RFD: authentication/access control

[Christophe Kalt <kalt@xxxxxxxxxxx>]

I've been thinking about this for a while, and need some
feedback on it before I think more about it.

currently, ircd supports

o 2 methods for authentication: DNS (for the hostname) and
  identd (for the username).
o 2 methods for access control: K (very flexible, but works
  based on user@host only), and R lines (better, but
  blocking!).

The most common use of R lines I have seen is to perform
additionnal authentication making possible to decide whether
access should be granted or not.

What I am considering is modifying ircd to easily allow for
more authentication mechanisms, and non blocking R lines
while limiting the number of processes spawned (R lines are
handled by an external program, non blocking => several
processes).

I'm interested in hearing from people who would need other
authentication mechanisms (what is it? public?  homebrewed?
..) In particular, I'd like to hear from people currently
using R lines.