On 21 Dec 1998, Niels [ISO-8859-1] Möller wrote:
> "J.H.M. Dassen (Ray)" <[EMAIL PROTECTED]> writes:
>
> > Please look at the PAM documentation. Using PAM means one doesn't have to
> > deal with a system's specific password setup (shadow or not, DES or MD5SUM,
> > NIS, NIS+). From http://www.kernel.org/pub/linux/libs/pam/modules.html it
> > looks to me that it isn't bound to password authentication.
>
> I have now looked at PAM, and I'm disappointed. The summary below id
> from the NOTES file of the December 21 snapshot.
IMHO we should only use PAM for password authentication. It can be easily
done, and doing this results in transparent shadow, md5 and NIS support.
That's less than PAM was intended to do, but it's more than nothing.
(though libc should transparently support the last two)
The _idea_ behind PAM is great, but the design is poor.
BTW: it comes from Sun, solaris has the same API.
--- Bazsi
