>>>>> "Ted" == Ted Lemon <mellon@hoffman.vix.com> writes:

Ted> Chris, I'm not asking these questions in order to be a Luddite.

Oh, I never got that impression at all.  You asked why we should want
to integrate a module-based authentication system.  That's a perfectly
reasonable question.  Sorry if I sounded nasty in my response.

Ted> I definitely would like to have new authentication mechanisms
Ted> available to NetBSD users.  Kerberos V is pretty high on my list.
Ted> The question is, do we need an elaborate, complex infrastructure
Ted> to support it, or do we just integrate it?

What I was attempting to say in my previous message was this:  I like
kerberos, but that's because it's just about the only decent option
for networked authentication.  I think we're going to see other auth
systems coming along, though, that will (hopefully) create some
healthy competition.  My opinion is that we should support these new
auth systems, to the degree that providing such support is
reasonable.

In other words, we should support kerberos via some module-based auth
system, rather than by integrating kerberos code directly into a bunch
of different native apps.  As a result, we will be making it much
easier to support future authentication systems.

Chris

-- 
-----------------------------------------------------cjones@math.montana.edu
Chris Jones                                          cjones@honors.montana.edu
           Mad scientist at large                    cjones@nervana.montana.edu
"Is this going to be a stand-up programming session, sir, or another bug hunt?"