Re: Database market share 2004

Paul wrote:

> Where was any mention of Sybase? Pervasive? IBM's Informix offerings?

Yes, in the actual Gartner report, not the eWeek diluted one. They break out Informix revenues. Oracle beats IBM by a few points if you don't count Informix.

> Borland's Interbase? Someone must buy these products! What about MySQL
> and/or Firebird? MySQL is not free for commercial use - what's its
> market share? Pervasive have also started to support PostgreSQL, as

MySQL *is* free for internal commercial use, it's just not supported. Companies like Google make a mint by modifying GPL software and not re-releasing the source because they don't ship a product, they offer a service. IF they do release components, it's out of a spirit of community giving ;)

To answer your question, Gartner does have an "others" category at around 6.6% revenue or $517.1 million, which about the same share as Sybase, Teradata, and Informix combined. This is down 5.2% year-over-year, BTW. I assume this includes revenues from all the smaller vendors, but I don't really know.

I'm not sure it matters that much: MySQL AB only had revenues of $5 million in 2002 -- at best they're $10-12 million now. (reference .. http://uk.builder.com/architecture/db/0,39026552,20276852,00.htm).

To give you a reference point, RedHat had revenues of $124 million in 2004.

> No mention either of any revenue streams for companies that have
> support contracts for any of the open source dbs.

This measures "new license revenue" -- not support contracts. This is a general problem with the enterprise software space, in that analysts (and Wall Street) determine the health of a company purely based on its "new licenses", not counting support or upgrades under maintenance.

I don't know how MySQL AB's accounting rules work, but it's hard to fight against an entire industry's accounting practices. They may suck, but they make sense given the value system of many enterprise software buyers.

But also note that if they measured database maintenance, support & services contracts, IBM and Oracle make billions off those too, so I'm not sure that's the answer to give OSS databases more "cred".

> That has to be market share in some shape or form, but was completely
> ignored in the article which only mentioned PostgreSQL, and even then
> only provided a link, with no mention of its market share!

Again, it's a measure of new revenue market share. I'd say it's pretty accurate on those terms, but can be misleading because these press articles never quite paint the whole picture (Oracle leads in UNIX, Linux, and is #2 on Windows -- yet IBM still is #1? Hmmmmmmm.)

Whether revenue is relevant, of course is debatable. But what's a market without dollars? Does unit share really make that much sense (especially in the embedded market)?

Look at it this way: most full time Linux or MySQL developers are subsidised by venture or public capital today (MySQL has 17 million EUR if I recall). They need license revenue to sustain that investment or it's effectively going to be supported and developed by hobbiests, college students, and engineers with an itch to scratch. That model works well enough for creating a good product (like Linux up to 1.2 arguably). But it doesn't quite give one warm fuzzies about their ability topple the big boys.

Anyway, back to making fun of Gartner. There's a nice disclaimer at the bottom of the report: "The information contained herein has been obtained from sources believed to be reliable. Gartner disclaims all warranties as to the accuracy, completeness or adequacy of such information. Gartner shall have no liability for errors, omissions or inadequacies in the information contained herein or for interpretations thereof. The opinions expressed herein are subject to change without notice."

yay.

Cheers
Stu Received on Wed Jun 01 2005 - 04:37:53 CDT