Dear Editor, we are really shocked by the report on the paper JHEP/019A/0401, which is extremely aggressive and denotes a personal adversion of the referee either against the subject of this work or (and) against the authors. Furhermore, from the large number of superficial statements and mistakes of the report, we think that the referee did not read the paper carefully and is not competent on many of the aspects of our study : all the objections raised by him/her turn out to be either wrong or generic statements about the Bayesian approach that do not have any connection with the present paper. In detail: - For Lattice QCD, in our Collaboration there are several experts in the field. We wrote many pages explaining in all detail how the final numbers were obtained. These numbers are in very good agreement with estimates made by other world experts such as Claude Bernard who gave a plenary talk at Lattice 2000 on $B$ physics. The results include unquenched calculations of the most important quantities. Only for $\hat B_{B}$ we have used estimates based on chiral perturbation theory (and not on the quark model as uncorrectly stated by the referee), which is a well established and theoretically founded method. Moreover the referee ignores that NRQCD has its specific systematic errors, which are not common to other lattice calculations. Thus we think that the discrepancy between the two results for the $B$ meson decay constant is a problem of NRQCD and not of the lattice approach. In this respect, it would be very instructive for the referee to read the recent literature on effective theories and power corrections before making meaningless statements. In general, the comments from the refere show that she/he has not followed the recent developments in Lattice QCD and uses five years old arguments. Furthermore, as explained in the paper, the parameter which the referee specifically addresses (f_B sqrt(B)), has a very little impact in the final result. We suggest to the referee to read more carefully our paper, besides the recent literature on the subject. - Vcb: We discussed several times with collegues who prefer to quote only results obtained with exclusive decays, Bd --> D*+ l- nu, on the basis that the value obtained from inclusive measurements, b--> l X , relies on the validity of parton/hadron duality. It is useful to recall to the referee that the analysis of exclusive decays is made in the framework of the HQET and that the size of the theoretical uncertainties with this approach has been the subject of very hot controversies in the recent past (see also our comment above on effective theories). Since the referee enjoys to quote names of famous and respectedd physicists, he could have also mentioned that I. Bigi, who was rapporteur on Heavy Quarks at the Osaka 2000 HEP conference, thinks that inclusive decays rather the exclusive should be used, that is the opposite of the referee opinion. - Vub : We have used the information coming from two channels: one being the exclusive measurement by CLEO and the other the result from the combined LEP experiments. There are no common systematics in these measurements and we have averaged them in the proper way. As explained previously, the argument against the validity of parton/hadron duality is not generally accepted, in particular by those theorists who have done the most detailed work on this issue (I. Bigi, N. Uraltsev). The value that we have taken for the inclusive determination of $V_{ub}$ comes from the LEP/SLD/CDF working Group. Hoecker et al. have decided to inflate the error, which is acceptable but certainly not compulsory. Instead, we find asbolutely unacceptable to be criticised as very optimistic by the referee, only because a group of four persons (Hoecker et al.) decided to double an error which was determined by a large community of experts in this field (the LEP/SLD/CDF working Group). In the Hocker et al. paper there is a sentence explaining how they have evaluated the final systematic error when combining exclusive and inclusive measurements of Vub: "The maximum of both single ranges was assigned as the final systematic theoretical uncertainty." This is a possible choice. We have done it differently and the referee should at least present a scientific argument to explain why the "Hoecker choice" is better justified for combining uncorrelated uncertaities. - Delta m(s) : the method used for Dms is that adopted by the LEP/SLD/CDF working Group, composed by all the world experts in the field . The method is on solid mathematical ground ( see H.G. Moser and A. Roussarie, Nucl. Inst. and Meth. A384, 491 (1997) ). The plot is public and everybody can judge if 7ps-1 corresponds to 4 sigmas..... The fluctuation at 17.5ps-1 is used as it should be, as a "fluctuation/signal" at 2.2 sigma. With this method, we are not subjective and we do not decide a priori whether it is a fluctuation or a signal. On the basis of some information which remains secret, the referee seems to know already that it is a fluctuation. We invite her/ him to attend to the next LEP/SLD/CDF Working Group meeting to explain her/his point there. Other methods which have been proposed are not based on solid mathematical grounds, rather they appear to us as ``ad hoc" recepies. Also in this case, in spite of the fact that there is a intense discussion on the subject, the referee knows already all the answers. We invite the referee to read all the concerned papers. The referee is also questionning our normalisation for the likelihood. It has escaped to him/her that, for delta(ms), we use a likelihood ratio and not a simple likelihood to cure this "problem". - Statistical Method : We have clearly and explicitly written that we have done our analysis in the framework of the Bayesian approach. Bayesian approach is rather natural and it has become to be used now in particle physics. Our analysis is absolutely logical and consistent and the referee was not able to disprove this statement. The results represent, in our views, the best way to combine the available information and it is valid under some clearly stated conditions. We have given a comparison between our results and those obtained in a different approach as scanning. This has been done for the first time in a quantitative way. No such a detailed study was done before. All the negative statements of the referee are only based on some phylosophical prejudice about the Bayesian-frequentist querelle, rather than on a careful reading of the paper and of cited works. The only relevant question is whether other approaches give substantially different results. This is what we did explicitly in our paper, in a devoted section, by comparing results obtained in the "BaBar scanning" approach and in our "Bayesian approach". This was done by: - using the same parameters as input for the two methods; - comparing 95% CL regions obtained in the two cases and not +-1sigma obtained with our approach with 95%CL contours obtained with the scanning. The result of the comparison presented in the paper is that the two 95%CL ranges are essentially identical. Thus the comments of the referee are simply WRONG! and not back-up by any study. The same conclusions can be drawn from the resukts of the analysis by Hoecker et al., as compared to our Bayesian analysis, if the comparison is made by using similar values for the input parameters (and errors) and the 95%CL intervals are confronted. The other issue is the choice for central values and uncertainties for the quantities taken from theory. We have discussed already this point for individual quantities. Theoretical parameters have been established for what concerns lattice QCD by several experts in the field. For the other quantities entering in the evaluation of Vub and Vcb we had also contacts with experts. We note that the referee believes in the validity of exclusive measurements to extract Vub and Vcb, other people rely more on inclusive results. We have used the two in a consistent way. We even made a quantitative study which consists in inflating individual and all theoretical errors by a factor two. The results are quoted in the paper where a complete table is presented . Also on this point, we got the impression that the referee has not spent much time in reading the paper. It may well be that this is due to the fact that ``The use of English in this paper is not good". Looking forward for the day when only papers by authors whose mother language is English will be accepted for publication, we apologize with the referee who tried (unsuccesfully) to understand the content of our paper. As final remark, we recall that this work has been presented at several conferences and seminars as invited talk: HEP 2000 in Osaka (August 2000), BEAUTY 2000 in Israel ( Sept. 2000), Workhop on CP Violation in Durham ( Sept 2000) , CP Violation Conference in Ferrara ( Sept. 2000) , Moriond 2001 ( March 2001) , TH-seminara at CERN, at Hambourg-DESY, SLAC, Southampton, Pisa, Parma.. Furthermore many speakers have decided to show and use our results at several conferences. Among these , A. Buras, I. Bigi, A. Ali, J. Ellis.... and in recent past R. Peccei, Quinn...... The paper had already an important impact in the community as demonstrated by the number of citations obtained in a few months, We are very upset for this referee report which we find superficial, confuse, aggressive and offensive. We hope that in the future the Journal will adopt better criteria to select the referees. We also think that the Editorial Board, when receiving such a report, should have reacted by choosing another referee, before sending offending comments on the work of collegues, who are also ``respected" theorists and experimentalists. Since it is cleary impossible to continue any scientific discussion with the present referee, we ask for a new referee, if this is possible. If the Editorial Board decides instead that our paper is rejected, please let us know quickly the answer since the first referee took already a very long time to produce such a careful report. Our Best Regards, The authors