Perl 6 - the future is here, just unevenly distributed

IRC log for #perl6, 2007-02-23

Perl 6 | Reference Documentation | Rakudo

| Channels | #perl6 index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
00:05 dduncan re the latest synopsis update, that option to have an extra short label for a longer named parameter looks useful
00:05 dduncan as I've found that the best internal name is often too long to be used repeatedly as a named argument
00:06 dduncan and hence I've been using shorter names throughout
00:06 dduncan on the other hand, trying to figure out a name that is both short and descriptive allows for some more thinking about what works well
00:16 YetAnotherEric left perl6
00:16 buubot joined perl6
00:18 buubot joined perl6
00:18 TimToady dduncan: it seems like it was the logical way to get at :g($global) adverbs as well, and I decided it shouldn't have too much impact if the long names are just the fallback.
00:18 TimToady and I wanted it for STD
00:19 dduncan ok
00:22 weinig|bbl is now known as weinig
00:23 Schwern joined perl6
00:24 CardinalNumber joined perl6
00:32 macli left perl6
00:43 svnbot6 r15329 | lwall++ | modifiers for regex, trans and p5regex
00:43 svnbot6 r15329 | lwall++ | long modifiers in :g($global) named parameter form
00:44 CardinalNumber is now known as ProperNoun
00:44 Limbic_Region hey TimToady - is hash/array slices in scalar context documented anywhere (perl5 and/or perl6)?
00:45 TimToady don't think it's documented in P5
00:45 TimToady if it is, probably documented to take the last value
00:45 * Limbic_Region couldn't find it in perldata
00:46 TimToady in P6 would always turn into [...]
00:46 Limbic_Region thanks
00:46 TimToady there's nothing in P6 that automatically returns the last value of a list except [-1]
00:47 TimToady well, any maybe pop
00:47 Limbic_Region reverse shift
00:47 TimToady but none of those are "automatic"
00:48 Limbic_Region yeah
00:48 Limbic_Region I was just thinking that you had to be wrong though
00:48 Limbic_Region since that would imply there is only one way to do it
00:48 Limbic_Region :P
00:50 TimToady I'm trying to think what's there that would only give you one way.  .WHICH and .WHAT maybe
00:50 TimToady though you could always write macros...
00:50 TimToady actually, those come in functional forms too, so I guess they don't work
00:50 tene could you manually dig them out of a symbol table or some such?
00:51 Limbic_Region speaking of which, what synopses speak to introspection?
00:51 Limbic_Region s/what/which/
00:51 TimToady mostly S12, some in S02
00:52 TimToady a lot of it is intentionally punted off to .HOWland, which is where MO and Moose come in.
00:53 Limbic_Region audreyt has indicated that 6.28 release is RSN so I will read S12 in detail RSN too
00:53 TimToady S12 won't talk about MO
00:53 * Limbic_Region has mostly held off
00:53 TimToady well, not much...
00:54 TimToady presumably .HOW.methods is returning something from MO
00:54 Limbic_Region who was the prime mover in the meta object work - stvn?
00:54 TimToady yes
00:55 tene "prime mover"
00:55 TimToady with a lot of help from audreyt and, putter was it?
00:55 * tene things of Thomas Aquinas.
00:55 Limbic_Region I think so
00:55 tene s/thing/think/
00:56 nipotaway is now known as nipotan
00:56 TimToady and of course nothingmuch is at one level the prime mover of MO
00:56 TimToady sort of the Moose porter
00:56 Limbic_Region so has perl 6 fully adopted the MO model?
00:56 Limbic_Region and if so, shouldn't S12 cover it in greater detail
00:57 TimToady no, Perl 6 is completel agnostic on the subject of metaobject models.
00:57 TimToady s:5th/ /y /
00:57 TimToady s:5th/ <sp> /y / rathre
00:57 TimToady er
00:58 TimToady "Pay no attention to the MOP behind the .HOW curtain..."
00:59 Limbic_Region ok
00:59 Limbic_Region well, I then don't much care about MO as the means by which pugs delivers an S12 perl 6
00:59 TimToady of course, the fact that P6 uses MO for its default MOP should likely be documented somewhere eventually...
00:59 Limbic_Region I just hadn't followed S12 too closely until it was something I could actually use
01:00 Limbic_Region perl 5 was my first dive into OO
01:00 Limbic_Region and while I speak Java relatively fluently
01:00 TimToady you can use most of S12 now, it's just a handrolled MOP that's not very introspectible
01:00 TimToady (in pugs)
01:01 Limbic_Region I don't quite grok things that are that abstract without some practical implementation to play with
01:01 Limbic_Region oh, I realize that now
01:01 TimToady 6.28 should change that part
01:01 Limbic_Region but not when I wanted to
01:01 Limbic_Region pugs OO was changing daily when I was interested
01:01 TimToady indeed, the abstraction barrier is precisely why MO is swept behind the .HOW curtain.
01:01 buetow joined perl6
01:02 TimToady but I admit that the distinction is firmer in my mind than in reality.  :)
01:02 bsb left perl6
01:03 Limbic_Region I think the flavor of my code will strongly be affected by perl 6 OO once I grok it completely
01:03 Limbic_Region I rarely resort to OO in perl 5
01:03 Limbic_Region mostly imperative with a strong functional flavor
01:03 TimToady well, if not, we didn't design it right... :)
01:03 Limbic_Region despite never having learned a truly functional language
01:04 TimToady just think of objects as strange monads and you'll be fine.  :)
01:05 Limbic_Region heh
01:05 Limbic_Region I kinda need to learn ML
01:06 mjk joined perl6
01:06 allbery_b haskell!
01:06 TimToady mooskell!
01:07 TimToady MoosekelL
01:07 Limbic_Region allbery_b - Alice is based of ML and since Alice has the features I need, it wins
01:07 TimToady well, what about Bob?
01:08 Limbic_Region http://en.wikipedia.org/wiki​/Alice_programming_language
01:08 Limbic_Region Bob - http://imdb.com/title/tt0103241/
01:08 lambdabot Title: What About Bob? (1991)
01:13 TimToady someone should port Perl 6 to run on top of Alice someday.  :)
01:13 TimToady futures in Perl 6 are generally spelled "==>"
01:14 tene Does that make "<==" "pasts"?
01:15 * TimToady is not sure whether he should cry or swear.  :)
01:16 tene Both!
01:27 TimToady decommuting &
01:33 putter joined perl6
01:37 putter can one create a A.pm which when used, causes a B.pm's import() to run as if it had been called directly?
01:38 putter if B has an actual import() sub, rather than getting it via inheritance, one could simply goto() it.
01:38 putter and if !UNIVERSAL::can(B,'import'), you are done.
01:39 putter but if B's import() comes via inheritance, perhaps from Exporter... (visions of special casing that pushed aside for just now), what can one do?
01:39 [particle] damian has a perl 5 module for that
01:40 putter !!! :)
01:40 devbot6 putter: Error: "!!" is not a valid command.
01:40 putter do tell
01:40 [particle] lemme see if i can remember, with cpan to aid me
01:40 [particle] http://search.cpan.org/~dconway/
01:40 lambdabot Title: Damian Conway - search.cpan.org
01:41 [particle] Toolkit
01:41 [particle] http://search.cpan.org/~dconwa​y/Toolkit-0.0.2/lib/Toolkit.pm
01:41 lambdabot Title: Toolkit - Keep your handy modules organized - search.cpan.org
01:48 putter :)
01:48 putter [particle]: my thanks
01:49 putter he took the source filter approach.  i tried that, but though I encountered problems when B was itself a filter.  but perhaps i was confused.  will try again.
01:49 putter tnx
01:49 putter s/though/thought
01:53 bsb joined perl6
01:53 Squern joined perl6
01:57 rashakil__ joined perl6
02:00 avar joined perl6
02:15 REPLeffect_ joined perl6
02:16 putter oh drat.  that commit note should have been [particle]++. ;)
02:16 lpwave joined perl6
02:27 Schwern joined perl6
02:35 svnbot6 r15330 | putter++ | misc/pX/Common/v6.pm - Created.  A v6.pm delegator.  This small module flexibly delegates "use v6-alpha;" requests to actual implementation modules.
02:35 svnbot6 r15330 | putter++ | It defaults to a Pugs::Compiler::Perl6::v6, where the current v6.pm could be moved.
02:35 svnbot6 r15330 | putter++ | First draft - largely untested.
02:38 xdg joined perl6
02:41 Schwern joined perl6
02:42 Schwern joined perl6
02:43 [particle] joined perl6
03:20 REPLeffect_ joined perl6
03:23 Aankhen`` joined perl6
04:03 putter how about yet another variable scope?  "nested local".
04:04 putter consider a regex / (...{local $a=3}...{X}... ) ... {Y} /
04:04 putter currently both X and Y can see $a=3.
04:05 putter the objective is a scope such that X can see $a=3, but not Y.
04:05 putter a "local" scope which respects the regex group nesting.
04:06 ashleyb joined perl6
04:07 weinig is now known as weinig|bbl
04:08 putter is there a spare twigil in the house....? :)
04:10 Patterner U+2325 is cute
04:11 Patterner -= as an ascii alternative :)
04:11 putter looking...
04:13 putter people are going to be writing some truely bizarre code soon.... (apropos unicode)
04:14 putter Patterner: :)
04:14 TimToady Â¥ just came available.  :)
04:14 putter lol
04:14 putter zipper gone?
04:14 TimToady Z
04:14 putter ah
04:14 TimToady even looks like "row major"
04:15 TimToady gotta run, though.
04:15 TimToady bbl &
04:15 * putter considers looing for a reflected Z, resists...
04:15 putter bye
04:17 * allbery_b doesn't see one (osx makes it easy...)
04:20 putter unicode isn't complex enough yet.  let's add reflection and rotation combinators.
04:21 Patterner I thought Unicode already supports reflection/mirrored characters.
04:21 Patterner putter: what about U+16CE/RUNIC LETTER Z? :)
04:25 putter or RUNIC LETTER JERAN J
04:26 putter I wonder if ( ) parens are a combining char.  for (+) analogy to $+x.
04:29 Patterner I wish I had a font to see those runic letters...
04:30 lpwave joined perl6
04:31 putter ᛃ ?
04:32 Patterner [] is probably not really the letter :)
04:32 putter :)
04:46 putter anyone know order of magnitude how expensive the method first-use, scan ISA, is? 10x sub call, 100x, 1000x?
04:49 * putter is contemplating an architecture which would have all of the Prelude, Perl.foo grammar rules, and maybe even a p6 runtime api, all ISA()ed into _every_ package (well, every one which is using p6).
04:50 amnesiac joined perl6
04:50 putter scary from a runtime performance perspective.  but nice from a development perspective.  want to do yet another p6 on p5 variant,  "package Foo; use base SomeExistingOne; sub mumble {...mutant mumbling...}"
04:54 putter @tell TimToady  re extensibility of the p6 grammar, one needs to be able to retract as well as extend, no?  So one can play games like starting with the quick and dirty, or stable, version of token/rule/regex x, and then others selectively override it with more real, or experimental, variants.
04:54 lambdabot Consider it noted.
04:55 ashleyb joined perl6
04:56 TimToady sure, but the overriding is based on longnames including longest token match.
04:56 lambdabot TimToady: You have 1 new message. '/msg lambdabot @messages' to read it.
04:56 putter I find myself writing   regex inherited_hack_rule_no_longer_needed { <fail> }  overrides.
04:57 putter no longest token
04:58 TimToady 'only' or 'proto' would cause this rule to hide all rules with the same short name
05:00 TimToady oh, $<foo> already scopes to the current (...)
05:01 putter class A { regex statement_control { if ...havent implemented else yet...} } class B does A { regex statement_control { if ...full version...} }   where does the only go?
05:02 TimToady in front of the 2nd "regex"
05:02 TimToady same place "proto" or "multi" would go
05:02 putter re $<foo>  !!!  :)  *pause*  user hat replaced with have to code it over the weekend hat.  ?!?! oh, sigh.
05:04 putter how would B get rid of if entirely?  regex statement_control { if...and whatever else might be contributing to the token left intact... <fail> }  ?
05:13 REPLeffect_ joined perl6
05:14 TimToady something like that, or use proto that intelligently calls SUPER somehow
05:15 cj 1~9ohgefgfffffffffffffffffffffffffffffffffff
05:15 cj (scarlet says hi)
05:18 * putter wonders if linux/X has a "recognize and filter cat on keyboard" program, like dos once did.
05:22 allbery_b doesn't look like it
05:23 putter :)
05:27 putter @tell audreyt To get a feeling for your current thinking on the p6 ast, where is the best place to look?  kp6?
05:27 lambdabot Consider it noted.
05:29 * putter ponders a p6  class A is reblessable {...}
05:33 putter re |,|| ...  "YACC is stealing my |, and I want it back!" ;)
05:34 putter people are going to be soooooo confused.  sigh.
05:39 putter end of day.  g'night all. &
05:40 zgh joined perl6
05:55 obra 'evening
06:13 BooK_ joined perl6
06:20 gaal mornin'
06:27 jisom joined perl6
06:28 jisom left perl6
06:28 jisom joined perl6
07:12 rfordinal_ joined perl6
07:21 marmic joined perl6
07:52 elmex joined perl6
07:53 devogon joined perl6
08:07 drrho joined perl6
08:30 rfordinal_ is now known as rfordinal
08:41 allbery_b joined perl6
08:52 dduncan left perl6
09:22 mj41 joined perl6
09:25 pdcawley joined perl6
09:33 IllvilJa joined perl6
09:35 Schwern joined perl6
09:37 IllvilJa joined perl6
09:41 lisppaste3 joined perl6
09:55 buetow joined perl6
10:04 chris2 joined perl6
10:12 ruoso joined perl6
10:29 ruoso_ joined perl6
10:31 ruoso_ is now known as ruoso
10:56 nipotan is now known as nipotaway
11:09 rindolf joined perl6
11:09 rindolf Hi all.
11:10 moritz hi rindolf ;)
11:10 shay hey moritz :)
11:10 moritz hi shay ;)
11:11 rindolf Hi moritz
11:11 rindolf moritz: what's up?
11:11 moritz nothingmuch ;)
11:12 moritz .oO(sorry for hilight-spamming ;)
11:12 shay hehe
11:13 Patterner we forgive, but we won't forget
11:14 moritz ;)
11:14 moritz my irclogs don't forget either
11:14 moritz unless the hds crash ;)
11:15 nipotaway is now known as nipotan
11:19 rindolf moritz: don't you have backups?
11:20 moritz rindolf: yes, but irregularly
11:20 rindolf "Only wimps use tape backup: real men just upload their important stuff on ftp, and let the rest of the world mirror it." -- Linus B. Torvalds
11:20 rindolf moritz: I see.
11:20 rindolf moritz: I'm backin up my hard-disk every month.
11:21 moritz rindolf: I don't administer my server and don't have too fast access to it...
11:21 rindolf moritz: rsync?
11:22 moritz rindolf: I'm currently in a student's hall with approx. 24kB/s downstream...
11:22 moritz rindolf: that's no fun...
11:22 moritz I mirror the most important data with unison
11:23 moritz (which is really two-way rsync)
11:23 rindolf moritz: OK.
11:24 rindolf moritz: why so low a bandwidth?
11:24 rindolf moritz: where do you live?
11:24 moritz normally I live in Würzburg, Germany but now I'm in Edinburgh...
11:24 moritz and the university doesn't seem to be interested in providing faster internet
11:26 moritz yesterday the http proxy start killing connections randomly :(
11:32 rindolf moritz: ouch. :-(
11:32 rindolf moritz: do you have a faster Inet-connection at home?
11:33 moritz rindolf: yes, normal DSL-1024
11:34 moritz wich gives about 124kB/s downstream via ftp
11:34 moritz not ideal, but much better ;)
11:36 rindolf moritz: OK.
11:36 rindolf moritz: I have more than that.
11:36 rindolf 200 KB or even more.
11:39 ruoso_ joined perl6
12:18 ruoso joined perl6
12:23 IllvilJa joined perl6
13:10 drbean joined perl6
13:15 Limbic_Region joined perl6
13:29 rindolf moritz: what are you doing in Edinburgh?
13:35 cookys joined perl6
13:44 stevan_ any of our chinese hackers familiar with "quad byte" format?
13:48 stevan_ oh wow,.. its basically UTF-32... how strange (at least to me)
13:48 stevan_ (Dan Kogai)++ # Encode handles it :)
14:14 moritz rindolf: studying physics ;)
14:22 rindolf moritz: are you studying in Edinburgh during your entire degree?
14:23 moritz rindolf: no, just one year... hopefully long enough to make my "master of physics"
14:23 mdiep joined perl6
14:26 [particle] joined perl6
14:28 moritz rindolf: and where do you live?
14:30 moritz when trying to upload smoke reports, I get an error message "no space left on device"
14:31 moritz is there anybody here who can fix that?
14:31 rindolf moritz: I live in Tel Aviv, Israel.
14:33 moritz rindolf: how much does the israel<->palestians conflict affect your daily life?
14:33 rindolf moritz: not much.
14:33 moritz rindolf: I only know isreal from TV, and I don't have a clue ;)
14:33 moritz that's luck ;)
14:33 rindolf moritz: it's usually quiet here.
14:34 Limbic_Region joined perl6
14:35 moritz rindolf: it's cool that the perl dev is so international ;)
14:36 moritz is anybody from africa here?
14:38 rindolf moritz: my homesite is http://www.shlomifish.org/
14:38 lambdabot Title: Shlomi Fish' Homepage
14:41 vel6608 joined perl6
14:44 moritz rindolf: quit interesting ;)
14:46 mdiep joined perl6
14:51 allbery_b aha, you're shomifish!  the world shrinks a bit more :)
14:55 rindolf allbery_b: Shlomi Fish.
14:57 allbery_b sorry, I'm good at typoes :)
14:58 rindolf allbery_b: typos. ;-P
14:58 * Limbic_Region hasn't been to a YAPC yet but I suspect name tags should include real name and then a list of online AKAs
15:02 moritz what happend to the default filehandles? IIRC they used to be $*DEF{IN,OUT,ERR}, but now none of the appear in the specs
15:02 [particle] STDIN?
15:03 moritz [particle]: in perl6?
15:03 moritz ?eval STDIN
15:03 evalbot_r15324 Error: No compatible subroutine found: "&STDIN"
15:03 [particle] 176:    use GLOBAL <$IN $OUT $ERR>;
15:03 [particle] 180:    #     my ($IN, $OUT, $ERR) ::= ($*IN, $*OUT, $*ERR)
15:03 [particle] S11.pod
15:04 moritz [particle]: thanks
15:05 [particle] ack to the rescue!
15:05 allbery_b moritz: isn't DEFIN ruby or something? :)
15:06 moritz allbery_b: I remeber that I wrote testcases for say and print using $*DEFOUT, it worked and nobody complained ;)
15:06 moritz allbery_b: and I thought I found it somewhere in the specs
15:07 moritz it's in t/builtins/io/say_and_print.t
15:10 ruoso joined perl6
15:13 REPLeffect_ joined perl6
15:13 baest joined perl6
15:18 moritz what is the recommended way to iterate over a filehandle for reading?
15:18 moritz while defined (my $line = $fh.readline()) is somewhat ugly...
15:19 [particle] for =$fh -> $line {...}
15:19 moritz [particle]: thanks
15:22 IllvilJa left perl6
15:24 bonesss joined perl6
15:36 svnbot6 r15331 | kudra++ | Late, but finally here
15:54 amnesiac joined perl6
15:54 svnbot6 r15332 | kudra++ | I always have to correct this line break in the completed summaries, so I'm fixing it in the prototype
15:57 ruoso joined perl6
15:58 svnbot6 r15333 | kudra++ | Full thread listing but no summaries yet
15:58 svnbot6 r15334 | kudra++ | Partial darft
15:58 svnbot6 r15335 | kudra++ | Placeholder; week isn't complete yet
16:02 devbot6 planet6: Audrey Tang: Weekly Perl 6 mailing list summary for 21-28 January, 2007 <http://pugs.blogs.com/pugs/2​007/02/weekly_perl_6_m.html>
16:10 REPLeffect_ joined perl6
16:14 REPLeffect_ is now known as REPLeffect
16:14 REPLeffect left perl6
16:14 REPLeffect joined perl6
16:16 bonesss joined perl6
16:27 turrepurre joined perl6
16:30 nipra joined perl6
16:39 ruoso joined perl6
16:39 falesca joined perl6
16:41 turrepurre joined perl6
17:02 iblechbot joined perl6
17:11 ashleyb joined perl6
17:28 ProperNoun joined perl6
17:36 falesca joined perl6
17:41 macli joined perl6
17:45 justatheory joined perl6
18:01 Xaduha joined perl6
18:15 nipra joined perl6
18:17 orafu joined perl6
18:23 jisom joined perl6
18:23 jisom left perl6
18:24 jisom joined perl6
18:30 kolibrie ingy: is it planned for the Perldoc module on CPAN to be able to process Perl 6 perldoc?
18:33 justatheory joined perl6
18:34 ingy kolibrie: yes
18:34 kolibrie oh, goody
18:34 TimToady probably want to wait till Damian comes out with his parser
18:34 ingy nod
18:35 TimToady and just embed that.
18:35 ingy it's not really in my brain at the moment, but yes
18:35 TimToady he's just about done with it, I believe.
18:35 ingy awesome
18:35 kolibrie very nice
18:35 TimToady I'll need to install a translated version of it into the P6 parser as well.
18:36 TimToady since P6 is supposed to be able to access its surrounding pod via %= variables
18:36 ingy oh cool. so perldoc is parser into the ast?
18:36 ingy parse*d*
18:37 TimToady yes, has to be an official subset of P6 grammar
18:37 TimToady which is of course extensible...
18:37 ingy :)
18:37 ingy moment of panic relieved
18:37 TimToady It occurred to me that you might take that wrong.  :)
18:37 ingy it was a very short moment
18:38 TimToady have to observe the kwidities, after all...
18:38 neonse joined perl6
18:39 ingy :)
18:39 TimToady I guess that should be spelled kwiddities
18:42 TimToady does anyone remember how to get a web svn diff of the synopses?  seems like something there should be a top-level page for somewhere...
18:42 TimToady wanted to mention it on perlmonks but couldn't find it...
18:43 beppu joined perl6
18:51 Limbic_Region TimToady - couldn't find what on PerlMonks?
18:51 * Limbic_Region pretty much has PerlMonks in memory at all times
18:52 Limbic_Region oh nevermind
18:52 Limbic_Region you wanted to put something on PerlMonks that doesn't currently exist (there)
18:56 IllvilJa joined perl6
19:02 baest TimToady: can you explain what %= does or is a specification available somewhere (couldn't find it in the synopsis). Very curious on precisly what it does
19:05 PerlJam baest: %= looks like mod-equals to me.  
19:05 PerlJam baest: i.e.,  $a %= b  is the same as $a = $a % b;
19:05 baest PerlJam: oh, yes, I'm not precise I was thinking about: 19:36:21 < TimToady> since P6 is supposed to be able to access its surrounding pod via %= variables
19:05 TimToady = is a twigil indicate a file-scoped document variable.  The = is meant to remind you of =begin/=end
19:05 Limbic_Region ?eval my $num = 100; $num %= 3; $num
19:05 evalbot_r15324 is now known as evalbot_r15335
19:05 evalbot_r15335 \1
19:06 PerlJam oh! POD context
19:06 TimToady no, term context
19:06 TimToady so @=DATA could be the new <DATA>
19:07 TimToady but it would be a =begin DATA/=end DATA podstream
19:07 [particle] so, =$filehandle becomes $=filehandle?
19:07 TimToady no
19:07 TimToady has nothing to do with filehandles anymore
19:07 [particle] 's confusing, though
19:07 baest that makes sense. So why would you access the pod info. Is that instead of repeating yourself? Not sure I understand the effects of it
19:08 PerlJam baest: literate programming comes to mind  :-)
19:08 TimToady you can think of it as a displaced heredoc
19:08 PerlJam baest: other forms of introspection too
19:08 TimToady and autodocs can refer to pod comments near them
19:09 TimToady so a function could have a method to return its documentation, for instance.
19:09 PerlJam runnable examples are always nice
19:09 TimToady some languages kludge this in with an extra declaration in the header of the function
19:09 baest TimToady: that's very cool. That is something that really bugged, that you had to repeat the options for a script twice or be _very_ creative
19:10 TimToady or by processing comments
19:10 baest bugged *me
19:10 PerlJam TimToady: are you making digs at python there?  :-)
19:10 TimToady not intentionally
19:10 TimToady but if the shoe fits
19:10 PerlJam If you'd said something about semantic whitespace I would have been sure you were targeting python
19:11 TimToady python is sort of self-targeting that way... :)
19:11 baest :)
19:11 PerlJam TimToady: so, how is your day going so far?  (What is it that you do for a living these days?)
19:12 TimToady I'm almost awake by now.
19:12 TimToady my blood pressure started too high this morning, so I'm taking it pretty easy, apart from posting to perlmonks, which can raise anyone's blood pressure
19:13 tene perhaps give functions an "is documented" trait?
19:13 TimToady I officially work for Netlogic Microsystems
19:13 TimToady but they pay me to work on Perl 6 a lot. :)
19:13 TimToady tene: that is precisely what I'd like to avoid.
19:13 PerlJam well (Netlogic Microsystems)++ then
19:14 jisom that'd require as "is documented where" trait
19:15 TimToady or if there is such an "is documented where" it would default to "the preceding pod block"
19:15 TimToady or the pod block links it the other direction instead
19:16 tene that could also work.
19:16 tene or support both
19:17 jisom and if you have some of your pod inside the function and some outside because you want the comments to be close to the code?
19:17 diakopter joined perl6
19:17 TimToady main point is, pod is a data structure navigable by the program.
19:17 TimToady everything else is convention
19:17 jisom how strict is the parsing going to be?
19:17 TimToady how strict do you want it?
19:18 Belaf joined perl6
19:18 diakopter TimToady: how would one go about making a pragma to embed interdependent tests in/among their corresponding methods/subs?
19:18 TimToady certainly the default for documentation can be a little more forgiving.
19:18 PerlJam TimToady: did you see chromatic's recent use.perl post?
19:18 jisom the stricter it is, the easier to write a parser(even if harder to write for beginners)
19:19 PerlJam jisom: perl is/has a parser
19:19 * [particle] agrees with jisom -- let's not go the direction of HTML 1.0
19:19 jisom except be strict in the release stuff.....perl5 sort of has that issue, bad pod in the released in the tarball
19:19 jisom *in the release tarball
19:19 diakopter TimToady: I mean, if pod will be runtime accessible by the module and vice versa ... why not tests as well
19:20 * PerlJam suddenly experiences some deja vu
19:20 TimToady diakopter: it's been mentioned before as a possibility, but for the moment I'm lumping that in the "conventions to be established at some point" category.
19:22 TimToady PerlJam: which post is that?
19:23 PerlJam TimToady: http://use.perl.org/~chromatic/journal/
19:23 lambdabot Title: Journal of chromatic (983)
19:24 PerlJam TimToady: you were talking about high blood pressure and the link in that article about "Why perl is a bad language" might raise yours if you read it.
19:25 PerlJam But the sad thing is is that even on perlmonks, you find people like David Cameron
19:25 jisom "why is perl a bad language"....I can't write my kernel in it?
19:27 PerlJam heh, http://www.adequacy.org/public/s​tories/2001.12.20.165434.32.html
19:27 lambdabot Title: Pearl vs. Python: A Technical Review
19:27 PerlJam That one *has* to be a joke :)
19:27 REPLeffect left perl6
19:28 PerlJam anyway ...
19:29 * PerlJam goes back to patiently waiting for perl6
19:29 [particle] this is more fun to read: http://www.barcodeart.com/a​rt/clock/launch_small.html
19:29 lambdabot Title: Scott Blake
19:30 * jisom just figured out where "TimToady" came from
19:30 PerlJam jisom: Did you ride the short bus to school?  ;-)
19:30 TimToady not to mention http://www.barcodeart.com/a​rt/clock/launch_large.html
19:30 lambdabot Title: Scott Blake - Bar Code Clock
19:31 CardinalNumber joined perl6
19:32 PerlJam I wish I could find a job that would a) pay me enough to keep my family happy and b) let me work on perl-related stuff all the time.
19:33 TimToady Cheop's Law: everything costs more and takes longer
19:34 jisom unless you make it open source and get people to work on it for free, then it just takes longer
19:35 PerlJam jison: the costs just change from monetary to something else.
19:36 TimToady mostly opportunity cost of letting others get ahead, which can raise your blood pressure.  :)
19:37 kolibrie PerlJam: my boss is looking for someone, but I'm not sure exactly what he's looking for
19:37 TimToady a lot of people are looking for someone...
19:38 * kolibrie never found his childhood kindred spirit -- too late now
19:39 TimToady some of us just stay children until we do.  :)
19:39 kolibrie hmm, have to think about that
19:43 xerox TimToady: that law seem to be related to the second principle of thermodynamics. There's space for developement (-:
19:44 Limbic_Region xerox - I know you from ##java land right?
19:44 xerox Limbic_Region: YUCK. Here and #haskell.
19:44 moritz xerox: the principle basically says that the universe will end as a homogenous mass - not a good place to live in ;)
19:45 Limbic_Region hrmmm
19:45 * Limbic_Region is trying to remember the helpful soul in ##java whose handle started with an x
19:46 jisom but the universe is mostly homogenous, empty space, we're an anomaly in the homogeneousness of the universe
19:46 jisom my spell checker says it's a word and spelled right!
19:47 xerox I think both are oversemplification, but then I don't have a better theory of everything to propose. I can reccommend Penrose stuff to read, and watch, though.
19:47 TimToady no, the universe is mostly a seething mass of virtual particles.  we are merely persistent accounting errors.
19:47 * [particle] *coughs*
19:47 TimToady some are more persistent than others...
19:47 moritz TimToady: that explains a lot ;)
19:48 * [particle] seeths
19:48 moritz xerox: does penrose lectures thermodynamics? I always thought he did some QM or unified theories...
19:49 TimToady well, the fact is whenever we say "an electron started here and ended up there", it's mostly wrong.
19:49 moritz TimToady: that's why we talk of wave functions ;)
19:49 moritz or at leas I do ;)
19:49 * [particle] waves
19:49 xerox moritz: some of his ideas are explained in a quite funny way in his special talks for the MSRI (<http://msri.org>)
19:49 [particle] electrons travel backwards in spacetime
19:49 justatheory joined perl6
19:50 TimToady nah, they just obey Spacetime Transactional Memory
19:50 moritz [particle]: or forward - it so hard to tell since the schroedinger equation is invariant under time inversion ;)
19:51 allbery_b there is that thing where an electron moving backward in time is indistinguishable from a positron
19:51 TimToady STM is what keeps the accounting errors persistent, lucky for us.
19:51 xerox Ladies and gentleman, here it is the Spontaneous Symmetry Breaking.
19:52 TimToady by the way, my son Aron just had his first physics paper published: http://www.arxiv.org/abs/hep-th/0702124
19:52 moritz Aron++ ;)
19:53 obra TimToady: congrats.
19:53 xerox What a title!
19:53 obra TimToady: What's his Erdos #?
19:53 TimToady dunno, I'll have to ask him
19:53 moritz obra: physicists don't care, only mathematicians do
19:53 TimToady I used to work with someone whose number was 1
19:54 moritz (or at least I think so)
19:54 obra TimToady: did you co-publish with him? ;)
19:54 TimToady but never published anything with him, alas
19:54 obra alas.
19:54 obra moritz: it varies
19:54 moritz obra: ok ;)
19:55 obra my girlfriend is a physicist by training and she cares :)
19:55 moritz I know a computer science guy with erdül 1 - I never met anyone else who was more crazy ;)
19:55 * obra ponders an April 1 hoax to get citeseer to add erdos as the author on every paper
19:56 TimToady I've just asked him if any of his coauthors know their Erdös number.
19:56 moritz he once hold a lecture in ancient greek and afterwards told his collegeas: "It did'nt make a differnce" ;)
19:56 moritz obra: cool idea ;)
19:56 CardinalNumber is now known as ProperNoun
19:56 obra http://www.oakland.edu/enp/ErdosA and http://www.oakland.edu/enp/thedata.html For those playing the home game
19:57 obra Only two Walls with a # <= 2.
19:57 obra Wall, Curtiss E.
19:57 obra Wall, James R.
19:57 TimToady I guess that's Erdős instead
19:57 obra ...now I want to know how many CPAN authors have co-authored with audrey.
19:58 moritz TimToady: your son isn't a professor by any chance?
19:59 moritz TimToady: usually the last author on a paper is the prof who did nothing but nod and said "good work, folks" ;)
19:59 moritz (or at least in germany, that is)
20:00 Caelum joined perl6
20:00 TimToady no, he's the grad student slave labor who wrote a lot of the paper.
20:03 obra moritz: in the US, usually that prof gets first billing.
20:06 elmex joined perl6
20:14 moritz obra: how sad ;)
20:16 TimToady yeah, we screw up the dates too.  :)
20:18 jisom and measuring distance too
20:18 TimToady at least the traditional European dates are consistently little endian.
20:18 jisom and weight
20:18 TimToady though I think I prefer big endian dates anymore
20:18 TimToady 2007/12/25
20:18 blueagnes joined perl6
20:18 jisom and if you type 01/02/03 you have no clue when it is :-p
20:19 jisom big endian, little endian, or mixed like the US
20:35 Xaduha joined perl6
20:38 ashelyb_ joined perl6
20:41 blueagnes left perl6
20:45 as joined perl6
20:46 moritz @tell rindolf on your homepage on the decss page there is at least one broken link
20:46 lambdabot Consider it noted.
20:48 miyagawa joined perl6
20:48 mncharity joined perl6
20:48 mncharity is now known as putter
20:50 TimToady @mess
20:50 lambdabot Maybe you meant: messages messages? msg
20:50 TimToady @moosages
20:50 lambdabot putter said 15h 55m 58s ago: re extensibility of the p6 grammar, one needs to be able to retract as well as extend, no? So one can play games like starting with the quick and dirty, or stable,
20:50 lambdabot version of token/rule/regex x, and then others selectively override it with more real, or experimental, variants.
20:52 buubot joined perl6
20:53 putter *startlement*  '@moosages" worked!?!?!
20:53 putter hack attack time.  let's see if the neurons are up for it.
20:54 amnesiac joined perl6
20:54 allbery_b sure.  lambdabot has(limited) edit distance heuristis
20:55 putter is there really no Filter::WithSub (or somesuch)?  it seems like an obvious good thing.
20:55 integral joined perl6
20:56 TimToady you mean grep?
20:56 putter ... use Filter::WithSub INSERT,sub{join(",",map{...}@mumble}; ...
20:58 putter ... use Filter::WithSub sub { filter::simple code to create a bit of abstraction in this file };
20:58 putter wraps nice #line around it.
20:59 putter use Filter::WithSub method=>'mumble', debug_to=>STDERR;
20:59 polettix joined perl6
21:01 putter there's Macro  http://search.cpan.org/~segv/Macro-0.2/Macro.pod
21:01 lambdabot Title: Macro - Simple code templating mechnism - search.cpan.org
21:03 putter basically I need some better metaprogramming for p5, or doing nontrivial ast's will make me run away.
21:03 putter TT: btw, for myself, feel free to use simple text macros in STD.  :)
21:04 weinig|bbl is now known as weinig
21:08 TimToady well, that's essentially what the PrecOp role is doing currently...
21:09 putter k
21:09 TimToady and the return-type coercions could be copied down to the {*} stub.
21:10 TimToady pugs supports neither parameterized roles nor &.() methods yet.
21:10 tene if you really need those in pugs, check out a revision from the future.
21:11 TimToady well, sure, but the checkout takes longer than just waiting.
21:12 putter (the more developers you have sitting around twiddling thumbs waiting for next release, the longer it seems to take...)
21:12 putter oo, there's an idea...
21:14 [particle] ...hire some monkeys to code haskell?
21:15 jrockway i don't think monkeys are able to inderstand polymorphic type classes
21:15 putter I was hoping for a unicode "thumbs up" glyph, so we could have a thumb twigle.  :)
21:17 putter Twiggling our thumbs.
21:19 tene Heh.
21:20 putter Ok, I hallucinate a fuzzy vision of a componentized p6 impl on p5.  so assorted folks can push things in assorted directions, without the past/current constraints of their being "one big thing" (which must not be broken, etc).
21:20 putter s/their/there/
21:21 putter not sure if it can be usefully brainstormed without a whiteboard.  :(
21:21 putter big downside of non-local development for me.
21:22 putter @tell stevan Was talking to someone at boston-pm about Moose.  any interest in doing the proverbial boston-area p6 users group meeting?
21:22 lambdabot Consider it noted.
21:22 putter (there would be a whiteboard there... *wistfully*)
21:23 moritz putter: like... a wiki?
21:24 putter :)
21:25 putter nifty ("nifti"?), but... for a complex conversation, you really need to be in the same room, to handwave at each other will waving at a whiteboard.  don't know of any substitute with current tech.  the video wall stuff I guess.
21:25 putter s/will/while/
21:38 putter btw, is it just me, or does this seem odd:  rule x / {env $x=3;return 'foo';} /; rule y /{env $x =2} <x> {$+x is now 3}/;
21:39 putter the "return isn't really exiting the stack frame" thing... cognitive dissonance.
21:42 Caelum joined perl6
21:44 stevan_ joined perl6
21:45 Aankhen`` joined perl6
21:47 * putter attempts to local()ize a symbol table entry.  eg, A::new.
21:48 putter local *A::new = \&B::new;
21:48 putter woot
21:50 putter so we can have a p5  Perl6::Match  for instance, which individual implementations can say "for Match, please use  Pugs::Junk::ImplB::Match'), which should @ISA Perl6::Match, and it all just works.
21:53 CardinalNumber joined perl6
21:54 KD-Misafir311 joined perl6
21:54 putter still, it would be nice to have a scoping mechanism local(), but only in some specified set of package/lexical scopes.  p7.
21:55 KD-Misafir311 is now known as nurii
21:58 putter sort of "just works".  really want to localize the ISA of Perl6::Match.  but that does fly with current p5 impl.
22:00 putter anyone see any problems with local()ing new() to permit a standard set of names (Perl6::Foo) to be used by everyone, despite there being multiple implementations?
22:01 putter Their .class will have to lie, but that was needed anyway.
22:07 putter In p6, a p5 package Foo is named perl5::Foo.  Any objections to a p6 Bar being named perl6::Bar in p5?
22:12 putter @seen audreyt
22:12 lambdabot audreyt is in #perl6 and #haskell. I last heard audreyt speak 1d 7h 8m 29s ago.
22:12 CardinalNumber is now known as ProperNoun
22:31 Psyche^ joined perl6
22:32 putter @seen gaal
22:32 lambdabot gaal is in #perl6 and #haskell. I last heard gaal speak 16h 11m 34s ago.
22:32 kolibrie putter: in november, clkao and audreyt at the prodding of cj? suggested a boston meeting in early May
22:32 kolibrie if we can get dates, I may swing up
22:33 kolibrie stevan also expressed interest
22:36 * kolibrie cycles home
22:43 putter kolibrie: awesome.
22:43 putter but... 2+months away...
22:45 putter beautiful sunset.  cloud cap 15deg up in the west, deep red, against windblown bare deciduous and waving tall pines.  :)
22:47 Psyche^ is now known as Patterner
22:48 putter its unfortunate, but the difference in visual complexity between   a.foo(...)   and  $a->foo(...)   (as a way of mixing sublanguages (here "a")),  has me wanting to make different architectural decisions. :/
22:49 SamB ???
22:49 jrockway joined perl6
22:49 SamB what does -> do?
22:49 tene putter: elaborate?
22:49 putter a.foo(b.bar("fribble",3),"up")   I can read, but
22:50 putter $a->foo($b->bar("fribble",3),"up")   not so much
22:50 SamB that just makes me think "Python"...
22:50 SamB for some reason...
22:51 putter sublanguages meaning domain specific languages.  eg, mixing PIL1 in with something else.  $pil1->stmts  vs pil1.stmts  pil.stmts  p.stmts
22:54 putter Hmm, mixed results in experiment re Tufte's "minimal ink" guideline: http://portal.acm.org/ft_gateway.cfm?id=​1125090&amp;type=pdf&amp;coll=&amp;dl=AC​M&amp;CFID=15151515&amp;CFTOKEN=6184618
22:54 lambdabot http://tinyurl.com/ywtjxn
23:06 lampus joined perl6
23:12 pjcj_ joined perl6
23:13 TimToady I suspect it's the > more than the - that interferes, since it's more vertical.
23:13 TimToady and it's more of an ascender
23:14 TimToady I'd say -> IS making THE same mistake AS putting your keywords IN all caps.
23:15 miyagawa joined perl6
23:16 TimToady infix operators should be huffman coded vertically.  :)
23:16 TimToady I don't think this applies so much to prefixes, however.
23:16 TimToady since you're not trying to connect two sides so much
23:17 TimToady (hence, I'm not too worried about sigils, which typically signal the start of a noun phrase.)
23:17 TimToady $x.foo.bar.baz()  has a nice shape.
23:18 TimToady especially when trying to pick out what will interpolate
23:22 putter :)
23:23 putter $a.foo($b.bar("fribble",3),"up")
23:23 putter a.foo(b.bar("fribble",3),"up")
23:24 mako132 joined perl6
23:24 putter could be.  interesting.  hadn't thought of it that way.
23:25 TimToady interesting that all our interpolaters are "strong" vertically
23:25 TimToady $ @ % & { \
23:25 TimToady and | for interpolating a caputre
23:25 TimToady *capture
23:26 TimToady though caputre is a cool word
23:26 putter lol.
23:26 TimToady caputré
23:26 putter and ./foo  blends away, atleast in my typeface
23:27 TimToady certain $.foo and .foo have a very different shape on the front.
23:27 TimToady s/<sp>/ly /
23:30 prism joined perl6
23:30 TimToady I was worried about the vertical visual parsing of 1,2,3 Z 4,5,6 until I realized that was artificial
23:30 TimToady and you usually have @foo Z @bar
23:31 Squern joined perl6
23:31 TimToady mixed with letters the Z stands out adequately, I think.
23:31 TimToady same for @foo X @bar
23:32 putter indeed
23:32 TimToady part of my unstating reasoning for going from XX to X
23:32 TimToady unstated even
23:33 TimToady I don't usually talk about half the things I think about...
23:33 TimToady but your bargraphs brought it back to mind
23:33 putter if you journal any of them, that might make for an interesting post/book/mumble sometime
23:34 TimToady I worry that those experiments are measuring the amount of time to produce a number, not an understanding.
23:34 putter yes indeed.
23:34 TimToady but it's hard to measure how well people integrate the areas under the "curve", and a lot easier to measure how long it takes them to say 12.2
23:35 TimToady the bars are there for "bulk" reasons that are more accessible to the backbrain than the front.
23:35 xinming joined perl6
23:35 putter hmm, there's a question.  one-laptop-per-child is getting closer to release.  what could one collect/write to facilitate graph _understanding_.  "graph as language".  hmm
23:36 TimToady much like some people still prefer analog clocks even though it's faster to say 6:36
23:36 TimToady but then they have trouble translating the numbers back to the "feel" of time.
23:37 putter and lots of folks liked my http://www.vendian.org/envel​ope/dir2/day_of_dots_clock/
23:37 lambdabot Title: A dot for every second in the day - a clock
23:38 krod joined perl6
23:38 putter objective was to help folks develop a feel for order of magnitude time.  ie, what's 10^5 seconds.
23:38 krod Could someone tell me the proper way of doing this: http://pastebin.ca/369643 ... I am trying to store the output of exec(uptime) somehow
23:38 buubot The paste 369643 has been copied to: http://erxz.com/pb/1512
23:39 putter that doesn't *look* like perl6...
23:39 allbery_b perl5 questions are best asked in #perl
23:40 putter `uptime`
23:41 jrockway maybe people think perl5.6 == perl6
23:41 putter anyone know if PIL2JS is still working at the moment?
23:42 putter "after all, the 5 is part of the language name, isn't it?  been there forever". :)
23:43 * allbery_b rolls his eyes
23:44 TimToady hmm, that does make me wonder whether foo1.2.3 should be consistenly parsed as a token.
23:44 TimToady v1.2.3 would then fall out of that
23:44 krod left perl6
23:45 TimToady maybe even foo.1.2.3
23:45 [particle] and v1.2.3b1.4_209f
23:46 putter strawman hypothesis:  if PIL2JS is working, or can be easily repaired, that seems a good way (the best way?) to drive development of a new front end.  take tests, emit old PIL1, if PIL2JS works on it no less well than with pugs -CPIL, you are golden.  can immediately use entire pugs t/.  without conflating runtime development (much).
23:47 putter thoughts?  questions?  "you forgots"?
23:47 * [particle] wonders what *-* is
23:47 xinming 06:32:52 < putter> objective was to help folks develop a feel for order of magnitude time.  ie, what's 10^5 seconds.
23:47 xinming 06:33:04 < krod> Could someone tell me the proper way of doing this: http://pastebin.ca/369643 ... I am trying to store the output of exec(uptime) somehow
23:47 buubot The paste 369643 has been copied to: http://erxz.com/pb/1512
23:47 putter re token, hmm.  no opinion.  don't really understand the space
23:47 TimToady the dumbell operator?
23:48 putter err, the design space :)
23:48 xinming Oops. sorry.
23:48 TimToady well .1 can never be a method call.
23:48 xinming Mis click
23:49 TimToady it either has a number on the front or something else.
23:49 putter hmmmm
23:49 TimToady so if it has a non-number, we could treat them all consistently.
23:50 putter .1  is 0.1 or $_.1 ?
23:50 TimToady could also be a canonical namemangling for things like infix:<+> somehow
23:50 TimToady at the moment bare .1 is outlawed.
23:50 putter k
23:50 TimToady er, no, that was 1.
23:51 TimToady .1 is 0.1
23:51 TimToady least surprise
23:51 stevan__ joined perl6
23:51 putter but a bit of a gotcha here
23:51 TimToady but foo.1 can't be a method call
23:51 TimToady a listop requires a space, foo .1
23:52 putter ^.^
23:52 putter (eyebrow non-method call)
23:54 xinming TimToady: I'd suggest 0.1 for number. using .1 form is not as good as it should be. At least. I don't think people will think It's tedious for just a '0' more
23:54 putter oh, right.  confused.  eyebrows down.
23:55 TimToady xinming: you could be right, if only to prevent people from wondering if .1 means $_.1 somehow.
23:55 putter no comments/questions on the strawman in the corner?
23:56 TimToady I don't understand really.
23:56 putter np
23:56 TimToady I just push arbitrary symbols around.
23:56 TimToady My mind is sort of a Chinese Room.  :)
23:56 putter :)
23:58 TimToady are you proposing a way of working on new frontends without knowing Haskell?
23:58 putter let's see...
23:58 TimToady and what do you mean by a "frontend"?
23:59 putter the objective is to have a "second" p6 parser.  mostly because

| Channels | #perl6 index | Today | | Search | Google Search | Plain-Text | summary

Perl 6 | Reference Documentation | Rakudo