Date: Wed, 23 Oct 1996 13:47:09 -0700 (MST) From: Terry Lambert <terry@lambert.org> To: jgrosch@sirius.com Cc: mrcpu@cdsnet.net, jkh@time.cdrom.com, hackers@freebsd.org Subject: Re: Possible Commercial app for FreeBSD. Message-ID: <199610232047.NAA10376@phaeton.artisoft.com> In-Reply-To: <199610230415.VAA01690@superior.truenorth.org> from "Josef Grosch" at Oct 22, 96 09:15:43 pm
next in thread | previous in thread | raw e-mail | index | archive | help
> I have seen the SCO version of Oracle running on FreeBSD. Granted I did'nt > try testing it to see how well it worked but it is a start. Perhaps we > should start a campaign to see if we can get Oracle or Sybase to do a port > to FreeBSD. No, I am not holding my breath. :-) I personally ran Sybase on FreeBSD 1.1.5.1 + patches + Soren & Sean's SCO ABI code + The Linux /dev/socksys code. It was the SVR3 Sybase, as sold for the AT&T StarServer. The local modifications were mostly FS modifications, since 8G of disk in one FS was not supported by the main line code at the time. It was sufficient to run a local copy of the Human Genome Project database (which requires Sybase), which is all I used it for. > Don't know about C/BASE so I will not comment. There is also C-Tree. I have > used it on a number of projects using FreeBSD. It's not bad. The down side > is that C-Tree is a library of ISAM routines not a RDB. It has also gotten > very expensive in the last year. It's now around US$900.00. When I bought > it it was US$200.00. I understand that company that makes C-Tree (the name > escapes me just now) has an SQL server but I have not worked with it and I > think it cost around US$2,000.00. Raima and or dbVista, I tink. You're right that it's not relational: it's associative. Associative is better than relational: you can do fuzzy searches based on incomplete sets of keys out of possible key categories. I used the dbVista code for a support database a number of years ago (I also paid $200 for the thing). Enter some keywords related to the error message and the customers complaint, and it would sort in hit probability order based on descending best match, a techinical support fix for the problem. Build the expertise into the system, not into the support operators, I say (but then, as support manager, it benefitted me to be able to hire college students instead of PHd's... they cost less). Ah, m Log(n) fuzzy searches, where m never gets larger than the number of potential key categories. 8-). As far as I know, the only other associative databases which are commercially available are sold by IBM and run on big IBM iron. Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199610232047.NAA10376>