************************************************
  * Late minutes of the EW task force    (9-may) *
  ************************************************
 
1) Henri stressed the point that for the calculation of the
   off-peak Bhabha cross-sections and asymmetries it is
   important to produce off-peak UNIBAB Monte Carlo.
   For the present results, in fact, the peak UNIBAB
   efficiency is also used for the off-peak points.
 
   Henri and Jean Claude have completed the analysis
   of the tau and muon asymmetries for all the years 90-95.
 
2) UNIBAB version 2.2 has been interfaced  to Kingal
   thanks to Pere and Brigitte.
   The Bhabha production with the latest UNIBAB version
   has in fact already started.
   In the previous minutes it was uncorrectly said that
   the acollinearity cut could be removed in the Bhabha
   generation. In fact such a cut is essential,
   otherwise the generator will spend
   most of its time generating events with a very low weight
   which will not be accepted.
   Consequently the MC production was started with the usual
   choice of max acoll.=30 degrees and |costheta_star|<0.91.
   This choice of the cuts is also suitable for Gerri's
   background studies and for Ioana and Edwige's analysis.
 
   Andrea should tell us if some more MC is also needed
   for the muon analysis.
 
3) The mismatch between the integrated luminosities given
   by scanbook now and those published in our 93 paper for
   the '90 peak point is now understood (thanks to Tom's
   help). It turns out that for the paper, in order to have
   a better stat. precision, relative luminosities were used.
   However some of the runs were not included (about 128nb-1)
   for the rel. luminosities calculation (for reasons that we
   do not know) and this explains the difference.
   The correct luminosities are the ones now given by scanbook .
 
4) Pere addressed the question of the "old" reprocessing
   of the 90-91-92 data. Also: can we use the current
   reconstruction to generate MC for earlier years?
   No major diffences among versions exist.
   However Pere pointed out that two different Gampex
   versions were used for 90-92 and 93-95.
   This should not be a prob. since the only difference
   between the two versions is the addition of new
   estimators which are not relevant for us,
   as pointed out by M.Noelle.
   There is also a bug in the Hcal hit finding
   corrected after alephlib 15.6 (see following table),
   therefore affecting the 90-92 data.
   This has no influence on the muon analysis but could have
   an effect on the taus via the eflow.
   The 91-92 MC has the same bug for.. compatibility with
   the data.....
 
  Latest Reprocessing Available on Data:
 
  year      Last        Kind      JULIA      ALIB      MINI
          reprocess.              vers.      vers.     vers.
 
  90      Mar 92      POT->DST    259.05     13.5       9
  91      Sep 93      POT->POT    264.14     14.4       9
  92      Dec 92      RAW->POT    264.05     14.2       9
  93      Sep 95      RAW->POT    276.04     20.7      10
  94      Dec 94      RAW->POT    275.04     20.4      10
  95      Dec 95      RAW->POT    279.02     21.1      10
 
5) Brigitte showed us some plots were it was apparent that
   for some runs there is an abnormally high background.
   The Sical group therefore requests to move three 95 runs
   from the EW to the EA selection (action-> Lee)
   They are runs:
                   37755 (P-2)
                   37550 (P)
                   37551
 
Next meeting:  may 15 at 15:00 room 1-1-045.
The time has been moved following a request from MNoelle
and Bolek. See you there.