1Fermer3
Lionel DebrouxLe 18/09/2008 à 10:51
There's no question that the fork will be opened smile
On the one hand, I'm unconvinced there's much case for posting the URLs of the fork (SVN, Trac, ML, forum) public, as long as we have only changes that are invisible to a wide majority of users, which is the current state of the fork.
On the other hand, as much as I hate us having to give in to Kevin's pressure, I think we'd better make the fork public before he closes the TIGCC CVS repository. That way, it will be harder for him to make us look bad for not working openly for a while (even if we have valid reasons not to "commit early, release often", IMO).
Note that if we want people to actually use the fork, there's no way we can ask them to compile the fork themselves: we'd have to spend the time to make a release (recompiling the binaries that need to be recompiled, etc.). It probably wouldn't be THAT long, and it would be less of a problem for us than it seems to be for Kevin, because we have more manpower, more Windows and Linux versions to compile and test on, though...

What do others think ?

If Kevin sabotages our infrastructure, he should indeed be banned. With no round of "next time you behave such a way, you'll be banned", because he knows it very well that that this kind of behaviour is unacceptable.
Wikis and SCM can be reverted to previous versions, so if he was stupid enough to wreck havoc, things could be straightened out quickly. But I think we'd be pretty foolish to give him write access to the SVN, after he threatened in topics/113677-kdewin-foutage-de-geule of introducing subtle bugs in upstream TIGCC so that the fork (which is sooooo eager to take up his own commits while not giving anything back, as he has written in topics/113677-kdewin-foutage-de-geule and topics/2-114905-tigcc-contributions-fork ) has bugs...

That said:
* Kevin talks much more than he actually acts (remember him threatening multiple times of poisoning in TIGCC's GCC the identifiers used by ExtGraph itself or its modified grayscale routine ?);
* I think he's too clever to sabotage actively GCC4TI: that would make HIM look bad, with no one (of good faith) on his side to excuse such an unacceptable behaviour.