S.Haugan had gone through data from all instruments supplied by the official archive to verify that data can be read, understood in relation to the physical measurements done with the instrument, and used at least in a qualitative sense (notwithstanding any calibration issues yet), with a reasonable investment of time searching for information on the data sets.
The results can be summarised as:
GOLF: FITS files contain descriptive labels in French, and this made it difficult to decipher the contents and relate them to the physical measurements taken. A working knowledge of the language may have changed the outcome. Simple tools or descriptions of file contents should be available.
VIRGO: Passes, with good descriptions through ESTEC home page on how to read data using standard tools. But: TFORM keyword is misspelled and TNULL values not defined in PMO6 data (easily fixed, though).
MDI: Well documented, no problems. No time-series files inspected.
SUMER: It is possible to read and understand the data, and there is quite a bit of documentation available. However, within a short time span 3 different sets of routines for handling data were found (Wilhelm, Warren and Carlsson were the three main sources). It is also very unclear how to apply the ``Wilhelm'' set of routines to the data set in the archive (if it is indeed possible).
CDS: Data can be read, understood and used. A true test of how an outsider would view the experience was impossible, due to the affiliation of the test person.
EIT: Useful, straightforward user's guide. No problems. Multi-image data sets not explored, but they are described in the user's guide.
UVCS: Information on how to read, calibrate, and manipulate data appears to be well documented (and the description worked as far as it was tested). It would presumably take quite some time to get useful information from the data (other than single exposure displays), but this comes with the nature of the instrument.
LASCO: Data can be read and interpreted, but it was impossible to use for e.g. identification of streamers etc, since no model subtraction had been done. It would have been very welcome with a software package that could do this (with the calibration data distributed through e.g. the SolarSoft Database).
SWAN: Data could only be interpreted after some ``dimensional analysis'' (relating the number of pixels to the sizes of data arrays). Also, only maps in ``mechanism encoder'' coordinates were possible, with no conversion algorithm to sky positions (even relative to SOHO). Amendments have been promised.
CELIAS: Files are in CDF format -- which are supported by stand-alone libraries, IDL routines etc. However, they are somewhat ``opaque'', meaning that very active exploration is needed to figure out what they contain. However, given the effort to do this, the files seem to contain enough information to enable an understanding of the data. Some help or explanation of file contents should be available.
COSTEP: Easy to read ASCII files, but not easily understandable. It was pointed out at the meeting that file documentation has been provided. Yours truly has now discovered the pointers to the file documentation - from the SOHO software page, in the SolarSoft tree!
ERNE: Easy to read ASCII files, but not easily understandable. It was pointed out at the meeting that file documentation has been provided. An investigation is under way to make this available.
It was agreed that S.Haugan (action 33-4) will construct a web page template for each instrument to fill in with links to locations with e.g. file descriptions and other relevant information. These pages will be pointed to from the archive.