Subject Re: Design of new built-in functions
Author paulruizendaal
Jim wrote:

In the time that Firebird has put out three releases
(counting 1.5 as a release), MySQL has gone from a handful of
people to a profitable company over over 300 people with tens of
millions of dollars in the bank, millions of customers, and
alliances with dozens of major products and companies.

Paul wrote:

This is factually challenged. In essence, MySQL finds it roots in
1995, when David Hughes invented LAMP, only months after browsers had
progressed enough to support the concept. David created mSQL and
Lite, a PHP-like embedded language. It took off like a rocket. mSQL
had one important feature missing: it did not support indexes. Monty
Widenius (legally) copied mSQL, added an indexed engine, and thus
created MySQL, which soon replaced mSQL in the installed base. So,
MySQL has a 10-year path, not a 5-year path.

Despite the 10-year effort, MySQL does not have millions of
customers. The latest info from MySQL marketing says they have 7,000
customers; it might be 10K now, who pay over $5,000 on average for
the honour. Half of the latest reported $40 mln in revenue may come
from a few million-dollar customers.

Yes, MySQL have an installed base of millions. MySQL marketing
calculates there are 1.000 users for every customer. My (consensus)
estimate would be around 5 million installations, having peaked late
2004. At the higher end it seems market share is lost to Firebird
(and to a lesser extent Ingres), on the low end share is lost to
SQLite. This 5 million installed base compares well with the
estimated installed base for FB/IB of 3 million.

Moreover, considering that MySQL is weighted towards Linux (60..70%)
and that Firebird is weighted towards Windows (70..80%), Firebird is
actually *the market leader* on Windows.

Jim wrote:

"Could be" the market leader, not "is". That's what so frustrating.
The Firebird project has a deep understanding of Windows where MySQL
engineers tend to pride themselves on their ignorance of Windows.
But without a strategy to exploit this advantage, it will go away.

Here are steps that Firebird must take to survive. There are
certainly others.

1. Provide sufficient on-line documentation for a database savvy
developers succeed with a trial project. Do it now!
2. Track and follow up downloads to find out what problems people
have and if/when they give up.
3. Start a certification program.
4. Find out why US developers are shunning the project and fix it.
5. Institute an architectural review process
6. Develop the culture to make predictable releases
7. Identify and exploit Firebird advantages over MySQL and
Postgres (Windows support, for example)

Paul replies:

Jim, check your facts & do the math: FB/IB *is* the market leader on
Windows. If you disagree, please provide the numbers showing that you
are right, but please do not babble - it does not fit you.

Although the points you make have merit, the facts are that FB is in
a good position and gaining momentum. You and I both know that
Firebird regularly wins over other databases in selection processes.
You may not like that, but it does not change the facts.

Item 4. is another factual error: US developers are not shunning
FB/IB. There is research showing that FB is the most popular among US
developers, outpacing SQLServer (MSDE) and Jet.,1759,1756736,00.asp?

Yes, the US is under-represented in FB development, but that is the
same for almost all O/S projects. The epi-center of O/S tends to be
in the EU, in LA, the FSU and China.

Jim wrote:

Ignore the details of MySQL economics ...

Paul replies:

Why? Is it easier to ignore the facts than to discuss them?