Re: PHP 5.0.0 Beta 1 : alternatives to MySQL?

Alle 15:00, lunedì 30 giugno 2003, Matthew Palmer ha scritto: > Erm, why does the customer have to now pay for a licence? Are you under > the mistaken impression that licence changes can be retroactive? Or, for > that matter, that the GPL requires you to but a licence? Neither is > correct. I think you're FUDing rather heavily. I apologize if I'm FUDding someone else. For sure, I have been FUDded (a lot!) by this new regarding the change in the license of MySQL. > Now, if you were referring to the licencing change from libmysqlclient10 to > libmysqlclient12 from LGPL to GPL, well, that's a slight other matter - but > it's nothing huge, honestly. Put the program under the GPL. Since it's an > internal-only webapp, it's not going to be redistributed, so you can link > whatever you like to whatever you like. If that can't be done because > you're selling this webapp, use the old version of the client library. In my personal case, I'm going to GPL everything. I was still thinking about this topic and this new pushed me definitely in this direction. Looks like there will be yet another web platform like PostNuke on SF, this fall :-) I'm just afraid that not all of the other programmers here around will be authorized to do the same and this will raise a few assorted problems :-( > The big question is, though, what kind of consultant would recommend > hanging their client's livelihood on a product and not getting a bit of > paid backup support for it? Yeah, right. I was just "thinking aloud" about the consequences of such a change.... > Hey, Hurd's not dead, it's just doing a very nice imitation of a Norwegian > Blue... <g> > No, actually, it's got a Debian port and there are people working on making > it dance. I did not know it! I'm happy to hear that it is still alive. HURD seemed to be an ambitious project, when it started, and it was sad to see it (almost) abandoned when Linux came up. Looks like we will have three kernels this winter to play with: Linux, Darwin and HURD. I will be pleased to choose among them :-) > > Regarding mSQL, little regret... It is so limited that I would have > > hardly taken it into consideration. > > It was my start to DB programming way back when (first version was 1.0.5). > Nice intro to SQL and whatnot, but I didn't play with it for long, and as > soon as I saw the direction it was taking with the series 2 releases, I > just gave it up and went with MySQL. I never took mSQL into consideration. It is too limited. I would have tried it only if I was forced by a shortage of MySQL and PostgreSQL supply. After MySQL and PGSQL, my choice would have been Firebird, most likely. > It'd be interesting to see your findings, since a performance OODBMS would > solve a whole raft of problems. Don't discount performance, though - as > much as a lot of webapps don't do very much most of the time, there's no > point hitching your wagon to a horse and then finding out that the horse > won't pull your wagon... Right. For sure this topic deserves some more investigation and this can be the right moment. As you say, a OODBMS would solve a whole category of problems. I'll let you know what I find out. BTW: I just finished to write down the n-th interface to a MySQL DB (in the usual form of a PHP Class). I'm getting really tired to play this game. An OODB would be a nice new toy to play with. CU ------------------ Alessandro Bottoni