Comments/Responses
on PRL draft 0.12
"B -> D** mu nu X"
4/4/2005
from G.Borissov
1. General. Most probably, you will have the problems with space.
Consider to replace 2-column formulas to 1 column. Also, recently
a very short detector description was distributed. Try it.
ok
2. You refer to B+/B0 paper on mu-D0 selection. It has an important
difference in the cut pt(D0)>5 GeV. From this the difference in the
number of events in these papers. Probably you should explain it.
ok, rephrased this piece
3. Figure 2 caption. Explain peaks shown in the figure
ok
4. Page 5 in the bottom. "To determine the number of events ...
the fitting function has been modified..." This sentence is not
clear. I understand that you fit N1 and N2 instead Ntot and r,
but there could be questions from the external reader.
after some additional discussions and checks we concluded that both pairs of
the variables that we were using (N_total; ratio) and (N1; N2) are equivalent.
it's important to use the corresponding variables during the fit so the errors
are correctly estimated. we changed the associated text so the reader is not confused.
5. This comment is from my experience with B+/B0 paper: explain
more the systematics. Statements like "Efficiency modelling systematic
error accounts for possible mismatch of kinematic variables..."
is quite obscure and can cause the referee's questions. Give more
details.
modified
6. I still think that the interference effect can occur if the efficiency
is not flat. It should be tested with Monte Carlo. The angular distributions
for these states is well known, it could be easy to check it.
yes,indeed. the helicity distributions for D1(2420) and D*2(2460) in MC are shown below.
the helicity angle alpha is defined as angle between pi* and pi** in the D* rest frame.
D1(2420) (1+3*cos^2(alpha))
D*2(2460) sin^2(alpha)
one can see that the efficiency is lower at cos(alpha)=-1 especially for the D1(2420) case.
this is mostly due to the momentum cutoff for the slow pion. we've checked that the other
analysis cuts do not influence the angle distributions. numerically the efficiency drop is 4.7%+-0.5%
for D1(2420) and is consistent with 0 for D*2(2460) (cos=-1 is suppressed in this case).
the interference effects should cancel out after integration over the angles but the efficiency effects
can prevent them from the full cancellation. it is clear however that 1) the effect cannot be more that 4.7%
mentioned above (=4.7% when efficiency is flat vs cos(alpha)). 2) since most of the efficiency drop happens
in the region where one of resonances is supressed (D*2(2460)) the effect will be less than 4.7%.
more coming here.. short summary : after much thinking we are convinced now that this systematic is very small.
in PRL draft 1.01 we assigned to it a 2% relative uncertainty. we will do the math (integration over angles for the distributions)
to prove this. at this level it will not change the results.
7. Again for the systematics from fitting procedure. Give more details.
e.g., did you vary parameter z in the Blatt-Weisskopf form-factor?
checked how results depend on z : changed z = 1.6 GeV by +- 0.2 GeV. Largest Br change was by relative 0.5%.
so this is a small systematic and can be neglected.
8. Order references in the right order. E.g. ref [22] should not be in
the end.
ok
3/29/2005
From Rick van Kooten
Physics comments indicated with a "*", English and topography with
a "o" symbol.
Abstract
----------
o Whenever writing a semileptonic Br, I believe it is clearer to
write, e.g., Br(B -> D0_1 mu nu X)
rather than the other order. This is also
how the PDG expresses it, and in formatting, we are supposed to try
to conform to the PDG style when possible.
corrected
o Colon after "equal to".
ok
o Systematic error on the second line in the text is 0.017 instead
of the indicated 0.012.
ok
o Space between the number and "(stat)" or "(syst)".
ok
Page 3
--------
o Throughout: "B" in "B mesons" should be in italics.
ok
o "narrow D** mesons" is jargony. Suggest: "decays of $B$ mesons to
orbitally excited states of the $D$ meson that have narrow decay
widths. In the simplest case, these states consist of..."
ok
o Many of your defined symbols have too much space before and after
the symbol. Check your definitions, use \defn instead of \defn\ .
o Refs. [1], [2], write it instead as \cite{ref1,ref2} to get the
two numbers inside of the same square bracket.
ok
o "are expected to be narrow" "are expected to have narrow decay
widths"
ok
o "to be wide" "to have wide decay widths"
replaced with "to have large decay widths"
o "Ratio of semileptonic" "The ratio of the semileptonic"
ok
o Colon after "branching rates of B meson to D1' and D2*".
ok
o Comma after R definition equation.
ok
* "least model independent predictions of HQET", but then the
prediction is between 0.4 and 0.7? This is a prediction with a
fairly hefty error; there are many other HQET predictions better
than this? Worth it to make a big deal that it should be so model
independent?
added "...is one of the least
model dependent predictions of the
Heavy Quark Effective Theory
(HQET) for the orbitally excited $D$ mesons."
Added also more to the motivation section.
o "it can be reduced to" "it can have a value of 0.4 - 0.7"
ok
o For a range of numbers like this, use "--" in LaTeX to get the
correct symbol.
ok
o "The narrow D** mesons", space after D** (and now can probably get
away with "narrow" for non-HEP readers...
ok
o "measured by DELPHI" "measured by the DELPHI"
ok
o "value of R, 0.37 ± 0.14" "value of R = 0.37 _/- 0.14"
ok
o "is a result of averaging" "is the result of averaging"
ok
* For Ref. [9], how do they get rid of the relative Br(D**->D* pi)
branching ratios? Surely they are in the same boat?
yes, they are and actually we
are using the same numbers for this Br when we calculate R at the end
as they do.
4/22/05 at the end we decided to use the PDG number for the ratio
Br(D_2^*->D* pi)/Br(D_2^*->D pi) = 2.3 +-0.6 (p.699)
to deduce the Br(D_2^*->D*
pi) = 30+-6%.
please see more on this topic at the very top of this link
http://d0server1.fnal.gov/users/nomerot/Run2A/Dstst/prl_comments.html
o Consider replacing the long, extended dance version mix of the
detector description with:
ok, replaced - saved some space indeed
o "D** meson production in the B->mu nu D* X decays" "D** meson
production in B->D* mu nu X decays"
ok
o Ref. [22] wildly out of order. Go through and check that the order
of citations in the bibliography matches the order in the text.
changed
o Figure 1
- clean up for publication quality:
- remove grey window - Add "D0"
- Add labels on both axes - remove title block in upper left-hand
corner
- consider making the fit line thinner, it is obliterating the data
points
- consider plotting points with error bars for the data points (but
okay to leave the same-sign as a hatched histo).
- in the caption, have to describe all the components, i.e., curve
is fitted, points with error bars opposite-sign data, hatched histo is
same-sign candidates.
Figure updated
* "where D** and D0 mesons were reconstructed" "where D* and D0 mesons
were reconstructed"
thank you
o "The search was started with a selection of the semileptonic sample"
"A selection of a semileptonic sample ... was obtained ..."
ok
o "c" should be math italics throughout: GeV/$c$ and GeV/$c2$.
ok
o drop "and pseudo-rapidity" (eta is defined earlier).
ok
o "invariant mass compatible" - define "compatible".
rephrased this paragraph, mention also the 5 Gev cut of pT(D0)
o "with the mass of D0" "with the mass of the D0"
ok
o Space between "Ref." and "[12]", i.e., Ref.~\ref{whatever} (with
the ~ to prevent line breaks across.
ok
o "resulted in selection of" "resulted in the selection of"
ok
o "D* candidates were searched selection and additional pion with the
charge opposite..." "D* candidates were selected by requiring an
additional pion with pT > 0.18 GeV/c and charge opposite to the
charge
of the muon".
ok
o "Mass difference \Delta M" "The mass difference \Delta M"
ok
o Space between "Fig." and "1", i.e., Fig.~\ref{whatever}.
ok
* Need to be more explicit with "sum of the exponential", i.e.,
it is more like p0/(p1 - p2*exp(p3*x)), right?
i think this does not add extra information since the numbers are not given anyway
o "corresponds to the wrong combination" "corresponds to the same-sign
combination of the muon and slow
pion charges"
ok
o "in the peak is equal to" "in the peak is determined to be"
ok
o Add "(stat)" after the 280 error.
o "and was defined as number of signal" "and was defined as the number
of signal"
ok
o Show the 0.142 to 0.149 mass window on Fig. 1 as vertical lines
at these values. Always nice in a paper.
ok, done
o "for normalization B candidates" "for normalization, B candidates"
ok
o Again, check that "B" italicized.
ok, done
o "had at least two" "had to have at least two"
ok
o "defined as distance" "defined as the distance"
ok
* Describe how the primary vertex found (or reference).
referenced
o "had to be better than 0.5 mm" "had to be smaller than 0.5 mm"
ok
o "and chi-squared" "and the chi-squared"
ok
o Clean up Fig. 2 for publication quality. See comments on
Fig. 1 above. More complete caption (should stand alone from the text).
figure updated
o You have a cut on significance, but don't define it anywhere.
ok, significance is defined now
* Your results are for B -> whatever, where B can be any of the B
mesons, including B_s. We don't distiguish the pion from kaon
(i.e., expected from B_s decay). Is there any concern at all that
"your B" is a different admixture of B mesons than other analyses
that for example may do pion identification? You cut on decay length
greater than 0.25 mm. This will also change the mixture of
mesons. How stable is your result to varying this 0.25 mm cut
by a few resolution multiples?
our result should be compared to other LEP and Tevatron measurements. we use for the normalization the Br b->D*lnu from the appropriate PDG section. the evolution of the sample composition will happen because of the difference in the lifetime of various B (Bu, Bd, Bs) but it will be small since the lifetimes are similar. Bs contribution to the sample composition is very small (~2%) so the confusion of pion and kaon should not be an issue.
added a phrase that we neglected the Bs and Lb contributions.
the most dramatic effect potentially could be from the ccbar contamination.
we checked that the efficiency of the cut (VPDL>0.25 mm) && (significance of DL > 3) to ccbar is 6%. so ccbar is indeed very suppressed by this cuts and can be neglected in the sample composition. before the cuts it has been estimated at 10+-7%.
we also redid the analysis for several representative values of the (VPDL>0.25 mm) cut and checked the variation of the total Br and ratio :
VPDL cut, mm total Br, % r
0.0 0.178 0.35
0.15 0.181 0.36
0.25 0.179 0.39
0.50 0.171 0.38
we conclude that the results are stable wrt to this cut
o* "The last selection removes the cc(bar) contamination to the " "The
last criteria reduces the cc(bar) contamination..."
Surely you can't remove all of it.
agree, see above.
o "After these selections the total number" "After these selections,
the total number"
ok
o Include "(stat)" after the 230 error.
ok
o Would it make more sense to show the D* peak at this point rather
than earlier? Sure the 200k D*'s of Fig. 1 are more impressive, but
a plot after the cuts would be closer to the analysis.
there were 55k D* in the peak. i don't have strong opinion which one to show. will ask EB.
o "The D* candidates were selected" - repeats selection from earlier,
remove.
ok
o ", at least two SMT" ", and at least two SMT"
ok
* Instead of the more complicated impact parameter w.r.t. the SV and PV
requirements, would it be cleaner, make more sense to simply vertex
the pi** with the D* or B?
the ratio of IP significances is less sensitive to the uncertainty of the tracking errors than vertexing.
* Would a vertex constraint in this case clean up the mass peak?
tryed at some point without success. the natural width of the resonances is large so the resolution does not play a dominant role in the width of the peaks.
o You refer to the D* vertex and the B vertex. The naive reader may
not realize that the D* decays strongly and essentially immediately,
i.e., they are the same vertex. Possibly clarify this.
ok, "D* vertex" replaced by "B vertex"
* How stable is the analysis (in numerical results) to varying these
impact parameter requirements by a bit?
to do
o "Sometime tracks from the primary vertex can point to the SV and,
thus, pass the" "Tracks from the primary vertex can sometimes point to
the SV and
pass the ...."
ok
o "the IP significance for the primary vertex be more than 1" "the IP
significance with respect to the primary vertex be
greater than 1"
ok
o "and applying the D* mass constraint" - this mass constraint is
not described yet?
mass constraint here means adding the D* mass from PDG to the measured D**-D* mass difference.
text modified
o "Contributions of D1 and D2* are shown also separately"
"Contributions of D1 and D2* to teh fit are also shown separately"
ok
o Space between "Eq." and "1".
o In Eq. 1, define x as the invariant mass.
ok
o "In the formulas above" "In the formulae above, "
ok
o Possibly "Gamma_i" instead of "Gamma_0"?
ok, better indeed
o "L=2" "$L=2""
ok
* Two Gaussian functions with the parameterization determined from MC
simulations. You have to give the parameters here to give the reader
a sense for our mass difference resolution. Have to justify where
you get the 20% scale factor. Later when you change the resolution by
±20%, is this enough? There must be a narrow and a wide
Gaussian.
What about varying the ratio of the two? Surely having more of a
wide component (very likely in the data...) has a large impact.
Now for the publication, it would be much more satisfying if a data
source for this resolution could be used (or at least a more
explicit check of the MC prediction).
the 20% difference in the mass resolution between data and MC is justified by the corresponding difference for the D* peak between data and MC - a sentence has been added to the text.
the scaling factor 1.2 was varied +-20%. the justification of this was that we know that we got it wrong by 20% in one direction so we assume the same in the opposite direction.
we checked how the result is sensitive to the contribution of the wide gaussian (the second gaussian is wider by 2.4 contributing 28%). the fraction and the width of the second gaussian were varied by 20%. the parameteres for D1 and D*2 were varied simultaneously
fraction, % sigma scale factor N(D**) N(D*2) / N(D1)
28 2.4 644 0.377
40 2.4 650 0.353
16 2.4 635 0.400
28 1.9 636 0.394
28 2.9 651 0.361
we think this should be considered as a double check rather than a separate systematic error.
this topic has been discussed also with EB, see
http://d0server1.fnal.gov/users/nomerot/Run2A/Dstst/prl_comments.html
o "'wrong sign'" "`wrong sign'" (to get single quote going the correct
way).
ok
o "but the overall normalization of the background was left
floating in the fit" ", fixing the function shape parameters, and
allowing the overall normalization of the background to float in the
fit."
ok
o "from PDG" "from the Particle Data Group"
ok
* You quote the 2002 PDG. You have to update this to the 2004
reference. Are the values for these parameters the same in the
two versions? You really should quote directly the values you
fix to along with their errors.
ok, values specified in the text now
o Include "(stat)" after the 38 uncertainty. State it explicitly as
N_D** = 644 ± 38.
ok
o "as an integral" "as the integral" - why use an integral? In
principle if you
have the B-W form, you can integrate ± inf? It will make only
a small difference, but what do you "miss" in event counts by
cutting off like this?
the right tail of BW is rather large so questions may arise what happens with a larger window. we checked that the integration window is large enough so the difference is very small. this has been checked increasing the window.
4/22/05 this sentence has been removed
* "number events in two narrow peaks separately" "has been modified to
have amplitudes" - this text
is very confusing. You have equation (1). You fit for
N_D** and r. The number in each peak is then derived from these. To get
N_D**, you _have_ to let r float to get a good fit, this is not a fixed
parameter. So what are you
modifying? Explicitly give r with its error as this is the
fit result, the N_D1, etc. are derived. Rewrite this section
accordingly.
revised, see above in responses
to Gennadi
* Put "stat" on the number of D1 and D*2's. These have to
be very heavily correlated. Possibly give the correlation
coefficient?
errors on D1 and D*2 were determined by Minuit which presumably takes the correlation into account.
* For Br(B -> D* mu nu X), either here or near beginning, perhaps
clarify that "B" is the B0/B+/B_s/b baryon admixture.
When checking for this Br in the PDG, the B0 -> D* mu nu
is known quite bit better. While on this topic, Br(B -> D* ell nu X)
= (2.75 ± 0.19) %, while
Br(B0_d -> D* ell nu X) = (5.44
± 0.23) %.
Uh, why such a huge difference? Are these defined differently?
Is the first assuming the production fractions of all the states?
Surely after your selection you should be very much dominated by
B0_d. I may be missing something here,
but I am confused.
Br(B->D* l nu X)=2.75% means in fact Br(b -> D* l nu X)=2.75% so indeed it includes the production fraction (b->B). perhaps the best is to include Br(b->B) to the results as a multiplier. modified the draft - will need to discuss with EB/group if this is an adequate interpretaion.
* How good of an assumption is it that the single pion modes
saturate the D** decays?
answer to the same question from
http://d0server1.fnal.gov/users/nomerot/Run2A/Dstst/prl_comments.html
is pasted below
- D** single- double pi decay modes. a saturation by a single pi decay mode was assumed.
Andrei will provide more justification for this.
double pion mode is for ex. narrow D**-> D* pi pi. As we don't reconstruct it the Br
we measure could be underestimated. however it is the usual practice to assume this decay rate is zero.
except from the lifetime ratio note 4280 describing the argument follows.
as one can see the DELPHI measurement is compatible with 0 (line 4 below)
with more data we should explore this mode and try to do the measurement ourselves.
o Add before equation 2: "We use the formulae:"
ok
* Use "r" in this equation instead of N_D*2, etc.?
<>
variables (Ntot,r) and (N1,N2) are equivalent, see also above
o Comma after Equation 2.
ok
o The line "D** stands for ... " is confusing. I see what you
are saying, i.e., it can mean one or the other , and in another
context, can be both.
will rephrase
o "47.2 ± 1.0%" "(47.2 ± 1.0)%" Include "stat only"?
ok.
* Is there any reason why these two efficiencies would be different?
With masses so close together, do kinematics change
enough to see any visible difference? Just average?
as two species of D** have different quantum numbers (and hence difference angular distribution) there may be differences in eff, see for ex. slow pion effect on the helicity distribution on the top of the page.
* What value do you use for eps_D** when you mean it as "both"?
weighted average
o Move references to MC earlier, to when MC first mentioned.
Only a quote for PYTHIA? How about one for evtgen? It is separate from
PYTHIA. Get name for evtgen correct (capitalization).
done
o Add colon after "were determined to be equal to"
ok
o In the results, the syst. error 0.019 is different from
abstract.
thank you, fixed
o There is an extra bracket in the numerator and denominator
of the ratio of Br's.
ok
* What about "X" in "B->D* mu nu D** X"? The R defines it
without any extra pions (and I guess kaons if B_s?).
the theoretical prediction is done for the ratio of Br(B -> D** mu nu) so when calculating R we assume no extra pions as stated in the paper by explicit discussion of the decay modes. all other Br are what they are: B -> D** mu nu X. the text was modified.
o "The contribution from Br(B->...) "The contribution due to
uncertainties on ..."
ok
o No space before \pm in front of 20%.
ok
** Again, I have some concern whether the ±20% on the
narrow and wide Gaussians is indeed enough. If the fraction
of the wide Gaussian were instead varied within its uncertainty,
it could have a larger effect.
answered above
* The line "If the width of D*2 was left free in the fit then.."
comes out of the blue. So? The fixed number used is not
given so hard to compare. Even so, is this a consistency test?
Used for a systematic error estimate?
yes, this was a consistency check, will need to rephrase indeed
4/22/05 sentence removed
* Describe the "mismatch of kinematic variables" in a bit
more detail and what you did.
modified
o "from this resonance was evaluated by fitting" "from this resonance
was evaluated and included as a systematic uncertainty by..."
ok
o Comma after "was not included in the fit"
ok
o "by the estimate of the wide resonance systematic error" "by the
estimated systematic error due to the possible
presence of a wide resonance."
ok
o Spelling of "intereference"
ok
** I may have missed an EB meeting or summary: it is still
my understanding that even if the acceptances are identical, there will
not be a complete cancellation of the interference effect unless you in
fact
do integrate over all angles. From the cuts, we know that
we are not integrating over all angles. Isn't there still a systematic
due to this? Even so, your efficiencies for
eps_D01, eps_D*2 are not equal (but consistent). You only
know that the acceptances are equal to some level. Impact
on possible interference effects?
see above in the responses to Gennadi
* You vary the functions describing the D* mass peak and
backgrounds. Do you mean varying the functions for the
D** background shape? Do you change the background shape
parameters within their errors?
variation of the bkg within
errors should be accounted in the statistical error on the signal
(coming from from the fit) since the bkg function is used in the fit.
to estimate the D** bkg systematic errors we tried a line and 2nd order
polynomial for the D** bkg so indeed we varied the shape.
* What about a systematic due to the remaining cc(bar) contamination?
ccbar contamination is small
after final cuts, see above
Table I - use "--" instead of "na"
ok
- "D* br" actually also includes
Br(B -> D* mu nu), right? Clarify source?
yes, 7% uncertainty is mostly due to Br( B->D*munu).
modified : error due to
D*->D0 pi is a separate line now .
* Instead of
saying "latter decays into Dpi in (71+/-7)%
of the cases" add what value you then get for Br(D*2 -> D* pi) as an
intermediate step.
ok, modified
o "1.8 standard deviations away" "There is a difference of
1.8 sigma
between this R value and the
previous determination of R" (since you use both errors; "away"
implies only one set of errors).
ok
References:
o are a mess. Use the accepted formats consistently. Update the
PDG to 2004., write out journals, use the correct order of page, date,
etc., make sure all the citations are actually used in the text (where
did you cite using d0mess?) Do you cite muon identification?
Do you refence [16]? Correct the order to match that of the text.
Give first author and then the collaboration name consistently.
Consistent on "et al." being italicized. Commas before hep- preprint
numbers. If published, don't need hep- numbers.
agree. hopefully much improved now
from Mike Strauss
First sentence: B should be italics.
ok
First sentence on page 4. Add the word "the". "...j is the
sum of the light quark spin..."
ok
You need a reference to HQET in the first paragraph on page 4.
ok
Page 4 second sentence, second paragraph. B should be italics.
Throughout the paper, make sure all references to B mesons
have an italics B. There are many places this needs to be
fixed.
ok
Page 4, third paragraph. There should be a colon, not a period.
"... is comprised of the following main elements: a magnetic..."
this paragraph is removed now
Page 4, fourth paragraph. Change the word order. "...followed
by two similar layers after the toroids..."
ok
Page 4, fourth paragraph. And the word "the" "... is defined
using the polar angle..."
ok
Page 4, fifth paragraph. "reconstructed" is misspelled.
ok
Page 4, fifth paragraph. In this paragraph and throughout the
paper there is too much space after the D**, D*, and D0.
Going from math mode to regular mode, you have too much space
in the text.
ok
Page 4, sixth paragraph. No space after D0.
ok
Page 4 seventh paragraph. Add some words. "D* candidates were
searched for by selecting an ..."
ok, modified
Page 4 seventh paragraph. Add the word "The" "The mass difference..."
ok
Page 4 seventh paragraph. I don't see a "hatched" histogram in
the figure. Maybe it is just my pdf copy of the paper.
figure updated
Page 4 seventh paragraph. You use the terminology "slow pion" but
have never defined what pion that is.
removed 'slow' - don't need it
Page 4, eighth paragraph. Add a comma. "..used later for normalization,
B candidates..."
ok
Page 4 eighth paragraph. Add the word "the" before the chi2.
ok
Page 5 second paragraph. Switch the word order "decay can also be
selected..."
ok, rephrased
Page 5 second paragraph. Question: Selecting an IP signficance
of 1 won't really remove the tracks, it will only remove some of
them. SHould you change the last sentence to "Most of these tracks
were removed..."?
agree, thank you
Page 5 third paragraph. You need space after the D* in the text.
ok
Page 5, fifth paragraph. change "...k, k_0..." to "...k and k_0..."
and add a comma before the word "respectively".
ok
Page 5, fifth paragraph. You use the abbreviation MC without
ever stating that it means Monte Carlo.
MC defined at the first occurance
Page 5, fifth paragraph. Add the word "the" before the last "MC".
ok
Page 5, sixth paragraph. First sentence needs to be rewritten.
Define 'Wrong sign" Don't say "... but the overall...". Instead
say somethine like "while allowing the overall normalization of the
background to float."
rephrased
Page 6, first sentence. How can they be found to be equal if one
is 467 and one is 176?
there was a word "correspondingly" there.. rephrased
Page 6 third paragraph. The first sentence doesn't make sense.
It should not say "...D**, D*...", but rather "...D** and D*...",
but it still doesn't make sense.
typo fixed, rephrased
Page 6 third paragraph. Again, don't say "D10, D2*0", but say
"D10 or D2*0"
ok
Page 6 fifth paragraph. The sentence about allowing the D* width
to float doesn't fit in the paragraph. Tell me how this affect the
systematic error, not just what the new value is.
indeed. this was a systematic check, rephrased
Page 6 fifth paragraph. I don't understand the last sentence at all.
sentence modified
Page 7 first paragraph. Any motivation for assuming the errors
are not correlated, or is it just easier? I think this needs more
explanation.
taking into account possible correlations is a considerable work. we looked up the published results for the B -> D** semileptonic modes by CLEO, OPAL, DELPHI and ALEPH and everywhere the systematics errors were added in quadrature.
Page 7, second paragraph. Is R "determined" or "calculated"? I think
calculated is a better word.
agree
from Brad Abbott
In the abstract, you don't mention the value R. However in the
text you mention R is a mostly model independent measure and you
conclude with this number. Is there a reason it is not mentioned
in the abstract?
the motivation was : the ratio of Br given in the abstract is our measurement with its errors. R is more relevant for the theory and is derived from the ratio of Br.
Typo in abstract: ... with the D0 detector, the (missing comma)
ok
First paragraph:
In the limit of large charm quark mass -> limit of a large charm quark
mass
ok
.. j is sum of the -> j is the sum
ok
.. orbital momentum L -> orbital angular momentum L
ok
.. are expected to be wide -> are expected to be wide and not measureable
at D0. (It would be nice to add a little more here about what "wide"
means as far as the analysis)
the sentence rephrased. the wide states could be measurable at D0 with more statistics.
Ratio of semileptonic -> The ratio (R) of the semileptonic ...
ok
second paragraph:
result of averaging over all measurements. -> all previous measurements.
(I assume the new D0 measurement is not in the average)
ok
I would like to see a little bit more motivation as to why this should
be a PRL. Now it only states that R does not have a lot of model
dependence. Also since the current value of R is 0.37 +- 0.14 and
our number has larger errors, what does our number contribute to
our understanding of D**
good point. more motivation added in the introduction
Figure 1 needs some work.
There are no axis labels. No D0 Run II. Should use a solid and
dashed line instead of different colors etc.
Explain what different curves are on the plot/caption
ok, figure updated
D* candidates were searched selecting an additional...
(this sentence needs some work)
sentence rephrased
To select a sample of B-> mu nu D* decays used later for normalizing
B candidates..
(This sentence also needs some english work)
sentence rephrased
and chisquare of the B vertex fit had to be less than 25.
-> and the chisquare of the...
ok
Figure 2. Also caption labels, no color curves, explain curves
D0 RunII etc.
figure updated
Page 5: top left
The last selection removes the ccbar contamination..
Does it remove all of it? Should this sentence be
removes most of the ccbar contamination..
or
removes all of the ccbar contamination ..
see same comment from Rick. "removes" -> "reduces"
Page 5: top right
D** decay can be also selected -> can also be selected
ok
Page 5: middle right
Contributions of D_1 and D_2* are shown also separately.
Since you have been using
the notation D_10 and D_2*0, you should use this notation here.
ok
Page 5: bottom left
R is the mass resolution function.
You have already defined R. I believe this is a different definition
of R so you should use a different symbol.
thank you. R -> Res
Reference 17 ??
references fully revised