...
(McEnery) I disagree that the data will be treated differently in inertially pointed and survey mode. The LAT is not pointing at a particular direction (in celestial coordinates) during survey mode, the direction of the LAT boresight is constantly changing. In pointed mode, there is no guarantee that the LAT will stay perfectly inertially pointed, it may move by a few degrees to satisfy contraints, and will move by a lot during an Earth avoidance maneuver. So even in a pointed observation RA_SCX, DEC_SCX, RA_SCZ, DEC_SCZ, OBJECT are not automatically defined, and a science analysis cannot assume that the pointing direction is constant.
(Foschini) My notes were only suggestions, I did not state that there will be a different treatment, but I suggested a different treatment. If I have understood well, although the LAT has no FIFO, that creates problems during slews in other satellites (see, for example, the XMM Slew Survey), it remains open the fact that during slews it is necessary to give the attitude for every single photons. The sum of these errors in the attitude reconstruction changes the PSF size and pattern, that is worse than the case of pointed observation.
With reference to the fact that a satellite is not perfectly stable during a pointing, this is normal, but the keywords RA_SCX, DEC_SCX, RA_SCZ, DEC_SCZ define the boresight of the intruments and deviations from the required attitude during a pointing are tagged as bad-time intervals (the complementary to good-time intervals) and data discarded (or later reanalyzed with different methods). In other words, the basic philosophy in pointings is to have certain key parameters taken as reference and to consider deviations from them.
5. MC_TRUTH keyword
(Digel) I think that the original intent was that this keyword indicate whether the data are entirely Monte Carlo truth values (i.e., actual directions, energies, etc.). We do not actually have files like this, and I propose removing this keyword. gtobsssim does add a column MC_SRC_ID to the files it generates, but this is easy enough to check for without the MC_TRUTH keyword.
...