IPAC 2MASS Working Group Meeting #70 Minutes
IPAC 2MASS Working Group Meeting #70 Minutes, 8/29/95
Attendees: T. Chester, R. Cutri, T. Evans, J. Fowler, T. Jarrett,
D. KirkPatrick, G. Kopan, C. Lonsdale, H. McCallon,
S. Terebey, J. White
Last week's minutes stated "the offset between KAMPhot and aperture
photometry became smaller in some (but not all) scans, while the dispersion
remained constant". It was subsequently determined that the offset became
smaller in all scans tested.
- New FDD Flowcharts
- Galaxy Analysis
- Lugosi Problems
- Photometric Calibration
- Flowcharting Options
- New FDD Flowcharts -- J. Fowler distributed
revised flowcharts for Figure 2 (2MAPPS Scan Processing) of version 3.0
of the 2MAPPS Functional Design Document (FDD). These
were made in postscript format for inclusion in the 2MASS web site
documentation. The new MAPCOR subsystem was included, along with the new
baseline approach to position reconstruction. Several corrections were
pointed out and will be incorporated.
- Galaxy Analysis -- C. Lonsdale presented a brief
overview of some work in progress to investigate the quality and
robustness of 2MASS galaxy magnitudes. She has selected
all galaxies in the May '95 Hercules data which have entries in the RC3 and
has derived H band growth curves for comparison to published Tully-Fisher
- Lugosi Problems -- R. Cutri reported that
lugosi had crashed again after having already gone
through a crash-and-restart cycle that day. Lugosi has been having problems
lately, shutting down CPUs for unknown reasons. Several hardware fixes have
been tried unsuccessfully, and work is continuing. Meanwhile, running on
lugosi is discouraged, and those who must do so are requested not to write to
/l6, since that is where core dumps are being directed and other debugging
information is being written as solutions are sought.
- Photometric Calibration -- R. Cutri reported that
the point sources from all the photometric nights of the April '95 run
have now been processed for photometric calibration. The
extinction correction slopes now look reasonable, and the night-to-night
zero-point shifts are now computable. Some discussion took place about
how airmass should be handled in 2MAPPS. In the current pipeline, airmass
is computed for each frame and carried in the FITS header, but the FITS
frames are not read downstream where the calibration is done, and the
airmass values used for all point sources in a scan are from the beginning
of that scan. This isn't necessarily too coarse, but it would be fairly
easy to do better in 2MAPPS, so it was decided that FREXAS would
generate an airmass history file for use by CALMON and any other
software that needs airmass information.
- Flowcharting Options -- So far, the experiences
some team members have had with the da Vinci flowchart software has
been disappointing. Those who have tried this package have
all found it probably unusable for our purposes. The following complaints have
Since no one had a positive experience with da Vinci, we will abandon all
official recommendation of it and concentrate on two alternatives. First, we
will order a copy of ABC Flowcharter to be installed under MS Windows on the
public 486 machine (this order has since been placed by R. Scholey); this is
the program used to produce the flowcharts discussed in item (1.) above.
Second, we will try the approach of taking hand-drawn flowcharts to the
graphic arts section in Spalding to be converted into postscript files. J.
White has some flowcharts that we will use as a test case. The postscript
files have to be usable on our network printers, something that G. Kopan
found in the past to be a problem with postscript files generated on
- There are not enough symbols. For example, there is no disk data set
symbol. There is supposedly a way to create "icons" to be used as symbols,
but this appears time-consuming and may not be worth the effort
because of other shortcomings, and the capability may not work any better
than the others. There appears to be no way to control connecting lines
and arrow heads.
- It is difficult (and maybe impossible in some cases) to force the
program to place the symbols vertically where one wants them. The
software demands a reason to go to a lower level; for example, there may
have to be a symbol above the one on a lower level.
- A kludge is necessary if one wants multi-line labels inside a symbol.
There appears to be no way to enter a multi-line label via the standard
dialog box other than to use special symbols like @@ to represent newline
symbols, save the graph to a da Vinci script file, run that through a sed
script to change all occurrences of @@ to \n, and then read the script
file back into the program.
- The documentation is very poor, and the terminology is confusing
(e.g., symbolized items like processing boxes and decision diamonds are
called "nodes", and connecting lines are called "edges").