Shibboleth Developers

[Walter Hoehn <>]

If you mean frozen for the 1.x line, then that sounds reasonable, 
especially since any release before 2.0 will probably be a 1.3.x 
release.  During the move to 2.0, any number of components might have 
to change, but of course we want to try and be sensitive to which 
changes might break these kinds of dependencies and avoid those when 
possible.  API stability is mostly dictated by usage requirements, 
though.  Just because something is labeled as a java interface, 
however, doesn't mean that you can count on it staying the same 
forever.

-Walter


On Apr 29, 2005, at 12:11 PM, Tom Scavo wrote:

> On 4/29/05, Walter Hoehn 
> <>
>  wrote:
> > You're right, I don't think there are many (if any) out there.  But, 
> > on
> > a philosophical level, I believe that at this stage in development we
> > still want to have the flexibility to evolve the interfaces between
> > major versions.
>
> Well, this is a potential problem.  From our (GridShib) point of view,
> the API should be frozen (or at least backwards compatible) from 1.3
> onwards.  Starting in summer, we will begin to deploy GridShib, slowly
> at first but eventually across the board (i.e., built into Globus
> Tookit).  Obviously, if you modify the Shibboleth name mapping API (or
> anything that it depends on), you break GridShib.  That would be a
> very Bad Thing.
>
> Tom

Attachment: smime.p7s
Description: S/MIME cryptographic signature