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

[Christopher St John]

On 5/11/07, Jonathan Daugherty <cygnus at janrain.com> wrote:
> # Good grief, chill out guys, people wanting to implement the spec is
> # a _good_ thing, it means OpenID is winning...
>
> And we need to remember that there are two different kinds of winning:
> the technology wins in terms of industry and developer acceptance, and
> the technology wins in terms of *user* acceptance.
>
> Having lots of implementations may imply that there is winning of the
> former sort (again, forgetting ethics of re-use), but when things
> don't work half the time, that's not winning with users, so we have to
> have some balance.  A clean-room implementation *is* good for the spec
> and the internal discourse on the technology.  Scaring users away from
> an RP because it runs a broken or buggy fly-by-night implementation is
> *not* good.
>

Users don't choose OpenID implementations. Platform or application
developers choose them. And they aren't totally stupid. They are
capable of using Google.

Assuming OpenID is a success long term, there will be:

 - One or two primary Open Source implementations per major
   computer language and/or framework. They will be high quality
   and differ in obvious and well documented ways, possibly in
   licensing. The Java one will be from Apache.

 - Possibly one major closed implementation per platform. Microsoft
   will have one integrated into the .Net stack, and we might see
   proprietary implementations from other big players. Or they might be
   the major contributors to one of the Open Source efforts. (Sun and
   IBM spring to mind)

 - Many experimental implementations, either in obscure languages,
   or for very specific contexts, or just for fun. (eg " "Chris's experimental
   Haskell implementation I did last week so I could play with monads"
   and "OpenID in Forth for PIC Microcontrollers")

Generally, it will be completely obvious which implementation to pick,
since it will be at the top of any Google search, followed by things like
my imaginary Haskell implementation.

Yes? I mean, in theory, it's possible it won't work out as above, but
it's certainly the way to bet. And the possibility for actual confusion
seems pretty minimal.

-cks


-- 
Christopher St. John
http://artofsystems.blogspot.com