OpenSAML user discussion

["Scott Cantor" <>]

> How far are those two projects to being stable?

Depends what you mean by stable. If you mean API-frozen, not until
Shibboleth freezes, which is at least two months. I won't personally be
stress testing or looking for leaks until late in that cycle either. Those
would be the obvious definitions of "stable" that I could think of.

If you're asking if they build, Windows almost always will from here on out,
Unix occasionally breaks when Makefiles get out of sync.

If you're not looking for guarantees, I would say that the only large API
changes that might occur are in the security code, and that's only iffy. I
don't think I'll have time to make big changes as long as it works.

I have a working SP implementation of SAML 1.1 via POST using all new code,
so that's a milestone of sorts that demonstrates basic functional adequacy,
I guess.

-- Scott