Hi, > If there is a real interest to use these classes outside the ALICE > context, we can discuss how to make them more "neutral". We are not > opposed, in principle, to an evolution in this direction. BTW, we are also > planning a TFLUKA class which allow the use of the FLUKA MonteCarlo > (http://fluka.web.cern.ch/fluka/) from the Virtual MonteCarlo classes, Is it only my private opinion that this would be VERY WELCOME ??? I think that an experiment independent TVirtualMC, TGeant3, TGeant4 and TFluka classes for ROOT would be VERY WELCOME. A good basis for them could be the current AliSoft. And if not ... then what ??? What other possibilities do we have ??? Would it be possible to get a common agreement inside of the ROOT community on how to proceed ??? > TGeantX. True these are still rather ALICE dependent, but with some > surgery they can be used in other contexts. The problem is that these classes are regularly modified/maintained by developers. Once one does the "surgery" oneself, one will be lost - no bug fixes, no new features, ... . This would make the problem that on a regular basis one would need to compare the "current" AliSoft with the one that was used for surgery. That's why I still hesitate to start it ... A much better idea would be to ask developers of these classes to make them experiment independent, so that also these "neutral" versions would be used inside of the AliSoft and would be maintained by them. On this occasion one should also mention some other parts of AliSoft which could be extracted into an "experiment independent" ROOT add-ons : MINICERNLIB (needed also by TGeant3, libHbook, ...) GEANT321 (needed also by TGeant3) ISAJET PDF PYTHIA PYTHIA6 As far as I understand it, all these packages are built as shared libraries by AliSoft. I think that making from these shared libraries some independent ROOT modules would again be VERY WELCOME. Best regards, Jacek.
This archive was generated by hypermail 2b29 : Sat Jan 04 2003 - 23:50:46 MET