00:00:05 <ehird> Let's see... Universal Records and Warner Bros.
00:00:16 <AnMaster> ok fun, freedb have three entries for this. All have slight misspellings, but the average would be correct...
00:01:01 <ehird> "HA HA, I thought this article would be about a bunch of crazed fanatics trying to make Ubuntu dependent on Haskell.
00:01:03 <ehird> http://www.fsf.org/news/dont-depend-on-mono"
00:01:07 -!- M0ny has quit.
00:01:09 <ehird> 1/10 F----- would not be trolled again
00:02:17 <AnMaster> musicbrainz has nothing on that cd however
00:02:38 <ais523> hmm... would Ubuntu depending on GHC bring out the same people who get angry at Mono dependencies, I wonder?
00:02:48 <ais523> or just the people who get angry at anything depending on GHC because it's so hard to compile?
00:03:09 <ehird> I don't think GHC has any supposed patent problems........................
00:03:17 <ehird> Or indeed any corporate interests.
00:03:27 <ais523> yes, but it's from Microsoft!!!!!!1one1
00:03:29 <ehird> Unless you think the Industrial Haskell Group is an evil illuminati.
00:03:33 <ehird> ais523: it's not even that
00:03:41 <ehird> some of the contributors happen to work for microsoft
00:03:45 <ehird> ghc isn't even their work
00:03:50 <ehird> and haskell and ghc have existed long before that
00:03:57 <ais523> yes, but you know how far people can take this sort of thing
00:04:27 <ehird> MONO: NOW YOUR FREE SOFTWARE GNU/LINUX OPERATING SYSTEM CAN GET STDS TOO!!
00:04:29 <AnMaster> ais523, is ubuntu going to require ghc as part of the core/system/whatever set of packages?
00:04:40 <ehird> AnMaster: Reading comprehension: 0%
00:05:04 <AnMaster> ehird, I wasn't sure if it is was something actually happening, or a "what if" scenario
00:05:29 -!- immibis has joined.
00:05:29 <ehird> Considering GHC is on 6.8.2 and 6.12 is almost out, I very much doubt it will happen any time soon.
00:05:53 <ais523> AnMaster: it's very unlikely, Ubuntu hardly ever puts compilers in core
00:05:59 <ais523> it ships with gcc, but only to compile kernel modules
00:06:10 <ais523> it's missing all the headers apart from kernel headers in a default install, for instance
00:06:15 <ehird> ais523: hey, Python is a bytecode compiler!
00:06:40 <ehird> [[More than 500 staff at Keihin Electric Express Railway are expected to be subjected to daily face scans by "smile police" bosses. ]]
00:06:43 <ehird> http://www.telegraph.co.uk/news/worldnews/asia/japan/5757194/Workers-have-daily-smile-scans.html
00:06:44 <ais523> ehird: the exception's interps for scripting langs, whether they're implemented in terms of compilers or some other ways
00:07:00 <ehird> Hmm, Brave New World might be a more appropriate reference.
00:09:20 * ais523 tries to figure out why 'Yahoo! Mail' writes its own name in single quotes
00:09:27 <ais523> generally speaking, web pages don't quote their own names...
00:09:33 <immibis> Because it's only pretending to be Yahoo! Mail?
00:09:46 * immibis wonders why GMail is still in beta after this long
00:10:02 <ehird> immibis: the google engineers get a cheap laugh out of it? :-)
00:10:13 <ehird> i remember in 2004, when we did scrabble for an invite, we did!
00:10:16 <immibis> Maybe you're supposed to shout 'Yahoo! Mail!' when you get mail?
00:10:18 <ehird> uphill! BOTH WAYS!
00:10:26 <ais523> at least there's no snow
00:11:40 -!- CESSMASTER has quit ("Computer has gone to sleep").
00:13:06 <pikhq> ehird: By "major label", you mean "member of the RIAA".
00:13:20 <ehird> No, I mean "major label" :P
00:13:24 <pikhq> There's only like 4, and they have a *lot* of subsidiaries.
00:13:27 <ehird> There are many non-major labels in the RIAA.
00:13:38 <ehird> pikhq: http://en.wikipedia.org/wiki/List_of_RIAA_member_labels
00:13:41 <ehird> there are far more.
00:13:41 <pikhq> Those are subsidiaries of the major 4.
00:13:48 <ehird> not just subsidaries
00:13:51 <ehird> those are marked hierarchically
00:14:11 <ais523> I suspect the label on my T-shirt is not a member of the RIAA
00:14:15 <ais523> but then again, it's hardly major
00:14:20 <pikhq> Oh, hey. That listing is from the RIAA website.
00:14:28 <pikhq> Which is known to be a blatant bunch of lies.
00:14:28 <ehird> ais523: wait, that's a joke right? :P
00:14:58 <pikhq> Oh, that lists the lies.
00:15:11 <ais523> ehird: yes, in a way that expresses my annoyance at people redefining words to mean something else
00:15:22 <ehird> ais523: are you fuckin' serious? :)
00:15:31 <ehird> damn those kids and their modern musicajig!
00:15:42 <ehird> The term "record label" originally referred to the circular label in the center of a vinyl record that prominently displayed the manufacturer's name, along with other information.[1]
00:15:43 -!- BeholdMyGlory has quit (Remote closed the connection).
00:15:46 <ehird> so it definitely is directly related.
00:15:59 <ais523> it's just the elision that annoys me
00:16:13 <ais523> sort of, it's like taking a specific sort of something, say "high-level language"
00:16:17 <ais523> and abbreviating it to just "language"
00:16:24 <ais523> that's just really misleading, and ridiculous
00:16:26 <ehird> it's called context
00:16:39 <ais523> I know; I just don't like languages to have to rely on context
00:16:48 <ais523> ok, so it's normally obvious what someone means
00:16:53 <ais523> it still takes more thought to parse, though
00:16:57 <ehird> ais523: are you serious? even lojban relies on context
00:17:07 <ehird> you want to exponentially inflate the length of every utterance?
00:17:09 <ais523> ehird: I'm not saying it shouldn't exist at all, it's useful
00:17:23 <ais523> I'm just saying that context is probably used a bit more than it ought to be atm
00:17:28 <AnMaster> out of 8 cds tried so far, muicbrainz had 4.
00:17:50 <AnMaster> freedb had all, but not as nicely formatted track titles and such.
00:17:58 <AnMaster> often several variants with misspellings
00:18:03 <ehird> AnMaster: your experience is highly abnormal. do not generalize it.
00:18:16 <AnMaster> so I guess, use musicbrainz if possible, fallback on freedb
00:18:32 <ehird> how about use musicbrainz or contribute?!
00:18:36 <AnMaster> ehird, translation: it isn't pop or mainstream
00:18:45 <ehird> i listen to plenty of obscure music
00:18:49 <ehird> musicbrainz is incredibly comprehensive
00:18:53 <ais523> ehird: AnMaster: both of you have a point here
00:18:53 <AnMaster> ehird, maybe. If I have time to.
00:19:12 <AnMaster> what with the long instructions for formatting the classical music entries
00:19:22 <ehird> i'm terribly curious where this busy AnMaster time goes to. he never seems to do much.
00:19:27 <ehird> sounds like an excuse to me.
00:19:36 <ais523> ehird: he may have a life completely separate from this channel
00:19:42 <AnMaster> ehird, atm? Reading up on theory for driving certificate
00:19:50 <AnMaster> that is where most of my time goes currently.
00:19:52 <ehird> ais523: he must carry around a portable IRC client at all times, then.
00:20:12 <ais523> ehird: no, he has a bouncer
00:20:14 <ehird> (well, I do too; it's called an iphone :P)
00:20:19 <ehird> ais523: i'm talking about talking
00:20:36 <AnMaster> I often live this "separate life" in the same room
00:20:43 <AnMaster> so I'm always within reach of the computer
00:20:43 <ehird> while talking. on irc.
00:20:50 <ehird> which is clearly a time that is possible to use musicbrainz too
00:20:55 <AnMaster> not always of course, but quite often
00:21:58 <AnMaster> ais523, what was this point we both had?
00:22:23 <ehird> nothing, AnMaster. you're 100% right.
00:22:24 <AnMaster> anyway, the more mainstream classical music is indeed on musicbrainz
00:22:33 <ais523> AnMaster: ehird's point is that most people don't have trouble, yours is that it's inappropriate for the way you use things
00:22:45 <AnMaster> ais523, right. That is what I have been trying to say
00:22:51 <AnMaster> yet ehird refuses to accept it.
00:23:02 <ehird> 00:22 ehird: nothing, AnMaster. you're 100% right. 00:22 ehird: utterly
00:23:05 <ais523> AnMaster: ehird's pedantically making correct but irrelvant statements
00:23:12 <ehird> i'm agreeing absolutely
00:23:25 <AnMaster> ehird, didn't you forget the "~"
00:23:29 <ais523> ehird: I know, but you're doing it in a way that makes you look like you're disagreeing
00:23:37 <ais523> which can only possibly count as AnMaster-baiting
00:23:56 <ehird> okay, so if i disagree with you you argue
00:23:58 <ehird> if I agree with you
00:24:04 <ehird> you argue about whether I agree with you
00:25:12 <AnMaster> anyway, my goal is reached. All the music is tagged, When all 30 cds were done, only 14 needed to be done with freedb due to musicbrainz lacking it.
00:25:56 <AnMaster> I haven't ripped the other ~60 classical music cds yet...
00:26:09 <ehird> http://www.reddit.com/r/funny/comments/8ypih/this_is_awesome_a_personal_ad_in_graph_form/c0auucv Wow. It's undownmoddable.
00:26:24 <ehird> ais523: You can't upvote it or downvote it.
00:26:30 <ehird> There are no buttons.
00:27:36 <AnMaster> ehird, there are buttons there?
00:27:44 <ehird> No up or down vote arrows.
00:28:28 <ehird> it usually doesn't
00:29:18 <AnMaster> ehird, other issue with both freedb and musicbrainz
00:29:29 <AnMaster> this cd was released with the same disc but two different covers
00:30:03 <AnMaster> one in Swedish (NAXOS 8.554777S) and one in English (NAXOS 8.554777)
00:30:13 <AnMaster> so I want the track titles correct for that ;P
00:30:19 <AnMaster> of course, there is no way to solve it.
00:30:23 <ehird> blame the artist for multi-naming shit
00:30:29 <ehird> kraftwerk are worse
00:30:33 <ehird> they made ENTIRE NEW VOCALS for the songs
00:30:37 <ehird> in both german and english
00:30:41 <ehird> for the two markets
00:30:48 <AnMaster> ehird, but with fingerprinting you can tell them apart
00:30:49 <ehird> literally, translate the lyrics
00:31:00 <ehird> AnMaster: i know, but it means there's two albums for every name!
00:31:08 <AnMaster> while here the actual cd doesn't differ at all
00:31:26 <AnMaster> just it says "Sinfonia i ciss-moll" instead of "Symphony in C-sharp minor"
00:31:28 <ehird> I wonder if they translated Autobahn.
00:31:56 <ehird> "We drive drive drive on the motorway" just doesn't have the same ring to it.
00:32:03 <AnMaster> "Overture in D minor" vs. "Uvertyr i d-moll"
00:32:30 <AnMaster> for music: moll = minor, dur = major
00:32:47 <ehird> hmm it seems to be "We're driving driving driving on the motorway"
00:32:50 <AnMaster> oh and for "sharp" in music we add "iss" or "ess".
00:33:05 <ehird> well okay it's actually "Wir fahr'n fahr'n fahr'n fahr'n auf der Autobahn" but I'm translatin' with the help of the interwebs.
00:37:40 <AnMaster> "Filling in your e-mail address is completely optional. However if you don't fill it in, the editing features [1] of the MusicBrainz service will not be available to you."
00:37:51 <ehird> Well, fill it in then.
00:37:57 <ehird> I assume it's to do with the moderation service.
00:38:13 <ehird> I would advise you to quit complaining, it's not like BugZilla is any better.
00:39:09 <AnMaster> ehird, still tl;dr for that classcial music formatting faq
00:39:30 <ehird> You have the attention span of a /b/tard.
00:39:53 <ehird> AnMaster: http://wiki.musicbrainz.org/Classical_Style_Guide is this page really too long for you?
00:39:56 <pikhq> /b/tards have more attention.
00:39:57 <ehird> it's like 3 fucking screens
00:40:04 <ehird> 3.5, I just measured
00:40:15 <ehird> The current official one
00:40:15 <ehird> http://musicbrainz.org/doc/ClassicalStyleGuide
00:40:17 <ehird> is the same length
00:40:25 <ehird> It's really simple.
00:40:54 <ehird> Buy a new monitor and an attention span battery.
00:41:08 <ehird> Don't you read Terry Pratchett?
00:41:20 <ehird> His books are quite long. Indeed, hundreds of screens.
00:41:27 <ehird> Yet you cannot manage 5.
00:41:36 <ehird> AnMaster: durr it is impossible to transfer information from books hurr.
00:41:43 <AnMaster> ehird, due to lack of interest.
00:42:23 <AnMaster> ehird, what are these "puids"?
00:43:01 <ehird> And a second in Google gives: http://musicbrainz.org/doc/PUID
00:46:03 <ais523> ehird: wow, you type fast if you can google that fast
00:54:43 -!- Judofyr has quit (Remote closed the connection).
01:01:03 -!- FireFly has quit ("Later").
01:01:39 <ehird> http://imagechan.com/images/4a1f16e12a8f69e53ef19798b535eeb1.png
01:03:58 -!- inurinternet has joined.
01:04:48 <AnMaster> ehird, editing on musicbrainz, have you done it yourself?
01:05:11 <AnMaster> ehird, if not, are you aware of that it is like picard... completely backwards UI
01:07:04 <AnMaster> I can't figure out how to do it
01:08:28 <ais523> AnMaster: I think ehird's gone to bed
01:14:53 -!- Sgeo has joined.
01:19:58 <AnMaster> ehird, for two of those cds the "puid" things were missing from the db, so it couldn't auto identify them
01:20:13 <AnMaster> ehird, the other are genuinely missing however
01:21:46 -!- CESSMASTER has joined.
01:37:13 -!- oerjan has quit ("Good night").
01:56:14 -!- coppro has joined.
02:25:55 <immibis> does anyone know of a good wad editor for windows?
03:25:58 <oklodok> god, stop having talked already
03:30:11 <oklodok> ehird: Gab flubb dirpmoglaaaaaaa tubadinoshçtok. <<< glio eglo flog balg nlo mlog
03:31:43 <zid> help help they're speaking in tongues
03:36:49 <ais523> AnMaster: isn't it 4am where you are?
03:37:00 <oklodok> AnMaster: ehird, atm? Reading up on theory for driving certificate <<< i don't believe a human can learn to drive a car as stably as people do.
03:37:14 <oklodok> that's one of the things that make me feel soliptistic
03:37:44 <AnMaster> I finally tagged the music perfectly
03:41:24 <oklodok> i just woke up, went to sleep at 21:00
03:41:31 <oklodok> or maybe a few seconds later
03:42:06 <oklodok> something like 23-5 would be nice
03:42:06 <MizardX> Go and sleep another 5 hours :P
03:42:26 <oklodok> i slept 15 hours just the other night
03:42:44 <oklodok> also there was a 1 hour sleep without artificial interruption
03:42:54 <oklodok> my brain has issues i think
03:46:43 * pikhq starts Project Eulering, in Haskell
03:46:54 <pikhq> main = print $ find (and . (\x -> [x `mod` y == 0 | y <- [1..20]])) [1..]
03:47:02 <pikhq> That's a... Pretty slow piece of code there.
03:47:59 <EgoBot> help: General commands: !help, !info, !bf_txtgen. See also !help languages, !help userinterps. You can get help on some commands by typing !help <command>.
03:48:02 <EgoBot> languages: Esoteric: 1l 2l adjust asm axo bch befunge befunge98 bf bf8 bf16 bf32 boolfuck cintercal clcintercal dimensifuck glass glypho haskell kipple lambda lazyk linguine malbolge pbrain perl qbf rail rhotor sadol sceql trigger udage01 underload unlambda whirl. Competitive: bfjoust fyb. Other: asm c cxx forth sh.
03:48:17 <immibis> !haskell main = print $ find (and . (\x -> [x `mod` y == 0 | y <- [1..20]])) [1..]
03:48:27 <oklodok> tell me when you surpass me
03:48:39 <oklodok> maybe i'll get interested again
03:48:48 <pikhq> That'll be a while.
03:48:58 <pikhq> Oh, right. import Data.List
03:49:18 <oklodok> i doubt i've played more than a week or two
03:49:29 <oklodok> well i guess if you don't know haskell
03:49:34 <pikhq> ./project5 295.32s user 3.41s system 94% cpu 5:17.56 total
03:49:40 <oklodok> i don't know whether you do
03:49:50 <pikhq> I'm using it as an excuse to code more Haskell.
03:50:05 -!- Sgeo has quit (Read error: 104 (Connection reset by peer)).
03:50:06 <pikhq> Well, returns the right answer (embedded in a Maybe).
03:50:41 <oklodok> a got into it because of this other dude, but i think he's in finnish top10 nowadays, i just didn't know enough math back then
03:52:05 <pikhq> Of course, I could have just done some smarter math. XD
03:52:05 <oklodok> also because python is 100 times slower than most languages, some problems are substantially harder for it
03:53:14 <oklodok> that's a pen and paper problem
03:54:04 -!- MizardX has quit (Read error: 104 (Connection reset by peer)).
03:54:36 -!- MizardX has joined.
03:54:36 <pikhq> Yes, but it was a Haskell one-liner.
03:55:18 <oklodok> sure, i was just agreeing with you about coulding to have done smarter math.
03:56:17 <oklodok> i mean i also agree there's no need to do smarter math, because it's a one-liner anyway
03:56:43 <pikhq> First few have been trivial.
03:57:19 <pikhq> Project 4 was a whole 9 lines. ... Because I needed to define a function to test if something was a palindrome or not.
03:58:59 <oklodok> i think i made some sort of merging generators thing for 4
03:59:45 <oklodok> or maybe i just wrote some sorta one-liner because it's #4, and i'm recalling some other prob
04:01:55 <pikhq> main = print $ last $ sort . nub $ filter (palindromeP . show) $ [x*y | x <- [100..999], y <- [100..999]]
04:02:17 <pikhq> palindromeP is an exercise for the reader, and I'm thinking last $ sort . nub $ filter is dumb.
04:02:34 <EgoBot> userinterps: Users can add interpreters written in any of the languages in !help languages. See !help addinterp, delinterp, show | !userinterps. List interpreters added with !addinterp.
04:02:41 <EgoBot> Installed user interpreters: aol austro b1ff bc bct bfbignum brit brooklyn chef chiqrsx9p choo cockney ctcp dc drawl dubya echo ehird fudd google graph gregor hello jethro kraut num ook pansy pirate plot postmodern postmodern_aoler redneck reverse rot13 sadbf sfedeesh sffedeesh sffffedeesh sffffffffedeesh slashes svedeesh swedish valspeak warez yodawg
04:03:07 <Warrigal> !swedish The quick brown fox jumps over the lazy dog.
04:03:08 <EgoBot> Zee qooeeck broon fux joomps oofer zee lezy dug. Bork Bork Bork!
04:03:29 <Warrigal> !yodawn Zee qooeeck broon fux joomps oofer zee lezy dug. Bork Bork Bork!
04:03:38 <Warrigal> !yodawg Zee qooeeck broon fux joomps oofer zee lezy dug. Bork Bork Bork!
04:06:48 <EgoBot> Installed user interpreters: aol austro b1ff bc bct bfbignum brit brooklyn chef chiqrsx9p choo cockney ctcp dc drawl dubya echo ehird fudd google graph gregor hello jethro kraut num ook pansy pirate plot postmodern postmodern_aoler redneck reverse rot13 sadbf sfedeesh sffedeesh sffffedeesh sffffffffedeesh slashes svedeesh swedish valspeak warez yodawg
04:07:21 <EgoBot> http://google.com/search?q=test
04:07:30 <immibis> well that was useful...not...
04:30:45 <immibis> !sffffffffedeesh Hello people, I am swedish.
04:30:45 <EgoBot> Hellu peuple-a, I em svedeesh. Bork Bork Bork!
04:31:06 <immibis> !swedish Hellu peuple-a, I em svedeesh. Bork Bork Bork!
04:31:06 <EgoBot> Helloo peoople-a-a, I im sfedeesh. Bork Bork Bork! Bork Bork Bork!
04:31:16 <immibis> !sweedish Helloo peoople-a-a, I im sfedeesh. Bork Bork Bork! Bork Bork Bork!
04:31:21 <immibis> !swedish Helloo peoople-a-a, I im sfedeesh. Bork Bork Bork! Bork Bork Bork!
04:31:22 <EgoBot> Helluu peuuple-a-a-a, I im sffedeesh. Bork Bork Bork! Bork Bork Bork! Bork Bork Bork!
04:31:31 <immibis> !swedish Helluu peuuple-a-a-a, I im sffedeesh. Bork Bork Bork! Bork Bork Bork! Bork Bork Bork!
04:31:31 <EgoBot> Helloooo peoooople-a-a-a-a, I im sffffedeesh. Bork Bork Bork! Bork Bork Bork! Bork Bork Bork! Bork Bork Bork!
04:33:51 -!- MizardX- has joined.
04:34:30 -!- MizardX has quit (Read error: 104 (Connection reset by peer)).
04:34:48 -!- MizardX- has changed nick to MizardX.
04:58:19 -!- Associat0r has joined.
05:46:41 <oklodok> i think i should glio a pizza
05:56:12 <Gracenotes> is it just me or do people throw around the term 'deconstruction' way too much
06:02:20 -!- Pthing has quit (Remote closed the connection).
06:02:55 <augur> Gracenotes: i dont see people use it hardly at all
06:03:14 <augur> are you reading much postmodernist/poststructuralist word and/or work influenced by derrida?
06:04:25 <Gracenotes> more like in discussions where people are considering whether or not a work deconstructed a genre, people tend to conclude the affirmative a bit too much
06:08:52 <Gracenotes> I don't see much use in the term personally
06:09:28 <augur> well, it does have a use, but people probably dont know what it means, so.
06:11:46 <immibis> ^run wget --bind-address=127.0.0.1 http://google.com/
06:13:46 <ais523> err, is ^run even a fungot command?
06:13:46 <fungot> ais523: suppose i have ( equal? ( convert3 4 5 6)
06:14:02 <ais523> in that case, you probably want a different prefix
06:14:08 <immibis> `run wget --bind-address=127.0.0.1 http://googlw.com/
06:14:13 <immibis> `run wget --bind-address=127.0.0.1 http://google.com/ 2>&1
06:14:14 <HackEgo> --2009-07-07 05:14:13-- http://google.com/ \ Connecting to 127.0.0.1:3128... connected. \ Proxy request sent, awaiting response... 301 Moved Permanently \ Location: http://www.google.com/ [following] \ --2009-07-07 05:14:14-- http://www.google.com/ \ Connecting to 127.0.0.1:3128... connected. \ Proxy request sent, awaiting
06:14:29 <immibis> `run wget --bind-address=123.45.67.89 http://google.com/ 2>&1
06:14:30 <HackEgo> --2009-07-07 05:14:29-- http://google.com/ \ Connecting to 127.0.0.1:3128... failed: Cannot assign requested address. \ Retrying. \ \ --2009-07-07 05:14:29-- (try: 2) http://google.com/ \ Connecting to 127.0.0.1:3128... failed: Cannot assign requested address. \ Retrying. \ \ --2009-07-07 05:14:29-- (try: 3) http://google.com/
06:18:28 * immibis slaps myndzi immibis with an immibissuffix
06:18:29 * myndzi slapmyndziyndzmyndzimmibimyndzimmibissuffix
06:18:29 * immibis slapimmibisyndzimmibismmibimyndzmyndzisuffix
06:29:48 -!- ais523 has quit (Remote closed the connection).
07:27:53 -!- coppro has quit ("The only thing I know is that I know nothing").
07:41:47 -!- coppro has joined.
07:43:27 <oklodok> http://www.vjn.fi/oklopol/music/sevenfold.mid
07:43:51 -!- coppro has quit (Remote closed the connection).
07:45:04 -!- coppro has joined.
07:48:10 <oklodok> i like it, but it kinda hurts my ears
07:48:31 -!- coppro has quit (Remote closed the connection).
07:48:59 -!- coppro has joined.
07:49:03 -!- coppro has quit (Remote closed the connection).
07:50:15 -!- coppro has joined.
07:59:59 -!- clog has quit (ended).
08:00:00 -!- clog has joined.
08:26:22 -!- GregorR-L has joined.
08:31:48 -!- M0ny has joined.
08:51:13 -!- MigoMipo has joined.
09:24:29 -!- olsner has quit ("Leaving").
09:44:13 -!- FireFly has joined.
09:55:29 <calamari> is it possible to have a multi line TextView in a TableRow?
09:55:40 -!- calamari has left (?).
09:58:24 -!- Judofyr has joined.
10:01:16 -!- MigoMipo has quit (Nick collision from services.).
10:01:17 -!- MigoMipo_ has joined.
10:26:19 -!- KingOfKarlsruhe has joined.
10:28:12 -!- BeholdMyGlory has joined.
10:29:49 -!- MigoMipo_ has changed nick to MigoMipo.
10:40:32 -!- immibis has quit ("A day without sunshine is like .... night").
11:07:16 -!- GregorR-L has quit (Read error: 110 (Connection timed out)).
11:12:20 -!- coppro has quit (Read error: 110 (Connection timed out)).
11:54:16 -!- oerjan has joined.
12:26:44 -!- KingOfKarlsruhe has quit (Remote closed the connection).
12:45:00 -!- M0ny has quit.
12:54:48 -!- GregorR-L has joined.
13:10:03 -!- KingOfKarlsruhe has joined.
13:20:07 <AnMaster> ehird, I had a series of 30 classical music cds based on theme, freedb has about 2/3 of them, musicbrainz has none. Oh and I'm not going to rip these, and it seems picard can't work directly from the cd. So I guess I can't add them.
13:20:29 <AnMaster> oh and a few other cds I'm not going to rip that it is lacking.
13:26:16 -!- MigoMipo has quit ("QuitIRCServerException: MigoMipo disconnected from IRC Server").
13:38:11 -!- oerjan has quit ("leaving").
13:39:10 -!- MigoMipo has joined.
13:48:44 -!- MigoMipo has quit (Nick collision from services.).
13:48:45 -!- MigoMipo_ has joined.
13:51:05 -!- nice_ has joined.
13:52:06 -!- nice_ has changed nick to nice_ka.
13:53:52 -!- KingOfKarlsruhe has quit (Nick collision from services.).
13:53:57 -!- nice_ka has changed nick to KingOfKarlsruhe.
13:54:26 -!- KingOfKarlsruhe has left (?).
13:56:53 -!- MizardX- has joined.
13:56:53 -!- MizardX has quit (Read error: 54 (Connection reset by peer)).
13:57:26 -!- MizardX- has changed nick to MizardX.
14:05:45 -!- BeholdMyGlory has changed nick to Wamanuz.
14:06:57 -!- Wamanuz has changed nick to BeholdMyGlory.
14:13:29 -!- BeholdMyGlory has quit (Remote closed the connection).
14:57:10 -!- GregorR-L has quit ("Leaving").
15:17:51 -!- MigoMipo_ has quit ("QuitIRCServerException: MigoMipo disconnected from IRC Server").
15:56:02 -!- ehirdiphone has joined.
15:56:55 <ehirdiphone> "Richard Stallman wrote emacs, gcc, gdb, glibc and the GPL. That is all." Wow, he's right. I guess we can't all be perfect. :-P
15:57:04 -!- ehirdiphone has quit (Client Quit).
15:57:45 -!- MizardX has quit (Read error: 54 (Connection reset by peer)).
15:57:55 -!- MizardX has joined.
15:57:56 -!- Pthing has joined.
16:04:37 -!- darthnuri has joined.
16:07:45 -!- inurinternet has quit (Read error: 60 (Operation timed out)).
16:25:12 -!- sebbu has joined.
16:29:24 -!- darthnuri has quit (Read error: 60 (Operation timed out)).
16:41:34 <pikhq> Project Euler is fun. Especially when you use the single most naive algorithms possible.
16:41:51 <pikhq> For prime factorization.
16:43:21 -!- augur has quit (Read error: 104 (Connection reset by peer)).
16:47:06 -!- KingOfKarlsruhe has joined.
16:48:56 -!- augur has joined.
16:50:17 -!- coppro has joined.
17:03:58 -!- myndzi has quit (Remote closed the connection).
17:16:31 -!- inurinternet has joined.
17:41:33 -!- KingOfKarlsruhe has quit (Remote closed the connection).
18:00:32 -!- inurinternet has quit (Connection timed out).
18:19:55 -!- inurinternet has joined.
18:24:26 -!- M0ny has joined.
18:28:32 -!- darthnuri has joined.
18:34:27 <AnMaster> pikhq, what is the most naive one? I can think of at least two naive ones. Plus a number of less naive ones. Oh and some ones that passed "naive" and went to "intentionally stupid and silly"
18:35:22 <pikhq> AnMaster: Take the list of all primes less than half of n, check and see if those are factors.
18:36:24 <AnMaster> the bloody stupid one would be "try every possible combination of above mentioned primes", trying first using only using two primes, then if no match found, try again with three and so on
18:36:44 <AnMaster> pikhq, how do you like that one? :)
18:38:36 <AnMaster> pikhq, like: foreach prime X < N/2 { foreach prime Y < N/2 { if (Y*X == N) return X,Y; } }
18:39:12 <AnMaster> could be done by making it a list of n-tuples
18:39:26 <AnMaster> taking two lists, generating every possible combination
18:39:26 <oklodok> forall i in n: prime(i); product l = n;
18:39:34 <oklodok> forall i in l: prime(i); product l = n;
18:40:53 <AnMaster> combine([2,7], [2,7]) -> [{2,2},{2,7},{7,2},{7,7}]
18:41:18 <AnMaster> except the input lists would already have such tuples
18:41:57 <AnMaster> combine([2,7], [{2,2},{2,7},{7,2},{7,7}]) -> [{2,2,2},{2,2,7},{2,7,2},{2,7,7},{7,2,2},{7,2,7},{7,7,2},{7,7,7}]
18:42:31 <AnMaster> then multiply all the elements in each tuple and check if they match N
18:42:49 <oklodok> depends on what combine is
18:43:01 <AnMaster> oklodok, the function described above
18:43:16 <oklodok> and what i mean is, [{2, {2, 2}}, ...
18:43:25 <oklodok> usually works that way in math tho
18:43:45 <AnMaster> oklodok, multiplication is commutative so it doesn't matter in this case.
18:43:47 <oklodok> also j has a few special cased things that lift tuples like that
18:44:10 <AnMaster> and I was using erlang syntax for lists and tuples
18:45:39 <AnMaster> pikhq, what language does one write the solutions in?
18:46:04 <AnMaster> or do you just provide the answer?
18:46:19 <AnMaster> pikhq, no other ones possible?
18:46:29 <pikhq> ... No, I mean I'm writing them in Haskell.
18:46:37 <pikhq> You just provide the answer.
18:46:59 <Deewiant> http://news.slashdot.org/story/09/07/07/0024224/Dont-Copy-That-Floppy-Gets-a-Sequel
18:49:05 <fizzie> There was a recent anti-piracy parody thing in that "The IT Crowd" TV series; http://www.youtube.com/watch?v=d82Lq2rVB_4
18:50:09 <oklodok> AnMaster: don't know why i said blah, my point was just combine must be somewhat smart to know when to lift tuples like that, but assuming you were just doing math in erlang notation, that's not important.
18:50:46 <oklodok> also i wanted to note j does that kinda lifting, although i don't remember what operators.
18:50:53 <Deewiant> "Recent" as in 2-3 years old, yes
18:51:08 <AnMaster> oklodok, note that combine is an invented function here
18:51:22 <oklodok> AnMaster: invented name for cartesian product yes
18:51:37 <AnMaster> oklodok, invented function for it yes for this purpose.
18:52:08 <AnMaster> there are better ways to do that in erlang iirc. Using list comprehensions comes to mind.
18:52:10 <fizzie> Deewiant: Recent in the sense that they've recently started showing that thing in Finnish TV. Or so I hear, anyway; we don't have one.
18:52:31 <oklodok> why would they start showing it
18:52:32 <fizzie> "In Finland, the show is broadcast by Yle TV2 since April 2009."
18:52:54 <Deewiant> And found it partially reasonably amusing
18:53:23 <oklodok> i'm just kinda tired of watching nerd humor without nerd content
18:53:47 <fizzie> Would you watch a "The #esoteric Crowd" TV series?
18:54:19 <oklodok> well i watch the it crowd.
18:54:41 <oklodok> but yes esocrowd would probably be awesome
18:54:44 <AnMaster> I mean, when it comes to technical details
18:55:02 <oklodok> where the nerd tells the chick about his code
18:55:18 <oklodok> there's a buzz so you don't hear what he says.
18:55:19 <fizzie> I've seen one (1) episode, and it didn't really go into technical details at all. It's more about the people, I guess.
18:55:28 <AnMaster> what I mean is, is it technobable or does the stuff make sense?
18:56:02 <AnMaster> Deewiant, and this is on TV? You are joking right?
18:56:33 <AnMaster> Deewiant, do you see any scrolling text on the face of a person in front of a computer?
18:57:10 <Deewiant> I doubt it, they don't spend much time sitting in front of their computers.
18:57:18 <AnMaster> Deewiant, then what on earth is the point?
18:57:30 <fizzie> Yes, it doesn't need technobabble when there's no techno to babble about.
18:58:04 <AnMaster> Deewiant, are the errors described actually plausible?
18:58:08 <Deewiant> One of them always answers the phone with "IT; have you tried turning it off and then on again"
18:58:37 <fizzie> Didn't they have an answering machine thing for the phone that suggested rebooting and the normal stuff?
18:58:47 <oklodok> all other characters are, well, very british.
18:59:07 <Deewiant> I found the manager quite British as well.
18:59:47 <oklodok> by british character i mean the kind you find in british sitcoms, bad :)
18:59:55 <oklodok> i'm not sure why i think that.
19:00:01 <Deewiant> That's what I meant too, apart from the bad
19:00:07 <oklodok> maybe i've watched the wrong wshows
19:00:20 <oklodok> well he has the scrubs like insane quality.
19:00:56 <AnMaster> if there is one thing I hate it is technobable when you know the stuff they are talking about.
19:00:58 <AnMaster> ...Once I learnt enough physics I stopped watching Star Trek...
19:00:59 <oklodok> i like character humor, and complex humor
19:01:18 <oklodok> i haven't seen much of the latter during my lifetime
19:01:34 <fizzie> Complex number humor; there's a definite lack of that.
19:01:47 <AnMaster> fizzie, I'm sure it has been done...
19:01:51 <oklodok> there's not much complex number humor that isn't just math related puns
19:02:06 <oklodok> math related puns are just puns, and puns are never funny
19:02:17 <fizzie> Z and X walked to a bar; but they're not orderable!
19:02:19 <AnMaster> oklodok, there isn't much math humour that isn't puns...
19:02:37 <oklodok> i was told that earlier on #math
19:02:52 <AnMaster> oklodok, oh? I was just speaking out of experience...
19:03:23 <oklodok> also i heard a math joke that wasn't a pun just after someone told me that
19:03:32 <AnMaster> oklodok, and what was that joke?
19:03:51 <fizzie> SGI's IRIX accelerated-math library thing (for FFTs and such) has a data type "complex" for pairs of single-precision floats, but the name for the double-precision variant is the hilarious "zomplex".
19:04:16 <AnMaster> fizzie, err, why is that hilarious?
19:04:16 <oklodok> it was about a branch of math that has very inexact bounds, something about a lecturer saying he didn't remember exactly, but something happened less than 10^10^10^10 years ago.
19:04:29 <fizzie> I don't know why, it just is.
19:04:37 <AnMaster> not that the name makes any sense
19:04:39 <fizzie> It's like some sort of zombie-complex.
19:05:02 <AnMaster> is there any explanation of the name?
19:05:14 <AnMaster> fizzie, zombies aren't really very funny
19:05:14 <oklodok> AnMaster: did you get the joke?
19:05:29 <AnMaster> not the 100rd (th?) time at least
19:05:34 <Deewiant> Watch Shaun of the Dead sometime
19:05:57 <oklodok> i didn't actually tell the joke, my interest in jokes is mostly theoretical
19:06:07 <oklodok> i can look it up if you can't laugh at it from that
19:06:14 <AnMaster> oklodok, well... Assuming the current estimate of the age of the universe...
19:06:40 <oklodok> clearly he used something less exact than that.
19:07:02 <oklodok> i don't see how that matters, the gist of the joke is it doesn't matter in whatever branch the lecturer does either
19:07:20 <ehird> 00:04 AnMaster: ehird, editing on musicbrainz, have you done it yourself?
19:07:37 <oklodok> the branch was named after a name of some sort, and i've never heard of it, so i'd need to read from the logs
19:07:40 <ehird> 12:20 AnMaster: ehird, I had a series of 30 classical music cds based on theme, freedb has about 2/3 of them, musicbrainz has none. Oh and I'm not going to rip these, and it seems picard can't work directly from the cd. So I guess I can't add them.
19:07:42 <ehird> 12:20 AnMaster: oh and a few other cds I'm not going to rip that it is lacking.
19:07:44 <ehird> you keep talking. are you fallaciously assuming I give a shit?
19:07:47 <AnMaster> ehird, adding cds with different composers for different tracks = pain
19:08:04 <AnMaster> ehird, compilations with themes
19:08:24 <Deewiant> AnMaster: It's twice the fun when you need to add almost every composer, as I have had to do in the past.
19:08:25 <ehird> 17:49 fizzie: There was a recent anti-piracy parody thing in that "The IT Crowd" TV series; http://www.youtube.com/watch?v=d82Lq2rVB_4
19:08:31 <ehird> is that the "you wouldn't steal a policeman's helmet"?
19:08:34 <ehird> can't view in this country
19:08:36 <ehird> anyway that's ancient.
19:08:36 <Deewiant> As well as the label that published the CD.
19:08:45 <AnMaster> ehird, read the rest of the log
19:08:54 <ehird> 17:52 fizzie: Deewiant: Recent in the sense that they've recently started showing that thing in Finnish TV. Or so I hear, anyway; we don't have one.
19:09:01 <ehird> http://www.theonion.com/content/node/28694
19:09:10 <fizzie> I think the name is because LAPACK's six-letter (for Fortran compatibility) function names start with a single-character data-type prefix; S = float, D = double, C = single-precision complex, Z = double-precision complex.
19:09:28 <fizzie> Of course that's a bit of a non-answer, because it doesn't explain why lapack chose Z there.
19:09:32 <AnMaster> <Deewiant> AnMaster: It's twice the fun when you need to add almost every composer, as I have had to do in the past. <-- yes I had a few of that. I just gave up. Too much work. And some said "Unknown, but probably Haydn or Mozart"
19:09:39 <AnMaster> Deewiant, that is where I gave up
19:10:01 <ehird> 17:55 AnMaster: what I mean is, is it technobable or does the stuff make sense?
19:10:01 <ehird> Moss: [picks up phone] Hello, IT? Yah-hah? Have you tried forcing an expected reboot? You see the driver hooks the function by patching the system call table, so it's not safe to unload it unless another thread's about to jump in there and do its stuff, and you don't want to end up in the middle of invalid memory.
19:10:09 <ehird> correct apart from being the wrong way around./
19:10:40 <AnMaster> ehird, that kind of could make sense assuming windows NT's design
19:10:47 <Deewiant> :-) I didn't remember that one
19:10:49 * AnMaster tries to remember how system calls worked
19:10:59 <AnMaster> yes I think it is possibly accurate
19:11:12 <ehird> AnMaster: it's fairly sane OS design.
19:11:28 <AnMaster> I remember there was/is a table you could patch to install rootkits. Mentioned on sysinternals iirc...
19:11:29 <ehird> (the correction is s/unless/if/)
19:11:44 <AnMaster> not sure if it was an internal table, or the system call one.
19:11:57 <oklodok> was just about to complain about unless
19:12:04 <ehird> 19:10 ehird: correct apart from being the wrong way around./
19:12:39 <ehird> unloading it is only safe if you're about to jump to it?
19:12:48 <ehird> THAT'S what will end you in the middle of garbage memory...
19:12:58 <AnMaster> I meant, could be make to work even if something is jumping to it
19:13:05 <pikhq> ata1: hard resetting link
19:13:18 <pikhq> That is cause for concern, I think.
19:13:29 <ehird> oh man, that sequel is official
19:13:45 <ehird> "A smug teen who's downloading files from 'Pirates Palace' and 'Tune Weasel' finds his world turned upside down when automatic weapons-toting government agents break down the door and take his Mom away in handcuffs. The teen finds himself in a prison jumpsuit forced to tattoo shirtless adult inmates who eventually turn on him, physically attack him, and make him run for his life back to his jail cell."
19:13:46 <AnMaster> just use a CAS instruction (CMPXCHG assuming x86), to swap it with the original function used
19:14:01 <ehird> it sounds more like a subversive ad for anarchism than against piracy
19:14:06 <AnMaster> then it doesn't matter if something is about to jump to it
19:14:31 <ehird> y'know what this has acheived?
19:14:38 <ehird> i wanna go pirate a bunch of software i don't want to use
19:14:47 <ehird> and delete it immediately
19:15:13 <pikhq> I'm running a smartctl test.
19:15:28 <ehird> i'm not giving them the money they rightfully should own as I would if I bought it then stomped on the disc
19:15:59 <ehird> wait, does it have the same mc? :D
19:16:22 <Deewiant> In a flashback, at least; watching it.
19:16:31 <AnMaster> oklodok, there are some non-puns in http://en.wikipedia.org/wiki/Mathematical_joke below the pun section
19:16:38 <AnMaster> see "Stereotypes of mathematicians"
19:16:49 <Deewiant> LOL at the prisoners' tattoos.
19:16:58 <oklodok> i'm also not that interested in mathematician jokes
19:17:11 <ehird> hahahahahah this must be a parody
19:17:22 <Deewiant> AnMaster: See /. link above, or http://www.youtube.com/watch?v=fHaAFqoVLtI for the lazy.
19:17:32 <oklodok> stereotype jokes are almost as easy as puns
19:17:54 <ehird> Deewiant: omg it's the same mc
19:18:16 <ehird> sounds like what what in the butt
19:18:19 <ehird> don't copy that, in the butt
19:18:31 <Deewiant> LOL at the crappy-looking and -sounding Klingons
19:19:04 <ehird> to copy data is a great dishonour? fucking L[123] caches and RAM!
19:19:14 <ehird> they should be illegal! (well, they were somewhere iirc :))
19:20:08 <ehird> well, that was delightfully bullshit
19:20:18 <ehird> lol, they're using Joomla on their website
19:20:21 <ehird> last I checked, it was open source
19:20:35 <ehird> i suspect them of copying it.
19:21:03 <ehird> *Joomla!; pedanticity must be applied even in the face of obnoxious exclamation marks
19:21:34 <AnMaster> I can't watch it, it is just too bad.
19:21:44 <ehird> "I bet if I showed this new video to the average 12 year old, they'd think it was some kind of internet sketch comedy thing."
19:21:51 <ehird> it actually is exactly like that
19:22:01 <ehird> if it wasn't on the SIIA website, I'd be laughin'
19:22:07 <ehird> AnMaster: no, it's real
19:22:21 <pikhq> ehird: Joomla! is GPL. ;)
19:22:29 <ehird> pikhq: "Copying data is a great dishonor."
19:22:50 <ehird> You know it's true because fake Klingons (…is there another kind?) said it.
19:23:56 <ehird> Deewiant: probably not the same actor; since the actor is like CEO of some enterprisey bullshit computer company
19:25:41 <oklodok> what's don't copy that floppy exactly?
19:25:56 <AnMaster> about that "patching system call table" quote... were there more like that in that series?
19:25:57 <ehird> oklodok: you have to watch it. to experience it.
19:25:59 <Deewiant> Something also available on youtube
19:26:05 <ehird> AnMaster: dunno, I only watched a few episodes.
19:26:08 <Deewiant> An early 90s anti-piracy video
19:26:11 <ehird> just torrent the damn thing and see :P
19:26:17 <oklodok> Deewiant: yes, but it takes about an hours to download something with my conn
19:26:46 <ehird> (it's okay, we brits paid for it with our yearly payment to the WE KNOW WHO YOU ARE SO PAY UP YOUR DAMN TV LICENSE commission)
19:26:58 <Deewiant> 10 minutes of fairly heavily compressed video shouldn't take /that/ long
19:27:05 <Deewiant> Unless they're bigger than I think
19:27:19 <ehird> Deewiant: they are, but it can play before downloading it all, so
19:27:20 <oklodok> takes 10 minutes to download a midi
19:27:33 <oklodok> some normal webpages time out
19:27:34 <ehird> (apparently the tv licensing people just track who has a TV, then harass anyone who doesn't have one)
19:27:50 <Deewiant> Webpages go up to megabytes these days, that doesn't surprise me
19:27:51 <oklodok> no i just have µtorrent on, and for some reason it kills http
19:27:53 <ehird> (we saw in your window that you have a tv and are watching it and are not paying us!* *note: we didn't actually look)
19:28:02 <ehird> oklodok: because your connection is saturated
19:28:05 <ehird> rate-limit utorrent.
19:28:12 <ehird> Deewiant: i've never seen a 1mb webpage
19:28:41 <ehird> Deewiant: Mu; I don't use them.
19:28:50 <ehird> Anyway, that wouldn't timeout the whole page.
19:28:52 <ehird> Just certain elements.
19:29:00 <Deewiant> If it's in a table it'll timeout the whole thing.
19:29:04 <ehird> oklodok: link to that sevenfold glio song thing? saw it in the logs ages ago.
19:29:12 <ehird> Deewiant: not eg an <img>
19:29:20 <AnMaster> <Deewiant> If it's in a table it'll timeout the whole thing. <-- huh?
19:29:29 <oklodok> ehird: sure, that's kind of a nobrainer, i just don't want to limit it, i prefer not using the net.
19:29:31 <Deewiant> ehird: No? Don't you need to know the size before you can flow it?
19:29:43 <ehird> Deewiant: modern browsers don't have that.
19:29:49 <ehird> Deewiant: netscape 4, I think, did that.
19:30:01 <oklodok> http://www.vjn.fi/oklopol/music/sevenfold.mid
19:30:02 <ehird> also, <img width=butt height=butt> :P
19:30:13 <Deewiant> I recall people complaining about as recently as in Phoenix
19:30:21 <ehird> oklodok: beautiful
19:30:22 <Deewiant> Granted, that's still a while ago
19:30:41 <ehird> "Their marketing department didn't even notice that they made an unauthorized reproduction and depiction of a well known anime character in their video..." —/.
19:30:51 <Deewiant> But anyway, it was quite obviously noticeable that tables didn't render the way divs did
19:31:01 <ehird> oklodok: i like the guitar/drums part
19:31:07 <oklodok> ehird: it's probably my only "published" piece that has completely random parts.
19:31:07 <ehird> very nice sandwiched with the noise.
19:31:31 <ehird> oklodok: i wonder if it's possible to play this irl
19:31:35 <ehird> i'm thinking the drums might be a bit hard.
19:31:44 <oklodok> usually the stuff people hear as noise in my songs is 100% thought through
19:32:45 <ehird> anicecreamymelody goes well after sevenfold.
19:33:09 <oklodok> ehird: if you want to hear, i do have some actual music too.
19:33:17 <ehird> oklodok: that's just hitting shit a lot though
19:33:27 <ehird> also you have to play the rhythmical part straight afterwards
19:33:34 <ehird> oklodok: you mean the metal stuff?
19:33:41 <ehird> i listened to that ages ago, didn't really like it.
19:33:57 <oklodok> you listened to the band stuff?
19:34:05 <ehird> months and months ago.
19:34:15 <ehird> yes, they sounded mostly identical ;P
19:34:43 <ehird> well i'm not really a metal guy you know?
19:34:47 <ehird> all sorta sounds the same.
19:34:55 <ehird> etudes, i haven't seen etudes before
19:35:25 <ehird> well i dunno what these .gt[45]s are i guess i could download the midi archive
19:35:35 <oklodok> anyway most of my songs are just on .mid
19:35:41 <oklodok> but, they are mostly metal
19:35:48 <ehird> well metal .mid is okay
19:35:48 <oklodok> i only have like 10 or so non-metal songs
19:35:57 <ehird> the music is fine, i just don't like how it sounds when performed
19:36:01 <AnMaster> oklodok, ehird: for drums, couldn't you use several people, playing on a round-robin schedule?
19:36:08 <ehird> AnMaster: have you listened to it?
19:36:20 <AnMaster> ehird, not yet, trying to find my headphones...
19:36:22 <ehird> it could work if you have instant, infinite communication and comprehension between everyone.
19:36:28 <ehird> AnMaster: you won't like it :D
19:36:41 <ehird> oklodok: well linky to mids?
19:37:00 <oklodok> i can privately up some stuff for you, but i don't like distributing them.
19:37:10 <ehird> i'll only give them to 5000 people max
19:37:22 <ehird> if it's a good day
19:37:40 <oklodok> oh 50,000? then we have a problem.
19:39:08 <Deewiant> Which reminds me, can anybody explain why 8000 got translated to 9000
19:39:14 <AnMaster> oklodok, I would like a copy too.
19:39:26 <AnMaster> Deewiant, hm? I thought 9000 was some meme
19:39:40 <AnMaster> 8000 I never heard of as a meme
19:39:50 <ehird> AnMaster: it was 8000 in the original japanese
19:39:53 <ehird> they translated it to 9000.
19:40:01 <AnMaster> I don't even know where the meme is from
19:40:10 <ehird> oklodok: i inferred from Deewiant
19:40:18 <ehird> it's clearly because americans are 1,000 better, anyway
19:40:20 <Deewiant> It's from an episode of Dragonball Z where Vegeta says it angrily.
19:40:36 <AnMaster> and this "dragonball z", is it manga or anime?
19:40:46 <Deewiant> Manga doesn't come in episodes.
19:40:48 <ehird> most animes are also mangae..
19:40:51 <fizzie> And it's a "power level" which should be over 9000.
19:41:00 <ehird> fizzie: shouldn't be.
19:41:03 <Deewiant> Dragonball Z is based on the Dragonball manga, though, where it was also 8000.
19:41:07 <AnMaster> Deewiant, ok. I'm not an expert on such stuff.
19:41:09 <ehird> the main line is expressing shock at said fact.
19:41:18 <Deewiant> And in the new Dragonball Kai which is sort of a remake of Dragonball Z it was also 8000.
19:41:19 <ehird> admittedly by Bad Guy(TM)
19:41:24 <ehird> i don't actually know anything about dragonball
19:41:31 <ehird> i'm just good at collecting info randomly and inferring.
19:41:49 <fizzie> Erzyklopedia dramatica has the dialogue, so:
19:41:50 <fizzie> Nappa: "VEGETA! What does the scouter say about his power level?"
19:41:50 <fizzie> Vegeta: "IT'S OVER NINE THOUSAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAND!" *crushes scouter*
19:41:50 <fizzie> Nappa: WHAT, NINE THOUSAND!?
19:42:03 <ehird> it has a total transcript
19:42:17 <AnMaster> quantum nano technology manager?
19:42:28 <ehird> http://qntm.org/?9000
19:43:42 -!- Slereah has quit (Read error: 60 (Operation timed out)).
19:44:53 <Deewiant> The original is quite clearly "hassen ijou da".
19:45:22 <ehird> Hussein is your dad.
19:45:37 <Deewiant> DBKai's version of it is evidently up at http://www.youtube.com/watch?v=g9oVNvRSlVk for the interested.
19:45:49 <Deewiant> The relevant phrase being at 0:34 or thereabouts.
19:45:53 <pikhq> My motherboard is starting to give up the ghost. YAY.
19:46:06 <ehird> Deewiant: You're obsessed with either over 9000 or Dragonball.
19:46:10 <fizzie> Apparently nine thousand isn't even all that much nowadays. (Source: http://tvtropes.org/pmwiki/pmwiki.php/Main/PowerLevels )
19:46:21 -!- Slereah has joined.
19:46:25 <ehird> Deewiant: also, how is that a remake? it looks exactly the same.
19:46:40 <ehird> oh. says hd remastered.
19:46:57 <Deewiant> And they removed some scenes to make it shorter and more in line with the manga.
19:48:26 <Deewiant> I'm mostly obsessed with accuracy. Your 1000s just reminded me. But I do know quite a bit more about Dragonball than most.
19:48:54 <ehird> I know a bit more about child pornography than most.
19:48:58 <ehird> I know a bit more about rape techniques than most.
19:49:05 <ehird> I know a bit more about assassinating the president than most.
19:49:24 <Deewiant> I don't mind admitting I have esoteric knowledge. Especially on #esoteric.
19:49:32 <ehird> …strangely, while I have no qualms about putting those in that template, I can't bring myself to put in things like "Dragonball"
19:49:37 <ehird> (I tried. My hands seize up.)
19:50:31 <ehird> I wish ais523 logread.
19:50:36 <ehird> "One thing I found puzzling was that the Brits consistently apologized for and/or denigrated Birmingham."
19:50:41 <ehird> —Bruce Eckel, http://www.artima.com/weblogs/viewpost.jsp?thread=261930
19:52:03 <AnMaster> ehird, what other things 'like "Dragonball"'
19:52:25 <ehird> "Scat porn" works.
19:52:38 <Deewiant> What's so bad about Dragonball vis-à-vis child porn anyway
19:52:46 <ehird> Rutanaloobeedoobeedoo
19:52:47 <AnMaster> Deewiant, was wondering that too
19:52:51 <oklodok> i've fairly sure i know more than most about all of those.
19:52:54 <fizzie> Deewiant: Aren't they synonyms anyway?-)
19:53:00 <AnMaster> ehird, what about inserting C++ there?
19:53:04 <ehird> it was re 19:48 fizzie: And you admit that?
19:53:11 <Deewiant> Not quite, no. In fact, not at all. :-P
19:53:21 <ehird> AnMaster: as much as it pains me to admit it, it's possible I'm going to willingly use C++ for something
19:54:07 <AnMaster> also what about "Plain English"? I think it would be true too. Sadly.
19:54:15 <ehird> Deewiant: game engine type stuff; lots of OOP stuff so not e.g. C, but needs a lot of assured speed and control over purity, so not e.g. Haskell
19:54:35 <ehird> Deewiant: i'd rather vomit
19:54:51 <ehird> gc would be quite nice but it's not really vital so
19:55:00 <AnMaster> ehird, are you going to work on a game engine? You know there are many good open source 3D engines already that can handle both directx and opengl?
19:55:02 <ehird> i am going to add a custom scripting language to it so i might just add refcounting to my object infrastructure
19:55:08 <fizzie> Deewiant: Wasn't there some sort of censorship thing about DBZ Finnish translation release? Or was it some other manga thing?
19:55:13 <ehird> AnMaster: what fun's that?
19:55:18 <ehird> i want something i can tweak
19:55:22 <pikhq> ehird: *cough* Haskell speed ~= C speed.
19:55:23 <AnMaster> crystalspace or something like that
19:55:36 <Deewiant> fizzie: Possibly... rings a bell but I can't remember any details
19:55:38 <ehird> pikhq: under ideal circumstances; but I also need a ton of libraries for shit Haskell doesn't really have a lot of
19:55:46 <ehird> i'd be in IO 90% of the time anyway
19:55:52 <AnMaster> ehird, could you insert "D" there instead of "C++"? ;P
19:55:55 <pikhq> ... Doesn't Haskell have glut bindings?
19:55:58 <ehird> since a whole lot of game logic will be in the scripting language anyway
19:56:04 <pikhq> I know I saw one in Hackage.
19:56:12 <ehird> 19:55 AnMaster: ehird, could you insert "D" there instead of "C++"? ;P
19:56:14 <ehird> 19:54 Deewiant: *cough*D*cough*
19:56:16 <ehird> 19:54 Deewiant: But yeah, toolchain etc.
19:56:18 <ehird> 19:54 ehird: Deewiant: i'd rather vomit
19:56:20 <ehird> I hate D as a language, also.
19:56:26 <ehird> → as in transition
19:56:28 <AnMaster> erlang has opengl bindings btw. Just in case anyone wants it...
19:56:42 <pikhq> ehird: So, I am actually getting that new system.
19:56:42 <AnMaster> there is even a 3D editor using them
19:56:45 <ehird> Deewiant: it's a gigantic hodgepodge
19:56:50 <Deewiant> Despite being a hodgepodgey mess I find it cleaner than C++.
19:56:50 <ehird> pikhq: THE $80K ONE? AWESOME!
19:57:02 <Deewiant> ehird: I mean, really. C++ isn't?
19:57:14 <ehird> Deewiant: i'm gonna be conservative in my use of C++ features, and at least it's a mess with a good toolchain
19:57:34 <AnMaster> the D toolchain definitely sucks
19:57:44 <AnMaster> even Deewiant has to admit that
19:58:10 <pikhq> It's a royal bitch to get set up, yes.
19:58:31 <AnMaster> <ehird> I know a bit more about assembler than most.
19:58:56 <ehird> more than most people in the world
19:59:08 <ehird> probably more than most people who have heard what assembly is
19:59:17 <ehird> but not more than most people who've written a program in asm
19:59:30 <Deewiant> pikhq: Once you've done it a few times it's not really much trouble at all, though. (On *nix. Windows is always a pain, but then it is so for almost any language.)
19:59:43 <AnMaster> ehird, you couldn't bring yourself to insert the phrase "D" there? Nor "dragonball"? What about other things on D? Maybe you have some sort of phobia against the letter D in that context?
19:59:55 <ehird> oh, that's what you meant?
20:00:03 <ehird> i thought "remove C++ from the project and insert D"
20:00:12 <pikhq> Deewiant: It's a consistent bitch on x86_64.
20:00:26 <pikhq> Since it *building* is a a gamble.
20:00:30 <AnMaster> Deewiant, remember that "wrong file, right line number" in debug info?
20:00:54 <Deewiant> Random remarks don't constitute a case
20:01:02 <Deewiant> I thought we'd settled that already
20:01:17 <ehird> haha, gmail isn't beta any more
20:01:20 <ehird> who was remarking on that yesterday?
20:01:41 <Deewiant> pikhq: LDC builds quite cleanly on x86-64.
20:02:06 <Deewiant> DMD and co I run in a 32-bit chroot like other 32-bit stuff.
20:04:40 <ehird> i like inventing practical esolangs.
20:09:02 <AnMaster> befunge is probably one of the most practical ones
20:09:10 <AnMaster> also make sure you don't stray into DSLs
20:09:14 <ehird> as in, an odd language designed for a practical purpose.
20:09:27 <AnMaster> a DSL out of context can easily look like an esolang and vice verse.
20:10:53 <AnMaster> ehird, sed is a special purpose language. If you never seen it before and then see an example of it used to implement a calculator (+-/* and square root) you would probably think "this is an esolang"
20:11:41 <AnMaster> of course this doesn't apply to for example malbolge, you can't mistake it as a dsl
20:12:04 <AnMaster> but some of those rewriting ones could probably be mistaken unless I misremember
20:12:15 <ehird> i wonder if anyone uses the extension .c++
20:12:21 <ehird> nobody seems to, prolly cause of windows
20:12:39 <ehird> Deewiant: it's invalid in filenames isnt it
20:12:47 <AnMaster> ehird, a year or two ago at least. Don't remember which open source project
20:12:56 <Deewiant> ehird: My statement was implying that I don't think it is.
20:12:58 <ehird> wikipedia's [[C++]] doesn't cover the file extension issue
20:13:03 <AnMaster> ehird, .C .cc .CC .cxx are more common though
20:13:10 <ehird> pretty sure it's invalid
20:13:11 <oklodok> AnMaster: see your pm btw.
20:13:17 <AnMaster> there was both foo.c and foo.C
20:13:23 <ehird> AnMaster: the former one sucks on case insensitive filesystems like HFS :p
20:13:29 <ehird> i hate it when that happens
20:13:38 <ehird> AnMaster: .cpp is quite common
20:13:39 <AnMaster> ehird, can't you manually fix it
20:13:40 <Deewiant> http://en.wikipedia.org/wiki/File_name doesn't appear to mention +.
20:13:41 <oklodok> or tell me here that you won't answer in pm
20:13:49 <ehird> .cc / .cpp > .cxx i would say for popularity
20:13:53 <ehird> AnMaster: also, i can but it's a pain
20:14:26 -!- FireFly has quit ("Later").
20:14:29 <AnMaster> ehird, it should be used more IMO
20:14:41 <AnMaster> doesn't POSIX mandate case sensitive file systems
20:14:47 <ehird> AnMaster: i don't think so.
20:15:01 <ehird> i'd say, ideally it'd be .c++, failing that, prolly .cpp, then .cc, then .cxx
20:15:08 <ehird> .c++ and .cpp are logical, .cc, .cxx and .C aren't
20:15:20 <ehird> also, we need to take into account headers
20:15:23 <ehird> i've never seen .hh
20:15:30 <ehird> people just use .h or .hpp
20:15:37 <ehird> certainly rare though
20:15:45 <AnMaster> ehird, never seen .H though...
20:15:52 <ehird> .hpp is the most common special h naming i've seen
20:15:56 <ehird> which lends more credence to .cpp too
20:16:04 <ehird> i'll ask stroustru
20:16:07 <ehird> maybe he has an opinion.
20:16:19 <ehird> AnMaster: shift typo
20:16:23 <ehird> i.e. hands rested one letter off
20:16:40 <AnMaster> Deewiant, I'd say .hpp > .h > .hxx > .hh
20:16:57 <ehird> http://www.research.att.com/~bs/pronounciation.wav strchstruwp
20:17:08 <ehird> .h is undesirable.
20:17:10 <ehird> if you do that, do .c too!
20:17:34 <AnMaster> <fizzie> O dpm <-- shifted to "I son"?
20:18:07 <AnMaster> fizzie, so the d was unshifted
20:18:19 <pikhq> ehird: String char string unsigned word pointer, in MS-speak. :P
20:18:20 <fizzie> It was just the right hand that was off-by-one. And the ' in the fi layout is next to enter, which is what I tried to press next.
20:18:35 <fizzie> Er, I mean, I tried ' but it came out as an enter.
20:18:41 <AnMaster> ehird, best way would be .C and .H clearly
20:18:44 <pikhq> They're Hungarian.
20:18:50 <ehird> AnMaster: best way would be .c++ and .h++
20:18:53 <ehird> .C/.H makes no sense
20:19:04 <AnMaster> ehird, ok, .c++ and .h++ would be better
20:19:15 <ehird> i never see .c++ either
20:19:22 <ehird> anyway, anyone got a windows box?
20:19:25 <AnMaster> I did see .c++, I think paired with .h
20:19:47 <AnMaster> ehird, where is oerjan when you need him
20:19:56 <ehird> AnMaster: his computer crashed :)
20:20:06 <Deewiant> I can't be bothered to reboot just for that
20:20:09 <fizzie> According to MSDN it should work: http://msdn.microsoft.com/en-us/library/aa365247(VS.85).aspx
20:20:21 <ehird> Deewiant: set up a vm pointed at the windows partition
20:20:22 <Deewiant> fizzie: Please link to the low-bandwidth one
20:20:22 <fizzie> The reserved ones are < > : " / \ | ? *.
20:20:34 <ehird> Deewiant: http://msdn.microsoft.com/en-us/library/aa365247(VS.85,loband).aspx
20:20:36 <ehird> sorry, you wanted fizzie
20:21:02 <fizzie> I can't seem to find a link to that, just the printer-friendly thing.
20:21:38 <fizzie> But I'm not going to link to it! Ha ha!
20:21:49 <ehird> NOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO
20:22:12 <ehird> [[See my C++0x FAQ. The aim is for the 'x' in C++0x to become '9': C++09, rather than (say) C++0xA (hexadecimal :-).]]
20:24:04 <fizzie> Of course there's the catch-all "Any other character that the target file system does not allow". And in fact the FAT short-name can't contain a +: "The following special characters are also allowed: $ % ' - _ @ ~ ` ! ( ) { } ^ # &" (does not include +).
20:24:32 <fizzie> But the long-name can contain any of + , ; = [ ] too. And I guess NTFS is a bit more flexible.
20:24:35 <Deewiant> ehird: You commented about big pages, btw; MSDN reminded me of that. www.fox.com is at 1.46 MB (and still waiting for something, it seems) according to Firebug.
20:24:54 <ehird> I question why you'd want to load fox.com
20:24:58 <ehird> But anyway, that's multiple things
20:25:01 <ehird> You won't get a timeout
20:25:05 <ehird> Just some broken images and stuff
20:25:31 <Deewiant> No, but if you're browsing two things at once the other might timeout.
20:25:44 <Deewiant> I don't care whether it's strictly part of the page, just how much it stresses the connection.
20:26:15 <Deewiant> Anyway, fox.com was just an example, I figured it'd be pretty huge.
20:27:52 <Deewiant> I actually tried cnn.com first but it consistently hangs my whole browser.
20:31:45 <Deewiant> (It did respond to Ctrl-W, though.)
20:33:12 <pikhq> Whoo. Seems today was a really good time to get stuff from Newegg.
20:33:39 <pikhq> Phenom? Screw that; a Phenom II FTW.
20:34:02 <ehird> pikhq: Buy FIVE MILLION of them.
20:34:07 <ehird> pikhq: Also, DDR3 prices are near DDR2 now.
20:34:20 <pikhq> Also, AM2+ motherboard was still cheaper.
20:34:20 <ehird> Mayhaps you could get 2x2GB of DDR3 on the cheap, I think.
20:34:42 <ehird> Deewiant: "Almost as low as"
20:34:51 <ehird> No, they've improved very rapidly
20:35:00 <pikhq> ehird: $42 bucks off in total.
20:35:04 <Deewiant> ehird: Numbers, please. Preferably ones that have something to do with their relative prices.
20:35:06 <ehird> Deewiant: 2GB for $27.99 of DDR3.
20:35:25 <Deewiant> Wow, I can get $27.99 of DDR3‽
20:35:54 <Deewiant> I have over 6 times that of DDR2
20:36:00 <ehird> Deewiant: Lowest 2GB cost on newegg is $21.99; the DDR3 I was talking about was Crucial - a respected brand.
20:36:04 <ehird> The lowest DDR2? "Allcomponents".
20:36:16 <ehird> Looking further, you're saving just a few dollars.
20:36:19 <ehird> Deewiant: they're huge...
20:36:32 <fizzie> I have here an old laptop I'd like a 1 gigabyte so-dimm for; but it eats DDR1 only, and for some reason a 1-gigabyte DDR1 thing is approximately 40€, while a 1-gigabyte DDR2 so-dimm is ~13€. Around here.
20:36:33 <ehird> mostly in the "aftermarket memory upgrades" market
20:36:44 <ehird> Deewiant: you have 8GB of ram right?
20:37:20 <ehird> Deewiant: How much did it cost, and when?
20:37:57 <Deewiant> Something like 60-70 € twice, IIRC. It was on sale around... November?
20:38:24 <ehird> You can get 4GB of DDR3 in 2x2 for $57.99; although you can save a whole cent by getting 2x2 separately = $57.98. 130 euros is $181.
20:38:41 <fizzie> I seem to have bought a 2-gigabyte DDR2 stick for 17.90 € in April 28th.
20:38:41 <ehird> $57.98 = 41.52 eur
20:38:50 <pikhq> Deewiant: Crucial is well-respected and not over-priced.
20:39:04 <ehird> fizzie: 2GB of DDR3 = 20 euros
20:39:23 <Deewiant> This was, of course, Mushkin's Redline RAM and thus considered somewhat extraneously quality.
20:39:36 <Deewiant> It was also approximately the cheapest DDR2 I could find, interestingly enough.
20:39:46 <Deewiant> (Among brands that have names.)
20:39:50 <ehird> Deewiant: Interestingly, the clock speed and the like on DDR3 don't change Core i7 performance much out of synthetic benchmarks.
20:40:05 <ehird> I think bsmntbombdood's 12GB of DDR3 RAM was like $200
20:40:16 <Deewiant> Perhaps those synthetic benchmarks don't stress RAM!
20:40:26 <pikhq> I'm going from 256k of cache to 6M.
20:40:28 <ehird> Deewiant: I meant non-benchmarks don't change
20:40:40 <ehird> Anyway, all I'm sayin' is, if you have the mobo support, DDR3 is the only sane option.
20:40:47 <pikhq> I'm getting 3 times the CPUs. And 4 times the RAM.
20:40:54 <ehird> pikhq: how much ram are you getting?
20:40:58 <pikhq> My system's going to be modern again!
20:40:59 <ehird> also, 3 times the cores
20:41:19 <ehird> pikhq: I'd say DDR3 performance is worth $42, btw.
20:41:36 <pikhq> I'd have to get a more expensive motherboard for that.
20:41:40 <AnMaster> three times as much ram as number of cores?
20:41:44 <ehird> pikhq: You said $42 more.
20:42:01 <Deewiant> ehird: Well yeah, if you have the mobo support DDR3 is your only option. :-P
20:42:09 <ehird> AM3 mobos support both.
20:42:11 <pikhq> ehird: No, I'm saying it was a total of $42 off, because the CPU was marked down by $30 and the motherboard by $10.
20:42:12 <Deewiant> Or are there DDR2+3 boards these days.
20:42:12 <ehird> Which is the relevant case for pikhq.
20:42:17 <pikhq> And the RAM by $2.
20:42:24 <ehird> pikhq: What is your mobo+RAM costing you?
20:42:37 <Deewiant> I really shouldn't talk about hardware when I haven't been researching it recently.
20:42:56 <pikhq> ehird: Maybe $100?
20:43:03 <ehird> pikhq: Look at the prices, plz :P
20:43:52 <ehird> pikhq: I'm just trying to prove that DDR3 wouldn't actually add much cost at all to your system.
20:44:32 <ehird> pikhq: $133.97 for 4GB of DDR3 RAM and a supporting AM3 mobo.
20:44:56 <pikhq> Okay. More but not much more.
20:45:07 <ehird> ...which can't be said about the performance.
20:45:22 <ehird> (Uh, rephrase that more... positively)
20:46:17 <ehird> AnMaster: Deewiant: fizzie: New evidence in the C++ naming debate — apparently cfront used .cc
20:47:36 <ehird> pikhq: FWIW, http://www.newegg.com/Product/Product.aspx?Item=N82E16813153149 and http://www.newegg.com/Product/Product.aspx?Item=N82E16820148149 x 2
20:51:24 <ehird> I wonder how the extensions fit with Objective-C++
20:51:40 <ehird> .mxx is really silly.
20:51:48 <ehird> Well, .mm or .mpp I guess
20:51:56 <ehird> So it's still down to .cc or .cpp in general
20:55:02 <fizzie> Maybe the ".mmm, marabou" extension.
20:58:15 <AnMaster> ehird, what about .m? I remember also seeing .c for C++
20:58:37 <AnMaster> was some shitty open source game
20:58:54 -!- oerjan has joined.
20:58:56 <pikhq> And gcc breaks on that.
20:59:39 <Deewiant> In general I don't call on gcc for C++ source anyway.
21:00:00 <AnMaster> pikhq, not if you want it to work
21:00:07 <AnMaster> you can use -x or something iirc
21:00:55 <AnMaster> gcj -x c++ foo.cxx should work
21:01:17 <AnMaster> for extra sillyness: gcj -x c++ -std=c++98 foo.cxx
21:01:30 <AnMaster> it certainly works for gcc and g++
21:01:34 <ehird> surely ou mean GNAT
21:01:40 <ehird> JGNAT is a GNAT version that compiles from the Ada programming language to Java bytecode.
21:01:43 <ehird> jgnat -x or whatever :P
21:02:00 <AnMaster> iirc GNAT is kind of special in general
21:02:53 <oerjan> <ehird> AnMaster: his computer crashed :) <-- that hasn't happened in a long while. although occasionally it refuses to start, saying "Operating system not found"
21:03:04 <ehird> oerjan: can you name a file butt.c++ ?
21:03:11 <ehird> well not necessarily butt
21:03:47 <fizzie> If he's on FAT32, he should be able too, as long as he doesn't care that the underlying 8.3-format short-name won't have a + there.
21:04:16 <fizzie> Well, FAT<anything>, I guess.
21:05:59 <ehird> Deewiant: tell that to ais523
21:06:12 <ehird> oerjan: did it work?
21:06:35 <ehird> argh i hate const correctness
21:06:50 <pikhq> (FAT<anything> with extra metadata to make it magically be proper UNIX)
21:07:01 <pikhq> Shame it's not in 2.6.
21:07:21 <oerjan> that's what i meant by "sure"
21:07:30 <ehird> hmm you don't have to write "return 0;" in c++
21:07:33 <ehird> i guess c99 stole that
21:07:36 <ehird> wonder if it's "best practice"
21:08:27 <oerjan> !haskell return 0 :: [Rational]
21:10:01 <oerjan> <AnMaster> and having a function combine
21:11:06 <oerjan> !haskell liftM2 (,) [2,7] [2,7]
21:11:10 <ehird> <AnMaster> and having a function combine
21:11:25 <oerjan> !haskell import Control.Monad; main=print$liftM2 (,) [2,7] [2,7]
21:11:27 <EgoBot> [(2,2),(2,7),(7,2),(7,7)]
21:11:39 <ehird> tellll meeeeeeeee oerjan
21:12:23 <oerjan> a discussion on stupid factorization in the logs, iiuc
21:13:01 <pikhq> Factorization, actually.
21:13:56 <ehird> pikhq: btw your solution to euler #1 is much longer than needs be
21:14:13 <oerjan> also list comprehensions are probably clearer for beginners
21:15:08 <ehird> that's what it was
21:15:17 <ehird> lemme find the code
21:15:29 <oerjan> i wasn't commenting on your comment, btw
21:15:48 <oerjan> but list comprehensions may be a good bet anyhow ;D
21:16:01 <ehird> sum [n | n <- [1..1000-1], n `mod` 5 == 0 || n `mod` 3 == 0]
21:16:20 <pikhq> Instead of my filtering stuff.
21:16:26 <ehird> [|] IS filtering stuff
21:16:40 <ehird> but yours filters through [1..] to find the answer or something, which is bizarre
21:16:41 <pikhq> Not using the filter function, though.
21:16:43 <ehird> also 1000-1 is bizarre
21:16:49 <ehird> pikhq: yes it does
21:16:51 <ehird> concatMap is isomorphic
21:17:02 <ehird> [n | n <- blah, a]
21:17:02 <pikhq> main = print $ sum (filter (\x -> x `mod` 3 == 0 || x `mod` 5 == 0) [1..999])
21:17:04 <pikhq> That's my solution.
21:17:12 <oerjan> !haskell sum [n | n <- [1..999], 0 `elem` map (n `mod`) [3,5]]
21:17:17 <ehird> pikhq: that wasn't what you said first
21:17:30 <ehird> main = print . sum . filter (\x -> x `mod` 3 == 0 || x `mod` 5 == 0) $ [1..999]
21:17:32 <pikhq> ... You're thinking of something different.
21:17:50 <AnMaster> <oerjan> <ehird> AnMaster: his computer crashed :) <-- that hasn't happened in a long while. although occasionally it refuses to start, saying "Operating system not found" <-- how do you fix that when it happens?
21:18:21 <oerjan> it's very intermittent
21:18:36 <AnMaster> oerjan, sounds like harddrive issues, if bootloader can't find the OS
21:18:38 -!- zid has quit (Read error: 60 (Operation timed out)).
21:18:39 <Deewiant> That's a strange error to get intermittently
21:18:42 <ehird> +/~.(3*i.334),5*i.200
21:18:54 <AnMaster> oerjan, I seriously hope you have backups...
21:18:55 <oerjan> and it only seems to happen at booting
21:19:11 <AnMaster> oerjan, I suspect it is due to a harddrive that is nearing it's end of life
21:19:27 <oerjan> it has happened occasionally since i got the computer :D
21:19:29 <oklodok> my computer has started sparkling and smoking
21:19:53 <AnMaster> oklodok, shut it off and unplug everthing?
21:20:05 <oklodok> no no just occasionally when i lift it
21:20:06 <AnMaster> and stand ready with something to put out any fire
21:20:18 <AnMaster> oklodok, err. lift it while running? It is a laptop then?
21:20:42 <oklodok> i don't mind it, except i guess i am a bit afraid of leaving it alone.
21:21:03 <oklodok> i have to put vlc on fullscreen if i don't touch the computer for half an hour, or it crashes
21:21:09 <oklodok> that's another fun thing about this
21:21:26 <AnMaster> oklodok, sounds like it crashes when screen blanking?
21:21:31 <ehird> pikhq: also "sum [3,6..999] + sum [5,10..999] - sum [15,30..999]"
21:21:33 <ehird> which is a fun solution
21:21:43 <ehird> interestingly, the most impressive solution is in PHP.
21:21:48 <ehird> echo 1.5*(int)(($x-1)/3)*(int)(($x+2)/3) + 2.5*(int)(($x-1)/5)*(int)(($x+4)/5) - 7.5*(int)(($x-1)/15)*(int)(($x+14)/15);
21:21:50 <AnMaster> oklodok, because vlc in full screen mode would prevent sleep and screen blanking
21:21:50 <ehird> using that wacky formula thing
21:21:56 -!- zid has joined.
21:21:58 <oklodok> it crashes, can't do anything anymore, have to reboot.
21:22:03 <AnMaster> so theory is that one of them cause it to crash
21:22:29 <AnMaster> oklodok, right, but does it crash in the moment it is about to turn of the monitor due to inactivity or such?
21:22:39 <AnMaster> or put disks into stand-by mode
21:22:44 <oklodok> i've turned all those features off
21:23:06 <oklodok> so something like that, but something vista doesn't explicitly let you control
21:23:19 <oklodok> which fullscreen disables anyway
21:23:23 <AnMaster> ehird, what is that formula supposed to do?
21:23:31 <ehird> AnMaster: see euler problem 1.
21:24:34 <AnMaster> where are solutions listed then?
21:25:01 <ehird> you have to solve it first.
21:25:10 <pikhq> ehird: Nice solution.
21:25:10 <AnMaster> ah, have to create an account and so on then
21:25:29 <ehird> yes, you have to take two seconds to choose a username and password.
21:25:38 <AnMaster> ehird, I don't understand how that formula solves that problem. Does anyone?
21:25:47 <AnMaster> ehird, and then solve the problems too
21:25:57 <ehird> AnMaster: Anyone who knows mathematics, yes. Also, oh god, you have to "Find the sum of all the multiples of 3 or 5 below 1000."
21:25:57 <AnMaster> sure I can solve that first one at least easily enough
21:26:01 <ehird> How could we do it?!
21:26:26 <AnMaster> ehird, it would be one way to solve it
21:26:32 <pikhq> There's a number of easier ways of doing it.
21:26:40 <ehird> It would be a very stupid way. Iteration is almost always retarded.
21:27:09 <oklodok> iterating to 1000 is kinda overkill yeah
21:27:25 <oerjan> ehird: now you're just trolling. even i used iteration, even though i perfectly well know how to calculate triangle numbers
21:27:27 <pikhq> Take the list of numbers below 1000. Remove all those that aren't multiples of 3 or 5. Add up.
21:27:27 <ehird> it's a sum plus a filter.
21:27:30 <ehird> biiiiiiiiiiiiiiiiiiiiig deaaaaaaaaaaaaaaaaaaal
21:27:34 <AnMaster> ehird, summing a list implies some sort of iteration. Even if it is hidden as a sum function
21:27:40 <ehird> oerjan: you used a for loop?
21:27:42 <ehird> is this in haskell?
21:27:47 <ehird> i'd say that's terribly stupid of you if so.
21:27:56 <oerjan> ehird: i consider sum [...] to be iteration
21:28:03 <oklodok> i think oerjan was thinking a different kinda iteration
21:28:03 <ehird> AnMaster: i can just as easily say that a foor loop is a gloss over a filter.
21:28:08 <oerjan> ghc would compile it down to iteration anyway
21:28:15 <ehird> you say it your way because you are cpu-biased
21:28:18 <oerjan> (even though i use hugs)
21:28:18 <ehird> it's theoretically bullshit
21:28:27 <ehird> and a functional expression isn't fundamentally an iteration
21:28:47 <AnMaster> ehird, what sort of computer doesn't implement it through iteration
21:28:59 <pikhq> sum [] = 0;sum (x:xs) = x + sum xs -- I didn't realise that was iteration.
21:29:04 <ehird> AnMaster: ooh, I'd love to see you across history
21:29:10 <ehird> "what sort of flying machine doesn't do it by being lighter than air"
21:29:20 <ehird> therefore, the airplane is bunk. QED.
21:29:26 <oklodok> pikhq: point is you can implement it without any iteration
21:29:29 <oklodok> because you can calculate it
21:29:34 <AnMaster> ehird, are there any examples of it
21:29:37 <ehird> AnMaster: i'm sure you'd have said that when the wright brothers started too
21:29:44 <oklodok> that's still the exact same algorithm
21:29:47 <AnMaster> sure it might be theoretically possible in the future
21:29:51 <oklodok> just less explicit ordering of the computations
21:29:55 <AnMaster> I'm just asking, does any such example exist today
21:30:12 <ehird> you're very stupid AnMaster. i'm surprised you program in anything but machine code.
21:30:42 <AnMaster> I fail to see how you think this insult would make sense.
21:30:44 * oerjan hands out some drama queen crowns |\/\/| |\/\/|
21:31:16 <AnMaster> only thing I'm asking is if there is any current computer that sums a list through anything but iteration
21:31:22 <oklodok> anyway i need to sleep now
21:31:43 <pikhq> AnMaster: Itanium.
21:31:49 <AnMaster> you could of course use for example vector instructions to sum chunks at a time
21:32:16 <oerjan> you could use binary branching too
21:32:25 <AnMaster> iirc you can load 4 32-bit integers in a SSE register and sum them. but what if you have an array not fitting in your vector unit, whatever size it is.
21:32:32 <pikhq> It's got somewhat silly vector instructions.
21:32:57 <AnMaster> pikhq, yes, but tell me of this case that doesn't need iteration to do it
21:33:42 -!- coppro has quit (Remote closed the connection).
21:33:49 <oerjan> sum each half recursively, then combining
21:34:18 <oerjan> if that's what it's called
21:34:36 <ehird> Huh, [1 | True] is [1]. I wonder what source it calls.
21:34:36 -!- coppro has joined.
21:34:38 <AnMaster> oerjan, don't know. But isn't mergesort basically the same? "sort each half, then combine"
21:34:40 <pikhq> Ah, sorry. It still iterates. It just does so in parallel.
21:34:48 <pikhq> With vector operations.
21:35:24 <AnMaster> you could maybe use summing memory
21:36:08 <AnMaster> like CAM but with hardware to sum all in parallel instead of hardware to match all in parallel
21:36:14 -!- ais523 has joined.
21:36:28 <AnMaster> doing that would however be rather silly
21:36:31 <oerjan> AnMaster: we managed to get lambdabot to calculate bigger factorials by using that method with products
21:37:00 <AnMaster> oerjan, which method do you mean?
21:37:26 <oerjan> recursing and combining
21:37:37 <oerjan> it obviously works for any associative operation
21:38:00 <AnMaster> oerjan, sounds like memoising(sp?)
21:38:09 <oerjan> AnMaster: good grief no
21:38:10 <ehird> it's not memoizing at all.
21:38:16 <ehird> wtf made you think that
21:38:51 <oerjan> it's just because it is better to multiply numbers of approximately equal magnitude, than to multiply large numbers by lots of small ones
21:39:07 <oerjan> so thus splitting up a factorial makes it faster
21:39:37 <AnMaster> oerjan, ah hm, maybe there is a reason for that. I don't know how bignums are implemented
21:39:43 <oerjan> we were using ordinary ghc Integers. i think it uses ... damn memory
21:40:35 <oerjan> i think it uses fast fourier transforms for really big numbers
21:41:04 <oerjan> although i'm not sure that applies much to this case
21:41:05 * AnMaster wonders what a slow fourier transforms would be
21:41:28 <ehird> speaking of bignum
21:41:32 <ehird> http://www.catonmat.net/blog/on-the-linear-time-algorithm-for-finding-fibonacci-numbers/
21:42:09 <fizzie> There are five multiplication algorithms; "Basecase", "Karatsuba", "Toom-3", "Toom-4" and "FFT"; they're all chosen by thresholds on the size of the numbers involved. http://gmplib.org/manual/Multiplication-Algorithms.html has the details.
21:42:11 <oerjan> we did something with fibonacci too
21:42:24 <fizzie> Five in GMP, I mean. It is rather unlikely there would be no others.
21:42:44 <ehird> i've always wanted something that's basically "like a float or a double if they were infinite"
21:42:49 <ehird> i guess Few Digits is basically that
21:45:12 <oerjan> ehird: there's some Computable Real library for haskell
21:45:30 <oerjan> those have some issues though, since comparison is undecidable
21:45:38 <ehird> oerjan: Few Digits == CReal
21:45:58 <ehird> !haskell print $ 2/3 :: CReal
21:46:02 <ehird> !haskell main = print $ 2/3 :: CReal
21:46:22 <oerjan> i doubt EgoBot even has that library, but who knows
21:47:03 <oerjan> lambdabot has a lot of extras
21:47:04 <ehird> oerjan: can you ask gwern to put \bot in #esoteric again? i'd feel pushy :P
21:47:18 <oerjan> i'm not even in #haskell you know
21:47:34 <oerjan> haven't been for a year or so
21:48:58 <ehird> Prelude Graphics.Gnuplot.Simple> plotFunc [] [0,0.1..10] sin
21:49:10 <ehird> pikhq: Throw out yer Maxima! Throw out yer MATLAB!
21:49:30 <ehird> that Haskell for Maths thing + cabal install gnuplot -fsplitBase -fexecutePipe
21:50:02 -!- AnMaster_ has joined.
21:50:20 -!- AnMaster has quit (Nick collision from services.).
21:50:22 -!- AnMaster_ has changed nick to AnMaster.
21:50:54 <AnMaster> <oerjan> we did something with fibonacci too
21:51:05 <ehird> oerjan: privmsg gwern then? :P
21:52:04 <AnMaster> ehird, anything directed at me?
21:52:37 <ehird> oerjan finally expressed his gayness by proposing to yo.
21:52:44 <oerjan> AnMaster: computable reals in haskell
21:52:54 <AnMaster> if anyone wanted something important to me they can just say it again.
21:53:06 <oerjan> and bitching about lambdabot not being here to test it
21:53:23 <oerjan> and ehird nagging to have _me_ ask them, when he is the current regular
21:53:34 <ehird> and oerjan proposing to AnMaster
21:54:06 <AnMaster> oerjan, why did that bot leave btw?
21:54:09 -!- amuck_ has joined.
21:54:17 <ehird> a #haskeller already!
21:54:19 <oerjan> AnMaster: it sometimes crashes
21:54:27 <ehird> AnMaster: installing few digits involves darcs and also it's easier.
21:54:36 <ehird> ok, AnMaster has a new thing: it's saying "suuuuuuuuuuuuure".
21:55:04 <oerjan> AnMaster: if it's still like when i was there, it has something like a memory leak issue that no one could find
21:55:08 <AnMaster> ehird, crashing is a side effect ;P
21:55:20 -!- randomity has joined.
21:55:26 <AnMaster> so you need a monad for crashing or somehting
21:55:32 <oerjan> but then it has been extensively changed, at least the @run part iiuc
21:55:40 <ehird> how come #haskellers always come in here in droves when I just mention #esoteic in there
21:55:44 <AnMaster> <ehird> oerjan finally expressed his gayness by proposing to yo.
21:55:46 <ehird> oerjan: it's mueval now.
21:55:57 <ehird> AnMaster: you forgot my correction!
21:56:04 <AnMaster> <ehird> and oerjan proposing to AnMaster
21:56:06 <AnMaster> <AnMaster> ehird, suuuuuuuuuuuuure....
21:56:08 <AnMaster> * Ping reply from oerjan: 57.94 second(s) <-- btw
21:56:22 <randomity> we're all searching for a language that's harder to learn than haskell. #esoteric seems like a natural choice...
21:56:27 <oerjan> ehird: and it no longer gives nice error messages for runtime errors, i noticed
21:56:36 <AnMaster> oerjan, isn't Haskell supposed to have a GC?...
21:56:39 <oerjan> well some errors anyway
21:56:50 <ehird> AnMaster: not if it spawns processes that leak...
21:56:59 <oerjan> AnMaster: GC doesn't prevent all memory leaks
21:57:18 <AnMaster> <ehird> AnMaster: you forgot my correction!
21:57:26 <AnMaster> the lag is so bad I'm over a minute out of sync
21:57:35 <ehird> stop talking until it fixes then
21:57:43 <pikhq> AnMaster: If you keep a reference to it but never use it.
21:57:44 <AnMaster> talking to anyone here would be faster using telegram!
21:58:06 <pikhq> randomity: Haskell was probably the hardest language to learn that I know.
21:58:07 <ais523> I doubt that would be faster than IRC
21:58:12 <AnMaster> <oerjan> AnMaster: GC doesn't prevent all memory leaks <-- true
21:58:13 <oerjan> ehird: #haskell is full of droves, obviously they come
21:58:21 <ais523> you /can/ send telegrams nowadays, although I'm not entirely sure how they're deliveired
21:58:24 <pikhq> Of course, I don't know IINTERCAL.
21:58:34 <ais523> ehird: it is harder to learn than Haskell, though, IMO
21:58:39 <ehird> randomity: i was about to say that
21:58:46 <ehird> if you're not a cryptographer, malbolge is the hardest, prolly
21:58:48 <ais523> I wouldn't say that Malbolge is learnt at all, more cryptanalysed
21:58:54 <ehird> intercal is just weird
21:58:58 <ehird> haskell has a bunch of mind-changers
21:59:00 <AnMaster> ehird, waiting for bouncer to join over 60 channels with freenode's rate limiting? Very funny
21:59:12 <ais523> ehird: well, I found Haskell easier to learn than INTERCAL
21:59:13 <ehird> AnMaster: you have nobody to blame but yourself.
21:59:26 <fizzie> http://en.wikipedia.org/wiki/Telegram#Worldwide_discontinuance_of_telegrams lists a rather sorry state of affairs.
21:59:36 <AnMaster> <ais523> AnMaster: telegram? <ais523> I doubt that would be faster than IRC <-- I feel for ehird. Sometimes.
21:59:38 <AnMaster> * Ping reply from oerjan: 70.49 second(s)
21:59:39 <pikhq> ais523: You must love functions.
22:00:01 <pikhq> Nobody knows all of C++. :P
22:00:05 <ais523> pikhq: disagree, C++ is easy to learn, just hard to learn well
22:00:12 <ehird> i wonder how i should represent irrationals
22:00:18 <AnMaster> ehird, on other servers it is done much much faster
22:00:20 <pikhq> ais523: A subset of C++ is easy to learn.
22:00:35 <pikhq> This is the commonly used subset of the language.
22:00:38 <AnMaster> it is just that freenode's rate limiting sucks so much
22:00:55 -!- Associat0r has quit (Read error: 104 (Connection reset by peer)).
22:01:01 <pikhq> If you want some mind-warping, do functional programming with the type system.
22:01:18 <ais523> AnMaster: I still guarantee you, freenode is faster than telegrams
22:01:34 <ehird> [[Hippasus, however, was not lauded for his efforts: according to one legend, he made his discovery while out at sea, and was subsequently thrown overboard by his fellow Pythagoreans “…for having produced an element in the universe which denied the…doctrine that all phenomena in the universe can be reduced to whole numbers and their ratios.”]]
22:01:34 <ais523> I'm not even sure who handles telegrams nowadays; possibly the postal service
22:01:38 <ehird> old mathematics are hilarious
22:01:42 <pikhq> randomity: As is the rest of C++.
22:01:49 <ais523> once they've been sent to near their destination electronically
22:01:55 <randomity> it's like haskell, without closures, where lambda is template <class T> struct { typedef ... }
22:02:14 <oerjan> randomity: unlambda is harder than INTERCAL, but not as hard as Malbolge. in my opinion.
22:02:26 <oerjan> (not that i've programmed in malbolge)
22:02:48 <AnMaster> ais523, it is joining over 70 channels
22:02:52 <pikhq> Unlambda's not all that hard to learn. Hard to use well, perhaps, but it is just combinators.
22:03:04 <oerjan> i made a variant though. not that anyone has programmed that either.
22:03:10 <AnMaster> ais523, I invite you to #feather-lang btw :P
22:03:20 <pikhq> And yes, people don't program in Malbolge.
22:03:36 <pikhq> They do genetic programming for it.
22:03:54 <ehird> plenty of people since have written in it
22:04:00 <ehird> one guy mastered it by poking it with a stick
22:04:08 <ehird> two other people basically cryptanalyzed it
22:04:22 <ehird> pikhq: there IS a looping 99 bottles of beer in it y'know
22:05:14 <ehird> pikhq: non-cooke material: http://www.lscheffer.com/malbolge.shtml, http://www.99-bottles-of-beer.net/language-malbolge-995.html
22:05:17 <ehird> http://esoteric.voxelperfect.net/wiki/Malbolge_programming
22:05:25 <AnMaster> sure telegram would be slower?
22:05:32 <ehird> lemme find the second person to do it
22:05:48 <ehird> http://www.antwon.com/index.php?p=234, darn it disappeared
22:05:56 <ehird> Hey there, party people. This is Antwon, checking in with his new and exciting web server from whence he can resurrect the full antwon.com experience.
22:05:56 <ehird> As you may have noticed, all of those nifty goodies like "features" and "content" aren't exactly displaying at this point in time. At the moment, all of my umpteen bazillion posts are stored away in a singular XML file - awesome from a "I did not lose this content!" standpoint, but less so from a "people can actually read my hard-fought years of maniacal writing!" vantage. Rest assured that I am working on things and will get something together Real Soon
22:06:03 <ehird> meh, web.archive.org away.
22:06:26 <ehird> http://lambda-the-ultimate.org/node/823
22:07:17 * ais523 reads discussion about Microsoft's announcement about Mono patents
22:07:27 <ehird> pikhq: there's a compiler to malbolge apparently
22:07:28 <ehird> http://www.is.nagoya-u.ac.jp/thesis/M2005/i/M350402019e.pdf
22:07:31 <ehird> http://www.sakabe.i.is.nagoya-u.ac.jp/~nishida/DB/pdf/iizawa05ss2005-22.pdf (japanese)
22:07:41 <ehird> from a C-like language
22:07:52 <pikhq> I'm fucking frightened.
22:07:59 <ehird> ais523: the responses from the trolls to that were great
22:08:29 <ehird> ais523: they basically consisted of "that's not a legal promise!" (actually, it was), "they won't hit you, perhaps... instead they'll rape you!" (paraphrased…slightly), etc
22:08:42 <ais523> yep, there are only two issues that look legit: a) it doesn't cover several common libraries, only C# and .NET itself, and b) Microsoft could sell the patents to someone else, and the someone else could then sue
22:08:57 <ehird> (a) doesn't matter for most mono apps due to using gtk# and stuff
22:09:03 <ehird> although asp.net mvc is kinda popular, whatever
22:09:12 <ais523> however, I don't think Microsoft particularly care about Mono + GTK#
22:09:24 <ais523> after all, it hardly lets people run Windows programs
22:09:38 <ehird> there isn't a windowsforms implementation either
22:09:43 <ehird> so really, just asp.net is the issue
22:09:49 <ehird> and it's not even an issue anyway
22:09:54 <ehird> people just like to complain :P
22:10:05 <ehird> estoppel was mentioned in the reddit comments, so…
22:10:06 -!- ehird has changed nick to estoppel.
22:10:14 <ais523> yep, that's what point b) is about
22:10:20 <ais523> Microsoft is estopped, nobody else is though
22:11:24 <estoppel> http://ciphersaber.gurus.org/WTCliberties.gif ← I want to make a motivational of this with just a title: "POLITICAL CARTOON"
22:11:35 <estoppel> ais523: I'm pretty sure such a case would be rejected under estoppel anyway; isn't it vague?
22:11:54 <ais523> estoppel: that would be a massive loophole
22:12:09 <ais523> if I promised not to sue you for pirating someone else's music, that wouldn't stop the RIAA doing it
22:13:46 -!- oklodok has quit (Read error: 104 (Connection reset by peer)).
22:13:48 <oerjan> this brings to mind that in norway at least you can put restrictions on what people can do with ground property, even when resold. isn't there anything similar in other countries?
22:13:52 <estoppel> [[On second glance, the C code is obviously just a Malbolge interpreter, so I guess there's not really anything useful to get from it (unless we can get it translated). --Rune 21:12, 1 Jun 2006 (UTC) ]]
22:13:57 -!- oklodok has joined.
22:14:07 <oerjan> i'm not sure if applies to anything but estate
22:14:20 <ais523> actually, it's quite hard to be /obviously/ a Malbolge interpreter
22:14:30 <ais523> surely nobody's memorized the encryption table?
22:14:46 <ais523> without knowing that, it would be hard to tell if something was implementing Malbolge, or a similar language with different encryption
22:15:02 <ais523> according to HackEgo, anyway
22:15:10 <oerjan> AnMaster: oklodok's pet word. be kind to it.
22:15:24 <AnMaster> <estoppel> [[On second glance, the C code is obviously just a Malbolge interpreter, so I guess there's not really anything useful to get from it (unless we can get it translated). --Rune 21:12, 1 Jun 2006 (UTC) ]] <-- source of citation?
22:17:30 * pikhq wants a Malbolge self-interpreter.
22:17:40 <pikhq> Erm. Malbolge interpreter in Malbolge.
22:18:06 <AnMaster> pikhq, could only interpret a smaller program
22:18:16 <estoppel> [[Irrational number]] doesn't have any ideas for representation :(
22:18:20 <AnMaster> estoppel, connect input to output
22:18:24 <pikhq> Oh, right. They made it arbitrarily finite.
22:18:42 <AnMaster> estoppel, I read about that somewhere else
22:19:07 <pikhq> Malbolge + PSOX is Turing-complete.
22:19:28 <pikhq> Assuming you can do the relevant PSOX support code, that is.
22:19:53 <AnMaster> pikhq, can't you just use PSOX right away as it currently is?
22:20:13 <pikhq> I think you misunderstood.
22:20:29 <pikhq> *Malbolge* code to actually talk with PSOX.
22:20:45 <AnMaster> pikhq, well that should be possible in theory
22:20:53 <AnMaster> if you can fit it in the length of the program
22:21:04 <pikhq> That's the tricky bit.
22:21:06 <ais523> wow at this statistic: fine against Jammie Thomas per song copied = $84000; compensation for family of each person who died in the Air France crash = $24000
22:21:30 <pikhq> ais523: Average yearly income ~= $40,000.
22:21:45 <ais523> that fits well with the other two
22:21:46 <oerjan> pikhq: the basic operations of malbolge are such that it is hard _not_ to make it arbitrarily finite. that's what i tried to fix in my Unshackled variant.
22:22:20 <estoppel> ais523: that $24,000 was the original payment
22:22:41 <ais523> it's statistics, it doesn't have to be correct
22:23:17 <oerjan> AnMaster: the program is loaded into memory, so same size bound
22:23:44 <oerjan> you didn't say the readon
22:23:58 <AnMaster> oerjan, I knew it, I just didn't spell it out.
22:24:41 <estoppel> oerjan: is there a way to represent the irrationals apart from as a computation (polynomials count as that)
22:29:19 <oerjan> some irrationals cannot be represented as computations >:)
22:29:45 <estoppel> oerjan: well, ok, but they're useless :)
22:29:51 <estoppel> oerjan: assume computable rationals
22:30:21 <pikhq> The computable irrationals may only be represented as a computation or an infinite series of digits.
22:30:35 <pikhq> (in Haskell, of course, the two are equivalent. ;))
22:30:48 <estoppel> pikhq: problem with digits is you have to pick a base, and that sucks.
22:31:12 <estoppel> and computations can't be compared.
22:31:48 <pikhq> An infinite series of digits may be compared, but if they're equal, the comparison will never terminate.
22:32:22 <oerjan> estoppel: oh. in that case use continued fractions. no base involved.
22:32:35 <fizzie> "Also the (==) function returns false if the two CReals are different, and does not terminate if the two CReals have the same value."
22:33:03 <estoppel> oerjan: hmm can continued fractions represent rationals?
22:33:12 <oerjan> estoppel: it may be using a cutoff?
22:33:19 <estoppel> oerjan: that would be Evil, I'm sure not.
22:33:27 <oerjan> estoppel: sure, they are _finite_ continued fractions
22:33:38 <estoppel> that's no continued fraction :D
22:33:52 <pikhq> oerjan: It's infinite, actually.
22:34:09 <pikhq> And an infinite number of them have 0 in the numerator.
22:34:12 <estoppel> oerjan: but (significand,exponent) is a bit more efficient than using continued fractions for rationals
22:35:36 <oerjan> estoppel: well you could use a mixed representation
22:36:04 <estoppel> well (significand,exponent) sucks anyway because it's quite close to picking a base
22:36:20 <estoppel> also it doesn't handle infinite-didget endowed rationals
22:36:27 <estoppel> i forgot now to spell digit...
22:37:01 <oerjan> erm rationals don't have infinite digits
22:37:23 <estoppel> oerjan: well that's true if you can pick a base
22:37:34 * oerjan swats himself -----###
22:37:51 <estoppel> 'cuz, like, you're a mathematician
22:38:04 <oerjan> i was thinking numerator/denumerator
22:38:58 <oerjan> i see 2 digits there, in total :D
22:39:39 <estoppel> anyway right, you can't represent that as (significand,exponent)
22:39:43 -!- FireyFly has joined.
22:40:14 <estoppel> (as a ... non-continued fraction)
22:40:33 <estoppel> continued fractions reduce to rationals
22:41:02 <estoppel> oerjan: is there a name for 0.01101110010111011110001001…?
22:41:19 <estoppel> that is, 0.{all integers from 0 up, in binary, concatenated}
22:41:59 <estoppel> I get how 1/x reduces obviously with a continued fraction
22:42:05 <estoppel> now i need to figure out how to do 2/3 that way
22:42:18 <oerjan> euclidean algorithm essentially
22:42:57 <estoppel> oerjan: that sounds dangerously Slow with a capital S.
22:43:22 <oerjan> it's the algorithm with divmod too
22:43:29 <estoppel> oerjan: calculating n/m is O(whatever euclidean is), not O(1)
22:43:40 <estoppel> with (n,m) representation, it's O(1)
22:43:58 <estoppel> oerjan: what about continued fractions, but instead of just 1 as the numerator,
22:44:05 <oerjan> bignums aren't really O(1) anyhow :D
22:44:10 <estoppel> then it just reduces to 2/3 = 0 + 2/3
22:44:15 <estoppel> indeed that's the subject of the post I linked
22:44:19 <estoppel> http://www.catonmat.net/blog/on-the-linear-time-algorithm-for-finding-fibonacci-numbers/
22:44:25 <estoppel> oerjan: but would that modification change anything?
22:45:22 <oerjan> i don't know how much it changes, it's not like i've tried implementing arithmetic operations on c.f. of either kind
22:45:34 <oerjan> those with only 1 are called simple, iirc
22:45:43 <estoppel> data CF = Add Integer CF | Div Integer CF | Zero
22:46:21 <oerjan> one thing it messes up is uniqueness of representation, obviously
22:46:32 <estoppel> oerjan: what's wrong with that if the representation isn't exposed?
22:46:52 <oerjan> it may make it harder to do comparisons?
22:47:06 <estoppel> oerjan: i'll just have a "simplify" function
22:48:41 <estoppel> oerjan: hmm an issue is that I can't `show` irrationals
22:49:10 <oerjan> a cutoff seems inevitable
22:49:34 <estoppel> oerjan: but entering "fullPi" in a REPL and getting an infinite output would be such fun :D
22:49:49 <estoppel> with infinite digits I can do that, with continued fractions not :<
22:49:59 <estoppel> infinite digits has less dependencies, essentially
22:50:06 <estoppel> whereas continued fractions have infinite dependencies
22:50:18 <oerjan> estoppel: you can have a showFull function, of course
22:50:32 <estoppel> oerjan: of type showFull :: CF -> String?
22:50:36 <estoppel> then why can't I have show :: CF -> String?
22:50:51 <oerjan> it's just that ordinary show is used for embedding things in larger data structures, so you need it to end if you want to show the rest of it
22:51:07 <estoppel> I thought you meant it was really impossible :)
22:51:11 <estoppel> oerjan: but I can't end it eg for pi
22:51:27 <estoppel> there'll never be a Read CF that covers everything
22:51:36 <oerjan> well you've got to make a choice then
22:51:37 <estoppel> oerjan: a cutoff is reasonable, though
22:52:00 <estoppel> oerjan: not sure how to encode it though; show is stateless
22:52:05 <estoppel> i guess show = myShow 0 rather
22:53:05 <estoppel> oerjan: i think a simplified continued fraction is best
22:53:11 <estoppel> oerjan: so I have one representation, except I won't
22:53:33 <estoppel> oerjan: because you can always make an infinite representation, can't you?
22:53:50 <oerjan> there is also the question that some numbers are undecidable to print to complete representation
22:54:24 <oerjan> actually the undecidability might happen already when calculating the c.f.
22:55:35 <estoppel> oerjan: this doesn't sound nice or fluffy to me.
22:55:46 <oerjan> estoppel: pikhq mentioned you could just append 0's. 1/(0+1/(0+...
22:56:00 <estoppel> oerjan: that makes > show 3 → "3<<HANG>>
22:56:15 <estoppel> oerjan: he was talking about unrestricted numerator.
22:56:32 -!- MigoMipo has joined.
22:56:34 <estoppel> oerjan: with restricted-to-1 numerator, you need a special base case
22:56:46 <estoppel> data CF = Add Integer OneDiv | Zero
22:56:51 <estoppel> ↑ pretty sure this representation is unique
22:57:14 <estoppel> oerjan: if I make sure that the Integer in Add is >=1, that representation is unique, isn't it?
22:57:21 <estoppel> all rational and irrational numbers have one representation
22:57:59 <oerjan> um you are missing numbers 0 <= ... < 1 there...
22:58:36 <estoppel> oerjan: but assuming numbers <1 don't exist
22:59:17 -!- olsner has joined.
22:59:51 <oerjan> you have infinity in there
23:00:00 <oerjan> i think Zero is in the wrong place
23:00:02 <estoppel> if it's impossible to get unique representations for all rational and irrational numbers, please do tell me, btw
23:00:40 <pikhq> estoppel: I strongly suspect it's possible to get unique representations for all rational numbers.
23:01:01 <pikhq> The irrational numbers cannot necessarily be represented.
23:01:03 <oerjan> but you also want to disallow one of 0 + 1/1 or 1
23:02:05 <pikhq> estoppel: Then it's obvious that there's a unique representation for them.
23:02:35 <estoppel> data OneDiv = OneDiv CF | Zero
23:02:35 -!- FireyFly has quit (Connection timed out).
23:02:57 <estoppel> oerjan: so from this what constraints do we use? i want to include negative numbers and 0 here
23:03:28 <oerjan> negative numbers needs sign somehow
23:03:59 <oerjan> that's just going to mess up things
23:04:20 <oerjan> but you already disallowed Add 0 _
23:04:30 <estoppel> oerjan: btw we can still represent 1/0
23:04:47 -!- M0ny has quit.
23:05:27 <estoppel> data CF = Add {-(/= 0)-} Integer OneDiv
23:05:27 <estoppel> data OneDiv = OneDiv CF | Zero
23:05:35 <oerjan> you seriously need a data type that is >= 1 only
23:05:46 <estoppel> oerjan: yes, but I'm not using bloody peano
23:05:58 <estoppel> oerjan: these constructors aren't public, anyawy
23:06:02 <estoppel> so nobody else can access them
23:06:15 <estoppel> oerjan: well, what would I do?
23:06:28 -!- MigoMipo has quit ("QuitIRCServerException: MigoMipo disconnected from IRC Server").
23:06:55 <oerjan> require >= 1 integer in CF there
23:07:12 <oerjan> and have another data type for all reals
23:07:26 <estoppel> oerjan: 23:06 oerjan: require >= 1 integer in CF there ;; with what? I don't really want to, as gmp is fast.
23:08:29 <oerjan> well you'll just have to have runtime assertion then
23:08:42 <estoppel> oerjan: what does that buy me? nothing at all
23:08:52 <estoppel> over just not writing (Add n) without checking n isn't 0
23:08:59 <oerjan> estoppel: i'm not saying literally
23:09:18 <estoppel> oerjan: "{-(/ 0)-} Integer" is my pet constraint description convention
23:09:44 <oerjan> i mean if you insist on using Integers, while c.f.'s are based on naturals, then you'll have to have runtime assumptions
23:10:18 <oerjan> also, you seriously don't want 1/(-1 + 1/...) to occur
23:10:52 -!- FireFly has joined.
23:10:57 <oerjan> no uniqueness whatsoever?
23:11:37 <estoppel> oerjan: can all negative reals be represented as -real?
23:12:20 <estoppel> data Add = Add {-(> 0)-} Integer OneDiv
23:12:21 <estoppel> data OneDiv = OneDiv Add | Zero
23:12:23 <estoppel> data CF = Positive Add | Negative Add
23:12:29 <estoppel> other constraints, i assume, must be applied.
23:12:51 <oerjan> CF needs a few more cases
23:13:06 <oerjan> you are missing all of (-1, 1) i think
23:13:30 -!- Pthing has quit (Remote closed the connection).
23:13:40 <estoppel> oerjan: right you are. that's why I have > 0 there.
23:13:46 <estoppel> unfortunately, solving one problem introduces another.
23:13:55 <estoppel> oerjan: do you mean like adding | MinusOne Add?
23:14:09 <estoppel> MinusOne (Add 1 (OneDiv (Add 2 Zero)))
23:15:00 <oerjan> that certainly destroys uniqueness again
23:15:14 <oerjan> 0.4 is logically contained in OneDiv
23:15:25 <estoppel> oerjan: no it's not, due to the Add dependency
23:15:53 <estoppel> "we've made it easy to re-enable the beta label for Gmail from the Labs tab under Settings"
23:16:48 <ais523> oerjan: yes, because 4.0 = 1 / (0.25)
23:16:49 <estoppel> oerjan: is 2.5 contained in Add, though?
23:16:53 <oerjan> also you need to be careful about negative zero
23:17:06 <estoppel> oerjan: so I don't miss (-1,1).
23:17:30 <estoppel> oerjan: 23:16 oerjan: 0.4 = 1/(2.5) ?
23:17:44 <oerjan> but 0.4 is _not_ contained in Add
23:17:53 <oerjan> Add and OneDiv are disjoint
23:18:10 <estoppel> oerjan: this is terribly confusing
23:18:38 <oerjan> that's because you're trying too hard to make a small number of elegant cases, i think
23:19:00 <estoppel> oerjan: i just keep having to restructure as you come up with more things :D
23:19:53 <estoppel> data Add = Add {-(> 0)-} Integer OneDiv
23:20:25 <oerjan> problem is you want something that is OneDiv except Zero
23:20:33 <estoppel> data OneDiv = OneDiv Add | Zero
23:20:49 <oerjan> otherwise you get a negative Zero when putting things together the obvious way
23:21:04 <estoppel> oh, you mean Negative (Add 0 Zero)?
23:21:10 -!- BeholdMyGlory has joined.
23:21:22 <oerjan> Add 0 Zero does not exist
23:21:31 <estoppel> oerjan: so we DON'T have negative zero
23:21:44 <oerjan> estoppel: you have not put everything together yet
23:21:55 <oerjan> you are still missing (-1, 1) in the final datatype
23:22:01 <estoppel> true. and i won't for, say, 12 hours. buh bye! thanks for your help.
23:22:13 * oerjan swats estoppel -----###
23:22:17 -!- amuck_ has quit.
23:22:50 -!- estoppel has changed nick to ehird.
23:24:37 <pikhq> IE usage is now down to 56% overall.
23:25:25 -!- coppro has quit ("The only thing I know is that I know nothing").
23:28:08 -!- olsner has quit ("Leaving").
23:29:32 -!- sebbu has quit ("@+").
23:31:45 -!- coppro has joined.
23:34:28 <ais523> hmm... latest browser news: statistics show IE market share dropped 8% in a month, nobody believes them
23:34:47 <ais523> the web statistic sites are assuming it's some sort of error in the way the data was collected
23:35:03 <oerjan> well 46% of all statisticians are liars, anyway