Juerd Still, do you think it hasin any way to do with Perl? 00:01
Or would it be coincidental?
wolverian I _like_ to think that Perl is a language that suits people who have unconventional tendencies 00:02
I'm not at all sure if it's true :)
(well, actually, I am. but me being true doesn't make it any more true, if you get what I mean. this is what you get for being relativist.) 00:03
er. s,me being true,me being sure,
I need to drink less mountain dew.
Juerd hehe
A year or so ago, my parents found my website and read it. I didn't especially mean to avoid using my real name there, but I did not use it anywhere. 00:04
They asked me "Do you want us to call you Juerd?" 00:05
I was very surprised "No, let's not go that far"
Though sometimes I wonder why not.
wolverian mm.
what is your real real name?
Juerd Jorrit J. Waalboer 00:06
The J. stands for Jelle
wolverian dutch names are nice. :) 00:07
Juerd They're not Dutch 00:11
They're Frisian
So is Juerd
wolverian oh, I'm sorry.
Juerd I'm not Frisian though
Neither are my parents
wolverian ah. educated myself a bit over at wikipedia
Juerd Good night 00:14
wolverian good night. 00:15
kolibrie seen nothingmuch 00:17
jabbot kolibrie: nothingmuch was seen 2 hours 39 minutes 56 seconds ago
wolverian I think I fell in love with ruby. (again) 00:46
svnbot6 r5131, Stevan++ | Perl6-MetaModel : Airport hacking; plane hacking; then some bus-back-to-home hacking; .... hmmm I think I need to install SVK tomorrow. Oh yeah, and I cleaned out the ::Class junk, changed the interface a bit to conform to the real Perl6 meta model (per AS12) added a test and started down the path to MetaRoles :) 01:33
putter wolverian: any interest in working on a p6 implementation of ruby core? 01:34
mugwump stevan: still in transit?
wolverian putter: hmm. maybe. I don't have much tuits though. 01:36
nothingmuch perl: "Easy things are easy, hard things are possible" 03:03
haskell "
haskell: "Hard things are easy, the impossible just happens"
-- autrijus
autrijus corrects: "hard things are easy, impossible things just happen" 03:04
autrijus++
i think i like the paraphrase "hard things are easy, the impossible just happenned" the most, since it's actually quoting, not paraphrasing GHC 03:05
wolverian I think that in perl, the easy things are more convenient than in haskell, but I might be biased. 03:08
putter wolverian: k. my current thought is once namespaces work a bit more, to do a dump of the core with {...} bodies, which can be incrementally fleshed out. tests, and core .rb files, can hopefully be, at least partially, automatically transliterated. 03:09
wolverian putter: hm. I'm a bit hesitant about using perl6, yet, actually. I hate running into pugs's limitations. 03:10
autrijus wolverian: sure, in haskell the easy things are incovenient ;) 03:11
wolverian autrijus: heh. :)
autrijus it's all about conservation of complexity
nothingmuch since haskell defines less 03:12
but that less is more reusable, and intercombinable
is that a word?
wolverian isn't combinable enough? :)
nothingmuch the hard things are easy, because there are less details to take care of
combinable with itself
there are less special cases in the language def
s/less/fewer/
wolverian do you mean the syntax or semantics?
nothingmuch semantics 03:13
syntax is nothing
wolverian right.
what do you mean with special cases, then?
nothingmuch autrijus has a wifi antenna on his head, so he does it need syntax
wolverian I think I have an idea what you mean, but it's hazy
nothingmuch every operator in haskell is a first class function
autocurrying is a generic approach that gives you both emulation of multiple params
svnbot6 r5132, chromatic++ | r5198@windwheel: chromatic | 2005-06-30 11:43:55 -0700
r5132, chromatic++ | Allow a plan to run zero tests.
r5133, chromatic++ | r5199@windwheel: chromatic | 2005-06-30 11:46:41 -0700
r5133, chromatic++ | Fix name collision between plan() method and $.plan attribute.
r5133, chromatic++ | Manually pass args to super new() as slurpy params don't yet work.
r5133, chromatic++ | Pass number of results to testplan footer() in DESTROY().
r5134, chromatic++ | r5200@windwheel: chromatic | 2005-06-30 11:49:41 -0700
r5134, chromatic++ | Launch an external test to test DESTROY(). 03:14
r5134, chromatic++ | Test plan setting.
nothingmuch and consistency with the fact that everything is 1 param
wolverian chromatic steals karma!!!
nothingmuch wolverian: you can do that too.
wolverian yes, I know :)
nothingmuch wolverian: anyway, my point is this: haskell is very simple
wolverian forgive me, I've been up all night, waiting for the morning to come so I can go to the doctor.
nothingmuch so simple that complex structures (like monads) are optimizable.
and that many things are safe 03:15
autrijus zzz &
nothingmuch so you can do crazier combinations
and things are sane
he's lying
wolverian nothingmuch: but it moves the complexity elsewhere.
nothingmuch wolverian: but complexity scales down slightly by that transformation
the curve of code complexity vs. problem complexity has a lower ratio
putter 'night autrijus. 03:16
obra evening lambdacamels 03:18
nothingmuch hola obra
obra how's canadia, nothingmuch
?
nothingmuch it's nice 03:19
putter wolverian: re limitations, agreed. a downside is drawing ruby community attention while a) oo pugs is still a work in progress, and b) grammars aren't up to doing ruby, so its "ruby-like p6", rather than a "p6 backend for ruby".
nothingmuch obra: are you already home 03:20
at home? 03:21
wolverian putter: yes. that's why I'm hesitant. I'm thinking about it.
putter upside is much of the grunge work of porting core could be done now, for some value of now, leaving, and clarifying, where things arent ready yet.
I'm tempted to just keep ripper-based transliteration, a ruby grammar, and a p6 core, "simmering" but not checked in, until oo matures enough that all 4 are "almost within reach". Ie, punt for now. 03:23
wolverian: doctor? 03:28
putter hopes it's nothing too serious...
obra nothingmuch: yes. drove 10 hours today 03:29
wolverian putter: just a test for a suspected slight thyroid gland thingie.
(sorry if that was too much information :) 03:30
QtPlatypus Is there anyway of doing :d on a file? 03:36
wolverian wolverian stares at QtPlatypus 03:38
lick your nose on a file?
QtPlatypus No dump the parse tree for this file (in pugs0 03:39
)
wolverian oh.
I dunno.
putter ah, np, hope it goes well. (standard one-liner advice #001: have someone along to serve as a patient (you) advocate. there is so much potential for... cruft/complications/etc, when there isnt a "project manager" along) 03:41
Daniel_Nee Hi, Autrijus: If you hear me, could you please let me know where are you and when will you be back, thanks. 03:47
putter 'night & 03:49
nothingmuch Daniel_Nee: autrijus says he will be back in 72 hours 04:09
he is still in toromnto
nothingmuch just wrote UNIVERSAL::isa 04:10
(a module, not a function)
=head1 NAME
UNIVERSAL::isa - Hack around stupid module authors using UNIVERRSAL::isa as a
function when they shouldn't.
wolverian haha. 04:11
what does it do?
nothingmuch on the cpan shortly 04:12
short answer:.
replaces UNIVERSAL::isa with a version that does a check on the first argument
if it is an object, and the object overrides 'isa', then a call to 'isa' as a method on the object is made
in all other scenarios the original 'UNIVERSAL::isa' is goto'd
sili_ interesting. 04:13
wolverian nice.
nothingmuch i hope chromatic will make Test::MockObject use it
wolverian can you make it output a warning like 'read perldoc -f eval' now and then randomly? 04:14
nothingmuch eval?
oh, like for eval { } ...?
wolverian yeah.
nothingmuch ah
well, they can just 'use UNIVERSAL::isa qw/isa/'
and then use it as a function
it's still safe
wolverian indirect object notation overrices? 04:15
er. RIDES.
nothingmuch btw, for fun regarding 'eval {}', read the source of pure perl Scalar::Util::bless and overload::StrVal
uh, i doubt it
if there is a symbol in the current package, it should be a function call 04:16
(i think)
it basically does a bless check for you
wolverian right.
nothingmuch so the function form is OK
wolverian I just read it as 'use UNIVERSAL qw/isa/' and was confused 04:17
sorry. :)
nothingmuch ah
that would also work
as long as UNIVERSAL::isa (the module) was 'use'd beforehand
wolverian why?
nothingmuch uh, sorry 04:34
i lost myself
because UNIVERSAL::isa is overridden
and the original is only stored in a lexical afterwords 04:35
it is no longer available by name.
i'm off to bed
read the code, it is up on the cpan
or actually 'it will be soon'
coke__ coke__ smacks Damian.
nothingmuch pause.perl.org/pub/PAUSE/authors/i....01.tar.gz 04:38
coke__: huh?
coke__ contextual::return 04:40
nothingmuch i so don't want to know what that means at this hour 04:42
svnbot6 r5135, geoffb++ | Added ext/DateTime and lib/Code-Perl to hack.pod 05:14
CLoCkWeRX o/t question, re chip [sorry guys, you're probably getting it alot]... how big of an effect does this trouble in his life have upon the release cycle of perl 6? 05:41
gaal_ good morning 05:55
geoffb . . . and good night 05:57
geoffb passes off the "awake" baton to gaal_
gaal_ gaal_ will accept it in a few minutes, after making coffee :)
gaal_ wonders if pugscc is something autrijus can fix while asleep... 06:20
who are all the people on ingy's slides? I recognize some... 06:33
wolverian I'm a recluse, I wouldn't know. :/ 06:36
gaal_ in yapc.kwiki.org/data-driven-testing/slide19.html - when does the http hit actually happen? when you say $block->url, or when you perform the like? 06:39
(does like now accept a closure to be run?)
hahahahahaa! the slides leading up to 26 are quite amusing 06:41
ingy++
wolverian yes. 06:42
:)
where does he get the like() sub from!!! 06:43
gaal_ his Test::Base, of course
wolverian the CPAN version only has run_like() 06:44
(or I'm blind)
gaal_ point
hhhmmm, no, search.cpan.org/src/INGY/Test-Base-...st/Base.pm and look at test_more_exports 06:45
wolverian I'd say that needs documentation!
s,!,, # too much caffeine
gaal_ looks like he's pulling the haskell reexport trick :)
wolverian what is that? I'm completely missing the point of the BEGIN 06:47
gaal_ he imports things on the use line of Test::More
so he needs @test_more_exports populated
at *compile* time 06:48
i didn't know you can pass an arrayref to T:M's import though!
wolverian oh, he's using the array there. yeah. 06:49
gaal_ one place lexical scope results in awkward usage is when you want to preload data in a BEGIN block. i've never liked how that forces you to declare the var outside the block. 06:50
i wonder if p6 will have something like our $OUTER::var :-)
wolverian heh. 06:51
gaal_ not that that's much prettier!
wolverian well, the fact that it works at all is somewhat mind-boggling.
"look, I'm using this variable now, but I promise I'll declare it later!"
gaal_ oh? no, the declaration happened before
look one line above the BEGIN block
remember my is compile time too 06:52
wolverian oh, I thought BEGIN is run before that happens
yeah. okay. :)
thanks for unconfusing me.
gaal_ if that like were to be placed after the BEGIN, you'd get a compilation error
actually, that's part of what makes-perfect-sense-but-gives-awkward-usage:
you can't just put your BEGIN block on the top of your code for everyone to see, you're forced to declare a bunch of stuff before it 06:53
wolverian that's the same as how you have to declare your subs before using them if you don't like to sprinkle your code with useless parentheses 06:54
(admittedly you don't have to _define_ them beforehand, but still..)
gaal_ in p5 i got to always using parens on sub calls because of that 06:55
wolverian will p6 fix it? :) 06:56
gaal_ it's a little early to write a perl 6 style guide though :)
well, there are no barewords in p6...
wolverian exactly my thoughts
gaal_ oooof, how to Show a TVar in a reversible manner? 07:00
wolverian wolverian hides under his admin desk and pretends not to hear the scary person 07:02
gaal_ gaal_ pulls out the Tesla coil 07:03
07:48 gaal_ is now known as gaal
dudley dudley is having a baby 09:37
Or, more accurately, my wife is ; )
meppl youre a harry potter character? :-P 09:45
QtPlatypus dudley: Congratulations. 11:01
dudley QtPlatypus: Thanks! 11:02
pjcj dudley: right now, or in six to seven months? 12:30
and congratulations again
dudley Right now. Actually, we're off to the midwife right now... I'll probably be back in a bit, though. 12:31
pjcj wow!
I seem to remember cwest giving realtime commentary on irc during the the event
though I'm not sure how good an idea that is 12:32
dudley I don't think my wife would appreciate it... I'll see, though ; )
pjcj good luck! 12:33
13:49 Yaakov_ is now known as Yaakov, coke__ is now known as Coke_, Coke_ is now known as _Coke_ 14:38 sbkhh is now known as Odin-
nothingmuch good morning 15:06
too many dreams in one night
kolibrie dreams of good things? 15:09
Limbic_Region is this news to anyone besides me? perlmonks.org/index.pl?node_id=471746 15:24
sili can't say i expected more 15:25
autrijus lexer and parser needs to be one-pass; that's one of the main points in my hw2005 paper 15:29
I don't think that's a problem really 15:30
but if damian promised otherwise, it's (again) wishful thinking
rehi, btw :) 15:31
kolibrie the compiler and interpreter are still quite separate, correct? 15:32
stevan hey kolibrie 15:34
nothingmuch++ # for UNIVERSAL::isa
_Coke_ _Coke_ answers somethign at the monastary.
Limbic_Region autrijus - what continent are you on today?
kolibrie hey stevan 15:35
autrijus Limbic_Region: still in toronto; 18hrs till going to airport 15:45
_Coke_ autrijus: gah. is it still 100 degrees in the shade? 15:48
nothingmuch morning 15:49
autrijus _Coke_: I don't know, I'm always in a room
meppl good morning nothinmuch 15:52
stevan I dont know about you guys, but sleep deprivation is hitting me like a brick 15:54
I am only just now realizing how little sleep I actually got over those 8+ days 15:55
autrijus stevan: //hand
kolibrie I did okay yesterday, but I'm having trouble today
autrijus stevan: 21 days for me
autrijus is having trouble waking up
stevan autrijus: you are young though, you will bounce back :)
autrijus very true :)
kolibrie stevan: hopefully we'll bounce back, too... 15:56
stevan kolibrie: yes, but not without permanent damage, we are too old :P
autrijus: I realized for the first time the true value of SVK, when I wanted to commit, but was in the airport/plane/bus 15:57
autrijus :)
stevan stevan is downloading it now
autrijus :D
stevan++
shower &
clkao clkao giggles at stevan 15:58
_Coke_ autrijus: doh. I forgot about eval::tcl 15:59
pjcj r5555, autrijus++ | some major submodule hacked whilst eating, sleeping and showering 16:00
stevan clkao: svk.elixus.org/ seems like it is down? 16:01
can I install from CPAN?
autrijus sure can 16:02
stevan cool, I will do that then :)
nothingmuch hola stevan
stevan guten tag nadda mucho
autrijus stevan: rt.openfoundry.org/Foundry/Project/...1.00-2.dmg 16:03
nothingmuch stevan: try again
it wasn't working for me a few times, and then it started
autrijus stevan: that's .dmg, use it :)
stevan autrijus++ # making my life much easier :)
nadda mucho
that work?
nothingmuch it worked for me 16:04
pdcawley Or possibly more complex. After all, what would you be doing if Pugs didn't exist?
stevan pdcawley: Pugs makes life interesting
complexity ~~ fun
_Coke_ perlbot pmichaud? 16:05
nothingmuch breakfast time 16:06
kolibrie um, lunch time? 16:09
stevan kolibrie: he is still in canada, eh
kolibrie stevan: believe so, till tomorrow 16:10
stevan: oops, forgot the 'eh'
stevan :) 16:11
pdcawley Not denying complexity ~~ fun, but I am questioning complexity ~~ easier
stevan pdcawley: I think pugs has definitely made my life more complex, but it also gives me a great amount of satisfaction, which makes me feel all warm and fuzzy inside, which makes life more bearable, which means that I can trudge through $work much easier (or at least with much less psychic wear-and-tear) 16:14
so in a very round about way ....
but in this particular case, I was refering to the link to the SVK dmg file :)
kolibrie kolibrie wonders if he must re-checkout before switching to svk 16:15
stevan pdcawley: BTW - the metamodel is now in fairly decent shape, if you have the time, comments appreciated
kolibrie: IIRC you are not re-checking out, but mirroring the entire repo 16:16
kolibrie so rm -rf pugs/; svk ...
autrijus svk co svn.openfoundry.org/pugs/
integral you don't have to mirror the whole lot, it takes ages. you want svk sync's --skipto option