Werner,

I somewhat believe in market / competition. The point what OSS has to offer to be successful was once detailed very nicely by Eric S. Raymond with his writeup about the fetchmail project that he started. It's called the cathedral and the bazaar. Basically he says, OSS needs to have at least one killer application, that is demanded by the market, where it outperforms other things and thus draws interest.

Now your problem is that castor does about the same job as jaxb, some better, some worse than it. Jaxb is a standard way. And you're competing in a marketplace where there's enterprise software ants working, who fear their decisions being challenged or micro-managed by their upper management. With a standard software package they're on the safe side.

To convince them, you (as in castor developer community, not you, werner, whose dedication for castor I seriously admire) have to come up with a real real KILLER application that makes castor outstanding over jaxb. You know, that kind where developer zines will pick it up and point out that even tho jaxb is available free of charge, standardized, included, etc. etc. etc., that you're way better of downloading and using castor.

Imho you have to ask yourself if you (as a team) have the dedication left to develop this killer application, and support it. And yes, I mean support as in a low enterprise level support definition. Yes for free. Because only if you invest that you will draw enough people that will want to help out that you can actually have dozens of people doing the work; helping with support; developing further; and ultimately even may make money out of it. But you need some sort of quantum leap that will make jaxb (and other comparable solutions) look lame. This is the crux of many an OSS project.

So, that's my view / opinion on things. I may be wrong of course. But I'm also advocating, using and contributing to OSS for a dozen years now -- I don't see the big flaw in thought above. If you do - feel free to point it out, take it off-list or whatever.

So.. thanks for your work so far & Regards,

-Martin

---------------------------------------------------------------------
To unsubscribe from this list, please visit:

   http://xircles.codehaus.org/manage_email


Reply via email to