[Python-Dev] Proposed schedule for Python 3.4

[Stephen J. Turnbull]

martin at v.loewis.de writes:

 > I wouldn't mind having alpha 1 in April 2013, and alpha 2 in October 2013.
 > I share Larry's skepticism, and actually fear that it may confuse users
 > (which find that they test something completely different from what gets
 > released).

I don't really think you need to worry.  The Mallory-style testers[1]
who download early alphas generally don't have specific expectations
that it will correspond to the eventual release in my experience.

I have little experience with testers who download alphas to get a
head start on integration work, but I suppose Larry is right, you'll
lose a few of those if an alpha is "too early".  I suspect that those
folks mostly listen to an internal clock, not so much to release
announcements, and download the freshest release when they are ready
to start that kind of work.  Those who decide *not* to download alpha
1 because "it's too old/incomplete, I'll wait for alpha 2" will be a
subset, I suspect a minority, of those ready to get an alpha for
integration testing at any given time.  OTOH, is it really that big a
loss?  If I'm right, they *will* get alpha 2.

In fact, you might find a larger than usual response from those folks
for alpha 1 because Nick will be out advertising "these changes are
big, guys, both you and we want testing early!"  Except that unlike
most advertising, Nick's pretty plausible.<wink/>  At least, that's
Nick's main point for doing this experiment *now* as I understand it.
(I gather he agrees with David about a trend to "continuous alpha",
but that's a separate though related issue.)

Footnotes: 
[1]  "Why did you download the alpha?"  "Because it's there."