GSS-krb5 and enctype lists, revisited

[Sam Hartman]

>>>>> "Steven" == Steven Michaud <smichaud at pobox.com> writes:

    Steven> So, before _too_ long, the problem that Ken Raeburn's
    Steven> talking about should disappear.  This seems to strengthen
    Steven> the argument for an interim solution -- one that doesn't
    Steven> permanently (publicly) change the MIT Kerberos API.


We should make an API change exactly if we believe the API was wrong
in the first place and the new API is more correct.

It is reasonable for a temporary problem to expose a long-standing
flaw in an API.