history of CADT

Posted May 1, 2016 19:19 UTC (Sun) by hitmark (guest, #34609)
In reply to: history of CADT by louie
Parent article: Devuan Jessie beta released

Not so much destroyed, as it being about rather than fixing reported bugs they were marked as invalid because the relevant component had been rewritten or replaced.

In essence, the originator of the term, JWZ, seems to hold the opinion that it is better to maintain existing code rather than rewrite from scratch at (ir)regular intervals.

to post comments

history of CADT

Posted May 2, 2016 16:31 UTC (Mon) by louie (guest, #3285) [Link]

Sometimes that's appropriate, sometimes it isn't. And sometimes the rewrite is such that previous tests, bug reports, etc., have a very low probability of being relevant. (In the case of GNOME 2, in most cases, it was both a significant tech rewrite and a more-or-less complete UX rewrite, so this was doubly true.)

So when developers have very little bandwidth (because they are typically volunteers), they can stop development for a year or two to review tens of thousands of low-value bugs, or they can ask tens of thousands of people to spend a very small amount of time to help out. If we really are community-driven development, the latter is a no-brainer.