFS and IBM Message-ID: <200002090135.SAA14768@usr08.primenet.com> In-Reply-To: <20000205034805.AC2DE5B@pinyon.org> from "Russell L. Carter" at Feb 04, 2000 08:48:05 PM
next in thread | previous in thread | raw e-mail | index | archive | help
> %> Look folks, is it too hard to understand that an OS/2 filesystem > %> released as "Opensource" might not make any sense at all to burn up > %> anybody's time, > % > %That's not clear. It might make sense, and I hope somebody takes a > %look. I personally do not think it is worth burning the time on, unless the license can be changed. A GPL'ed FS is useless as a boot file system in anything that anyone might want to incorporate into a commercial product. It has small merit as an interesting directory structure implementation that may be emulated in BSD (it's rather trivial to do, backing out some changes from a while back that merged the directory code in a little too solidly for my tastes). It is only technically really useful as a boot filesystem, since its primary merit (for which you trade a cleaner and potentially a defragger requirement) is to permit fast reboots. > %> Does it look like the IBM mamaship is finally clueing into > %> marketing? > %> > %> Why yes it does! > % > %There are some very good reasons to start with the OS/2 version. It's > %a lot easier to integrate into Linux, since it doesn't have all the > %hooks for the IBM VM system. The performance might still be > %acceptable on Linux. And I could be wrong, and it might even be very > %good. > > Hmm, maybe your optimism is realistic. > > Ok, when something like this shows up, I like to place a money > bet. Let's make it fun. I bet $50 (to Greg) that this is going to go > nowhere and that 2 years from now (Fri Feb 4 20:45:24 MST 2000) > you will not be able to download or buy a Linux OS (or FreeBSD OS) that > can sustain "large" 10 MB/s read or write performance to the > native JFS-based filesystem using some native, derived version of JFS to any > commercially available persistent hardware endsystem not composed of "RAM". > I'll be gracious on the definition of "RAM". > > As always, I hope I lose :-). I have personally been working on trying to get access to the AIX JFS, as opposed to the toy one on OS/2. You may want to be a little bit less gracious in your definition of "JFS". 8-). I now know the process for getting access to the source code, and who to club over the head to try and get it released. I'm actually very surprised on the code release. If I want to use GPL'ed code, I have to go to an IBM class on how I have to handle it in order to prevent intellectual property dilution, after which I am placed on a "blessed individuals" list, after which I can download the code from IBM approved mirror sites. In other words, the marketing people may have discovered how to use it, but the lawyers are still treating it as hazardous waste, so far as I can tell. It would be nice if we could get more BSD people into IBM to help spread the gospel. Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200002090135.SAA14768>