[Twisted-Python] major changes, release engineering, and learning cost

[Glyph Lefkowitz]

On May 27, 2010, at 4:27 PM, Laurens Van Houtven wrote:

> I have diagrammed the quantum-transmogrifier example that I tried to
> explain in the last email.

OK.  With this diagram in mind, I can see that what you're proposing is nearly identical to what I've already proposed, except that you are being very vague as to the *requirements* on when and whether branches get merged.  I'm not concerned with the "level of confidence" that you describe (we already have that), but with a modification to the *requirement* that branches be fully reviewed before a merge to trunk, and that the reviewer can block that merge.

What I've suggested, simply put, is that we can have branches that land on trunk without being fully code-reviewed, *provided that each commit to that branch was itself code-reviewed*.  There are a few fiddly details beyond that, but we seem to be in agreement on that broad picture.

So, sorry for a terse response to a message that obviously took a long time to write, but I don't think this merits further discussion :).

-glyph

-------------- next part --------------
An HTML attachment was scrubbed...
URL: </pipermail/twisted-python/attachments/20100527/09135128/attachment.html>