Front page | perl.perl5.porters |
Postings from February 2012
* Tom Christiansen <tchrist@perl.com> [2012-02-26 18:15]: > Aristotle Pagaltzis <pagaltzis@gmx.de> wrote > on Sun, 26 Feb 2012 06:32:41 +0100: > > > 5.16 *must* be released whatever the state of this issue. > > Apparently I wasn't paying attention when it was Decided to stop using > releases as actual milestones, and start using them to rubberstamp > some meaningless periodic snapshot of the development cycle. Good > thing you got rid of the idea of release blockers, too: wouldn't want > to risk anything interfering with Real™ Progress®, now would we? Sorry I disagreed with your assessment of the situation. I actually concur that that complex of issues you brought up is gravely important and in need of fixing. I disagree *only* about whether it needs to hold up 5.16: it is not a regression from any previous perl. 5.12 shipped with the excision of UTF8-flag-dependent Unicode semantics half finished, and left the rest for 5.14 to clean up. That was not the end of the world then either. Because 5.14 is *here*! It’s not a snapshot of the entire development cycle that gets shipped either. It’s a snapshot of the *stabilised* (and landed) features from this development cycle. That is the primary instrument for not shipping something with blocking bugs while at the same time not having them delay it until their resolution. And patting down the projected release product for blockers begins long before the ship date. That leaves an infinitesimal chance for the release to be delayed, which ensures that the work of porters and contributors can reach the hands of users in a timely manner, instead of getting mired in a tar pit for years. Do you want to go back to how things used to be? Your characterisation of what was “Decided” assumes the most malicious interpretation possible. Did you really think what you wrote is how it works? Your reactive sarcasm does not become you. Regards, -- Aristotle Pagaltzis // <http://plasmasturm.org/>Thread Previous | Thread Next