Date: Fri, 20 Jun 2008 00:37:54 +0200 (CEST) From: Krzysztof Kurek To: Marcin STOLARSKI Cc: Eva-Maria Kabuss , Gerhard Mallot , Krzysztof Kurek Subject: Re: questions/comments to my talk in US Dear Colleagues, In the light of comments obtained by Marcin it is realy important - in my mind - to be able to answer the questions (for future speakers). 1. Stratmann remarks: intrinsic charm - question of taste but according to my knowledge our x_g range is safe. To test it we can use CTEQ parametrization where intrinsic charm is taken into account and do extra systematics for chamr analysis. NLO are probably important but the result extracted from our data are kind of asymmetry which is independent on approximation. This is why I was protesting against to use word LO in the paper if not needed. The interpretation is indeed in LO but there is nothing wrong in the data analysis and it would be nice to say it to the community (and our potential speakers). second: resolved photon was checked and is zero. third: high-pt. We use LO based model and the tunning is related to fragmentation. Potential NLO is tested via PS on/off and data are in good agreement with MC. The model is not LO QCD only - model contains structure function, LO hard matrix element and fragmentation function with the assumption of the factorization. All these ingradients are our model As far as data/Mc agreed we can extract DG/G from this model. Fullstop. 2. Elke. Arguments about LO - see above. tunning is related to the fragmentation and has nothing to do with gluons. Moreover we are using standard Lepto version (no tunning) and the effect is taken into account in the systematic error. This is why we have the most important part of our systematic errors from MC. If she is not understand - then there is her problem but we cannot allow her to say stupidities publicly in every possible places. We should answer her becuase she is well known and people listen to her. By the way - The HERMES high-pt new analysis - again according my understanding - is a patchwork of LO and NLO K-factors cooked together with the way which even author of this analysis (Particia, PhD) was not able to defence and answer the questions. This is not the argument but it means that HERMES was able to convince the community that their analysis is in NLO and is correct. It is not the case but it means that HERMES propaganda is perfect. My general remark is - start to defence our analysis! No questions without answer. It cannot be that one theorist and Elke are attacking us every time without reaction. These questions and objections were posed many times. All of them were answered and they are comming back like boumerang. Please react every time when needed. This also means that senior physicist from COMPASS present on the conferences should help the speaker. Also it would be good if our spokepersons will sometime react. Best regards Krzysztof > > Dear Eva, > > Concerning my talk there where fallowing comments: > > > W. Vogelsang, > > asked about why for D0/D* scale of the process > was ~13 GeV^2. He is also worried that our > changes sign that this maybe a complication > as they expected that DG/G has a node X_g~0.1. > > I said that it doesn't matter we 'know' > on the event by event basis. > > > M. Stratmann, > > He made comment that open charm is a bad channel > to study PGF. a) because of intrinsic charm > b) NLO correction are very large (I guess for ) > c) resolved photon contribution is large. > In the same direction he doesn't like 2h-analysis > from high_pt since proper NLO correction do not exist. > He would like to see 1h analysis > (a'la RHIC, where all theoretical ingredients in NLO exists. ) > > Elke, > > As usually what we are doing is all wrong. It makes > no sense at all. Hermes did it correctly. > And COMPASS is ignorant collaboration which doesn't > listen peoples who has right. > > I had impression that she agrees with M. Stratmann why > D0 is wrong. Concerning hi-pt in she doesn't like > that it is analysis in LO + that we tune MC in the > region which we are using later for DG/G extraction. > > (I could only say that we are doing this test PS ON/OFF, > COMPASS/STD tuning, so largely this is included in > the systematic but she was not convenient.) > > She was about to start another stuff but chairman of the > session got a bit pissed off and said that we should cut > the discussion. > > BTW. Before and after talk, privately, we had nice conversations. > > > ----- > > Unofficially > > B.Surrow asked when we will have A_1_incl for proton. > He needs that for Higher Twist analysis. > (I said that rather next year) > > There was another guy, young theorist, unfortunately I have > forgotten his name, who would like to measure gluon Sivers > function (he expect effect ~10%). > This would be done by measuring D0(*) transverse asymmetries. > I have tried to convinced him that with our limited D0 statistic > in transversity we cannot achieve such a precision (However, I don't know > how looks like for this process) > > > > I think that is moreless all > > Best Regards > Marcin