The discussion was orientated on 2 points : how to validate the VINCI
data and which parameters should be monitored to keep track of the
"health" of the instrument.

Validation of the VINCI data
----------------------------
-given observation bracketed by calibrator observations
-consistency in the Kappa coeeficients (influence of the Color)
-mu^2 distribution (remove outliers scans)
-correlation between the 2 distributions in the 2 Interferometric Channels
-SNR in the photometric channels

How to interpolate the TF to calibrate the scientific Visibility points
-----------------------------------------------------------------------
In FLUOR : linear interpolation of the TF, some TF on calibrators will
be removed if the variation is too large.
We should be careful in the case of the VLTI, because of the uncertainty
in the knowledge of the diameters of the calibrators.

Observations and calibrations needed for the MIDI instrument 
------------------------------------------------------------
(tentative list of parameters to control to be implemented in the pipeline)

-the background should be monitored between 2 sky measurements
-we should see slow variations of the background due to the tunnel and the warm optics
-the detector should also be monitored ( electronics, zero point)
-overlap of the 2 beams