Kim-sensei and I have read your NIM draft for Jet Energy Scale
calibration (cdfnote 7543) and have some questions.
Can you answer them?

  1.  The (Ecem+Echa)/p plot in figure 6 converges to ~0.8 with high
momentum tracks at ~20 GeV/c.
       Is this because of the energy escaping from the crack in central
calorimeters?
       Seeing figure 15, the crack region looks to cover only about ~10%
of the area.  Does the different response
       to EM and Hadron components of the shower contribute in the
E/p~0.8?
       Dijet balance in figure 17 typically have values around ~1.1, and
looks to be in good agreement with the test beam
       results.

   2. Test beam plot for CHA in figure 8 converges to ~0.9, while PHA
converges to 1 in figure 10.
       The CHA test beam only used one calorimeter tower, and we thought
E/p for test beam would be smaller than
       in single track data.  Is the shower properties (especially
composition of shower particles) different in different
       particle momentum range?

   3. Monte Carlo in (Ecem+Echa)/p plot in figure 6 stays very flat with
~0.8 in 10-20GeV/c region.
       Can we expect this will smoothly connect to test beam result with
values ~0.9?
       Or is the difference considered to be due to the test beam setup
only using one tower?

   4. In figure 17, Monte Carlo description is not very good in high eta
regions for dijet balance.  We understood this is mainly
       due to the limited statistics of data for tuning Gflash.  You
performed tuning of Gflash independently for central and plug
       calorimeter regions.  Is that correct?

I am sorry we did not have a chance to read the draft earlier, but can
you include comments on these issues in your publication
if it is not too late?  We consider such comments would be nice to be
included.
Anyway, congratulations for the nice description for your tremendous
amount of work!

Thanks in advance,
Koji Sato
University of Tsukuba