I guess the website is really good. The only thing I'd do is to add a
section listing the core features of PostgreSQL - I think this could be
an important point.
In my opinion MySQL is not a competitor and we should not benchmark
PostgreSQL and compare it with MySQL. Those features which are really
important are not supported and so it is just not possible to make a
serious comparison unless you want to benchmark databases containing 300
records or so ...
Speed is not the only thing and it can lead to false results. Let's
thinks of an example:
Let think of a table containing 300 records and you want to :
SELECT cosh(x) FROM y HAVING cosh(x) > z;
How can anybody implement cosh for MySQL??? In this case the benchmark
cannot be done - in the case of MySQL the problem has to be solved on an
application level. These are the REAL ADVANTAGES of PostgreSQL and this
the reason why people are using it. I think it is not work talking about
small databases and simple queries. We should focus on stability and
extensibility and not on "SELECT 'micky mouse' FROM smalltable".
PostgreSQL is the most advanced database on earth and that counts.
These days we are negotiating with a potential customer who wants to
substitute Oracle on AIX for PostgreSQL on AIX because of costs and
extensibility. We did NOT choose MySQL because he has the impression
that MySQL people focus on speed rather than on reliability and
extensibility. I think that shows what things are all about.
I am not happy about the Windows port - maybe it will cause a lot of
troubles in real applications due to problems related to Windows.
Let's do enterprise computing and don't let us build a database for
minor Web databases.
Hans
Mark Radulovich (by way of Jean-Michel POURE ) wrote:
>This mail was sent to pgadmin-hackers list by Mark Radulovich
><[EMAIL PROTECTED]>. It is quite interesting:
>
>*************************************************************************************
>I've been following these mailing lists for over two years, and I guess now
> is the time to chime in. I agree with Tom that an organized effort is
> necessary. As such, I'd recommend the following:
>
>1 - Revamp the website. It's not bad, but it should be better designed to
>highlight things for new users as well as all of the documentation that is on
>it (I usually go to google instead of clicking around the website). I think
> the basic site should have four key sections - Application Developers,
> Database Admins, PG Core Developers, News & Downloads.
>
>2 - Get a list of people who can help with benchmarking efforts. This should
> be for magazines/websites that want to benchmark PG against the competition,
> as well as benchmarking PG on various hardware with various options. This
> could even start out as a simple "Benchmarking PG FAQ"
>
>3 - Revamp the "Developer's Corner". I'm a web developer, not a PG developer,
>but I still went here looking for info on building apps in Java, PHP, Perl,
> etc that need to connect to PG on the backend. This is probably a simple
> rename, but application developers need a more prominent area.
>
>4 - Reach out and talk to authors & developers. We have a great database here
> - let's tell the world. This can be simple - identify the major magazines &
> web sites, rank order them by relevant audience. Then, make sure we contact
> someone at each site once a month, and that they get press releases via
> email. (email is essentially free, so why not send them out to all the
> magazines/web sites?)
>
>5 - Show off PGAdmin!!! You'd think it was just an afterthought when looking
>around the web site. We should promote that as a great tool to manage PG, so
>that MS users can get the courage to try it out. We can't market it like MS
> can (unless someone around here dhas $40 billion lying around), but we can
> sure make PGAdmin more prominent on the site.
>
>6 - Improve the Windows port. I am convinced that mySQL is popular because a
>windows user can download Apache, PHP, and mySQL onto his machine and learn
> how it works. When he's ready, he can move to *nix. PG doesn't have that
> advantage (no newbie is going to mess with cygwin setup on his Windows 98
> machine). Also, just because Windows is not an optimal database platform
> doesn't mean we shouldn't serve it better - a lot of people (myself
> included) cut their teeth on Windows computers, simply because they cannot
> afford the time or money to learn another OS just to be able to use a
> database.
>
>7 - A simple thing, really. Can someone change the order of the months on the
>mailing list archive home page? Scrolling down for 66 months, just to click
> on the "by date" or "by thread" link for the current month just bugs me.
> Whether this is possible or not, I don't know - I just wanted to comment
> about it because I'm sure there are others with the same complaint.
>
>
>Anyway, these are just a couple of ideas I have. I have used PG since 7.0,
> and have been incredibly happy with it. As for any competition with MySQL,
> so what? Let's learn from what they do better than us, and use that to
> increase our visibility.
>
>
>
--
Cybertec Geschwinde &. Schoenig
Ludo-Hartmannplatz 1/14; A-1160 Wien
Tel.: +43/1/913 68 09 oder +43/664/233 90 75
URL: www.postgresql.at, www.cybertec.at, www.python.co.at, www.openldap.at
---------------------------(end of broadcast)---------------------------
TIP 2: you can get off all lists at once with the unregister command
(send "unregister YourEmailAddressHere" to [EMAIL PROTECTED])
