Pervasive, EnterpriseDB, and Making Money on Open Source

So, I was reading one of the press releases for EnterpriseDB’s [http://www.linuxelectrons.com/article.php/20060801175203576 latest round of venture funding], and contrasting it with an article on Pervasive [http://www.internetnews.com/dev-news/article.php/3623936 pulling out of the market], and something popped into my head. A lot of us in the open source world scoff at companies like IBM and Oracle for giving away free, limited / “crippled”, versions of thier database software, and then charging a premium for more enhanced offerings. The funny thing is, this is basically what all of the current PostgreSQL companies are doing. Greenplum and EnterpriseDB are both obvious examples, the extra oracle compatability and data warehousing are the upsell products, which in fairness to them it isn’t really obvious that these upsells would be accepted back into the community anyway, though probably more would be than not. But looking at the next big players in the market (and he’ll love me for calling him this), Command Prompt essentially does the same thing, with it’s Mammoth Replicator product. Fujitsu is also following this path, with it’s special blend of [http://www.fastware.com/postgresql.html enhanced postgresql offerings] it is pushing in Australia and south east asia. Sun is still a pretty new player, but I think they are going to follow this trend as well. So far the two big items to come from them are the the thumper based data warehouse appliance, and a proposed package of a dtrace enabled postgresql running on Solaris. This isn’t really as closed off as say a Bizgres MPP, but it’s not exactly open either. And this is where Pervasive was somewhat interesting. As far as I know, they didn’t really try to sell a proprietary version of Postgres, even thier pervasive postgres package was really just a rewrapped community edition. They did have some claims to improved drivers and some other tools (I think this was more of just an uber distribution), but by and large they really seemed to be competing on a service and expertese model. That can’t be easy in this market. I will note that SRA seems to be the other company that is trying to survive on the service and expertise model, although thier activity right now is a little nebulous (read as unfocused?) , and it isn’t as if they have tried the closed format as well (remember [http://sraapowergres.com/en/ powergres]?). Note I am not really doing a judgement call here, just throwing thoughts out there. I’ve always thought it’s good to be aware of your surroundings, and getting them down on “paper” never hurts. Hopefully it gives you some things to think about too.