xml-general mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Stefano Mazzocchi <stef...@apache.org>
Subject Re: [vote] A native XML database project under Apache
Date Sun, 21 Oct 2001 17:30:08 GMT
Martin Stricker wrote:
> +1 for the general idea (but I'm not a committer), *but* before we can
> really vote we'll have to get several things clear (others already
> mentioned a lot of them).
> My personal opinion about the programming language: C/C++! Java is far
> too slow.

Ok, I'll state this once again: the vote is *NOT* about taking over
dbXML and making it an Apache project, but is about creating a
development community that will create, maintain and reserach a native
XML database and in order to bootstrap it, the dbXML codebase is
proposed as a seed.

So, if any of you have a better seed to propose, I'm all ears.

If not, please, avoid making technical discussions on something that is
not even accepted (even if, so far, we have more votes that needed).

If you want to discuss technical details, I welcome any of you to
partecipate to the development of that newly created subproject and I
the more follow my suggestion and the more ideas get discussed, the
better the software will turn out to be.

This concludes the official response.

                             - o -

Now, talking personally, I believe that Java makes it a perfect choice
because even if it's not recognized as the fastest language in the world
(even if Sergio is right: there is no such thing as a fast language and
a slow one, it always depends on the situations and on the written
software), it *IS* recognized as an easier language to deal with
compared to C/C++ and friends and allows to research algorithmic
solutions faster than any other language yet maintaining a rock solid
architectural foundation (sometimes, even better than what results out
of C/C++)

And if development speed and ease-of-reserach were not enough for you,
there are examples (Sergio's is one of them, but there are many) were
java on the server side proves itself as fast as native solutions, even
because hotspot JVMs work on server software much better because they
have more cycles to use to optimize a program that normally stays up for
days or even months.

And if hotspotted java performance matching native performance was not
enough for you, I suggest you to take a look at the new NIO
(non-blocking IO) that is very likely to turn Java into a very powerful
enviornment even for I/O intensive usages as databases, file servers or
media servers.

Finally, if you think java is slow, you can always volunteer to rewrite
dbXML in C++. When you're done, the new released version of the software
with better algorithms, new JVM optimizations on the server side (which
is where Sun is heading) and hardware advancements might make the entire
port useless.

But if this still doesn't stop you, I'll welcome your submissions and
let the community decide the future of this eventual native port.

Stefano Mazzocchi      One must still have chaos in oneself to be
                          able to give birth to a dancing star.
<stefano@apache.org>                             Friedrich Nietzsche

In case of troubles, e-mail:     webmaster@xml.apache.org
To unsubscribe, e-mail:          general-unsubscribe@xml.apache.org
For additional commands, e-mail: general-help@xml.apache.org

View raw message