foster comments, continued, page 3:

--------------------------------------------------------------------------
                           "NEAR-TERM" OBJECTIVES:

  * TIME-GCM (release tgcm23):
    o  Port to non-cray platforms, following tgcm14 methods.
    o  Incorporate new top-level code from tgcm14 (including netcdf histories).

  * TIEGCM (continued development of tgcm14):
    o  Experiment with double spatial (1/2 temporal) resolution of tiegcm.
       (Distribute over 24, or up to 64 processors)

  * New platform/parallel method combinations, e.g., OpenMP on IBM,
      MPI on Cray and SGI, hybrids on non-Compaq platforms. 
    Port to to multi-processor Sun platform (e.g., SCD's k2 system). 
    Experiment with multi-processor systems at HAO as they become available.

  * Continue parallization in effort to improve scalability
    (e.g., parallel i/o, multi-tasked dynamo).

  * Implement new Richmond dynamo code (tiegcm first), and parallelize.
    Enable magnetic coordinate fields on secondary output histories,
    (and subsequent import and visualization by post-processors).

  * Other improvements and additions as directed by Roble, Richmond, et.,al.


- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
             POST-PROCESSING, DIAGNOSTICS, and VISUALIZATION

  * Port tgcmproc to hao Sun workstations, and new SCD systems with 
    NCAR graphics (e.g., IBM).

  * Improved on-line documentation (User's Guide) for post-processing.
    Make it as easy as possible for a new user to obtain results quickly.

  * Test and make available locally (hao machines) various netcdf 
    processing and visualization utilities.

  * Test compatability of TGCM netcdf histories with new CCM post-processing
    codes (e.g., NCL scripts that read netcdf files).

  * Resurrect v5d animations with SCD vis lab.

--------------------------------------------------------------------------