Perl 6 - the future is here, just unevenly distributed

IRC log for #perl11, 2014-12-05

| Channels | #perl11 index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
01:46 davido__ joined #perl11
04:21 travis-ci RPerl build failed. Will Braswell says 'Grammar, Critic Tests Part 4'
04:21 travis-ci http://travis-ci.org/wbraswell/rperl/builds/43060172 https://github.com/wbraswell/rperl/compare/fad9aa2ca7ee...14e3d697ba23
05:35 davido_ joined #perl11
05:54 willthechill aww snap
06:05 willthechill another case of Perl error messages changing
06:25 travis-ci RPerl build passed. Will Braswell says 'Grammar, Critic Tests Part 5'
06:25 travis-ci http://travis-ci.org/wbraswell/rperl/builds/43064779 https://github.com/wbraswell/rperl/compare/14e3d697ba23...3a15f975ee4a
06:26 willthechill there we go
07:25 travis-ci RPerl build failed. Will Braswell says 'Grammar, Critic Tests Part 6'
07:25 travis-ci http://travis-ci.org/wbraswell/rperl/builds/43067403 https://github.com/wbraswell/rperl/compare/3a15f975ee4a...cf5a3a8b93a4
07:25 willthechill dang it
07:35 willthechill totally different TYPES of errors this time  :P
07:50 travis-ci RPerl build passed. Will Braswell says 'Grammar, Critic Tests Part 7'
07:50 travis-ci http://travis-ci.org/wbraswell/rperl/builds/43068407 https://github.com/wbraswell/rperl/compare/cf5a3a8b93a4...0848634777cf
07:52 Reini joined #perl11
07:52 willthechill okay good
07:53 willthechill Reini: I'm saying good to the Travis build success 2 mins before you joined
07:53 willthechill am writing many hundreds of grammar tests now
08:14 Reini excellent
08:15 Reini I'm still fighting with one more critical blocker
08:17 willthechill oh?  which blocker?
08:44 Reini in potion it is numcmp in the jit (compare 2 floats), in parrot is a regression with the new fast arrays
09:18 willthechill so I have heard you say that parrot is now "ready", what does that mean?  you are using parrot as a VM instead of potion?
09:29 travis-ci RPerl build passed. Will Braswell says 'Grammar, Critic Tests Part 8'
09:29 travis-ci http://travis-ci.org/wbraswell/rperl/builds/43074365 https://github.com/wbraswell/rperl/compare/0848634777cf...3570c1919a69
09:31 willthechill great
10:29 travis-ci RPerl build failed. Will Braswell says 'Grammar, Critic Tests Part 9'
10:29 travis-ci http://travis-ci.org/wbraswell/rperl/builds/43078679 https://github.com/wbraswell/rperl/compare/3570c1919a69...43890e5822ee
10:36 willthechill yes I know
10:49 willthechill Reini: my question about parrot?
10:49 willthechill also, the type 'scalar' must be named 'scalartype' to avoid conflict with 'scalar' operation builtin reserved word in Perl 5
11:42 Reini I rewrote the parrot arrays, which needed 1 and a half week. This is now ready and 20% faster
11:47 willthechill yes, but are you going to use parrot at the VM or potion?
12:24 ilbot2 joined #perl11
12:24 Topic for #perl11 is now Perl5 + Perl6 == Perl11; # http://irclog.perlgeek.de/perl11/
12:42 Reini nonsense
12:42 Reini lot of false rumors around
12:43 Reini moar is much better, but the critical part, threads, is worse. moar cannot scale with the number of cores. it locks on all objects
12:44 Reini parrot is essentially lockless. it's just a lot of work to undo all the damage done in the last years.
12:45 willthechill so we're back to having 2 totally different and competing virtual machines for Perl 6?  great
12:45 Reini 3
12:46 willthechill yes potion I know
12:46 Reini and parrot still has the least failures of all 3
12:46 Reini no, jvm
12:46 willthechill oh jvm can go to hades
12:46 Reini potion is just for p2, and p2 is independent of rakudo. rakudo will never be fast enough
12:46 Reini their boostrapping process is mind blowingly stupid
12:47 willthechill LOL I'm sure they appreciate your opinions  :P
12:47 Reini I will not tell them. they cannot accept criticism in this stage
12:48 Reini but they need to get at least their goal finished, to ship a product.
12:48 willthechill yeah don't we all
12:49 Reini but I won't use it
12:49 Reini it looks like a very early java to me
12:50 Reini has potential, but needs a major overhaul
12:52 willthechill RPerl gives a roadmap for maintaining 100% XS & CPAN compatibility, while eventually rewriting the entire Perl 5 core to be compatible with whatever Perl 6 will exist in the future
12:52 willthechill thus, Perl 11
12:52 willthechill it will take a decade
12:52 willthechill but we CAN fully merge Perl 5 and Perl 6 eventually
12:52 willthechill without breaking XS or CPAN
12:52 willthechill or darkPAN
12:52 Reini I'm sceptical
12:52 willthechill yeah it will be super hard
12:53 willthechill but possible with RPerl
12:56 willthechill the point is that all of the Perl 5 core will have to be rewritten in P
12:56 willthechill *in RPerl
12:56 willthechill same use as RPython and NQP
12:56 willthechill RPerl is connected to the Perl 5 core, so it can be used to rewrite the Perl 5 core
12:57 willthechill without breaking any XS compatibility or CPAN or darkPAN, as I said
12:57 willthechill PONIE will ride yet again
12:57 willthechill ;)
12:59 Reini I'll doubt that it will help for perl5 as they hate static optimizations. perl6 can do types, so this possible
12:59 willthechill their hatred will be overcome by working code
13:00 willthechill same as always
13:04 Reini well, I will not discuss it at all anymore. I'll do it just for myself
13:05 willthechill same here
13:08 willthechill RPerl is passing 1,429 tests now
13:08 willthechill more to come!
13:18 basiliscos willthechill++
13:20 willthechill haha thanks  :)
13:24 Reini you are just passing your own tests. B::C is passing all of the perl5 core testsuite
13:24 Reini plus all the CPAN tests of the all >500 modules we use
13:24 willthechill pfft B::C is obsolete
13:24 Reini haha :)
13:25 Reini Still the best I have for now
13:25 willthechill only cPanel cares about a non-optimizing source-code-hiding "compiler"
13:25 Reini It optimizes a lot of startup-time and end-time
13:25 willthechill yes true
13:25 Reini and uses 10% less memory
13:25 willthechill haha 10% might as well be 0% as far as I care
13:25 willthechill and startup time doesn't matter to me
13:25 Reini we care a lot about memory, more than time
13:28 willthechill I care about memory and runtime, not startup time
13:28 willthechill although startup time is super fast in RPerl as well
13:28 Reini sure
13:28 willthechill but mostly I care about runtime
13:29 Reini runtime is something you will not be able to discuss with p5p sensibly
13:35 willthechill then my code will speak for me
13:37 Reini My code also spoke, and I even gave YAPC talks about the optimizations I did. Still silence and nonsense.
13:39 willthechill you need more money and momentum and followers
13:40 willthechill you only have cPanel following you
13:40 willthechill cPanel is awesome but still small compared to the entire Perl community
13:40 Reini Still the biggest perl shop around
13:41 Reini With most servers, users and devs
13:41 willthechill yes and I don't use cPanel because it doesn't support Ubuntu  :P
13:41 willthechill Webmin instead!  haha
13:41 willthechill tell Nick he needs to support Ubuntu
13:41 Reini yes, that's a shame. not even coreos or freebsd anymore
13:42 Reini ubuntu not for servers. but debian stable
13:42 Reini ubuntu is a joke
13:43 willthechill haha I have used ubuntu on stable servers for years, you don't know what you're talking about
13:50 Reini I know why ubuntu is a joke. just ubuntu users do not know
13:50 Reini on the desktop its ok
13:52 willthechill well I'm an expert at ubuntu on both server and desktop, so PLEASE enlighten me oh wise one, why is ubuntu a "joke"?
13:53 Reini security
13:53 willthechill FALSE: daily security updates
13:53 willthechill please, try again!
13:54 Reini no basic security features. they learned a lot from redhat lately, but I still no admin trusts them
13:55 Reini e.g. this list looks pretty good now: https://wiki.ubuntu.com/Security/Features
13:55 willthechill you're imagining things
15:36 Reini joined #perl11
17:46 travis-ci RPerl build passed. Will Braswell says 'Grammar, Critic Tests Part 10'
17:46 travis-ci http://travis-ci.org/wbraswell/rperl/builds/43118495 https://github.com/wbraswell/rperl/compare/43890e5822ee...b4d2af602efb
21:11 basiliscos joined #perl11

| Channels | #perl11 index | Today | | Search | Google Search | Plain-Text | summary