Hi Michael,

As I have written before, I'm currently working exclusively with PostgreSQL. I will also be recommending this for any new project I start (which does not seem to be contested by at least some of the MySQL people here). I do, however, think that some of your remarks are a bit far reaching.

Ben-Nes Michael wrote:

I hope no one will shoot me :) but MySQL remind me Windows. they built
something simple and now trying to expand its capabilities without changing
the foult of the past.

Its not true that I didnt heard about MySQL new capabilities , its just that
they are special and not part of the basic package. PG was designed to be
like it is now from the Begining . It is Relational DB which is much more
sophisticated.


Then again, if a database has modular data storage, and one of the data storages offer all that you need, and STILL be pretty fast, why shouldn't you? Then again, I get the feeling that measuring speed is a bit dependent on who is doing the measuring. For example, the benchmark quoted before that gave MySQL and Oracle the same speed score did not benchmark Postgres at all.

for example why MySQL is lame:
Sagi Wrote:


One thing, though. Unlike some other DB's, MySQL data types are not
totally strict - if you try to insert data of a wrong type it will
usually try to convert it instead of throwing an error. This may lead to
data loss if your application is poorly written



PG does not accept poorly written codes, this is part of the itegrity.
coders do mistakes which need to fix and you cant allow loosing data becouse
of this.


Then do help me to solve http://archives.postgresql.org/pgsql-hackers/2004-06/msg00681.php, please.

This is a discussion between two open source groups, not a commercial fight where we try to find fault in someone else's arguments. MySQL has faults. Well, duh? Of course it does. It's a software package. Show me one that doesn't (yes, even qmail has faults, just none with security implications). The thing for me is not to understand whether I need to drop PosgreSQL and switch to MySQL right now. I know PostgreSQL, and it's a good DB to use for new projects. I will have to look into Firebird, because no one else seems to know it on this list. Then again, maybe that's a pretty serious argument against it. The thing for me is, however, to figure out whether an existing client/friend/whatever who already has MySQL (or who is insisting on using it) should be convinced otherwise, or whether MySQL can deliver what said client/friend/whatever needs.

There way too much to write about the difrences but one thing is sure.
almost all the SQL gurus which i had the honer to speak with stated that PG
is a very serious DB which worth working with.


I don't think anyone here claimed otherwise. There were such claims against MySQL, however, and they seem to be either exaggerated or dated.

I incourge those who intrested in the area to search the net, then decide. I
myself will not work with MySQL if I could avoid it.


Any specific links you can point out? I'm talking about "I did this with MySQL, and these are the bad things that happened", or about "don't do such and such, as MySQL can't handle it". I.e. I'm looking for specifics.

Cheers


         Shachar

--
Shachar Shemesh
Lingnu Open Source Consulting ltd.
http://www.lingnu.com/


================================================================= To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]



Reply via email to