[OpenID] JanRain library licensing (was: Re: On OpenID 2.0)

[Josh Hoyt]

On 5/11/07, Christopher St John <ckstjohn at gmail.com> wrote:
> I believe the implication was that multiple implementations made
> the spec itself better, and helped solidify the interpretation of
> ambiguities and identify holes.

That's true to a point, but *only* if you write an implementation that
actually matches the specification and then try to do interoperability
testing with other libraries and then report bugs to the other library
authors or request specific clarifications to the specifications. Most
of the implementations that I have seen just do the minimum that they
can to get it to work for the common case. This does not help the spec
or the community.

I have a visceral reaction to unnecessary redundancy. I think that in
general, cooperation is better than competetion, and that unless you
have a specific, solid reason to duplicate code, you shouldn't. I'm
not going to say any more about the proliferation of implementations,
because I think that topic's pretty much been covered now.

Josh