←2013-11-16 2013-11-17 2013-11-18→ ↑2013 ↑all
00:00:33 <oerjan> <Taneb> After an incident a couple of weeks ago <-- ic, *opens caffeinated soft drink*
00:01:19 <oerjan> this + the pain killers better remove my headache.
00:02:35 * oerjan imagines Taneb running frantically around on a caffeine high
00:04:02 -!- prooftechnique has joined.
00:05:10 <Taneb> oerjan, it was more that I was visibly shaking and felt like my entire digestive system was trying to go on holiday
00:05:25 <Phantom_Hoover> oh i felt that way yesterday
00:05:38 <zzo38> I am typing the Dungeons&Dragons game recording; not finished yet but will be soon, hopefully.
00:05:39 <Phantom_Hoover> then i ate some food and it went away
00:05:42 <Taneb> And then I stopped visibly shakin
00:05:43 <Taneb> g
00:05:51 <oerjan> Taneb: ok i know that second part. if i take more than about two cups of coffeee.
00:05:56 <Taneb> And couldn't stop smiling and giggling
00:06:16 <Taneb> And I couldn't feel my hands
00:06:25 -!- Sgeo has quit (Read error: Connection reset by peer).
00:06:33 <oerjan> how f much did you drink
00:06:51 <Taneb> Two pints, on an empty stomach
00:07:00 <oerjan> `frink 2 pints -> l
00:07:10 <HackEgo> 473176473/500000000 (exactly 0.946352946)
00:07:12 <olsner> 2 pints of coffee?
00:07:13 <zzo38> I managed to get an extra 300 XP for an idea I had about what to do next, just at the end of the session, and this was just enough to advance to 27 ECL (20 HD), so maybe these two new spells might help a bit.
00:07:16 <Taneb> olsner, of cola
00:07:32 <oerjan> was it sugary as well, i expect that doesn't help
00:07:37 <olsner> hmm, that's only like half a 2l bottle
00:07:39 -!- Sgeo has joined.
00:07:40 <myname> i do think, writing java makes me urinate more
00:07:54 <Taneb> oerjan, yeah, it was coca cola
00:08:01 <Taneb> Very sugar, much caffeine
00:08:33 * oerjan drinks coca cola zero
00:09:01 <Taneb> I drink coke zero now when I haven't eaten much
00:09:56 <oerjan> myname: fascinating!
00:10:14 <myname> oerjan: indeed
00:10:38 <myname> it may be SOAP, though
00:11:08 <shachaf> oerjan: did you logread the whole space nonsense from last night
00:11:25 <oerjan> shachaf: i quite firmly skipped most of it.
00:11:38 <shachaf> oerjan: h8r
00:11:50 <shachaf> imo read it and gimme your insights
00:12:00 <olsner> was that the thing about infinite velocity black holes and whatnot?
00:12:02 <oerjan> my concentration would never last through all of that these days.
00:12:22 <doesthiswork> It's always disappointing when someone cites the laws of form seriously.
00:12:23 <oerjan> olsner: no, it was chu spaces and category theory and adjunction and a horrible mess
00:12:30 <oerjan> *adjunctions
00:13:38 <oerjan> i've never read the laws of form, but i vaguely recall some claim it's just an encoding of propositional logic or thereabouts.
00:13:59 <doesthiswork> yes it is
00:14:13 <doesthiswork> but people cite it for the metaphysics
00:14:14 <Taneb> Hang on, I've got a vague feeling I may have triggered this discussion
00:14:19 <doesthiswork> nope
00:14:30 <Taneb> Oh thank god
00:14:31 <doesthiswork> unless you mean chu spaces
00:14:40 <Taneb> That's what I meant
00:14:46 <oerjan> oh.
00:14:46 <doesthiswork> ok then yes
00:15:27 <Taneb> OK, I sincerely apologize
00:15:58 <oerjan> Taneb: you know, don't start going all Sgeo on worrying about what you're doing tdnh
00:16:34 <Taneb> oerjan, weren't you banned from variants of "hth"?
00:16:44 <Taneb> Also I don't know what you mean by tdnh
00:16:45 <oerjan> yes, but i have cheating methods
00:16:49 <oerjan> `
00:16:51 <HackEgo> ​/home/hackbot/hackbot.hg/multibot_cmds/lib/limits: line 5: exec: : not found
00:16:53 <oerjan> oops :P
00:16:58 <oerjan> `? tdnh
00:17:01 <HackEgo> tdnh does not help
00:17:13 <doesthiswork> `? hth
00:17:13 <oerjan> (that was me forgetting to use it)
00:17:15 <HackEgo> hth is help received from a hairy toe. It is not at all hambiguitous.
00:17:38 <doesthiswork> that explains it finally
00:17:54 <oerjan> Taneb: i wasn't banned, elliott wrote a censoring script.
00:19:18 <Taneb> I could do with an ice cream
00:19:35 <Taneb> I wonder if a) the petrol station next door is open and b) sells ice cream
00:19:49 <zzo38> What kind of ice cream did you want?
00:19:51 <oerjan> hm i was going to complain to Gregor that the logs broke the windows zooming shortcut, but then i realized i'd been pressing the neutralize zooming shortcut instead
00:20:01 <Taneb> zzo38, vanilla, preferably
00:20:07 <oerjan> (also that he'd just laugh evilly at me)
00:20:43 <doesthiswork> why did tunes.org choose to log esoteric?
00:21:06 <oerjan> i dunno, it happened before i got here
00:21:12 <oerjan> fizzie might know?
00:21:25 <olsner> Taneb: why not tar?
00:21:53 <Taneb> olsner, I am not a huge fan of tar flavoured ice cream
00:22:19 <oerjan> miksi ei tar
00:22:31 <oerjan> i'm suspicious on that last word, google translate
00:22:40 <oerjan> *of
00:23:44 * oerjan just realizes that fizzie might be the only active one here who's been here longer than him...
00:23:55 <oerjan> oh wait
00:24:00 <oerjan> Gregor, surely
00:26:40 <Taneb> Right, I'm seeing if the petrol station's shop is open
00:26:52 <oerjan> good luck on your quest
00:27:27 -!- yorick has quit (Remote host closed the connection).
00:28:13 <olsner> `quote petrol
00:28:15 <HackEgo> 345) <oklofok> what would you ever need petrol for <oklofok> newsflash: it doesn't actually taste that good \ 346) [on petrol] <ais523> oklofok: it's actually poisonous, so I advise against drinking it <Phantom_Hoover> ais523, also contains benzene, my carcinogen of choice.
00:28:37 <oerjan> `quote gasoline
00:28:39 <HackEgo> No output.
00:29:02 <shachaf> ais523 was on petrol?
00:29:11 <shachaf> #drugz
00:30:08 <oerjan> kmc: you should swat shachaf, he made a drugz joke not involving you
00:30:41 <oerjan> or he could if he weren't hopelessly idle
00:31:41 <shachaf> he's in korea and/or japan right now
00:32:05 <oerjan> @tell ais523 <ais523> based on comments he made in #esoteric, he used to be with some if he disliked <-- i'm too noise sensitive to have flatmates, period.
00:32:06 <lambdabot> Consider it noted.
00:32:40 <oerjan> samsung digital city, the small autonomous state
00:33:55 <Taneb> Well, that ice cream was certainly ice cream
00:34:12 <oerjan> "Samsung Town is twinned with Disneyland Paris and Sony World (Tokyo, Japan). The "twinning" is only symbolic and has no legal significance."
00:34:22 <oerjan> if you say so
00:34:39 <Taneb> I don't think twinning has any legal significance in any case
00:35:07 <oerjan> i recall trondheim has a twin city in israel and also one in palestine.
00:35:38 <oerjan> as well as several in europe
00:36:22 <oerjan> wait, that's friendship city, but presumably essentially the same thing.
00:37:19 <oerjan> Taneb: i suppose it might not look good if samsung and sony had an antitrust case against them...
00:40:35 <Taneb> "The streets had been laid down, but no houses were built; merely a row of cardboard boxes housed the first few residents who had migrated east from the slums of Ashburton to enjoy a better standard of living, only to be bitterly disappointed with the estate agents who promised milk and honey but instead delivered a mosquito infested swamp. The Ashburton immigrants, while disappointed all agreed that the swamp was a major step up from their former homes."
00:40:51 <oerjan> `addquote <ais523> my sleep schedule is not actually random, but I'm not convinced it compresses well
00:40:55 <HackEgo> 1134) <ais523> my sleep schedule is not actually random, but I'm not convinced it compresses well
00:41:50 <oerjan> Taneb: wtf
00:42:08 <Taneb> oerjan, from the Wikipedia page of the suburb I lived in when I lived in Australia
00:42:48 <Taneb> http://en.wikipedia.org/wiki/Mount_Waverley,_Victoria
00:44:46 <oerjan> monash, the ancient grounds of agora nomic
00:46:28 <Taneb> I could have met the founders of agora nomic!?
00:46:32 <Taneb> When I was 4!?
00:46:38 <coppro> I've met waggie
00:47:49 <oerjan> me too!
00:48:14 -!- ^v has quit (Quit: http://i.imgur.com/MHuW96t.gif).
00:48:22 <oerjan> Taneb: well Steve i suppose.
00:49:41 <oerjan> agora looks disturbingly quiet.
00:55:38 <oerjan> `unidecode ™"
00:55:41 <HackEgo> ​[U+2122 TRADE MARK SIGN] [U+0022 QUOTATION MARK]
00:56:06 <oerjan> very useful, this command
00:56:48 <oerjan> <prooftechnique> Wolfram Language will be the last programming language we ever need™ <-- i assume because it will cause the end of the world somehow.
00:57:47 <Bike> `unicode TRADE MARK SIGN
00:57:49 <HackEgo> ​™
00:57:51 <Bike> cool
00:58:04 <shachaf> how ™cool™
00:59:30 <myname> alt gr+shift+8
01:00:27 <Bike> ¾?
01:00:40 <myname> it's ™ here
01:05:16 -!- prooftechnique has quit.
01:06:52 <Phantom_Hoover> compose-t-m
01:07:16 <Phantom_Hoover> note to self: make a better compose setup
01:08:30 <oerjan> `unicode REGISTERED TRADE MARK SIGN
01:08:32 <HackEgo> Unknown character.
01:08:39 <oerjan> soo fickle
01:08:50 <oerjan> `unicode REGISTERED SIGN
01:08:52 <HackEgo> ​®
01:12:52 <doesthiswork> `unicode COPYRIGHT SIGN
01:12:53 <HackEgo> ​©
01:13:03 <shachaf> copyright ⓒ
01:14:17 <doesthiswork> shachaf™
01:14:52 <oerjan> `unidecode ⓒ
01:14:54 <HackEgo> ​[U+0020 SPACE] [U+24D2 CIRCLED LATIN SMALL LETTER C]
01:17:29 <shachaf>
01:19:40 <oerjan> `run unidecode Ⓒ #oh no, shachaf is onto me
01:20:01 <Phantom_Hoover> `unidecode
01:20:02 <HackEgo> Traceback (most recent call last): \ File "/hackenv/bin/unidecode", line 4, in <module> \ print u" ".join("[U+{0:04X} {1}]".format(ord(c), unicodedata.name(c, "DUNNO")) for c in " ".join(sys.argv[1:]).decode("utf-8")).encode("utf-8") \ File "/opt/python27/lib/python2.7/encodings/utf_8.py", line 16, in decode \ return codecs.utf_8_decode
01:20:07 <Phantom_Hoover> `run unidecode
01:20:08 <HackEgo> Traceback (most recent call last): \ File "/hackenv/bin/unidecode", line 4, in <module> \ print u" ".join("[U+{0:04X} {1}]".format(ord(c), unicodedata.name(c, "DUNNO")) for c in " ".join(sys.argv[1:]).decode("utf-8")).encode("utf-8") \ File "/opt/python27/lib/python2.7/encodings/utf_8.py", line 16, in decode \ return codecs.utf_8_decode
01:20:21 <Phantom_Hoover> ...am i on that stupid encoding still
01:20:30 <shachaf> Phantom_Hoover.......................................................................
01:20:35 <shachaf> `unidecode ©
01:20:37 <HackEgo> ​[U+00A9 COPYRIGHT SIGN]
01:20:39 <Bike> c, DUNNO
01:20:45 <Phantom_Hoover> yes
01:20:49 <Phantom_Hoover> fuck's sake xchat
01:20:54 <Phantom_Hoover> `run unidecode
01:20:56 <HackEgo> Traceback (most recent call last): \ File "/hackenv/bin/unidecode", line 4, in <module> \ print u" ".join("[U+{0:04X} {1}]".format(ord(c), unicodedata.name(c, "DUNNO")) for c in " ".join(sys.argv[1:]).decode("utf-8")).encode("utf-8") \ File "/opt/python27/lib/python2.7/encodings/utf_8.py", line 16, in decode \ return codecs.utf_8_decode
01:21:04 <Phantom_Hoover> ok fuck this shit
01:23:06 -!- Taneb has quit (Ping timeout: 245 seconds).
01:23:29 -!- Taneb has joined.
01:29:27 <Phantom_Hoover> i like how useless the aleph numbers are
01:32:03 -!- Phantom_Hoover has quit (Quit: Leaving).
01:35:05 -!- ggherdov has quit (Ping timeout: 252 seconds).
01:35:28 <Sgeo> /nick LordEnglish
01:36:33 -!- lambdabot has quit (Quit: move to better server, be right back).
01:38:18 <oerjan> NOOO LAMBDABOT, IT'S A TRAP!
01:41:05 -!- lambdabot has joined.
01:42:08 <int-e> @botsnack
01:42:08 <lambdabot> :)
01:42:31 <int-e> @let fibs = fix ((0:) . scanl (+) 1)
01:42:32 <lambdabot> Defined.
01:42:36 <int-e> > fibs
01:42:37 <lambdabot> [0,1,1,2,3,5,8,13,21,34,55,89,144,233,377,610,987,1597,2584,4181,6765,10946...
01:42:48 <myname> huh
01:42:54 <myname> someone explain
01:43:08 <myname> i know the 0:1:zipwith-way
01:43:25 <int-e> > scanl (+) 0 [a,b,c]
01:43:26 <lambdabot> [0,0 + a,0 + a + b,0 + a + b + c]
01:43:59 <shachaf> myname: Isn't it more fun to figure it out yourself?
01:44:51 <myname> it kinda makes sense
01:44:58 <myname> what does fix do?
01:45:22 <int-e> > fix f
01:45:23 <lambdabot> No instance for (Debug.SimpleReflect.Expr.FromExpr a0)
01:45:23 <lambdabot> arising from a us...
01:45:31 <int-e> aw.
01:45:35 <int-e> @src fix
01:45:35 <lambdabot> fix f = let x = f x in x
01:45:45 <myname> aaaah
01:45:58 <myname> Y combinator?
01:46:05 <int-e> right
01:46:27 <myname> funny
01:47:38 -!- KingOfKarlsruhe has quit (Quit: ChatZilla [Firefox 25.0.1/20131112160018]).
01:47:43 <oerjan> > fix (expr . f)
01:47:44 <lambdabot> f (f (f (f (f (f (f (f (f (f (f (f (f (f (f (f (f (f (f (f (f (f (f (f (f (...
01:47:59 -!- CADD has joined.
01:48:17 -!- ggherdov has joined.
01:53:03 -!- KingOfKarlsruhe has joined.
01:53:04 -!- CADD has quit (Quit: leaving).
01:53:19 -!- CADD has joined.
01:53:22 -!- CADD has quit (Client Quit).
01:53:49 -!- CADD has joined.
01:54:37 -!- oerjan has set topic: The channel of the chimæric hellos | The most corum, clargoint chait you could ever loofefl your slance in. | Magnus!!! | Koirammekokaan ei lennä? :( | https://dl.dropboxusercontent.com/u/2023808/wisdom.pdf | logs: http://codu.org/logs/_esoteric/ or http://tunes.org/~nef/logs/esoteric/.
01:57:21 -!- CADD has quit (Client Quit).
01:57:38 -!- CADD has joined.
01:57:38 -!- CADD has quit (Client Quit).
01:58:24 -!- CADD has joined.
02:01:03 -!- CADD has quit (Client Quit).
02:01:36 -!- CADD has joined.
02:02:28 <oerjan> CADD: FIX YOUR CONNECTION
02:03:07 -!- CADD has quit (Client Quit).
02:04:05 -!- CADD has joined.
02:04:29 -!- CADD has changed nick to Guest59952.
02:04:30 <Guest59952> oerjan: its not my connection, im fixing my .irssi
02:04:55 -!- Guest59952 has quit (Client Quit).
02:05:41 -!- CADD_ has joined.
02:05:56 -!- CADD_ has quit (Client Quit).
02:09:53 <int-e> huh
02:10:01 <int-e> does CADD know about /reload?
02:14:17 -!- ^v has joined.
02:24:08 -!- Taneb has quit (Quit: Leaving).
02:27:22 <oerjan> POSSIBLY NOT
02:27:47 <oerjan> pizza!
02:33:33 <shachaf> oerjan: norwegian pizza?
02:33:37 <shachaf> i hear that's good
02:33:59 <oerjan> OKAY
02:47:12 -!- ^v has quit (Quit: http://i.imgur.com/MHuW96t.gif).
02:53:01 -!- ^v has joined.
03:11:53 -!- trout has changed nick to constant.
03:23:49 <doesthiswork> I've got a new book for the esoteric list http://www.amazon.com/exec/obidos/ISBN=0195079515/principiacyberneA/
03:25:20 <doesthiswork> it will change your life!
03:27:59 <doesthiswork> Kauffman's recent work translates his biological findings to the http://en.wikipedia.org/wiki/Mind_body_problem and issues in neuroscience
03:28:27 <oerjan> wait, your second line makes me suspect you are being sarcastic.
03:30:03 <Bike> heard enough bullshit like that in one life time thanks
03:30:24 <doesthiswork> I've actually read the book, it's interesting but not particularly meaningful
03:30:36 <Bike> oh, wait, kauffman. so he's not like, insane, just... right.
03:30:56 <oerjan> not insane, just right, sounds like a good place to be.
03:31:45 <doesthiswork> it points out that if you make the fitness landscap complex enough, fitness doesn't really matter
03:32:11 <oerjan> @djinn f :: Maybe (b -> Either a b)
03:32:11 <lambdabot> Cannot parse command
03:32:15 <oerjan> wat
03:32:21 <oerjan> @djinn Maybe (b -> Either a b)
03:32:22 <lambdabot> f = Nothing
03:32:32 <Bike> doesthiswork: like how
03:32:32 <oerjan> lambdabot: you're boring.
03:33:08 <doesthiswork> and that if you collect a set of more than K Boolean functions it is universal
03:33:09 <shachaf> oerjan: Your question is boring.
03:33:38 <doesthiswork> (by universal I mean you can construct any boolean function)
03:33:54 <oerjan> doesthiswork: that's technically incorrect.
03:34:04 <doesthiswork> and that is pretty much the entire 2 pound book
03:34:07 <oerjan> (see my beloved Post Lattice)
03:35:05 <Bike> well, that reminds me i should pick up some systems biology stuff.
03:35:15 <oerjan> hm i wonder if it _is_ possible to have a set of infinitely many boolean functions where none can be expressed in the others.
03:35:35 <oerjan> hm no.
03:36:10 <doesthiswork> Kauffman wrote (January 2012) "No entailing laws, but enablement in the evolution of the biosphere," which aims to show that evolution is not law entailed, as is physics, and that, without selection, evolution enables its own future possibilities.
03:36:27 <doesthiswork> which I can't tell what the hell it means
03:36:29 <oerjan> because all the classes are finitely generated, and you must be able to express the generators in finitely many of them, which means you can express everything.
03:36:46 <oerjan> however is there really an upper bound K...
03:36:53 <doesthiswork> post lattice is cool
03:37:20 <oerjan> well, technically _if_ you generate everything, your set has 5 which do it.
03:37:34 <oerjan> *5 functions
03:39:01 <doesthiswork> which 5?
03:39:51 <Bike> oh, it's on arxiv.
03:40:13 <oerjan> doesthiswork: for each of the 5 maximal subclasses, find a function that doesn't belong to it.
03:40:37 <oerjan> the same function can work for more than one, so it doesn't have to be the full 5.
03:40:37 <doesthiswork> bike: can you recommend me some systems biology books or papers?
03:40:50 <doesthiswork> ok
03:41:40 <Bike> http://arxiv.org/abs/1201.2069 yeah i don't know what the hell they're on about
03:41:40 <oerjan> that is, find (1) a function which doesn't send all 0 to 0 (2) a function which doesn't sent all 1 to 1 (3) a function which isn't monotone (4) a function which isn't linear/affine (5) a function which isn't self-dual.
03:41:51 <oerjan> *send
03:42:18 <Bike> doesthiswork: i've been recommended http://www.amazon.com/dp/1584886420/
03:42:23 <oerjan> you will note that nand and nor don't do any of those, and so generate everything alone.
03:42:59 <doesthiswork> oerjan: thank you, now I know more than I did this morning
03:43:18 <oerjan> you're welcome!
03:43:52 <Bike> the first one star review of that has some oter links, eheh
03:44:40 <doesthiswork> the university library has it, I'll take a look
03:45:58 <doesthiswork> oh wow, the abstract actually makes sense to me.
03:47:29 <doesthiswork> it says with physics you can predict the future at pretty well (some of the time). With biology who the hell knows
03:49:21 <oerjan> looking at the lattice diagram again, i cannot see any node other than the top which has 5 edges down - so it probably _is_ the case that if you have 5 functions none of which can be expressed in the others, they must generate everything.
03:50:25 <oerjan> although i'm not at the spot quite sure if each of the 5 classes misses a function that belongs to all of the others.
03:51:25 <Bike> constant zero, constant one, identity, ... i guess and and or would let you build xor and negation
03:51:41 <oerjan> in which case it would simply be impossible to find 5 such functions.
03:51:41 <constant> Bike: I refuse to identify myself
03:52:54 <Sgeo> I miss the List of Things That Don't Exist
03:52:55 <Sgeo> :(
03:53:19 <doesthiswork> Sgeo: Your true love.
03:53:28 <oerjan> Bike: your list includes nothing non-monotone yet
03:53:30 <Bike> wisconsin
03:53:52 <oerjan> so you definitely cannot build xor and negation
03:55:16 <Bike> so... is that five functions that aren't complete
03:55:20 <oerjan> identity is assumed by default.
03:56:09 <oerjan> like, it's what you get if you start with a variable and don't use any functions.
03:56:33 <oerjan> so it's expressible without anything.
03:56:35 <shachaf> Sgeo: Like gruncheons and pritons?
03:56:51 <Bike> aw.
03:57:23 <doesthiswork> bike: http://www.amazon.com/review/R7DHOCB2N9PYT/ref=cm_cr_dp_title/177-0981632-5731719?ie=UTF8&ASIN=0231075650&nodeID=283155&store=books
03:57:28 <oerjan> constant zero and constant one together take care of three of the classes, however.
03:57:47 <doesthiswork> Critics of reductionism have pointed to Kurt Goedel's 1931 "incompleteness theorem" (which shows that in any axiomatic formulation of, say, number theory there will be true theorems that cannot be established) as a contrary example, but this paradigm-shattering result has been largely ignored the scientific community, which has blithely persisted in its reductive beliefs.
03:58:12 <shachaf> http://english.lem.pl/home/bookshelf/how-the-word-was-saved
03:58:14 <Bike> doesthiswork: time to misunderstand everything at once
03:58:24 <oerjan> Bike: hm i think you _may_ have detected a flaw in my argument, anyway.
03:58:50 <Bike> brute force is the method of kings, oerjan
03:59:00 <shachaf> nolars, nightzebs, nocs, necs, nallyrakers, neotremes and nonmalrigers :'(
03:59:38 <Bike> "Why are living things alive? As a theoretical biologist, Robert Rosen saw this as the most fundamental of all questions-and yet it had never been answered satisfactorily by science. " god damn it people
04:00:08 -!- CADD has joined.
04:00:13 <Bike> "Since both the atom and the organism can be seen to fit that description, Rosen asserts that complex organization is a general feature not just of the biosphere on Earth-but of the universe itself."
04:00:27 -!- ^v has quit (Remote host closed the connection).
04:01:18 <Bike> "One of his results is to show precisely why physics (including molecular biology) has little to say about life (and non-life)."
04:01:56 <Bike> seriously jesus fucking christ unpredictability doesn't mean indeterminism
04:02:07 <shachaf> why are you reading this Bike
04:02:34 <Bike> because i guess i'm an evil reductionist somehow
04:02:40 <oerjan> Bike: hm i think my argument that (5 independent functions => generate everything) holds
04:02:46 <doesthiswork> because I like to torture him :D
04:04:17 <Bike> "He concluded, based on examples such as this, that phenotype cannot always be directly attributed to genotype and that the chemically active aspect of a biologically active protein relies on more than the sequence of amino acids,," and he's doing the "REVOLUTION: THING NOBODY ACTUALLY BELIEVES IS WRONG" thing that is so, so fucking irritating
04:04:33 <oerjan> because he(*)'s a sucker ((*) everyone)
04:04:34 <doesthiswork> this is really failing to convince me that there is anything to systems biology
04:04:48 <Bike> well it's not just this guy
04:05:05 <Bike> "Certain questions about Rosen's mathematical arguments were raised in a paper authored by Christopher Landauer and Kirstie L. Bellman which claimed that some of the mathematical formulations used by Rosen are problematic from a logical viewpoint. It is perhaps worth noting, however, that such issues were also raised long time ago by Bertrand Russel and Alfred North Whitehead" lol
04:05:15 <zzo38> Reductionism is a useful tool, but not always!!!
04:05:32 <doesthiswork> Dr. Rosen was one of those unfortunate scientists who worked on problems, that to the rest of his community were non-existent.
04:05:41 <Bike> pst i'm not really a sole reductionist
04:06:04 <oerjan> "REVOLUTION: THING NOBODY ACTUALLY BELIEVES IS WRONG" <-- you could make that into an onion headline
04:07:44 -!- CADD_ has joined.
04:08:00 <oerjan> i think it just needs an "IN SCIENCE" after the REVOLUTION
04:08:37 -!- CADD has quit (Read error: Connection reset by peer).
04:09:28 <Bike> i mostly think of systems biology as being like this one mostly informal book i read tbh
04:09:41 <Bike> which was about cells-as-computers, which is the opposite of this rosen guy so fuck 'im
04:10:06 <doesthiswork> But he goes on and shows how the recursion in a function can be seen as constraints between (differential) equations, which he calls "chronicles".
04:11:37 <doesthiswork> ok bike what did you read?
04:12:28 <Bike> http://www.amazon.com/dp/0300167849
04:12:31 <Bike> dunno if it 'counts'
04:12:34 <oerjan> hm i think maybe you _cannot_ have 5 boolean functions that cannot be written in terms of each other.
04:13:10 <oerjan> because if one is self-dual and sends all 0 to 0, it will automatically send all 1 to 1.
04:13:11 <Bike> goes over gene regulatory networks and protein stoichiometry equations and shit
04:13:18 <Bike> for non-scientists
04:14:51 <oerjan> so you cannot get a function that is in not in P_1 and yet in all the others. which means whatever function avoids P_1, avoids one of the others, and so you need at most 4 of them to avoid all the maximal classes.
04:15:14 <oerjan> meanwhile bike's example shows that you _can_ have 4, so this is optimal.
04:15:35 <Bike> just as i planned
04:15:48 <oerjan> congrats
04:17:24 <doesthiswork> yay!
04:23:56 <zzo38> According to this paper, the Pentagon once asked if nuclear bombs can be made to explode in the past.
04:24:16 <doesthiswork> that sound right
04:24:59 <Bike> possibly my favorite 'systems' stuff is in ethology though. it's not really 'systems biology' but like lorentz had this adorable machine model made of waterwheels
04:30:29 <Bike> he actually built it at one point iirc
04:31:57 <doesthiswork> konrad lorenz?
04:32:23 <Bike> yea
04:33:23 <doesthiswork> or edward lorenz
04:34:10 <Bike> i said ethology, didn't i
04:34:51 <doesthiswork> http://www.ace.gatech.edu/experiments2/2413/lorenz/fall02/
04:35:08 <doesthiswork> but this lorenz has waterwheels and everything
04:36:05 <Bike> that looks cool, actually.
04:36:22 <Bike> there's a double pendulum set up in one of the physics buildings, it's fun to fuc with
04:47:28 <doesthiswork> why is it that interesting things are also interesting to crackpots?
04:48:41 -!- nisstyre has quit (Quit: Leaving).
04:52:32 <zzo38> I suppose because it is interesting in general.
04:53:54 <doesthiswork> AI is interesting but the AI channel here is a wasteland
04:56:52 <oerjan> the ais like to keep it that way >_> <_<
05:07:05 <shachaf> Is that what we're calling him now?
05:09:48 <oerjan> `? ais523
05:09:53 <HackEgo> Agent “Iä” Smith is an alien with a strange allergy to avian body covering, which he is trying to retroactively prevent from ever evolving. On the 3rd of March, he's lawful good.
05:10:16 <oerjan> shachaf: no, we have no evidence that he is plural yet
05:17:36 <Sgeo> `? Sgeo
05:17:38 <HackEgo> Sgeo is a language nomad. (Not to be confused with a language monad.) He invented Metaplace sex.
05:19:30 <zzo38> 3rd of March of what year? In what timezone?
05:19:52 <Sgeo> Of the year 0/0
05:20:16 <oerjan> zzo38: i have always assumed it's every year.
05:20:30 <oerjan> and probably british standard time.
05:20:32 <Bike> and whatever timezone ais is in, probably
05:20:56 <oerjan> which is probably the same as UTC that time of year
05:21:33 <myname> so... what is metaplace sex?
05:22:40 <Sgeo> Two people repeatedly fainting without clothes
05:23:33 <myname> what
05:24:30 <Sgeo> That's how I implemented a 'sex' option in Metaplace: Caused the participants to loop the "fainting" animation
05:24:56 <myname> should i know metaplace?
05:25:13 <Bike> isn't it Yet Another Second Lifey Thing
05:25:31 <Sgeo> http://en.wikipedia.org/wiki/Metaplace
05:25:40 <Sgeo> Bike: yeah, in 2d, Flash-based
05:25:41 <myname> as long as there is no ncurses ui, i am not interested
05:25:54 <Sgeo> It's dead now
05:26:07 <Sgeo> It died less than a month after I invented metaplace sex
05:26:12 <myname> surprise surprise
05:26:18 <myname> lol
05:30:13 <myname> i'd love to have something like snow crashs metaverse with a terminal interfacee
05:34:27 <oerjan> wait so the part about Sgeo inventing metaplace sex is _true_?
05:34:51 <Sgeo> Yes.
05:34:52 <zzo38> And what did ais523 invent, then?
05:35:07 <oerjan> zzo38: it doesn't say.
05:35:40 <Sgeo> He invented the ability to learn multiple esolangs at once. Isn't there that news article about him knowing 25 languages?
05:35:54 <myname> define "knowing"
05:35:59 <zzo38> OK add that in, then.
05:36:11 <myname> can't be that hard
05:36:25 <myname> just learn any bf derivate and you are way ahead
05:36:29 <Sgeo> No one said it was an accurate news article
05:36:34 <Sgeo> erm, misread you
05:37:29 <oerjan> `run sed -i 's/sex/sex, thus killing it within a month/' wisdom/sgeo
05:37:34 <HackEgo> No output.
05:38:08 <Sgeo> fwiw, the part about killing it within a month isn't true, it's correlation-therefore-causation fallacy
05:38:27 <Sgeo> It died within a month, but I didn't cause it
05:38:28 <myname> oerjan: :D
05:38:39 <myname> Sgeo: can you prove that?
05:39:20 <oerjan> myname: the entry was breaking #esoteric tradition with its accuracy.
05:39:43 <myname> hth
05:40:41 <oerjan> Sgeo: we prefer to believe.
05:41:07 <Sgeo> `quote what sex looks like
05:41:09 <HackEgo> 313) <elliott> sgeo do you actually know what sex looks like i am just checking here <Sgeo> I think so
05:42:18 <zzo38> O, that is Sgeo, too. So I suppose it is related, then.
05:42:39 <Sgeo> Yeah, elliott said that in response to me talking about metaplace sex
05:45:12 <myname> so, did you make a language about virtual sex?
05:46:19 <kmc> sex can look like a lot of things
05:46:23 <kmc> that's one of them
05:46:27 <oerjan> interestingly, the only mention of "sex" on the wiki is Taneb's userpage.
05:46:37 <myname> :D
05:46:49 <myname> maybe i should make a page only with the word sex in it
05:47:41 <oerjan> <oklopol> you make the awesomest palindromes, you don't need sex and you don't need sleep. you are a fucking superhuman :|
05:47:44 <oerjan> <oklopol> oh and you can transform into an elf
05:47:47 <oerjan> User:Oklopol (note: only one of these is true)
05:48:11 <ion> That palindrome sucked.
05:48:17 <oerjan> i think the "you don't need sex" is the true one.
05:48:34 <oerjan> i cannot tell whether oklopol believes it's the palindrome part, though.
05:48:36 <zzo38> Probably that is correct.
05:48:47 <myname> there is a great german palindrome i heard recently
05:48:51 <oerjan> Taneb has stated being asexual.
05:49:39 <oerjan> otoh maybe he's just on his way to the big breakdown.
05:50:24 <coppro> ^
05:52:06 <oerjan> "York student has psychic breakdown and starts fucking everything that moves, and some things that don't." you heard the headline here first.
05:52:18 <oerjan> wait this is britain.
05:52:28 <Bike> psychic breakdown? like carrie?
05:52:30 <oerjan> "York student has psychic breakdown, starts fucking everything that moves, some things that don't." you heard the headline here first.
05:52:58 <oerjan> Bike: wait what is the correct phrase.
05:53:02 <oerjan> oh nervous
05:53:24 <oerjan> "York student has nervous breakdown, starts fucking everything that moves, some things that don't." you heard the headline here first.
05:53:33 <ion> https://lh3.googleusercontent.com/-P6h3xdZ34rE/ULVLrUmc3eI/AAAAAAAAZo0/JA2D-oN8Zjg/w506-h788/carrie_sriracha.jpg
05:53:38 <shachaf> ...Did this person just go through my old stackoverflow answer and add Unicode quotes everywhere?
05:53:51 <shachaf> http://stackoverflow.com/posts/13538351/revisions
05:53:59 <oerjan> ion: XD
05:54:19 <zzo38> However, I can tell you what I invented at least; I invented matrix accounting (while sitting in an accounting class in school, having nothing else to do, trying to think of how to use complex numbers in accounting in a way that doesn't violate GAAPs; I concluded that this is impossible). Yes, you can make up lies about it if you want, I suppose.
05:55:17 <Sgeo> I seem to be one of only a handful of people who know how to do a certain thing with Varnish
05:55:25 <shachaf> *And* he moved punctuation inside quotes?
05:55:27 <shachaf> That's too much.
05:55:29 <myname> shachaf: i like how it states "typo"
05:55:32 <zzo38> What certain thing with Varnish?
05:55:53 <shachaf> myname: Well, apparently «`s -> (a,s)» was missing a `.
05:56:04 <Sgeo> Passing data computed in the VCL to ESI subrequests
06:00:12 <oerjan> is &ndash; in "world-passing" even plausibly correct?
06:00:42 <oerjan> shachaf: ^
06:01:04 <ion> My intuition says no, but you probably shouldn’t trust it.
06:01:41 <shachaf> I don't know if it's correct, but I don't like it.
06:02:05 <shachaf> I think I'll roll this back and add the missing ` myself.
06:03:18 <oerjan> same with error-prone, i guess.
06:05:19 <oerjan> `unidecode -- —
06:05:20 <HackEgo> ​[U+0020 SPACE] [U+002D HYPHEN-MINUS] [U+002D HYPHEN-MINUS] [U+0020 SPACE] [U+2014 EM DASH]
06:05:39 <oerjan> changing -- to mdash is an improvement in my mind, though.
06:06:02 <shachaf> I don't like it.
06:06:15 <shachaf> I didn't write - -. I wrote --.
06:06:28 <oerjan> you don't like _anything_, shachaf.
06:06:36 <shachaf> I liked it the way it was!
06:06:44 <oerjan> OKAY
06:06:58 <oerjan> it's just the one thing that's definitely better in my mind, though.
06:07:28 <oerjan> and i'm neutral on the quote marks.
06:07:29 <shachaf> I think correct usage of "—" dictates that you don't put spaces around it.
06:07:31 <shachaf> Which is even worse.
06:08:12 <oerjan> shachaf: yeah wikipedia says you should use no spaces around mdash but spaces around ndash
06:10:28 <shachaf> Oh, he's actually changing semantic meaning too!
06:11:25 * oerjan didn't notice that.
06:11:45 <shachaf> "and also quite error-prone (if we accidentally try to read too far into stdin, the program will just block until the user types more in)."
06:11:48 <shachaf> into
06:11:51 <shachaf> "and also quite error-prone. (If we accidentally try to read too far into stdin, the program will just block until the user types more in.)"
06:12:01 <shachaf> That's some sort of semantics, anyway.
06:12:39 <oerjan> it's a small detail, like all the rest.
06:13:27 <shachaf> ...Is he using an en-dash in "as–is"?
06:14:41 <oerjan> hey, i'm not actually stopping you from rolling back.
06:20:30 -!- conehead has quit (Quit: Computer has gone to sleep.).
06:50:35 -!- nisstyre has joined.
07:02:21 * oerjan suddenly realizes that hangman must be hell to play in vietnamese.
07:03:26 <oerjan> _ _ _ (1 incorrect: ở)
07:09:23 -!- CADD_ has quit (Ping timeout: 260 seconds).
07:09:25 <lifthrasiir> oerjan, or korean :)
07:10:22 <oerjan> is korean that bad if you use "letters" instead of "characters"?
07:10:42 <oerjan> hm i guess you'd have the problem of where to place the _'s
07:10:57 <lifthrasiir> that's why we use only the initial letters (jamos, to be exact) in hangman-like quizzes...
07:11:09 <shachaf> copumpkin: Where were you yesterday when we needed your help with foo-spaces?
07:11:14 <oerjan> wat
07:11:18 <copumpkin> ?
07:11:46 <lifthrasiir> oerjan, it therefore feels like a hangman but without vowels
07:12:22 <oerjan> lifthrasiir: hm and that would be hebrew or arabic
07:12:56 <lifthrasiir> haha
07:13:56 -!- Sprocklem has quit (Ping timeout: 245 seconds).
07:14:12 * oerjan wonders if indic languages do the same thing as korean, since vowels are like "accents" there
07:15:44 <zzo38> I can think of a few possible ways to make up a variant of hangman game for languages having much more letters of alphabet, whether it is due to accents, or other purposes.
07:16:38 <oerjan> the first page of google hits for hangman game "hindi" makes me suspect they only use it as an english learning tool
07:17:28 <oerjan> i suppose i might need to know the hindi words to search for the real thing.
07:17:31 -!- asie has joined.
07:35:47 -!- asie has quit (Quit: I'll probably come back in either 20 minutes or 8 hours.).
07:35:57 -!- doesthiswork has quit (Quit: Leaving.).
07:41:33 -!- zzo38 has quit (Quit: zzo38).
07:58:24 -!- asie has joined.
08:05:55 -!- Slereah has quit (Ping timeout: 272 seconds).
08:06:12 -!- Slereah has joined.
08:09:58 <oklopol> i dead 'since vowels are like "accidents" there', and read quite a bit of history in case something would explain what on earth you mean.
08:10:04 <oklopol> yes, i dead.
08:18:00 <mroman_> Need some suggestions for weird instructions
08:18:04 <mroman_> well... I don't need them
08:18:11 <mroman_> but I'm always looking for other ideas :)
08:19:52 <Bike> mroman_: http://en.wikipedia.org/wiki/Apollo_Guidance_Computer#Instruction_set definitely CCS
08:22:06 <Bike> or how about 'execute instruction in register'
08:23:23 <mroman_> Bike: Neat
08:23:26 <mroman_> That would work
08:23:37 <mroman_> as I have 16bit Registers and the longest instruction is two bytes
08:24:38 <Bike> maybe something really silly, like have a register for how much the instruction pointer moves each cycle
08:24:41 <Bike> execute code backwards
08:25:37 -!- asie has quit (Quit: I'll probably come back in either 20 minutes or 8 hours.).
08:26:06 <Bike> good for debugging! or reliving the days of barrel memory
08:37:53 -!- asie has joined.
08:39:39 -!- nisstyre has quit (Quit: Leaving).
08:40:28 <mroman_> hm
08:40:41 <mroman_> I think I'm going to add Rational Number Support
08:40:53 <Bike> i don't know if you're going for "weird instructions" as in crazy dumb shit or what though
08:41:09 <Bike> rationals in hardware?
08:41:49 <mroman_> Yeah.
08:42:03 <mroman_> 8bit Quotient, 8bit Denominator
08:42:24 <mroman_> Bike: Not really "crazy dumb shit"
08:42:31 <mroman_> just "crazy but still useful shit"
08:42:34 <Bike> damn but i like the barrel thing :(
08:43:14 <Bike> i guess rationals aren't too hard though, mostly you need gcd right
08:43:14 <mroman_> Since I don't have barrel memory reversing the control flow
08:43:17 <mroman_> would be crazy dumb shit
08:43:20 <shachaf> i was in a barrel once for a play
08:43:26 <shachaf> it was good
08:45:00 <Bike> i guess it depends on what the processor's for. i mean presumably the vax had that evalpoly thing for a reason.
08:45:29 <Bike> possibly just because horn evaluation is super easy but w/e.
08:48:18 <mroman_> What's that thingy called in english
08:48:26 <mroman_> where you do 4/8 = 1/2?
08:48:45 <Bike> "reducing a fraction to lowest terms"
08:48:57 <mroman_> ok
08:49:00 <mroman_> so...
08:49:03 <mroman_> RDC then :)
08:49:03 <Bike> or just "reduction" i guess.
08:49:06 -!- asie has quit (Quit: I'll probably come back in either 20 minutes or 8 hours.).
08:49:08 <Bike> sounds good to me!
08:49:22 <mroman_> One has to do that manually
08:49:41 <mroman_> because it's probably slow if the CPU does it automatically all the time
08:50:53 <Bike> i uh, dont' think i've ever cared about speed when i was using rationals, really. dunno how good the algorithms are.
08:51:04 <Bike> plus like, when i use rationals i usually want infinite precision...
08:51:09 <mroman_> Although with 8bit/8bit I don't have a sign flag
08:51:19 <mroman_> Hm.
08:51:30 <Bike> so i'm really wondering the use case here i guess?
08:51:32 <mroman_> unless I treat both 8bit values as signed 8 bit
08:52:40 <mroman_> Bike: To allow < 1 pixel calculations
08:52:50 <mroman_> i.e if you wan't to move 1/10 of a pixel per frame
08:52:53 <mroman_> *e.g.
08:53:04 <mroman_> Of course
08:53:04 <Bike> mm, don't you usually use fixedpoint for that
08:53:09 <mroman_> You could also do that with integers
08:54:24 <mroman_> (It's a processor for some weird game console I guess)
08:55:08 -!- asie has joined.
08:55:20 <Bike> suddenly wonders how that one snes coprocessor on... starfox? worked
08:55:52 <mroman_> The Super FX?
08:56:09 <Bike> dunno. let it 3d render.
08:59:34 <kmc> hi
08:59:49 <Bike> yo
09:00:09 <kmc> hangul is the best
09:00:57 <shachaf> what's an ul
09:01:11 <shachaf> is it a many-appendaged creature for the korean version of hangman
09:01:25 <oerjan> that would explain things.
09:01:48 <kmc> shachaf: see one of the many benefits of hangul is that you can't mis-parse a word into syllables like that
09:02:10 <shachaf> kmc: i know, hangul is p. good
09:02:18 <kmc> glad u agree
09:03:00 <shachaf> hang̈ul
09:37:14 -!- Taneb has joined.
09:49:20 -!- ais523 has joined.
09:54:27 <oerjan> Taneb: hi did you know the esolang wiki's sole mention of the word "sex" is on your user page hth
09:55:23 <Taneb> Heh
09:55:31 <Taneb> I did not
09:57:34 <ais523> I also didn't know that
09:58:00 <Taneb> It's almost ironic
09:58:04 <fizzie> Clearly time for the "Sex" language, eh? (It's a brainfuck derivative with lewd words replacing the instructions.)
09:58:12 <Taneb> In that way that makes people think it's ironic but really not at all
09:58:13 <ais523> fizzie: no
09:58:23 <Taneb> fizzie, there's already at least one language that is almost exactly thaty
09:58:43 <int-e> . o O ( Ook! )
09:59:03 <oerjan> http://esolangs.org/wiki/Fuckfuck
09:59:31 <int-e> ah, 'sex' was too short for that one.
09:59:48 <fizzie> Ø.: V. good.
09:59:50 <oerjan> that explains it.
10:02:13 -!- nooodl has joined.
10:08:23 <int-e> "[Ook] represents the first, although unfortunately not the last, in a long line of trivial Brainfuck command substitutions." Amen. ("Look, I came up with 8 tokens and now I'm the creator of an esolang!")
10:09:43 <int-e> (whitespace deserves a honorable mention though)
10:10:00 <mroman_> I hate brainfuck derivatives with an exact 1:1 mapping to brainfuck
10:11:53 <myname> int-e: i thought of a whitespace derivate with zero width non-breaking space and zero width space
10:16:46 <int-e> I guess one could play with control characters like U+202A (left-to-right) and U+202B, too.
10:16:59 <mroman_> are there matrices with only integers whose inverses also consist solely of integers?
10:17:04 <mroman_> except for the identity matrix
10:17:04 <int-e> but it's sort of boring.
10:17:14 -!- MindlessDrone has joined.
10:17:40 <int-e> mroman_: of course. A big class is all the triangular matrices with 1 on the diagonal.
10:17:50 <myname> int-e: yeah, the only interesting part is you had to use a hex editor to write
10:18:18 <int-e> myname: ordinary brainfuck + some trivial sed, awk or perl.
10:18:19 -!- MindlessDrone has quit (Client Quit).
10:18:20 <mroman_> The inverse of 2x2 has a factor of (1/(ad-bc))
10:18:36 <mroman_> So I guess a,b,c and d must all be multiples of (ad - bc)
10:18:51 <int-e> ad-bc = 1 is not that hard to satisfy
10:19:07 <mroman_> No. That should be easy
10:20:27 <mroman_> although
10:20:42 <mroman_> what's a singular matrix?
10:21:06 <int-e> You can start with the identity matrix and then swap rows, columns, or add a multiple of a row or column to another row and column; finally you may negate rows and columns.
10:21:17 <mroman_> "Singular matrix, a matrix that is not invertible"
10:21:18 <mroman_> ok
10:21:28 <mroman_> so not every Matrix with ad-bc=1 is invertible
10:22:18 <mroman_> oh wait
10:22:24 <mroman_> I accidentally did ad-bc = 0
10:22:34 <mroman_> which is exactly the criteria for it not being invertible
10:22:36 <ais523> yeah
10:23:34 -!- carado has joined.
10:25:32 <int-e> for the sake of completeness: ad-bc = -1 is also allowed
10:31:46 -!- nooodl has quit (Quit: Ik ga weg).
10:33:41 <mroman_> Good point
10:35:09 -!- MindlessDrone has joined.
10:36:24 <oerjan> <int-e> (whitespace deserves a honorable mention though) <-- whitespace is not brainfuck-based.
10:39:02 <oerjan> mroman_: ad - bc is also known as the determinant of the matrix.
10:39:32 <oerjan> and yes, integer matrices with det M = +-1 have integer inverses.
10:42:47 <oerjan> that's if and only if, actually, because the inverse as determinant 1/det M which must be an integer for M^-1 to have integer entries.
10:42:53 <oerjan> *has
10:43:45 <int-e> oerjan: oops. thanks!
10:44:57 <mroman_> Yeah.
10:45:12 <mroman_> But only integer matrices with M = +=1 have integer inverses.
10:45:22 <oerjan> *det M
10:45:36 <mroman_> That's at least what my math tells me
10:45:48 <oerjan> that's what i am saying!
10:45:48 <mroman_> a|(ad-bc) ^ b|(ad-bc) ...
10:45:58 <mroman_> adbc|(ad-bc)
10:46:05 <mroman_> and so on
10:46:13 <mroman_> which yields to 1 = x/(bc - ad)
10:46:39 <mroman_> oh
10:46:41 <mroman_> Yeah
10:46:47 <mroman_> Didn't read your 11:42
10:51:04 <oerjan> https://en.wikipedia.org/wiki/Adjugate_matrix#Inverses
10:52:04 <oerjan> "From this formula follows one of the most important results in matrix algebra: A matrix A over a commutative ring R is invertible if and only if det(A) is invertible in R."
10:52:32 <int-e> mroman_: your notation is off though; (ad-bc) | a. (The symbol | can be read as "divides").
10:52:37 <oerjan> note that +-1 are the invertible elements in the ring Z of integers.
10:58:00 -!- asie has quit (Quit: I'll probably come back in either 20 minutes or 8 hours.).
11:07:04 -!- oerjan has quit (Quit: leaving).
11:12:38 -!- impomatic has quit (Ping timeout: 245 seconds).
11:24:53 -!- nooodl has joined.
11:26:07 <mroman_> int-e: I know nothing about math notation :)
11:26:18 <mroman_> almost nothing.
11:27:40 <mroman_> Luckily converting between binary and hexadecimal is still considered by my university to be a very difficult task for students
11:28:09 <mroman_> so my lack of math skills does not matter
11:30:55 <mroman_> Do any of you guys know the C167?
11:35:16 <mroman_> anyway... to make a point short: it's apparentely hard to read specs and find out how to figure out the physical address of stuff
11:36:02 -!- yorick has joined.
11:45:17 -!- nooodl_ has joined.
11:49:04 -!- nooodl has quit (Ping timeout: 264 seconds).
12:00:51 <oklopol> "From this formula follows one of the most important results in matrix algebra: A matrix A over a commutative ring R is invertible if and only if det(A) is invertible in R." from that formula, the "if" certainly follows, for the other direction i guess you need to extend to commutative ring to a field or something?
12:01:32 <oklopol> and then because there's an inverse if it's nonzero and inverses are unique, the only if follows
12:02:07 <oklopol> *if the determinant is
12:02:08 <int-e> oklopol: if A is invertible in R, then det(A) and det(A^-1) are both in R, and their product is one, so the determinant is invertible in R.
12:02:21 <int-e> s/in R/over R/g
12:02:57 <oklopol> i'm just wondering why it follows from that formula
12:03:17 <oklopol> but err
12:03:36 <int-e> det(A)*det(A^-1) = det(A*A^-1) = det(I) = 1 is an instance of that formula
12:03:38 <oklopol> like you said i guess, take the determinant from both sides
12:04:02 <oklopol> oh err
12:04:10 <oklopol> apparently it's explained right after
12:04:18 <int-e> oh :)
12:04:59 <oklopol> "<int-e> det(A)*det(A^-1) = det(A*A^-1) = det(I) = 1 is an instance of that formula" how is it an instance of that formula?
12:05:20 <oklopol> (although i'm not sure this question is very useful)
12:06:14 <oklopol> the formula was "A adj(A) = det(A) I"
12:07:34 <oklopol> my point is that indeed it should follow from that formula that A has an inverse in an extension of the ring where there is an inverse for det(A)
12:08:32 <oklopol> and err... okay dunno. maybe i should forget about this.
12:11:37 <Taneb> Ikea donated more to the Philippines recovery effort than China
12:12:15 <oklopol> :P
12:12:43 <int-e> Maybe Ikea is the larger country ...
12:16:15 <int-e> oklopol: I think you're right, it's not *that* formula; at least every argument that I see for invertibility of A implying that det(A) is a unit uses det(AB) = det(A) det(B), and we've seen that this identity is all one needs.
12:16:42 -!- Phantom_Hoover has joined.
12:24:00 <oklopol> perhaps after seeing that there is little incentive to try much else, because that's sort of obvious
12:24:37 <oklopol> although i don't think it's quite trivial to prove det(AB) = det(A) det(B) (maybe because i've never tried to prove it myself and it's quite an index hassle in most proofs i've seen)
12:25:26 -!- asie has joined.
12:27:11 -!- S1 has joined.
12:27:20 <S1> What is the opposite of bootstrapping
12:27:59 <Taneb> A safe shut down?
12:28:49 <S1> nah I mean in terms of programming paradigms and stuff
12:29:33 <Taneb> So, bootstrapping here means building a program in small components, right?
12:30:40 <S1> I understand it as bringing things up to more abstract layers. as in from machine code via assembler to script languages to modern programming languages
12:31:33 <S1> Look here: http://en.wikipedia.org/wiki/Bootstrapping#Software_development
12:31:41 <Taneb> I don't think it really has an opposite, then, only an absence
12:31:50 <Taneb> Which would be writing an agda compiler in machine code, I guess
12:32:14 <S1> k
12:35:43 <Phantom_Hoover> <oklopol> although i don't think it's quite trivial to prove det(AB) = det(A) det(B) (maybe because i've never tried to prove it myself and it's quite an index hassle in most proofs i've seen)
12:35:56 <Phantom_Hoover> if either A or B is singular then so is AB
12:36:13 <Phantom_Hoover> who cares what the determinant is if it isn't 0
12:36:49 <oklopol> and how do you prove the determinant of AB is 0 iff that of A or that of B is?
12:37:45 <oklopol> or what do you mean by "singular"
12:38:13 -!- S1 has left.
12:38:22 <Phantom_Hoover> 'has determinant 0'
12:38:44 <oklopol> and how do you prove that if the determinant of A is not 0 and that of B is not 0 then also the determinant of AB is not zero
12:39:03 <oklopol> and how does that help with showing what we were trying to show anyway?
12:39:15 <Phantom_Hoover> because if A and B are bijective then so is their composition?
12:39:28 <oklopol> okay i have a hunch you have no idea what we're talking about
12:40:23 <oklopol> we are trying to show that the invertibility of the determinant of A has something to do with the invertilibity of A
12:40:40 <oklopol> determinant not being 0 is _not_ what characterizes invertilibity of A
12:41:14 <Phantom_Hoover> yes it... is, isn't it
12:41:37 <oklopol> "A matrix A over a commutative ring R is invertible if and only if det(A) is invertible in R."
12:41:45 <oklopol> this is what we're talking about
12:41:53 <oklopol> not fields
12:41:58 <Phantom_Hoover> ahhhhh
12:42:16 <Phantom_Hoover> well yes i joined well after any of that was said
12:42:25 <oklopol> EXCUSESSSSS
12:43:45 <Phantom_Hoover> should you not call that something other than matrix then
12:43:55 <Phantom_Hoover> like they call vector spaces over a ring modules don't they
12:44:21 <oklopol> perhaps
12:45:11 <int-e> oklopol: To calculate det(A*B) one can use column operations on A (which are column operations on AB) and row operations on B to transform both A and B into upper triangular matrices, with the determinant unchanged or multiplied by -1 (from swapping rows or columns). This works over fields, e.g. Q. So now we have det(AB) = det(A)*det(B) over the rationals. We may regard the entries of A and B as variables, whence det(AB) =...
12:45:17 <int-e> ...det(A)*det(B) becomes an identity of two polynomials of degree n in 2n^2 variables. The identity holds over the rational numbers, which means that the polynomials must be identical. They also have integer coefficients. Therefore, the identity carries over to arbitrary commutative rings.
12:45:23 <oklopol> usually you say "matrix A over a commutative ring R" i guess, but then matrix after you've said it a few times.
12:46:19 <int-e> oklopol: one still needs to show that row and column operations preserve the identity, but that's *far* less messy than doing the full calculation for det(A*B).
12:46:31 <oklopol> let me think about that for a sec
12:48:23 <oklopol> okay i guess that works
12:48:46 <oklopol> i have only read proofs in the case of R and C anyway
12:48:49 <oklopol> R as in reals
12:49:00 <oklopol> C as in complicated numbers
12:49:33 <oklopol> i imagine row and column operations work pretty easily just from the permutations and signs and stuff definition
12:50:02 <int-e> I really like the idea of using polynomial rings for transferring results from Q to commutative rings.
12:50:21 <oklopol> yeah
12:51:47 <oklopol> that's probably not something i would come up with using in a proof, or at least i would then be like "does it really work like this daddy?"
12:52:02 <oklopol> gtg play tennis
13:12:21 -!- impomatic has joined.
13:27:06 -!- yorick has quit (Remote host closed the connection).
13:32:30 -!- yorick has joined.
13:40:22 <Taneb> Is there a way to get a strike-through to render on text-based browsers, especially links2
13:44:06 -!- yiyus has joined.
13:44:41 <mroman_> why won't lex accept my #include <string.h> o_O
13:45:20 <Phantom_Hoover> Taneb, i assume it's like italics
13:47:27 <ais523> mroman_: is it in the wrong part of the file? lex files are divided into sections, it'd need to be in the section appropriate to C includ files
13:51:47 <mroman_> wait
13:51:53 <mroman_> strdup is not actually in string.h
13:52:10 <mroman_> there's no strdup in C
13:53:02 <ais523> indeed
13:53:10 <ais523> it's kind-of trivial to write, though
13:53:53 <mroman_> strdup is gnu99
13:53:56 <mroman_> but not c99
13:53:58 <mroman_> i see
13:54:15 <mroman_> I did not know that
13:54:23 <mroman_> that's kinda surprising
14:10:00 -!- ais523 has changed nick to ais523\unfoog.
14:26:35 -!- mroman_ has quit (Read error: Operation timed out).
14:28:00 -!- mroman_ has joined.
14:38:44 -!- conehead has joined.
14:42:29 -!- Frooxius has quit (Read error: Connection reset by peer).
14:49:34 -!- asie has quit (Read error: Connection reset by peer).
14:51:18 -!- asie has joined.
14:53:58 <fizzie> There are very few (if any) standard C library routines that'd return allocated memory, except for malloc/calloc/aligned_alloc themselves.
14:54:54 <fizzie> And strdup-except-without-allocating-storage is basically strcpy.
14:57:25 <kmc> an allocating sprintf would be nice... I think GNU or BSD has that
14:58:53 <fizzie> Yes, asprintf.
14:59:18 <fizzie> In both GNU and BSD, I believe.
15:00:54 <fizzie> GNU also has the 'a' flag for *scanf conversion specifiers, so that e.g. %as allocates a buffer of sufficient size, and returns a pointer to that.
15:01:28 <fizzie> (And getline, of course.)
15:07:25 -!- Frooxius has joined.
15:28:43 <olsner> ah, strdup is posix at least
15:35:27 -!- asie has quit (Quit: I'll probably come back in either 20 minutes or 8 hours.).
15:39:59 <nooodl_> fizzie: whoa. i never knew about those
15:42:39 -!- asie has joined.
16:41:43 <mroman_> I still have enough space for 15 Instructions...
16:41:49 <mroman_> But I'm running out of ideas.
16:43:13 <mroman_> hm
16:43:20 <Bike> evaluate generalized hypergeometric
16:43:26 <mroman_> distribution?
16:44:06 <mroman_> Hm
16:44:13 <mroman_> "Count Bits"
16:44:17 <mroman_> That oughta be usefull
16:44:26 <Bike> like count one bits?
16:44:46 <mroman_> Yes
16:44:51 <mroman_> 101 -> 2
16:44:59 <Bike> yeah, that's always nice to have
16:45:21 <Bike> how about findfirstset while you're at it
16:45:46 <mroman_> index of the first set bit?
16:45:53 <Bike> mmhm
16:45:57 <mroman_> FSB
16:46:02 <mroman_> :)
16:46:08 <mroman_> a touch of russian
16:46:18 <mroman_> I'm not opposed to it
16:46:24 <mroman_> but where would one need that?
16:47:30 <Bike> it's floor of log
16:47:51 -!- asie has quit (Quit: I'll probably come back in either 20 minutes or 8 hours.).
16:48:09 -!- asie has joined.
16:49:29 <FireFly> Apparently the linux kernel's scheduler uses it to implement a priority queue as a bitfield
16:49:52 <mroman_> ok
16:50:05 <mroman_> then bsr and bsl it ist
16:50:11 <mroman_> (starting from right, starting from left)
16:53:45 <Bike> http://dl.acm.org/citation.cfm?id=1592232
16:54:17 <mroman_> although I currently have no matrix or vector instructions
16:54:38 <mroman_> and no floats
16:54:40 <mroman_> floats are bad
16:55:37 <Bike> what kind of instructions would you have just for vectors? if not just do X in parallel i mean
16:56:12 <mroman_> I don't know yet
16:56:18 <mroman_> also.. that paper seems to be paywalled
16:56:34 <Bike> huh, is that so.
16:57:23 <Bike> "Purchase article: $19" damn bloodsuckers
16:57:59 <Bike> i got a copy if you want it.
16:58:58 <mroman_> I'd like to have a look at it
16:59:02 <mroman_> unless it's too illegal
16:59:05 <mroman_> then I rather don't.
16:59:17 <Bike> i think intel will survive one copied paper.
16:59:30 <Bike> lemme try dcc
17:00:50 <Bike> bam.
17:00:56 <mroman_> nvm
17:00:58 <mroman_> I googled it
17:01:01 <Bike> haha k
17:01:07 <mroman_> inurl:*.pdf
17:01:08 <mroman_> ;)
17:10:52 -!- oklopol has quit.
17:16:16 <mroman_> alright
17:16:24 <mroman_> Bike: Well, I have three register sets
17:16:29 <mroman_> with each 4 Registers
17:16:38 <mroman_> So that would easily allow 2x2 matrices
17:16:53 <mroman_> although 2x2 is probably not really helpful for 2D Graphics?
17:17:04 <Bike> yeah...
17:17:12 <mroman_> all 2D transformation matrices I know are 3x3
17:18:27 <mroman_> however
17:20:24 -!- doesthiswork has joined.
17:21:09 <Bike> with such small matrices would you even get any advantage from a dedicated instruction rather than multiplies and adds, though
17:21:13 -!- yiyus has quit (Read error: Connection reset by peer).
17:21:55 <fizzie> You can multiple a point with a 2x2 matrix to rotate it, if you count that as "helpful for 2D graphics". (Maybe not.)
17:21:56 <Phantom_Hoover> <mroman_> all 2D transformation matrices I know are 3x3
17:22:30 <Phantom_Hoover> can't you represent all euclidean motions of an n-dimensional space with an n+1xn+1 matrix
17:23:49 <doesthiswork> I think you need 4x4 for 3d rotation (quatranons)
17:23:55 <mroman_> Phantom_Hoover: So?
17:23:58 <mroman_> That's exactly the point
17:24:04 <Bike> isn't the extra dimension to let you do translations (which aren't linear)
17:24:04 <mroman_> 2D transformations require 3x3?
17:24:12 <Phantom_Hoover> Bike, yeah
17:24:13 <Bike> nb i don't know shit about grafix
17:24:27 <mroman_> It's the same as vector geometry
17:24:28 <mroman_> ;)
17:24:31 <Phantom_Hoover> (it works because projective geometry)
17:24:39 <mroman_> Instead of points you move pixels
17:24:42 <Bike> but i don't know projective geometry :(
17:25:13 <Bike> well, with 2x2 you can do any linear transform, i guess
17:25:21 <Bike> still not sure i see the computer-side point
17:26:38 <Bike> i mean matrix×vector is gonna be four mults an two adds no matter what, right.
17:27:19 <mroman_> I don't think you can do translations with 2x2 matrices
17:28:06 <Bike> no, you can't. translations aren't linear.
17:28:27 <mroman_> yep
17:30:19 <Bike> http://en.wikipedia.org/wiki/Freivalds%27_algorithm wow this is the easiest algorithm ever
17:36:59 <mroman_> Nope.
17:37:02 <mroman_> That's [-]
17:37:52 -!- asie has quit (Quit: I'll probably come back in either 20 minutes or 8 hours.).
17:40:25 <mroman_> !bfjoust [-]
17:40:25 <EgoBot> ​Use: !bfjoust <program name> <program> . Scoreboard, programs, and a description of score calculation are at http://codu.org/eso/bfjoust/
17:40:31 <mroman_> !bfjoust loop [-]
17:40:41 <EgoBot> ​Score for mroman__loop: 10.5
17:40:46 <mroman_> wow
17:40:49 <mroman_> two digit score .
17:41:29 <mroman_> !bfjoust loop [-]<[+][>--[>+]]
17:41:32 <EgoBot> ​Score for mroman__loop: 9.1
17:42:18 <mroman_> !bfjoust loop [-][>->-<<]
17:42:20 <EgoBot> ​Score for mroman__loop: 10.4
17:42:23 -!- zzo38 has joined.
17:42:41 <mroman_> !bfjoust loop -
17:42:45 <EgoBot> ​Score for mroman__loop: 6.8
17:44:41 <mroman_> !bfjoust ++[[]+][>+][-]
17:44:42 <EgoBot> ​Use: !bfjoust <program name> <program> . Scoreboard, programs, and a description of score calculation are at http://codu.org/eso/bfjoust/
17:44:53 -!- asie has joined.
17:44:53 <mroman_> !bfjoust saladdressing ++[[]+][>+][-]
17:44:57 <EgoBot> ​Score for mroman__saladdressing: 6.4
17:46:35 <ais523\unfoog> mroman_: [[]+] is the suckiest triplock ever :)
17:46:47 <ais523\unfoog> !bfjoust suckiest_triplock_ever [[]+]
17:46:51 <EgoBot> ​Score for ais523_unfoog_suckiest_triplock_ever: 6.4
17:47:30 <mroman_> I thought it'd loop until it drops to zero
17:47:31 <mroman_> increment it
17:47:34 <mroman_> and then wait again
17:47:53 <mroman_> which should've prevented my from being killed
17:48:13 <mroman_> !bfjoust saladdressing [[]+++]
17:48:17 <EgoBot> ​Score for mroman__saladdressing: 6.4
17:48:19 <mroman_> hm.
17:48:47 <int-e> [] takes a cycle to detect the zero, and then it's been there for two cycles and you lost.
17:49:14 <ais523\unfoog> mroman_: it does do that, but it's a cycle too slow to avoid losing
17:49:37 <ais523\unfoog> this is intentional in the rules, that sort of program would be unbeatable if it actually worked
17:49:46 <ais523\unfoog> well, not 100% unbeatable
17:49:47 <ais523\unfoog> but pretty close
17:50:35 <mroman_> !bfjoust saladdressing >+++>+++>+++[]<+++[]<+++[]
17:50:39 <EgoBot> ​Score for mroman__saladdressing: 1.4
17:50:42 <mroman_> Yeah
17:50:45 <mroman_> I can go suckier
17:52:38 -!- Phantom_Hoover has quit (Ping timeout: 245 seconds).
17:52:44 <mroman_> !bfjoust salad (>-)*9([-])*21
17:52:48 <EgoBot> ​Score for mroman__salad: 0.2
17:52:51 <mroman_> Hoho
17:53:04 <mroman_> !bfjoust salad (>-)*9([-]>)*21
17:53:08 <EgoBot> ​Score for mroman__salad: 5.0
17:53:36 <mroman_> How can that be worse than [-]
17:54:13 <mroman_> !bfjoust salad (>->+)*4([-]>)*21
17:54:16 <EgoBot> ​Score for mroman__salad: 6.2
17:54:28 <int-e> !bfjoust ten (>+)*9([[-.]]>)*21
17:54:30 <EgoBot> ​Score for int-e_ten: 11.8
17:54:37 <mroman_> !bfjoust salad (>->+)*4([-]>[+]>)*11
17:54:39 <EgoBot> ​Score for mroman__salad: 8.2
17:54:49 <int-e> !bfjoust ten (>+)*9([[-]]>)*21
17:54:50 <mroman_> what's .?
17:54:52 <EgoBot> ​Score for int-e_ten: 6.1
17:54:54 <int-e> nop
17:55:03 <int-e> takes a cycle :)
17:55:17 <int-e> !bfjoust ten (>+)*9([-.]>)*21
17:55:20 <EgoBot> ​Score for int-e_ten: 11.3
17:58:40 <int-e> [-] defeats [+] almost all the time, and [-] half the cases, I think.
17:59:44 -!- oerjan has joined.
18:00:14 <int-e> (but it's still interesting)
18:00:15 <ais523\unfoog> int-e: but it loses to .
18:00:31 <ais523\unfoog> anyway, that's what I call a "probabilistic lock", I need to add it to the wiki
18:01:24 <ais523\unfoog> against [+] it forms a perfect lock 100% of the time; against [-] it forms a perfect lock 50% of the time, a detectable lock 25% of the time, and loses 25% of the time
18:02:16 <int-e> "detectable" meaning that [-][.this.code.runs.]?
18:02:52 <ais523\unfoog> int-e: yeah
18:03:00 <mroman_> !bfjoust salad (>>--)*4>([-]..+>])*21
18:03:03 <EgoBot> ​Score for mroman__salad: 0.0
18:03:07 <mroman_> yeah
18:03:12 <ais523\unfoog> mroman_: unbalanced []
18:03:14 <mroman_> !bfjoust salad (>>--)*4>([-]..+>)*21
18:03:17 <EgoBot> ​Score for mroman__salad: 6.3
18:03:20 <mroman_> damn
18:03:22 <mroman_> !bfjoust salad (>>--)*4>([-].+>)*21
18:03:25 <EgoBot> ​Score for mroman__salad: 6.3
18:03:35 <ais523\unfoog> mroman_: you don't need that . after the [-]
18:03:41 <ais523\unfoog> the - sets it to 0, that's one
18:03:47 <mroman_> !bfjoust salad (>>--)*4>([-]->)*21
18:03:47 <ais523\unfoog> the ] checks it's 0, that's two
18:03:50 <EgoBot> ​Score for mroman__salad: 7.9
18:04:16 <mroman_> !bfjoust salad (>>--)*4>([-]->)*22
18:04:19 <EgoBot> ​Score for mroman__salad: 7.9
18:04:21 <mroman_> !bfjoust salad (>>--)*4>([-]->)*23
18:04:24 <EgoBot> ​Score for mroman__salad: 7.9
18:04:28 <mroman_> ok
18:04:29 <mroman_> hm
18:04:32 <mroman_> 30 cells at max
18:04:32 <mroman_> so
18:04:43 <mroman_> 22 should get me to cell 30
18:05:29 <mroman_> !bfjoust salad (>>-->+>-)*2>([-]->)*22
18:05:31 <EgoBot> ​Score for mroman__salad: 9.5
18:06:04 <int-e> !bfjoust one >+[]<[-]
18:06:08 <EgoBot> ​Score for int-e_one: 6.2
18:07:02 <int-e> !bfjoust one >+[]<.[-]
18:07:06 <EgoBot> ​Score for int-e_one: 5.7
18:07:30 <mroman_> !bfjoust salad (>>-->+>-)*2>([-][<]+[>]>)*22
18:07:33 <EgoBot> ​Score for mroman__salad: 1.7
18:07:45 <mroman_> [<] finds a non-zero cell, right?
18:07:46 <int-e> hmm. need two more?
18:08:03 <int-e> mroman_: no. it finds a zero cell, skipping over non-zeros
18:08:15 <int-e> !bfjoust one >+[]<..[-]
18:08:18 <EgoBot> ​Score for int-e_one: 6.8
18:08:22 <int-e> !bfjoust one >+[]<...[-]
18:08:25 <EgoBot> ​Score for int-e_one: 7.9
18:08:30 <int-e> funny.
18:08:47 <mroman_> !bfjoust salad (>>-->+>-)*2>([-][<]+[>]>)*220
18:08:50 <EgoBot> ​Score for mroman__salad: 1.7
18:09:07 <int-e> !bfjoust zero .[-]
18:09:11 <EgoBot> ​Score for int-e_zero: 9.8
18:09:12 <mroman_> !bfjoust salad (>>-->+>-)*2>([-]<[<]+[>]>>)*220
18:09:15 <EgoBot> ​Score for mroman__salad: 3.5
18:09:19 <mroman_> !bfjoust salad (>>-->+>-)*2>([-]<[<]+[>]>>)*22
18:09:22 <EgoBot> ​Score for mroman__salad: 3.5
18:09:29 <mroman_> hm. wait
18:09:51 <mroman_> !bfjoust salad (>>-->+>-)*2>([-]<[<]+[>]>)*22
18:09:53 <EgoBot> ​Score for mroman__salad: 8.6
18:10:40 <mroman_> !bfjoust salad (>>-->+>-)*2>([-]<[<]-[>]>)*22
18:10:43 <EgoBot> ​Score for mroman__salad: 8.0
18:10:52 <mroman_> !bfjoust salad >(>>-->+>-)*2>([-]<[<]-[>]>)*21
18:10:55 <EgoBot> ​Score for mroman__salad: 5.2
18:12:37 <FireFly> !bfjoust pancake ->>>-->>+++>*10([-]>)*0
18:12:41 <EgoBot> ​Score for FireFly_pancake: 0.8
18:13:55 <fizzie> http://sprunge.us/DJaG MATLAAAAAB
18:13:55 <mroman_> !bfjoust salad >>->+([>]->[-])*30
18:13:58 <EgoBot> ​Score for mroman__salad: 5.3
18:14:47 <mroman_> wait
18:14:52 <mroman_> .[-] is suicide?
18:15:03 <mroman_> why does that give you more points than my non suicid ones :(
18:18:07 <oerjan> i'm going to guess it's because .[-] can get a tie or even win against a fast attacker
18:18:50 <oerjan> while the other ones have a severe chance of going out of bounds
18:18:57 <fizzie> Apparently an X server is required for "matlabpool local" to operate correctly. Possibly because (from what I recall) the JVM thing is somehow tied to the "desktop" system.
18:21:12 <zzo38> What is "matlabpool local" supposed to do?
18:21:38 <int-e> !bfjoust zero (-)*128[-]
18:21:41 <EgoBot> ​Score for int-e_zero: 10.5
18:22:29 <oerjan> !bfjoust zeroer (-)*-1
18:22:34 <EgoBot> ​Score for oerjan_zeroer: 11.5
18:26:10 <fizzie> zzo38: It's related to the MATLAB Distributed Computing Server, which is a system for parallelizing MATLAB code. It has a concept of a "pool" where the workers come from; "matlabpool open local 4" tells it to make up a pool that just involves running up to 4 workers on the local machine. (Other pools may distribute the work into nodes in a cluster, and so on.)
18:38:07 <mroman_> !bfjoust salad >+++>(>(+)*4[<---]+++)*1000
18:38:11 <EgoBot> ​Score for mroman__salad: 0.0
18:38:15 <mroman_> what
18:38:34 <ais523\unfoog> mroman_: it's the [<---] loop
18:38:43 <ais523\unfoog> you have nothing but nonzero value between there and your flag
18:38:48 <ais523\unfoog> so you end up falling off your own end of the tape
18:39:30 <mroman_> well
18:39:31 <mroman_> no
18:40:02 <mroman_> not on egojsout
18:40:18 <ais523\unfoog> ah right, no, cell 2 is clear
18:40:30 <ais523\unfoog> you fall off your own end of the tape when the opponent unclears it
18:41:26 <mroman_> yeah
18:42:25 <int-e> huh.
18:42:32 <ais523\unfoog> and by 2 I mean 3
18:42:43 <ais523\unfoog> because I use 1-based indexing for BF Joust for some reason
18:42:50 <ais523\unfoog> maybe I should change my mental indexing system
18:43:58 <zzo38> I found something which is called SWIG says is for wrapping and interface generator for several programming languages, but it doesn't mention Haskell and SQL.
18:44:47 <mroman_> !bfjoust salad >+++>(>(+)*4[<---]+++>)*100([-]>)*21
18:44:50 <EgoBot> ​Score for mroman__salad: 1.6
18:47:44 <int-e> ais523\unfoog: but it needs a 3 on the tape - there is one in cell 2, but if that gets modified (and all bots on the ladder will eventually attack), then it trips over the left end very quickly.
18:48:57 <int-e> !bfjoust zero >+[]<<
18:49:01 <EgoBot> ​Score for int-e_zero: 1.9
18:49:27 <int-e> !bfjoust zero >+++[]<<
18:49:31 <EgoBot> ​Score for int-e_zero: 3.6
18:49:41 <fizzie> The definition of persistence, from #scheme@freenode: http://sprunge.us/SOgT
18:51:17 <int-e> aah. <chare> i was hoping you would do the work
18:51:20 <ais523\unfoog> fizzie: I remember reading somewhere that one of Blizzard's early RTSes, I think it was Starcraft, was delayed while they tried to reach feature parity with something that one of their rivals had demoed
18:51:43 <ais523\unfoog> and it turns out that their rivals didn't have the feature at all, the demo was just a video recording rather than something actually playing
18:52:28 <mroman_> hu
18:52:35 <mroman_> why does [>+][-]> run over the tape?
18:52:43 <mroman_> [>+] should find a non zero cell
18:52:50 <mroman_> then zero it
18:52:55 <mroman_> and then move to the next one
18:53:03 <fizzie> After the +, pretty much any cell is nonzero, so the loop goes on and on.
18:53:08 <ais523\unfoog> mroman_: [> moves to the next cell, + sets it to 1, ] goes back to the start of the program
18:53:19 <int-e> for [>+] to stop, the cell below the pointer must be zero.
18:53:30 <mroman_> so
18:53:35 <mroman_> how does one find a non-zero cell?
18:54:10 <ais523\unfoog> mroman_: (>[program here])*29
18:55:16 <myname> anybody here knows where to find this regex example of why ruby sucks at parsing with whitespaces?
18:55:34 <oerjan> ais523\unfoog: well _did_ blizzard reach feature parity anyhow
18:55:51 <ais523\unfoog> oerjan: yeah, they were the first to impl the features
18:56:57 <fizzie> I think that's called the "bait-and-switch".
18:57:43 <fizzie> (The "demo" of a nonexistent feature, that is.)
18:57:56 <oerjan> <fizzie> The definition of persistence, from #scheme@freenode: http://sprunge.us/SOgT <-- did they manage to ban him, alternatively did shachaf quit in disgust
18:58:04 -!- Deewiant has quit (Ping timeout: 264 seconds).
18:59:27 -!- myndzi has quit (Ping timeout: 240 seconds).
19:03:04 -!- myndzi has joined.
19:03:39 <fizzie> oerjan: I wasn't really following. My grep sees no bans, though no other signs after that 15th-day appearance either. Still, I'm sure that wasn't the last we'll see of them.
19:04:43 -!- augur_ has changed nick to augur.
19:05:12 <oerjan> mhm
19:06:10 <mroman_> !bfjoust nop []+
19:06:14 <EgoBot> ​Score for mroman__nop: 6.5
19:06:48 -!- asie has quit (Quit: I'll probably come back in either 20 minutes or 8 hours.).
19:07:54 <oerjan> !bfjoust zeroest (-.)*-1
19:07:57 <EgoBot> ​Score for oerjan_zeroest: 10.2
19:09:12 <mroman_> !bfjoust salad >+>+[]<<[+]
19:09:16 <EgoBot> ​Score for mroman__salad: 3.1
19:13:53 <mroman_> !bfjoust salad >+>+>+[]<[+]<[]<[-]
19:13:58 <EgoBot> ​Score for mroman__salad: 0.3
19:14:25 <mroman_> !bfjoust salad >+>+>+[]<[+]<[]<[-](>[-])*30
19:14:30 <EgoBot> ​Score for mroman__salad: 1.0
19:16:49 -!- Deewiant has joined.
19:17:03 <FireFly> fungot: hey, have *you* tried bfjoust?
19:17:03 <fungot> FireFly: ( i remember something like rest or tail). new hardware is required usually for games, its fpu ran much slower than realtime to figure out how
19:17:26 <fizzie> fungot: I don't think you'd really need new hardware for bfjoust.
19:17:26 <fungot> fizzie: these people really _are_ on crack. fnord::' demi-thread. if you have to code an attempt at adding " l'" to the wall with bug listings
19:18:49 <mroman_> fungot: You're crack.
19:18:50 <fungot> mroman_: and that's a neat new language
19:18:57 <mroman_> Wow. Gee. Thanks
19:19:03 <mroman_> I always new all my languages were neat.
19:19:05 <mroman_> *knew
19:19:27 <mroman_> fungot: What about your languages?
19:19:27 <fungot> mroman_: the primary documentation for emacs is done.
19:19:58 <mroman_> fungot: I'd rather have vim...
19:19:58 <fungot> mroman_: i have a midland accent ( or in most cases it is best not to hang around those cobol programmers in their late thirties and early fnord
19:20:00 <FireFly> fungot: elisp isn't "your" language
19:20:01 <fungot> FireFly: that is the fucking point. i really appreciate it. as i type. how nice.
19:20:39 <FireFly> I'm not sure what point fungot is trying to make..
19:20:39 <fungot> FireFly: y " is not a great essay writer the sentence choices seem odd.
19:23:09 <int-e> fungot writes essays?
19:23:09 <fungot> int-e: so much about coding, as i've been giving them some stuff to do. you don't have to
19:31:01 <nooodl_> the sentence choices DO seem odd
19:32:56 <mroman_> how dare you question that
19:39:18 <doesthiswork> this is a funny quiz about undefined behavior in C http://blog.regehr.org/archives/721
19:41:52 <doesthiswork> !bfjoust [>+][>-]
19:41:53 <EgoBot> ​Use: !bfjoust <program name> <program> . Scoreboard, programs, and a description of score calculation are at http://codu.org/eso/bfjoust/
19:42:08 <doesthiswork> !bfjoust blah [>+][>-]
19:42:11 <EgoBot> ​Score for doesthiswork_blah: 0.0
19:43:27 <doesthiswork> !bfjoust blah2 [>-]
19:43:30 <EgoBot> ​Score for doesthiswork_blah2: 0.0
19:43:44 -!- AnotherTest has joined.
19:44:14 <olsner> doesthiswork: I'm disappointed that some of the questions are about specific implementations rather than C
19:46:09 <doesthiswork> he said that no one uses the standard, they all use implementations of the standard so questions about implementations are more meaningful
19:47:41 <doesthiswork> for the kinds of things you need to watch out for
19:53:25 <oerjan> i assume that is why he made me fail question 5, the bastard.
19:53:59 <oerjan> of _course_ i'd know the answer if he'd said only actual implementations counted.
19:54:03 <oerjan> *known
19:55:50 <fizzie> I remember reading that thing and finding a number of things to nitpick on.
19:57:05 <fizzie> Like in the explanation of Q4, it speaks of "promoting" an unsigned int to long, but that's (at least in nitpick mode) not a promotion, it's just part of the usual arithmetic conversions.
19:57:23 <olsner> it's a nice touch that the introduction with all those assumptions gets hidden when you start the quiz
19:59:03 <fizzie> The signedness of char wasn't in the introduction, anyway.
19:59:15 <doesthiswork> that's what makes it fun :)
19:59:23 <fizzie> Oh, it says as much in the explanation.
20:00:08 -!- asie has joined.
20:02:57 <fizzie> I also disagree with Q10 explanation, which claims that "other values can be safely left-shifted", while in fact no negative values can.
20:03:52 -!- CADD has joined.
20:05:03 <ais523\unfoog> !bfjoust attack1 [>[-]+]
20:05:06 <EgoBot> ​Score for ais523_unfoog_attack1: 3.8
20:05:47 -!- asie has quit (Quit: I'll probably come back in either 20 minutes or 8 hours.).
20:15:54 <shachaf> fizzie: Aw, I thought it would be about persistent data structures.
20:18:32 <fizzie> I also think Q18 is unambiguously wrong. It claims that, for int x, "(short)x + 1" is defined for all values of x, while in fact it's undefined, as acknowledged in the comments.
20:20:00 <fizzie> And the "I’ll stand by my answer since I’ve never heard of an LP64 compiler that doesn’t just truncate" cop-out in the comments is utter nonsense, given the "Make no assumptions about undefined behaviors" requirement of the introduction, and the way other questions (like overflowing a signed integer) are handled.
20:21:23 <fizzie> Oh, conversion to narrower integer was in fact implementation-defined? How weird. I guess in that case it's a little more defensible.
20:21:31 <olsner> I was thinking that if sizeof int == sizeof short, then (short)x+1 overflows if x+1 overflows
20:21:54 <Bike> c is fuked up
20:22:21 <fizzie> olsner: "sizeof (int) == sizeof (short)" is more clearly disallowed by the assumptions in the introduction, though.
20:22:54 <olsner> yes, that was before I figured out which assumptions I was supposed to be making
20:24:14 -!- nooodl_ has quit (Read error: Connection reset by peer).
20:24:40 -!- nooodl_ has joined.
20:26:42 -!- asie has joined.
20:26:46 <zzo38> Do you know of any free graphics plotting software with language bindings for calling from SQL?
20:27:10 <ais523\unfoog> there's a thread on comp.lang.c about whether "f() + f()" is defined behaviour for all int-returning functions f (when f itself always has defined behaviour)
20:27:25 <ais523\unfoog> anyone want to take a guess at the currently accepted answer?
20:27:37 <shachaf> I assume it isn't?
20:27:53 -!- yorick has quit (Ping timeout: 246 seconds).
20:28:08 -!- yorick has joined.
20:28:45 <zzo38> ais523\unfoog: It probably works for most functions on most computers, I would think; maybe there are some cases allowed by the C standard where this doesn't work but I don't know how obscure these are
20:29:45 <ais523\unfoog> the answer sees to be "it's always defined in C11; it's also always defined in C89 as of Defect Report 87 and C99 as of Defect Report 287, the actual text of the standards is silent on the issue"
20:30:18 <zzo38> O, so it depends on the standard.
20:30:33 <olsner> what happened to signed int overflows? are those not-undefined now?
20:30:50 <shachaf> http://www.vex.net/~trebla/humour/tautologies.html
20:30:51 <fizzie> olsner: I think it's not "about that".
20:31:05 <fizzie> olsner: At least based on what those defect reports address.
20:45:26 <ais523\unfoog> olsner: I think they're still undefined
20:46:00 <ais523\unfoog> let me check n1570
20:48:11 <ais523\unfoog> yeah, C11 says that overflow is undefined, and unsigned operations cannot overflow
20:50:18 * oerjan is pretty sure olsner meant in the f(x) + f(x) context
20:50:32 <oerjan> er, f() + f()
20:51:00 <olsner> yes, that's what I was talking about
20:51:27 -!- KingOfKarlsruhe has quit (Quit: ChatZilla [Firefox 25.0/2013102400]).
20:51:33 -!- oerjan has quit (Quit: Good that's settled, good night!).
20:51:53 <ais523\unfoog> oh, I wasn't thinking
20:52:35 <ais523\unfoog> bleh, are there any symmetrical binary operators that are defined no matter what their arguments?
20:52:51 <ais523\unfoog> bitwise operators aren't fully defined on negative numbers
20:53:07 <ais523\unfoog> + - * can all overflow, / and % can divide by zero
20:53:19 <ais523\unfoog> (and / can also overflow if you do INT_MIN / -1)
20:53:56 <ais523\unfoog> we can at least change the problem to "is f() + f() always defined if f() is always defined and returns an unsigned int?"
20:54:17 <olsner> or simply assume that f doesn't return something that causes an overflow
20:54:41 <ais523\unfoog> yeah
20:55:22 <zzo38> In LLVM, at least, I think addition and subtraction are defined to wrap around unless you explicitly specify that overflow is undefined, in which case it is undefined.
20:55:59 <ais523\unfoog> in Verity, there aren't signed versions of + - * because it's always two's complement and thus the unsigned versions work just fine
20:56:04 <ais523\unfoog> so there's no issue with signed overflow
20:56:45 <ais523\unfoog> except in signed cast-to-fewer-bits, which is defined to simply take the low bits
20:59:38 <zzo38> Yes, that would work too (actually LLVM does the same thing; there is no separate signed and unsigned add command, although there are parameters to tell it to be signed or unsigned if you want overflow to be undefined)
20:59:59 -!- asie has quit (Quit: I'll probably come back in either 20 minutes or 8 hours.).
21:03:23 <zzo38> In Z-machine all arithmetic operators are signed and there are no unsigned versions. Overflow is supposed to be an error, but making it not error for addition and subtraction would help so that you can make address arithmetic.
21:05:40 <zzo38> But I did not think of division overflowing! Now I know. Therefore, DIV -32768 -1 S 0 (Frolg syntax) in Z-machine would also be error.
21:05:45 -!- carado has quit (Ping timeout: 252 seconds).
21:06:28 -!- MindlessDrone has quit (Quit: MindlessDrone).
21:06:29 <quintopia> sup
21:07:09 -!- asie has joined.
21:08:50 -!- ajf has joined.
21:24:23 -!- yiyus has joined.
21:24:55 <ion> http://steamcommunity.com/sharedfiles/filedetails/?id=195992856
21:25:48 -!- doesthiswork has quit (Quit: Leaving.).
21:25:55 <fizzie> "Human Revolution" sounds like an appropriate name.
21:26:11 <fizzie> (Now everyone knows your Steam Name.)
21:30:51 -!- asie has quit (Quit: I'll probably come back in either 20 minutes or 8 hours.).
21:33:06 <ion> So?
21:33:44 <fizzie> Well, they can...
21:33:50 <fizzie> I'm sure they can do something dastardly.
21:33:52 <Taneb> THEY CAN STEAM YOU
21:33:54 <fizzie> Friendly you in Steam.
21:33:56 <Taneb> STEEEAAAAAM
21:33:59 <fizzie> No "ly" in there.
21:34:14 <fizzie> There's no "lye" in your friend, at least under normal circumstances.
21:34:20 <olsner> steamly you in friendship
21:35:27 -!- lambdabot has quit (Ping timeout: 252 seconds).
21:46:22 -!- prooftechnique has joined.
21:55:43 -!- AnotherTest has quit (Ping timeout: 252 seconds).
21:57:43 <Taneb> Oh god I watched like a whole anime this weekend and suddenly I'm noticing loads of people reblogging it on Tumblr
21:58:00 <prooftechnique> What anime?
21:58:08 <Taneb> Puella Magi Madoka Magica
21:58:27 <prooftechnique> Neat choic
21:58:28 <prooftechnique> *choice
21:58:45 <quintopia> Taneb: it is quite a popular one yes
21:59:15 <prooftechnique> MOE
21:59:49 <Taneb> I enjoyed it
22:00:15 <Taneb> One of my friends tried to get me to cosplay it but then we realised how expensive that'd be
22:01:46 <Taneb> (really expensive)
22:01:52 <Taneb> (also I'd probably have to shave)
22:03:59 <prooftechnique> Taneb: Go as Snake Quiet. Much cheaper
22:04:16 <prooftechnique> (Quiet Snake, I guess)
22:04:43 <Taneb> I already have a Jake English cosplay
22:04:49 <prooftechnique> http://nerdreactor.com/wp-content/uploads/2013/09/metal_gear_solid_v__big_quiet_boss_xd_by_zombiesandwich-d6lio0e-424x600.jpg
22:04:57 <prooftechnique> nsfw
22:04:58 <prooftechnique> ?
22:05:12 <Taneb> prooftechnique, I don't think I have the muscle tone for that
22:17:45 -!- asie has joined.
22:19:22 -!- Phantom_Hoover has joined.
22:22:49 -!- asie has quit (Ping timeout: 265 seconds).
22:24:09 -!- ajf has left.
22:40:10 <shachaf> int-e: lambdabot :☹(
22:42:47 <fizzie> `unidecode :☹(
22:42:49 <HackEgo> ​[U+003A COLON] [U+2639 WHITE FROWNING FACE] [U+0028 LEFT PARENTHESIS]
22:43:00 <fizzie> Oh no, some kind of recursive frown.
22:43:07 <shachaf> Just nested.
22:43:22 <fizzie> If you zoom and enhance, I'm sure there's more frowns where that came from.
22:44:38 <shachaf> > let zoom [a,b,c] = [b]; let enhance "☹" = ":☹(" in (enhance . zoom) ":☹("
22:44:45 <shachaf> int-e: lambdabot :☹(
22:45:23 <fizzie> Don't worry, I'm sure there would've been some kind of Unicode problem.
22:46:00 <shachaf> Yep.
22:49:44 <nooodl_> `run echo 'zoom [a,b,c] = [b]; enhance "*" = ":*("; main = print ((enhance . zoom) ":*(")' | runghc
22:49:48 <HackEgo> ​":*("
22:50:07 <nooodl_> * is practically a smiley face. if you squint
22:50:41 <FireFly> Or a massively huge tear
22:55:19 -!- ais523\unfoog has quit.
23:19:05 -!- Slereah has quit (Remote host closed the connection).
23:19:17 -!- Slereah has joined.
23:19:21 -!- Oj742 has joined.
23:25:22 <Taneb> I DON'T HAVE ANY MILK
23:25:49 <Sgeo> I despise CAOS, but there's still an interesting property. In one way, it's like REBOL
23:26:18 <Sgeo> a b c could be interpreted as a (b c) if needed
23:26:34 <Sgeo> setv name "my variable" 5
23:26:46 <Sgeo> setv takes a variable and an int and puts the int into the variable
23:27:00 <Sgeo> name "my variable" refers to the variable on the targ called "my variable"
23:27:20 <Sgeo> (targ is a piece of global state, blech, but the REBOL-like processing is still interesting)
23:31:41 -!- yiyus has quit (Read error: Connection reset by peer).
23:42:30 <shachaf> Taneb: I DO
23:42:58 <Taneb> shachaf, IS IT FULL FAT
23:43:32 <ion> http://heh.fi/tmp/mpiling_data.jpeg
23:44:02 <shachaf> Taneb: IS THAT THE SAME AS WHOLE
23:44:11 -!- Oj742 has quit (Quit: irc2go).
23:44:15 <Taneb> shachaf, YES
23:44:41 <shachaf> Taneb: THAT'S WHAT IT SAID ON THE COW
23:45:31 -!- nisstyre has joined.
23:45:58 <Taneb> shachaf, CAN I HAVE SOME OF YOUR MILK PLEASE
23:46:11 <shachaf> Taneb: SURE
23:47:27 <shachaf> THAT MIGHT GET TO ME
23:47:56 <shachaf> Hmm, I bet you could figure out how to send mail to me.
23:50:48 <shachaf> @google tuscan whole milk
23:51:00 <shachaf> int-e: lambdabot :☹(
23:53:05 -!- Sprocklem has joined.
23:55:27 <nooodl_> :😸3
←2013-11-16 2013-11-17 2013-11-18→ ↑2013 ↑all