Re: PHP 5.0.0 Beta 1 : alternatives to MySQL?
| From: | Alessandro Bottoni | Date: | Mon, 30 Jun 2003 12:29:07 +0000 |
| Subject: | Re: PHP 5.0.0 Beta 1 : alternatives to MySQL? | ||
| References: | 1 | Groups: | php.pear.general |
| Request: | Send a blank email to pear-general+get-6338@lists.php.net to get a copy of this message | ||
Alle 11:31, lunedì 30 giugno 2003, Matthew Palmer ha scritto:
> How does PHP 5 hurt people who develop commercial software?
OK, let say you started to develop a program for a small customer some months
ago. At that time MySQL was LPGL'ed and you decided to use it for avoiding to
charge the customer of the added cost of a RDBMS license. Now your work is
almost completed and you are going to install the software but... alas, the
installed MySQL _now_ requires a commercial license (that is: $$). Who will
pay for it? For sure not you customer: he/she signed a contract for a
license-free installation. Maybe you... :-(
BTW: of course you did _not_ use PEAR::DB for not affecting the RDBMS
performances with a DAL... :-(
MySQL does not hurt, of course, but a LGPL to GPL change in the licence policy
can do (mostly because you depend on your hosting companies for the type and
release of the RDBMS installed on your server).
> > Firebird (ex Borland/Imprise Interbase): is Firebird LPGL'ed?
>
> Dunno.
I looked at the license. It is so unreadable that I'm almost sure it is not
LPGL'ed (and maybe non even Open Source ;-).
Wolf's law: unreadable documents are a sure symptom of the fact that you have
to pay for something, even if it impossible to uderstand why, how, how much
and when.
> mSQL? Hahahaha. So badly non-free it doesn't enter the equation. I
> didn't know GNU did an SQL server.
Little surprise: I discovered GNU SQL Server just this morning, while looking
for an alternative to MySQL. Its status looks very like the one of HURD:
dormand or deceased. Too bad... :-(
Regarding mSQL, little regret... It is so limited that I would have hardly
taken it into consideration.
> Perhaps, if that's what gets you going. There are definite advantages to
> going to an OODBMS, but I don't know if the benefits (ease of interfacing
> to code) outweigh the performance problems that all OODBMS that I've seen
> suffer from.
I'm going to investigate this hypothesis a little bit further. These days,
many web applications are so niche-oriented (and so little used) that
performances are hardly a problem (in particular in the case of a dedicated
server of the last generation).
CU
------------------
Alessandro Bottoni