On Fri, 2005-10-14 at 12:18, Ron Mayer wrote: > Marc G. Fournier wrote: > > On Wed, 12 Oct 2005, Joshua D. Drake wrote: > > > >> PostgreSQL doesn't suffer from that. Our only real, substantiated > >> concern that I can see is the potential for the Software Patent crap. > > > > > > Stupid question here ... if Oracle came at us with "the Software Patent > > crap", > > Personally I think it's quite unlikely Oracle would try attacking > any F/OSS project on patent grounds. They've pretty much bet > the company on Linux (http://news.zdnet.com/2100-3513_22-5825433.html > "Within the next 5 years, half of Oracle's customers may be running > Linux, company President Charles Phillips has predicted". With > the sensitivity in the community that SCO/Baystar-and-friends created > I think that it'd be suicidal for Oracle to start any sort of > patent-vs-F/OSS war. Imagine the speculation of whether they'd > go after Linux itself next......... But that's the beauty of buying innobase. Oracle can attack MySQL and still be the F/OSS hero! It's simple. Release the innodb code base under GPL only, no commercial license. Anyone who wants it can use it, but they have to only use it in GPL projects. Then, MySQL has a hard decision, do they continue to make a GPL version of MySQL with innodb and remove the innodb handler from their commercially licensed version of MySQL, or do they pull it altogether. If they leave it in the GPL version then they are encouraging their customers to rexamine their usage and try to use the GPL version. If they pull it, they encourage a fork of the code base by another group who might want to keep innodb and doesn't mind it being all GPL. Of course, this group could then architect a LGPL connect library on their own, and then MySQL would be freely usable with commercial software like it once was. Boom, MySQL loses large amounts of their funding, and Oracle wins a public relations coup by releasing innobase under the GPL and hosting the project on their servers. > My guess is that Oracle simply recognized that the Innobase guys > were solid database engineers with a product with a growing customer > base in a niche (low end databases) that Oracle didn't have a large > presence. There are, I believe, exactly ONE innobase guy, the primary developer. Nice guy, but there are still a lot of issues to be worked out in innodb. Hopefully, Oracle can provide the funding to make it work. It would be great if Oracle paid to fork MySQL to a pure GPL product, producing a new connection lib under LGPL, and hosting the whole thing as a version of MySQL that ONLY uses innodb. By forcing it to use one and only one table handler, they would then focus development effort on SQL compliance, proper operation, and adding features that work with innodb, like full text searching. Since the basic database would retain a good amount of interoperability with the old MySQL, this new database could easily take away a fair share of their market. Then, Oracle could sell support contracts, not licenses, and back them up with their rather large corporate infrastructure. > On those grounds I could certainly see Oracle buying successful > postgresql-based companies -- again, for both the talented people > and the proven market for those products. But rather than > harm the project, I imagine that would simply create incentives > for other talented database developers to join the project. Yeah, I'd see it something like that, with the focus being on selling consulting services for folks using PostgreSQL... ---------------------------(end of broadcast)--------------------------- TIP 2: Don't 'kill -9' the postmaster
