From: Jonathan A. <jon...@gm...> - 2009-11-30 12:38:40
|
I am going to keep meddling in this discussion (let me again say that I have not been personally involved in the JChempaint development and I can only give the view from the outside) and offer my suggestion of the source of the trouble. Could it be that this is the situation: When the Uppsala team ran into trouble with keeping track of all branches and work that was taking place in different places they introduced Git as a new solution for better handling the branches. But the EBI developers did not get sufficiently informed about why this was needed. For them things were working fine and Git was in fact not a solution to a branch management problem but instead one hell of a road bump. What for the Uppsala team was a salvation the EBI team saw as a strange and complicated set of new rituals hindering them from doing their job. Could it have happened somewhat like that? After having learned to work with Git myself I can easily see how this is a possible situation. Git is a powerful tool but it is actually not very much like SVN at all and learning to use it and taking the step from SVN to Git can be very painful. (Ask me, I know all about the pain of learning new tools) So could the source of this fork be about the step from SVN to Git? If that is the case then maybe we could come up with a solution where the EBI people keep using SVN and the Uppsala people uses Git. This is entirely possible in my eyes. I feel that it would be sad if this joint venture failed because it can not decide which version control system to use... On the other hand there might be more profound differences between the now two separate JChempaint projects than only the version control system they are using. Is instead this the case? -- // Jonathan |