Winner takes all in open source: The MySQL example

Dave Rosenberg once wrote that the winner in a given open-source category (ERP, CRM, etc.) takes all. There's no room for second place. The first mover to get critical mass tends to horde the community and media resources.

I didn't believe Dave at the time, but after looking through the data from Alfresco's Open Source Barometer survey, I'm becoming more and more convinced.

The Barometer now comprises a data set of 35,000 enterprises, a significant percentage of which hail from the Global 2000. In the case of open-source databases, these enterprises are overwhelmingly voting for MySQL, as Ian Howells writes:

As the previous Open Source Barometer showed, in open source at each level of the stack there is a clear leader. If you say open source linux, database, app server, enterprise content management, crm most people will be pushed to come back with more than one name. For open source database that name is MySQL.

The numbers speak for themselves, with 60 percent of the 35,000-strong Alfresco community running MySQL.

(Credit: Alfresco's Open Source Barometer)

It would be easy to discount these results if these were no-name mom-and-pop SMBs, but they're not. Alfresco's customers include Kaplan, Electronic Arts, US Federal Aviation Administration, Boise Cascade, and a wide swath of the world's largest financial services, media, education, and government organizations. 60 percent of which are happily running MySQL.

I'll be posting more data from the most recent Open Source Barometer report tomorrow, but in the meantime it's worth considering the what's at stake in coming in second place. It may well be the difference between $1 billion or nothing.

That said, I don't think it's all doom and gloom for everyone but the front-runner. Instead, I think it's clear that the "also-rans" need to run a different race. Their community outreach must differ from that of the front-runner. Their go-to-market strategy needs to differ, too. Basically, they need to find a different way to sell their products, and to different types of customers, than the front-runner has chosen.

[tristanbob]

I agree with the network effects of open source software. There is a point of critical mass where the adoption of open source increases exponentially.

People don't use open source software because:

1. They don't know about it - "You mean I don't have to pay for this office software suite?"
2. They don't know anyone else running it - "I don't want to be the guinea pig!"
3. They can't run their applications on it - "They don't (yet) make a Linux version of Photoshop"

Once these problems are addressed by critical mass and the network effect, the barriers to adoption are removed. Keep your eyes open for this to happen, it will be fun to see!

[debwoods_nc]

Interesting to see still such a large number of Alfresco downloads on non-open source databases. With 29% on Oracle ,Sql server and DB2, Data shows a need for enterprise capable solutions. Ingres at the time of this report was not certified to run with Alfresco.

I concur with your views on finding a different way to reach out to users, tighter integration is key. Finding ways to reduce costs while simplifying the solution will be the way forward.

[joncweaver]

There are always more than one way to accomplish any goal. The also rans just have to figure out how to make their implementation/strategy seem more compelling. Once they accomplish that they stand a viable chance of making inroads on the market place. Of course, finding that compelling story is easier said than done.

Jon Weaver
XAware.org

[billxa]

Venture Capital investors seem to agree that there is only room for one big winner in each open source category. I invested a lot of time over the last year talking with top software VCs about open source investments. They seem to have converged on the "room for only one" , and "all or nothing" theory as a major risk factor for making open source software bets. This shared believe is a contributor to a slowing rate of VC investment in OSS. I'm not too sure this is a rational analysis. As open source grows into a larger overall share of the software market, there will be more than one significant winner per category. Still, even if its not an "all or nothing", the market leader can expect an exponentially bigger pay day. This is not unique to open source software. It is well worth playing to win your category. Bill Miller, XAware.org