I already knew my way around the open source database scene when I started preliminary research for my diploma thesis last December. However writing this paper has been a great opportunity to dive into this topic really indepth, dispelling some misconceptions of my own along the way. Once I finished my last courses at university in March, I began to focus on this topic. Obviously I was still doing some open source and commercial development on the side during this time. So all in all I probably put in around 6 months of time in this.
On Wednesday I picked up the nicely bound copies of my 173 page diploma thesis. I dropped off at one copy of it at the university yesterday. So today I am releasing the paper to the public. The fancy title is as follows:
Synergies and Opportunities:
Open Source and Commercial Vendors
A study of the relational database market.
The paper gives an economic analysis of the open source involvement in the relational database market. I tried to make the paper accessible to economists but also open source developers interested in the economics behind it all. After the introduction follows a chapter on the theory of information goods and in particular software. The subsequent chapter focuses on how open source came about and how it relates to the economic theory. The next chapter talks about the relational database market. The real "meat" is in the fifth chapter, where I examine several OSS databases in the form of 8 case studies.
Several people have also made themselves available for interviews. It all started with Christopher Jones and Mark Townsend from Oracle, who I spoke to via a phone conference. I also conducted several email interviews with people in the open source database scene. While unfortunately not everybody was able to allocate the time to reply, I am still very grateful to the many people who did reply: Gavin Sherry (Alcove Systems Engineering), Holger Klemt (HK-Software), Josh Berkus (Sun), Joshua D. Drake (Command Prompt), Kaj Arnö (MySQL AB), Lenz Grimmer (MySQL AB), Luke Lonergan (Greenplum), Richard Hipp (Hwaci), Rick Hillegas (Sun), Ulf Wendel (MySQL AB).
Aside from that many people were helpful in making this paper a reality. Either they helped me with finding solid literature to base my research on or they provided me with people of relevance to contact. Some provided feedback as I discussed the topic with them. I would particularly like to thank Anika König, who was most helpful when I started my research back in December last year. Obviously a paper like this needs to help of many more people: Andreas Nitsch, Anika König, Armin Ortmann, Prof. Dr. Christian Wey, Christopher Jones, Dan Scott, David Fetter, Eike Schilling, Georg Richter, Jan Strotmann, Josh Berkus, Jutta Horstmann, Matthias Bärwolff, Meike Rademacher, Ralf Lippold, Rodney Smith, Tobias Erbsland, Torsten Logemann, Zak Greant.
Anyways I hope that by releasing this paper I can help improve the understanding of the open source phenomenon. Something that I personally believe in very much, but where a lot of people are also learning that its not all shiny and happy. I guess this is both good and bad. So I hope that who ever spends the time reading through it will provide feedback on this blog, sends me an email or lets me buy them a beer as we talk about things at a conference.
Could you put it on cafe-press for printing? I'd like to have it on paper to read it, and I'm willing to pay a bit for it. Just send me an email, and I'll order it :)
Wow, great work, Lukas. Thanks for putting this together. A very valuable analysis. I blogged about it, I definitely think this is worthy of attention.
Congratulations on getting your thesis published!
One note: s/Grant/Greant/
Doh! .. I was afraid of that, so I copied all names from my email client. Guess I thought I knew how to spell your name by heart :(
I will correct it in the pdf in the next version.
There is actually a common Silicon Valley term for what you call "Proprietary Extension": "Secret Sauce". It's widely recognized in the SV tech press that there are three ways to make money from an OSS project:
1) Support, Services & Training
3) Secret Sauce
Interestingly, while Dual Licensing was grabbing all the press a year ago (as "Open Source 2.0") many Dual Licensed companies have begun sliding towards a Secret Sauce model (e.g. MySQL Enterprise). This may mean that the Secret Sauce model is the more durable in the long run -- we'll see!
I don't really see MySQL sliding towards "secret sauce" if it is the same thing as I referred to as "proprietary extension". Of course MySQL Enterprise is somewhat "proprietary" or "closed", but its not like there is a difference as you can see between EnterpriseDB and PostgreSQL. It is essentially the same source code, just compiled for more platforms, more past versions and a bit more conservative choice of patches. So I guess the "secret sauce" in this case is on the packaging level and not so much on the code level. Since the code itself will be available first in the Community Editions.
Page 28 (pdf page 36) mentions updates to Postgresql's 7.6 branch which does not exist.
Directly below that it mentions the Seakmoney project.
Page 39/47 it says in the second paragraph, "... MySQL AB is know as" and should be "known".