Perl 6 - the future is here, just unevenly distributed

IRC log for #phasers, 2010-12-14

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

All times shown according to UTC.

Time Nick Message
04:40 Tene_ joined #phasers
04:40 Tene_ left #phasers
04:40 Tene_ joined #phasers
04:40 TimToady_ joined #phasers
04:52 spinclad left #phasers
04:52 cognominal left #phasers
04:52 sorear left #phasers
04:52 Tene_ left #phasers
04:52 Util left #phasers
04:52 dukeleto left #phasers
04:52 moritz_ left #phasers
04:52 eternaleye left #phasers
04:52 TimToady_ left #phasers
04:53 pmichaud left #phasers
04:53 PerlJam left #phasers
04:53 TimToady left #phasers
04:53 Tene left #phasers
04:53 [particle]1 left #phasers
04:54 dukeleto joined #phasers
04:54 Util joined #phasers
04:54 Tene_ joined #phasers
04:54 spinclad joined #phasers
04:54 [particle]1 joined #phasers
04:55 TimToady_ joined #phasers
04:55 PerlJam joined #phasers
04:55 pmichaud joined #phasers
04:55 moritz_ joined #phasers
04:55 eternaleye joined #phasers
04:56 cognominal joined #phasers
04:56 sorear joined #phasers
04:57 moritz_ left #phasers
04:57 eternaleye left #phasers
04:57 TimToady_ left #phasers
04:57 pmichaud left #phasers
04:57 PerlJam left #phasers
04:58 TimToady_ joined #phasers
04:58 PerlJam joined #phasers
04:58 pmichaud joined #phasers
04:58 moritz_ joined #phasers
04:58 eternaleye joined #phasers
04:58 TimToady_ left #phasers
04:58 TimToady joined #phasers
06:10 spinclad left #phasers
06:10 spinclad joined #phasers
15:42 smash joined #phasers
16:18 spinclad_ joined #phasers
16:23 spinclad left #phasers
18:11 diakopter joined #phasers
18:20 mberends joined #phasers
18:28 diakopter mberends: hi :)
18:28 mberends ahoy!
18:49 colomon joined #phasers
18:59 diakopter my report: took a break from the -rx portion of clrnqp-rx .. but when I pick it up again, I'll be blocked on questions for pmichaud
19:02 mberends testing 6model/dotnet atm on Mono 2.4.4, no failures yet :)
19:02 diakopter cool
19:03 colomon o/
19:03 diakopter 05-comments.t fails sporadically on .Net, we have no idea why
19:03 sorear hi
19:03 TimToady .oO(if it's random, it must be the GC :)
19:03 sorear no #phasers today?
19:03 TimToady sezhoo?
19:03 mberends this is it!
19:04 pmichaud \o/ #phasers
19:04 sorear the lack of o/ spam at 11:01 local time
19:04 mberends \o/ greetings pmichaud!
19:05 sorear o/ anyway
19:05 diakopter TimToady: :P actually it fails in the parrot portion (parsing), so YES. :D
19:06 diakopter so I guess I meant, parrot-on-windows and not .Net
19:07 PerlJam greetings!
19:07 * PerlJam phases in
19:10 diakopter sorear: your update?
19:11 sorear done:
19:11 sorear @foo[*-1] and @foo[1,2,3]
19:11 sorear INIT
19:12 sorear require
19:12 diakopter \0/
19:12 sorear starting to play with the new codegen
19:12 sorear end
19:13 sorear earlier, I had Perl code generating C#
19:13 sorear I'm trying to replace this with C# code using SRE
19:14 sorear which should be quite a bit faster, also, a bit more flexible
19:16 colomon replace the perl with C#, or the C# with C# using SRE?
19:17 sorear the former
19:17 * moritz_ did some work on the advent calendar. Solved some of masak's coding contest problems. No further Perl 6 hacking.
19:22 diakopter pmichaud: did you see my mention of you above
19:22 pmichaud diakopter: yes, I did.  Let me know when you pick it up again, I'll try to squeeze out some time.
19:22 diakopter it's probably a quick q&a
19:22 pmichaud my status report:  I continue to be distracted from p6-hacking by other things going on around here.
19:26 * Util made no progress on anything but masak's 5-problem contest.
19:28 TimToady 'tis the season to be crawly...
19:28 * colomon is frustrated by his progress on masak's contest.
19:29 colomon I did work around .ord and add .ords and .chrs at TimToady++'s behest over the weekend.
19:29 colomon And an Advent post.
19:29 TimToady colomon++
19:30 Util colomon++
19:32 colomon Planning on one and maybe two more Advent posts, and a lot more hacking on the contest in the next week.
19:32 colomon .eor
19:35 PerlJam TimToady: any chance you'll submit a post to the advent calendar?
19:36 PerlJam pmichaud: what about you?
19:36 PerlJam there are still several days open
19:37 TimToady mebbe, it's a little hard for me to type at the moment, because my spacebar thumb is sore
19:38 pmichaud PerlJam: I'll look at the calendar and see if it meshes with mine at all
19:38 TimToady getting old sucks, sometimes
19:38 PerlJam TimToady: You /were/ blessed with two thumbs  :-)
19:40 TimToady yes, the other thumb is how I'm typing at all :)
19:41 diakopter probably it hurts just to keep it lifted off the touchpad/spacebar
19:41 * diakopter guesses
19:46 pmichaud maybe.you.should.find.some.other.key.that.you.can.use.instead.of.space.to.type.:-)
19:46 pmichaud then.you.don't.need.to.use.spacebar.thumb.at.all.
19:47 pmichaud if.only.there.was.some.tool.that.could.then.be.used.to.translate.this.other.character.into.spaces...
19:47 pmichaud what's.the.opposite.of.Acme::Bleach? :-P
19:48 diakopter Acme::Sharpie
19:48 pmichaud I suppose we'd want Acme::Etch -- which bleaches the unmasked portions of the template leaving only the desirable traces
19:48 pmichaud Acme::EtchingSolution
19:49 PerlJam but then you'd need an ultraviolet light or some other catalyst.
19:49 pmichaud TimToady++ is bright enough to be his own catalyst :-)
19:53 pmichaud I like Acme::Sharpie, though.
19:54 * TimToady has this habit of derailing productive conversation...
19:54 mberends :)
19:54 diakopter as long as it's onto another set of productive rails...
19:55 PerlJam though the talk of etching made me wonder if there were already a "line-drawing" version of Camelia.
20:01 mberends regarding 6model/java, it's now clear to me that the generated-goto compiler requirement rules out Java as a target language. The Java subproject will have to become bytecode based, say 6model/jvm. If anyone feels like working on that with me, please volunteer. I must first recuperate from my $work induced tuit deprivation though, expected to be better in January.
20:03 sorear Alternatively, you could just design and implement a tiny JVM language with gotos
20:04 diakopter heh
20:04 diakopter 'just'ified by works
20:06 diakopter mberends++ # having lots of $work
20:06 mberends I'll try whatever works soonest.
20:07 diakopter sorear: i'm sure there are other structural/imperative languages for the JVM with labels/goto
20:07 diakopter ok, I'm not sure
20:07 diakopter I highly suspect
20:13 sorear Is 6model/dotnet bytecode based yet?
20:16 mberends no, it generates C# afaik
20:17 tylercurtis joined #phasers
20:19 mberends yes, verified, the Linux oriented try.sh script writes to the fixed name RakudoOutput.cs to pass on to Mono.
20:22 sorear tylercurtis!
20:24 diakopter ... at this stage it wouldn't be too difficult *for me* to write a CIL backend (interfacing with/wrapping runsharp), but it probably wouldn't correspond to how jnthn's imagining it... so I was going to discuss it thoroughly with jnthn beforehand.
20:25 sorear diakopter: How is runsharp better than System.Linq.Expression?
20:27 diakopter sorear: I don't know whether System.Linq.Expressions.GotoExpression is implemented in mono (maybe it is)
20:27 sorear it has to be, the DLR critically relies on it
20:28 diakopter it may have relied on System.Linq.Expression in .Net 3.5
20:28 diakopter but not System.Linq.Expressions.GotoExpression  b/c that doesn't exist in .Net 3.5
20:28 diakopter b/c the DLR definitely works on just 3.5
20:29 diakopter oh.. but in different namespaces
20:29 diakopter (not System.Linq)
20:29 diakopter (now I remember)
20:29 sorear the DLR uses "expression trees 2.0", which were cored in 4.0
20:30 diakopter yeah; afai remember a namespace change was involved.
20:32 diakopter so, to answer your question, runsharp *could be* better than the Linq.Expressions b/c it's a much smaller dependency, and I have no idea how good its generated code is... maybe it's decent.
20:33 sorear well, afaik 6model already depends on C# 4.0
20:33 sorear for "var" declarations
20:33 diakopter no :P
20:33 diakopter var was in 3.0
20:33 sorear so System.Linq.Expressions is a null dependency
20:34 diakopter I was speaking of load-time dependency, not just on-disk
20:49 diakopter sorear: I'm gun-shy of the DLR (and its Linq.Expressions) mostly b/c of the severe perf problems msft had with it
20:49 diakopter (when implementing ironruby & ironpython)
20:50 diakopter sorear: of course, such things don't matter once we're working in an App::Persistent-like setting
20:51 diakopter (similar to what .Net does on Windows with its ".Net Runtime Optimization" windows service)
20:52 sorear "severe perf problems", hehehe
20:52 diakopter yeah, I think it's fair to call them that
20:52 sorear all the docs like to gloat about how "fast" it is
20:52 diakopter bah
20:52 diakopter in some benchmarks, yeah.. but startup time always kills them
20:53 diakopter when compared to all the other pythons and rubys
21:42 spinclad_ is now known as spinclad
21:51 tylercurtis left #phasers
21:54 spinclad .oO( meeting done?  or just phased out? )
21:55 sorear it never really got started.
21:56 diakopter I don't see how my report at -1 minute wasn't a real start
22:00 spinclad a few other reports sprinkled in, and some relevent talk derailed to other relevent talk...
22:02 spinclad sounds like an ordinary meeting, just quieter than usual
22:05 sorear *much* quieter
23:23 smash left #phasers

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