Matt Zimmerman wrote: > I didn't say "always", but so far we have done this with every package > modified by Ubuntu. However, the situation with X.org seems quite different > to me, given your explanation that the packages were created independently. My x.org example is somewhat theoretical, FWIW. > As I explained in the message to which you replied, it depends on the > details of the situation, and both outcomes are possible. If the new > packages were to meet Ubuntu's needs as-is, or with little modification, > then the benefits of sharing the code would likely outweigh the cost of > making the switch. If they don't, then this might not be the case. There certianly seems to be the potential here for Ubuntu to leapfrog Debian and then for Debian to catch up and go off in a non-Ubuntu compatible direction, and for Ubuntu to decide it's not worthwhile for it to reconverge. At that point, as I said, we have a fork, in all the bad senses of the word. I suppose we will just have to wait and see. FWIW, I'd still appreciate an answer to the following question from someone involved in Ubuntu: | Let's assume that starting tomorrow, every patch that is applied to | Ubuntu, that is not Ubuntu-specific, and not something too trivial to | worry about (such as the python or c++ transitions), is accompnied by a | handwritten email to the Debian BTS containing the patch and an | explanation. Writing such an email would surely only take a few minutes; | how many new such patches do you really create each day? How could this | not be feasable? Wouldn't this tend to highlight cases where Ubuntu is | making many useful changes to Debian and tend to get the Ubuntu people | direct commit access to those packages? People involved in Ubuntu seem to have been going out of their way to do this for packages I maintain in the past few days, which I appreciate, even if it's a cooincidence, but I still don't understand why it cannot be done in general. -- see shy jo
Attachment:
signature.asc
Description: Digital signature