While I find Zed's choice a good one -- GitHub (git) repos feeling more accessible, it pains me to think hg doesn't have as great a domination in this space. I feel Zed has a stronger affinity towards Python and the simplicity / extensions of Mercurial, which comes out in his post; it'd be great to have him more involved in that space.<p>But before we get into some "holy-war discussion" over VCSes, it's clear Zed just wants himself and his committers to get shit done.
The author of fossil (and SQLite), Richard Hipp, has posted an analysis of Zed's problem:<p><a href="http://www.mail-archive.com/fossil-users@lists.fossil-scm.org/msg04699.html" rel="nofollow">http://www.mail-archive.com/fossil-users@lists.fossil-scm.or...</a>
Interesting from another perspective. How do you deal with software that just broke your work? You could just say it's broken and switch. Then find another project and hope that it will behave better. Pretty much an automatic and understandable decision.<p>On the other hand, what proves that some other project is more stable, since those situations are exceptional anyways? So what's the better option - switch as soon as your work is broken, or try to resolve the issue if it can be reproduced and hope that a bug like that will not happen again?