AIPS:  2003 Quarter 4

Key Developments

1.   Rolled the 31DEC03 version over to 31DEC04 and froze the 31DEC03
release.  Issued an AIPSLetter to describe the changes in the last six
months.

2.   Added various suggestions from Australian users to enable them to
run the midnight job.

3.   Made a variety of improvements to the VLBA calibration task APCAL 
including the use of the available weather data, improved 
interpolation options, and better opacity fitting and plotting.

4.   Upgraded the FITS readers to support the new World Coordinates
System nomenclature where possible and allow FITLD to read more than
one VLBA correlator data file from disk.

5.   Oversaw and verified the inclusion of calibration data from the
VLA when a single VLA antenna is used with the VLBA.  Corrected the
handling of weather tables while doing this.

6.   Solved a variety of problems with the MacIntosh OS/X installation
including getting very nice performance using an IBM compiler on G4
and G5 processors.

7.   Updated the CookBook and other documentation files for 31DEC03.

8.   Major bugs handling sub-arrays when recomputing projected
baselines (UVFIX) and flagging bad solutions (SNSMO) were corrected.

9.   In the period May 18 to December 31, 2003, 88 sites downloaded
the 31DEC02 version of AIPS and 416 downloaded the 31DEC03
(development) version.  A total of 551 different sites (separate IP
addresses) made some use of the AIPS cvs facility, either during
installation of 31DEC03 or running the "midnight job" to update their
copy of 31DEC03.


Goals for the First Quarter 2004

1.   Continue user support and bug fixes, as the major portion of AIPS 
effort.

2.   Investigate the use of overlapping solution intervals to minimize 
lobe ambiguities in fringe fitting.

3.   Continue the moderate-term project to explore improved troposphere 
calibration.

4.   Add weight-based flagging to be able to use flag tables.

5.   Begin investigations of new/improved imaging algorithms,
including those dealing with spectral index and multiple pointings.

6.   Consider methods to include source models with AIPS
distributions.