Pan 0.11.0.91, "Alethiometer", has been released.
It can be found at http://pan.rebelbase.com/.
 
What is Pan?

        Pan is a newsreader, loosely based on Agent and Gravity, which
        attempts to be pleasant to use for new and advanced users alike.
        It has all the typical features found in newsreaders and also
        supports offline newsreading, article filtering, multiple connections,
        and a number of extra features for power users and alt.binaries fans.

        It's also the only Unix newsreader to get a perfect score on the
        Good Net-Keeping Seal of Approval evaluations.  See
        http://pan.rebelbase.com/features.html for the rest of the
        sales pitch.

About 0.11.0.91

        This is a bugfix release that adds no new features.

        A revision number of >= 90 is used to flag this release as a beta
        without confusing RPM's feeble brain with a name like "0.10.0-pre1".

What's New in 0.11.0.91

    UPDATES

        * Better error-checking, handling, and writing-to-the-Pan-log
          when the group header cache can't be saved.

        * Better data recovery in the case of a Pan or system crash.

    BUG FIXES

        * Significant memory leak when saving articles fixed.  Thanks to Rick
          for noticing this and tracking own the problem.

        * Fixed bug that caused article bodies to be saved improperly when
          the user specified the filename from the save dialog.  Thanks to
          Andreas Ollinger for reporting this bug and to Joseph Turian for
          providing a patch.

        * Fixed bug that caused tasks to get stuck in the task manager.
          Thanks to Alex V Flinsch, Alan Young and Gabi Davar for reporting
          this bug.

        * Fixed awful bug that caused the server header cache directory to
          not be created, which caused new users to not be able to read
          articles in 0.11.0.90.  D'oh!

        * Fixed 0.11.0 bug that, when an empty group was activated, Pan would
          go to the server twice for new headers instead of once.  In addition
          to pointless load, if you were really unlucky there would be one or
          two new articles in the second load, marking the ones from the first
          run and leaving you scratching your head "how did I just download a
          batch of old articles from a new group?" :)