Date: Mon, 9 Aug 1999 21:30:08 +0000 (GMT) From: Terry Lambert <tlambert@primenet.com> To: garbanzo@hooked.net (Alex Zepeda) Cc: tlambert@primenet.com, BMCGROARTY@high-voltage.com, freebsd-advocacy@FreeBSD.ORG Subject: Re: Oracle for FreeBSD Message-ID: <199908092130.OAA16268@usr06.primenet.com> In-Reply-To: <Pine.BSI.3.95.990808215401.21271A-100000@fish.hooked.net> from "Alex Zepeda" at Aug 8, 99 09:58:09 pm
next in thread | previous in thread | raw e-mail | index | archive | help
> > There is already a FreeBSD port of Oracle. It runs on Oracle's > > (NCI's) FreeBSD based Network Computer server. > > > > It also runs on FreeBSD-current, if you tar it up and untar it, > > as someone who shall remain nameless has done to see if there > > were OS specific hooks built into the NC server FreeBSD kernel to > > support the database (there weren't). > > > > > > It's a matter of leveraging a big company that needs it (e.g. Yahoo) > > to get it released as a supported product on FreeBSD in general, > > instead of merely as a component of their NC server. > > > > It needs a corporate sponsor. > > Yes, I've heard you mention this *numerous* times before; but that's why I > think a boxed version of this would be cool. It wouldn't be hard to > actually create, and would create something for VARs to sell that actually > includes added value. The thing is, there isn't going to be the possiblity of boxing both FreeBSD and "Oracle for FreeBSD" together, unless "Oracle for FreeBSD" is made available first. There is a crying need for a large Oracle user, FreeBSD user, or both, to commit to a large enough unit purchase to cause the product to be supported. Since Oracle uses the product internally, this is clearly an issue of amortization of support costs, particularly including the increased supoort costs associated with active platform developement taking place (Oracle can be guaranteed that, on their release of FreeBSD, that the product will remain relatively stable). This is clearly an API mutability and product life cycle issue. The tack to take seems to me to be to ensure that the "big user" that the advocates get to come forward specify a particular FreeBSD -RELEASE on which the software is supported, and guarantee that there is no support that Oracle will have to put up with for any more recent platform releases and/or -STABLE versions which later occur. By far, the largest exposure for a commercial company considering a port to any platform is the risk that the platform will change, and cause the software not to run. I rememebr a concrete example of this, when SCO changed the tty ring buffers from 32 to 24 characters on Xenix 3.1; this change broke a lot of software that was running the buffers at greater than 24 characters per lbolt. 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-advocacy" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199908092130.OAA16268>