testing errors

[Cantor, Scott]

> Hmm, I suppose this ship has already sailed but that is the strongest
> argument I've heard to-date for a Java SP.

There's no longer widespread demand for Java as a development language. The main reason for that functionality is to embed it into the IdP to get proxying support. I think a stand alone Java SP would have very little appeal to most people at this point, that ship has indeed sailed for Java. It's not of no value, but it's not enough to justify the time it would take. We'd be better off collaborating on the Spring Security module for that and getting it moved to OpenSAML 3 and enhanced a bit. Not totally out of the question.

Java's problem is what it's always been, there's simply no standardized app framework to fit into that has any widespread adoption, so it's inherently fragmented to build anything. It's also simply not worth it. Very few people run Java containers without a front-end in practice (I do, but I'm in a minority). The SP works fine for that already, but no, it's certainly not great for testing.

Of course, some kind of one-off SP for testing purposes is also not out of the question, that's probably more tractable for us to consider. But it's hard to get people to accept that something isn't suitable for real world use.

-- Scott