- From: Otis Gospodnetic <otisg@panther.middlebury.edu>
- Date: Tue, 17 Sep 1996 20:22:40 -0400 (EDT)
- To: Phoebe.Couch@library.ucsf.edu (Phoebe Couch)
- Cc: www-rdb@w3.org
> Do you mean 1 - 10 records, or 1-10 Meg? > 1-10 is trivial and perl will do it fast enough to serve out. ah, I realized my mistake too late. it's 1-10+ million rows, not 1-10 :) I wish... > There are a lot of free databases out there such as msql,postgres,etc. yeah, I tried using both. mSQL is just too limited, the author doesn't support it really well, doesn't respond to email, and the thing isn't even free. I have no problem paying for things, but I don't think mSQL is worth the money the author wants for it. postgres95 I have, too, but it sounds pretty scary, hard to maintain and manage, and from what I read on the mailing list, not very stable, and causing all kinds of weird things ... > I use illustra here with the C API and it is sufficent here. > I support multiple databases with 10+ tables and 1000+ records accessible > from the web. I use the http server for the front end security which > is in turn controlled by the database. I use transactions to protect We would need something that would allow users to login to their entries in the database, change the info in those entries, and logout. > data during updates, etc. They have a web product -webblade that I find slow > (compare to the C API) and inflexible. > It's not free, and the price is going up since Informix bought them. Oracle, > Sybase, Informix are not cheap but can do the same things. > postgres is huge and I find difficult to set up and use in a production > environment. msql is for smaller db, I have not used it myself but my > co-worker has. Naviserver(http) comes with a limited illustra license > and is free. I've tried using Naviserver (GNN, right ?) but unfortunately it doesn't run on a BSDI and that is all we have right now. > The other alternative is to use regular text retrieval tools, such as > wais, isite, etc, if you don't need to do joins between tables. I'm afraid text retrieval is not what we need here :( It needs to be a real database this time... Thanks, Otis
Received on Tuesday, 17 September 1996 20:25:54 UTC