From: Alec.Cawley
Date: May 8 2002 11:23am
Subject: Re: Help! Deciding on Java/C++ ?
List-Archive: http://lists.mysql.com/mysql/108321
Message-Id: <OF459DA90C.54A3C229-ON80256BB3.003D7C06@Research.quantel>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii




> I agree.
> But i make reference about how it is a software developer permissive to
risk
> to have problems ?

Software will always have risks. There is no such thing as bug-free
software.
You cannot eliminate risk, you can only reduce it.

The only question is whether the software is tested enough that the
possibility
of bugs in the facilities you intend to use and on the paths you intend to
use
is sufficiently low.

In the final instance you are dependant upon your own testing. If you test
it
yourself, you know it works. But at some level we always it for granted
that some
of our tools are good enough to use without testing. For example, I don't
think
many people test their C compilers. And we don't tend to test our OSes - we
just
accept a consensus of "users like us" that a given OS is or is not up to a
certain standard.

The original query was attempting to ascertain this sort of general
consensus for
MySQL and surrounding tools. If a tool has a widely varied corps of users,
most of
whom are satisfied, this bodes well for another user, even if none of the
existing
users is exactly comparable to the new user.

To answer the original question:
MyODBC and MM.Mysql both seem to be well established with a significant
body of
satisfied users  - larger for MyODBC than for MM.MySQL.

C is widely believed to be faster at run time than Java, though Sun claim
that
this is down to only a factor of two.

Java is widely believed to be faster to develop than C and to lack certain
failure
mechanisms to which C is prone (e.g. heap leaks).

After that, it is down to you.

     Alec

mysql, query, database