MIT and the Novell LDAP Contribution

[Sam Hartman]

Hi, folks.

There has been a lot of discussion over the last couple of weeks about
the LDAP plugin.  AS people have tried to use it, they have run into a
lot of questions and some initial bugs.  So far, there has been one
important thing missing from that discussion: thanks to Novell for
their excellent work.

Every time some new feature is introduced, particularly some
complicated feature or change, there's going to be a lot of
discussions as people find problems, ask questions and propose
alternative ways of doing something.  Remember how long it took us to
get enctype handling right.  Remember the long discussions surrounding
TCP.  We're still having periodic discussions of threading.

The fact that the Novell code is drawing such discussion is a sign of
the strong demand for that code.  If Novell hadn't done the hard work,
we wouldn't be hear wishing for various improvements in the code.

Novell has done a great job.  They have also had tremendous challenges
working with a complicated code base like MIT Kerberos that they were
reasonably unfamiliar with, working with the limited resources we
could give them discussing design, and yet they managed to deliver.
Yes, there are problems that need to be fixed.  Yes there are
improvements we all want.  Both MIT and Novell are committed to
improving the LDAP code.


I'd also like to thank Ken Raeburn for tireless hours helping to
integrate the Novell code into the release.  Integrating a large third
party contribution is never easy. Ken has done an excellent job of
taking components such as the DAL plugin interface and generalizing
them into a plugin architecture

Thanks also to all those who have been giving suggestions to Novell
and MIT for improvements in the code or who have been helping test and
find problems.  Your efforts are greatly appreciated.


Sam Hartman
Manager, MIT Kerberos Team