Migrate MySQL... or not...

[Madison Kelly]

William Muriithi wrote:
> Clifford,
> 
> Please, I have NO attachment to either of the database, so do not take
> this as criticism.
>> I've written on the subject of MySQL vs. PostgreSQL on other lists. I'll
>> refer you to two examples.
>>
>> <http://omnis-dev.com/pipermail/omnisdev-en/2009-February/005142.html>
> This quote is from the above link.
> 
> If someone told me they wanted to use MySQL for a dinky web site with
> light traffic, I'd tell them, "If you insist, go ahead. It won't make
> much difference but, why bother if you have the choice?" Particularly if
> your dataset is so large and your queries so complex that it will tax
> *any* database and you don't have the resources of a Yahoo or a Google
> to co-develop MySQL, you should be looking at another database.
> 
> I find above quote contradicting. If you may allow me to paraphrase
> it, it essentially claim google and yahoo uses mysql just because they
> have resources. That does not sound convincing enough to me. Even with
> a lot of resources, I would expect them to choose use the better of
> the two as that would offer them a competitive advantage.
> 
> 
> I think its extensive use of mysql is  akin to why Cisco is widely
> used that Juniper. That's just a guess, and hopefully Madi will have a
> couple of slides of why mysql tend to be more favored in corporate
> world and heck, by even most of the open source project seem to offer
> mysql over PostgreSQL
> 
> 
> Again, the above is a polite opinion and in no way intended to be a flame.
> 
> William

As was earlier compared; MySQL vs. PostreSQL often stirs up emotions on 
par with the Vim vs Emacs, Linux vs Windows vs Mac, Canada vs. USA, 
Honda vs. Ford and many other "versus" debates. It's very difficult to 
be objective because each side is so convinced that they are right.

Unless one side presents reproducible, empirical evidence along with 
their arguments, it's hard to believe any one side in any such argument. 
This evidence is fleetingly rate. If I do indeed give this talk, it 
would only be after coming up with some evidence to back it up.

To do that though, I'd first need an idea of what each claims is their 
strength and investigate what each claims is the other's weakness. I'd 
need to come up with some code to actually test these assertions and 
make the code available for others to reproduce. I'd need to find a set 
of ways to fail each system and gauge their ability to recover.

This will not be easy or quick, but as I said in my first most on this 
topic, I think it would be a useful exercise for me, as I of course have 
my own biases (PostgreSQL, Linux, Canada, Honda). I'd like to see if I 
can indeed create a neutral talk and presentation. Hearing the arguments 
from both sides would just be the first step I took.

Madi
--
The Toronto Linux Users Group.      Meetings: http://gtalug.org/
TLUG requests: Linux topics, No HTML, wrap text below 80 columns
How to UNSUBSCRIBE: http://gtalug.org/wiki/Mailing_lists