|Problems? Is your data what you think it is?|
Hockey Sticksby raiph (Chaplain)
|on Jan 16, 2012 at 10:48 UTC||Need Help??|
Initially written Jan, 2012. Latest update Nov 2015.
(If you are what some call a "global warming denier", please bear with me. For the purposes of this post, let's imagine that global warming is real, and that the famous hockey stick graphs from a decade ago used good data.)
There is now increasing acceptance of the idea that the Perl Planet is again warming up after a brief cooling period. Perl 6 is a reinforcing factor in this warming, with an accelerating impact...
On #perl6 tonight, we see:
05:57 Coke # 01/16/2012 - niecza at 100.02% 05:57 BOOYAH! 05:57 sorear \o/ 05:58 Coke gist updated, google doc not. zzzzz. 06:04 moritz \o/ 06:05 sorear++ Coke++ colomon++ 06:05 congratulations everybody
What did Coke++, a long time contributor to many Perl things, push to achieve tonight before he succumbed to exhaustion? I've been watching Perl 6 for over a decade, and to me no moment has looked more symbolically significant than this one.
Niecza, a Perl 6 compiler, overtook Rakudo, another Perl 6 compiler, as measured by the number of "spectests" passed. (This does NOT mean that Niecza is better than Rakudo. The 100.02% figure does NOT mean that Niecza is magically doing everything Larry specced, plus reassigning the colon. A comparison of the two compilers' features based on fairly recent data.)
Nov 2015: Niecza is now dead/dormant. Rakudo no longer supports Parrot. Rakudo on MoarVM passes ~48k of ~50k spectests. I'm ignoring the recently introduced 70k or so Unicode tests. As chromatic emphasized in some of his original responses to this post, don't read too much in to these numbers. I think they prove that Perl 6 isn't dead as some believe, but ymmv.
Outsiders might now be able to see that, although the effect is gradual -- so gradual many deny it is even happening -- Perl, and Perl 6, have been developing consistently for years now, and this development is subject to self-reinforcing acceleration, i.e. the Hockey Stick effect.
There are at least TWO Perl 6 Hockey Sticks: Rakudo and Niecza, with corresponding coding heroes jnthn++ and sorear++. (The new audrey++s. Not to forget pmichaud++, but he's being a hero of a different sort at the moment. And not to forget other Perl 6 compilers and their authors such as Perlito and fglock++.) Not only is development of each of these Perl 6 compilers accelerating individually, each is reinforcing the pace of development of the other.
Nov, 2015: jnthn continues to drive Rakudo dev. sorear wrote the plumbing needed on the JVM in mid 2013 that allowed jnthn to start building out the P6 concurrency story on Rakudo/JVM ... and hasn't been seen on #perl6 since. Larry Wall began landing Rakudo commits in 2014 and has continued that through 2015.To get a sense of how the lead developer of a Perl 6 compiler sees things, check out Jonathan Worthington's post yesterday that summarized how 2011 went, and how 2012 should go, from the Rakudo perspective.
But better still, to check out how it feels from the perspective of the rest of us Perl 6 fans, check out Solomon Foster's Perl 6 post from yesterday in which, while describing the 400th Perl 6 Rosetta Code entry, he notes "the awkward situation with Rakudo Star has helped obscure the fantastic good news in Perl 6."I don't want to raise expectations unduly. There is clearly still plenty to do before 6.0.0. But I was a contributor to Mozilla 2002 thru 2004, and I recall how most people had more or less written off the whole project by then. And then came Firefox. So, if you know how to be nice to people, and to butterflies, why not come play hockey on irc chat at #perl6 on freenode?