On the topic of code reliability, my understanding is that FreSSH is
substantially smaller than OpenSSH.  I don't know if that's just because
the FreSSH authors haven't had time to add features, though.  If it's
smaller because of a better design, though (e.g., due to rethinking of
basic issues in light of how ssh has evolved), then even if it gets the
same feature set, it may remain smaller.  (There might also be features
that would never make the cut to FreSSH.)

Assuming that FreSSH is smaller, then all else being equal, we should
expect FreSSH to be more reliable.  There would be fewer places for bugs
to hide (in terms of lines of code) and probably fewer relationships
between parts to hide more subtle bugs.  This is, of course, a qualitative
argument.  (^&

Of course, all else seldom is actually equal...


On the other hand, what is included in NetBSD shouldn't be based on
whether some component may be over-hyped, but whether it is the best
component for NetBSD's functional needs.  If someone wants to zing OpenSSH
developers over this (with or without justification), there's probably a
slashdot thread going on that; or there's always [ab]Usenet.


Bringing back to the actual topic of the thread (FreSSH): I was interested
when FreSSH was first announced.  If someone (or group) returns to work on
it, that would be good to hear.  (I don't recall why I stopped using
FreSSH.  It may have simply been that I upgraded to a NetBSD version that
included OpenSSH and I didn't bother to grab FreSSH from pkgsrc.)


  ``I probably don't know what I'm talking about.'' --rauch@math.rice.edu