Dear all,
Please find below the comments that I received from Patrick, with some small additions by me. If you still like to add something, please respond before Thursday afternoon. It is a short paper.
Thanks, Wouter
-----------------------------------------------------------------
General: We disagree with the abstract and conclusions. You claim "indication" of a suppression (enhancement) of Y(1S) in the forward (backward) region. You measure 0.89 and 1.19, which both deviate from 1 by 0.8 sigma. We think the only fair statement is that the measurement is compatible with no deviation from one, but also compatible with theoretical models that predict a deviation.
The same applies for the FB ratio that is 1.4 sigma off. This is compatible with 1, compatible with predictions and compatible with the J/psi measurements.
The fact that all three deviations go in the expected direction is an interesting thing to point out at conferences, but cannot be defended in a paper.
More comments line-by-line:
L.1 what are the "early stages of ... collisions"? We think of a collision as something that happens at one given moment in time. You mean the evolution afterwards?
L.7-8: Remove "The list of" and write "include"
L.23: Remove "to be clear"
L.27-31: This sentence is too long. Move "based on... [21]" elsewhere
L.32 and other places: capital S in R^nS/1S.
L.35-36: We suggest "From the theoretical point of view, prediction for bottomonia are more reliable than those for charmonia due to the heavier quark masses and lower quark velocities.”
L.39: It looks like you’ll present a whole list of independent measurements, while the only thing you really measure are 6 production cross-sections: the rest is inferred.
L.41: remove "in pp collisions" (or replace by "in the laboratory" frame).
L.43: ‘respects’ —> ‘refers to’
L.44: insert ‘,respectively,’ in between ‘rapidity’ and ‘with rapidity'
L.78: number of events -> integrated luminosities
L.82: "Upsilon productions are" —> “Upsilon(nS) production is"
L.89-91: You use Prob(mu), which also uses information from the tracking system. Please ask Chris for the appropriate description.
L. 91: “1000 MeV/c” —> “1 GeV/c”
L.93-5: An unbinned extended ... mesons, in a fit range 8400 <….
L.96-102: We don't think the CB needs a mathematical definition any more. You can replace all that by "Gaussian with a power-law tail on the left side, whose parameters have been fixed using the high statistics J/psi..." There's also no need to cite Tomasz.
L.101: Doesn’t the upsilon mass resolution depend rathe strongly on Upsilon kinematics? If so, how do you account for that? (We cannot find it in the systematic error section.)
Fig1: add more space above the caption.
L.110: remove above
L.112-3: "candidate using the sPlot technique [40] with the dimuon..."
L.113: “discriminant variable” —> “discriminating variable” or eventually just “discriminant”
L.116: We think we understand what you do, but readers may be wondering what you do with background events (w_i<0).
L.119: p_T and y could be spelled out in words
L.121: latter. But what does it refer to? There are three efficiencies mentioned above.
L.130: systematic uncertainties due
L.148-150: Why half? If there's a significant difference between the trigger effs and you understand why, you should correct you numbers. If not, then either take the uncertainty on the trigger eff, or the full difference (depending on how large the uncertainties are).
L. 151-156: why does the result so strongly depend on the binning? is this purely due to MC statistics (in which case it should be labeled as such)? how did you choose the bin size?
L.157-160: Re-order: The systematic uncertainties due to ... have been studied in the J/psi analysis and determined to be 1.5%. Table 1: The values for Y(1S) in the ... are given in parentheses (make sure specially gets removed).
L.168+: Do you allow your signal PDFs to go negative? Judging the appendix C of you anay you do. You should probably say that in the fit section. In Table 1 you give the uncertainty for the fit model in percent (90%). But it's more likely to be an absolute number (imagine you had obtained -0.1 instead of +5 nb. What would you have quoted?
L.170: “error” —> “uncertainty"
L.171: insert "and the variation of the signal-to-background" after 'efficiencies'. (If the signal to background is the same everywhere, then your statistical uncertainty can NOT decrease by reweighing.)
L.180: "albeit with large uncertainties". Please quote them, or refrain from making any statement.
L.194-196, 221-225: See our general comment above: The measurement does not backup these statements.
L.214: Measurements and ... agree within... "
Fig 3: The top left plot in unreadable in BW. Caption: b-hadron decays
Fig2 and 3: It seems a bit exaggerated to add 8 plots to the paper to show just 3 LHCb measured data points. We propose to show only one plot from Fig2 and one from Fig3, either by choosing the model, or superimposing them. The original plots can be added as additional material for speakers.