←2014-04-27 2014-04-28 2014-04-29→ ↑2014 ↑all
00:00:09 -!- tromp has joined.
00:02:36 -!- ^v has quit (Read error: Connection reset by peer).
00:02:53 -!- ^v has joined.
00:05:57 -!- not^v has joined.
00:06:19 -!- ^v has quit (Read error: Connection reset by peer).
00:31:32 -!- nisstyre has joined.
00:37:06 -!- not^v has quit (Quit: http://i.imgur.com/DrFFzea.png).
01:04:07 <zzo38> I have idea: data Program = Halt Expression | Continue | Assign Expression Expression Program | While Expression Program Program; data Expression = Zero | Successor Expression | Dereference Expression | With Expression Expression | Call Program | Oracle Program;
01:05:11 <zzo38> If you have such halting oracle, what is the powerfulness?
01:05:36 -!- Vamadeus has quit (Ping timeout: 240 seconds).
01:07:33 <oerjan> zzo38: about like brainhype, i think
01:10:07 <FreeFull> zzo38: Then you have to worry about metahalting
01:10:16 <zzo38> What would be the powerfulness of proofs in such a system (where the halting oracle is still part of the system, and programs that halt are theorems)?
01:10:23 <zzo38> FreeFull: What is metahalting?
01:11:24 <Taneb> What ever happened to itidus20
01:11:43 <Bike> couldn't escape the matrix of [reference] [sarcastic sad emoticon]
01:12:43 <FreeFull> zzo38: It can figure out if any Turing machine will halt, but can't determine if it will halt itself
01:13:28 <pikhq> zzo38: Basically, for any superset-of-Turing system, the union of that + the halting oracle for it cannot determine if it *itself* halts.
01:13:48 <pikhq> So, Turing machine with a halting oracle has its own halting problem.
01:14:26 <Bike> you'll get, what's it called, post degrees
01:14:50 <zzo38> pikhq: Yes, although it might be paradox if it can, isn't it?
01:15:30 <pikhq> The proof that you can't determine if a Turing machine halts with a Turing machine is trivially generalizable to any super-Turing machine.
01:16:39 <zzo38> You can still determine if such a thing halts with a higher oracle, though, I think
01:16:44 <pikhq> Yes.
01:16:45 <Taneb> `quote 897
01:16:46 <HackEgo> 897) <shachaf> Taneb: STOP TRYING TO GET LENS INTO EVERYTHING <shachaf> Bike: You should use lens! <Taneb> NEVER <Bike> shachaf: i'm getting mixed messages here
01:17:00 <Taneb> shachaf, you will be glad that today I told someone not to use lens.
01:17:08 <pikhq> It's just that each individual such system would need an oracle to determine if it halts.
01:20:01 <shachaf> Taneb: when will i be glad about that?
01:20:11 <shachaf> wait, not glad about that. just glad that.
01:21:19 <Taneb> Because I did not try to get lens into his Hakyll site!
01:21:30 <zzo38> But it can be nested such as Halt (Oracle (While (Oracle x) Continue (Halt Zero))) meaning if it is true when halting then that makes it true when not halting, something like that anyways
01:21:44 <FreeFull> Use lens to implement lens
01:22:02 <Taneb> `uote 1059
01:22:02 <HackEgo> ​/home/hackbot/hackbot.hg/multibot_cmds/lib/limits: line 5: exec: uote: not found
01:22:05 <Taneb> `quote 1059
01:22:06 <HackEgo> 1059) <Taneb> I would like to learn how to use a sword <Taneb> And also how to ride a unicycle <Taneb> Perhaps not at the same time
01:22:08 <Taneb> :(
01:22:13 <Taneb> I have not learnt either
01:22:25 <zzo38> Can it be extended to allow an arbitrary amount if a program is reading in from the input
01:22:33 <Taneb> Although I did go to an Iaido taster session!
01:23:44 <FreeFull> Taneb: Buy a unicycle!
01:24:13 <Taneb> FreeFull, I actually have one
01:24:31 <FreeFull> Try to learn it!
01:25:01 <Bike> learn you a unicycle for great good
01:26:48 <Taneb> `addquote <Bike> learn you a unicycle for great good
01:26:49 <HackEgo> 1189) <Bike> learn you a unicycle for great good
01:28:09 <Bike> important question: is there a program that, given a number, gives you a bunch of pointless facts about it, a la http://en.wikipedia.org/wiki/593_%28number%29
01:30:38 <oerjan> yes
01:31:32 <oerjan> the proof is a simple exercise in writing a wikipedia scraper hth
01:31:42 <Bike> perhaps i need a rephrase: can you point me to a etc
01:31:55 <oerjan> picky, picky
01:32:37 <Bike> hm, i know a few programs to come up with an algebraic definition, i think
01:43:16 -!- BeingUntoDeath has joined.
01:46:37 -!- BeingUntoDeath has quit (Client Quit).
01:59:06 -!- tertu has joined.
02:01:36 -!- tertu has quit (Read error: Connection reset by peer).
02:01:42 -!- ter2 has joined.
02:02:19 -!- ^v has joined.
02:06:06 -!- ter2 has quit (Read error: Connection reset by peer).
02:06:32 -!- ter2 has joined.
02:08:20 -!- FreeFull has quit.
02:10:23 -!- ter2 has quit (Read error: Connection reset by peer).
02:10:46 -!- ter2 has joined.
02:23:21 -!- tertu3 has joined.
02:26:41 -!- ter2 has quit (Ping timeout: 264 seconds).
02:33:41 -!- ^v has quit (Ping timeout: 252 seconds).
02:34:14 <zzo38> In which MUD is it possible to use a zoog teeth you have stolen to learn to be dentist and make a false teeth in case other zoog don't have any more teeth?
02:38:29 -!- nisstyre has quit (Quit: WeeChat 0.4.3).
02:38:38 -!- tertu3 has quit (Ping timeout: 252 seconds).
03:01:23 -!- tertu3 has joined.
03:07:53 <kmc> fungot: if I knew now what I knew then…
03:27:52 -!- Sorella has quit (Quit: It is tiem!).
03:35:09 <kmc> fungot: why the long lag
03:41:09 <shachaf> https://plus.google.com/+DanPiponi/posts/djyKdnSGRch dpiponi.kmc.moed++
03:44:05 <ion> :-)
03:48:57 -!- edwardk has quit (Quit: Computer has gone to sleep.).
03:54:01 <oerjan> fungot: are you stonewalling kmc
03:54:02 <fungot> oerjan: don't i know you from somewhere? but, i mean, t-rex?
03:54:13 <oerjan> yes, yes you do.
03:57:37 <kmc> fungot: who will fungot the fungots?
03:57:37 <fungot> kmc: hey, t-rex, i've a long road ahead of me, it would be flattering. i like to look at them! i expect nothing less than...
03:58:04 <oerjan> fungot: nothing less than what?
03:58:05 <fungot> oerjan: have i, perhaps, a hint, utahraptor. i'm a cyborg, then when i enter it, my body is the other half. we join, marrying the machine and flesh, working together to ensure my own creation. i'm responsible for getting my own food and for making a bomb threat, delaying its maiden voyage!
03:58:31 <oerjan> those are important responsibilities.
03:59:50 -!- edwardk has joined.
04:01:58 <oerjan> hm low-friction bread
04:02:10 -!- kronion_ has joined.
04:02:19 <kmc> read "low-friction beard"
04:02:30 -!- kronion_ has quit (Max SendQ exceeded).
04:02:39 <oerjan> also a good concept
04:02:59 <oerjan> (the bread almost slid off my plate when i grabbed it)
04:03:32 <oerjan> i hope the onion hasn't started spamming
04:27:18 -!- nycs has joined.
04:29:41 -!- `^_^v has quit (Ping timeout: 264 seconds).
04:50:08 <Taneb> I have been awake for faaaar too long
05:01:30 -!- Deewiant has joined.
05:03:17 -!- tswett has quit (Ping timeout: 252 seconds).
05:07:19 -!- Tod-Autojoined has changed nick to TodPunk.
05:12:46 -!- tromp has quit (Remote host closed the connection).
05:12:54 -!- Sgeo has quit (Read error: Connection reset by peer).
05:13:18 -!- tromp has joined.
05:18:01 -!- tromp has quit (Ping timeout: 265 seconds).
05:25:27 <fizzie> fungot: Stop making bomb threats, they'll trace you back to my place.
05:25:27 <fungot> fizzie: oh, that may be a fact; it is impossible to know if my dream, everything had turned into plasticine! do that and now your sweetie, turn to panel 5. forget it! it's wasted time.
05:27:35 <oerjan> and then fizzie will be toast while fungot gets away on an insanity plea
05:27:35 <fungot> oerjan: but in a metaphorical sense. what metaphor would that be, exactly?
05:27:56 <Taneb> I am going to sleep now goodnight
05:27:58 * Taneb away
05:28:14 <oerjan> fungot: something cell-based.
05:28:15 <fungot> oerjan: as a man concerned, t-rex, that going shopping was your default activity? how is that weird, is it? you're just adding the same. but it's just so impossible, so doomed. he's got a stupid face, so at his own request that you give a raccoon a sugar cube, it'll wash it, i have!
05:28:17 <Taneb> Wait that isn't I away
05:28:40 <oerjan> Taneb is too tired to leave the channel
05:28:47 <fizzie> fungot: Stop calling my face stupid, too.
05:28:47 <fungot> fizzie: i don't have a very good student!
05:29:05 -!- Sgeo has joined.
05:32:36 <zzo38> Was "quit while you are ahead" invented by a gambler?
06:07:46 <zzo38> [P1] Clean your room! [C1] Therefore, clean your room or burn the house down! The error is that, "therefore, you have to either clean your room or burn the house down" is wrong, but, you can write instead, "either you have to clean your room or you have to burn the house down"
06:12:14 <zzo38> Isn't it?
06:41:14 -!- tertu3 has quit (Ping timeout: 255 seconds).
06:46:53 -!- oerjan has quit (Quit: leaving).
07:30:57 -!- slereah has joined.
07:31:53 -!- Sprocklem has quit (Quit: leaving).
08:01:25 -!- drdanmaku has quit (Quit: Connection closed for inactivity).
08:13:33 -!- HackEgo has quit (Remote host closed the connection).
08:28:25 -!- Patashu has joined.
08:34:04 -!- Patashu_ has joined.
08:34:04 -!- Patashu has quit (Disconnected by services).
08:38:26 <kmc> `coins
08:38:47 <kmc> gotta get paid
08:40:43 <fizzie> Huh, esolangs.org went down too, and the silly thing doesn't answer to SSH either.
08:41:14 <Jafet> Does fungot have coins
08:41:14 <fungot> Jafet: it's a very tricky to sound natural, then you can't sleep at all, then you go demented and then you die, we're giving dinosaur comics to the marmaduke guy. sincerely, the man, you just wait! kids will be boarding the train to scarytowne at my own pace, dromiceiomimus
08:42:02 <Jafet> Is fungot less volatile than btc
08:42:02 <fungot> Jafet: i think that you're being too strict about this, t-rex! geez!! i'm leaving. escape to panel 5. forget it! it's wasted time. prepare to lose your job for gibbering madness all the time!
08:42:20 <fizzie> You can pay your fungot usage with funcoins.
08:42:20 <fungot> fizzie: are you in this house, god? i am a different, but i assure you
08:42:49 <Jafet> I think fungot just fired me
08:42:49 <fungot> Jafet: oh, i must have put that in your pipe and smoke it? friday? is it friday? or is it..."
08:43:27 <Jafet> itt fungox
08:44:01 <fizzie> --- www2.codu.org ping statistics ---
08:44:02 <fizzie> 42 packets transmitted, 11 received, 73% packet loss, time 42027ms
08:44:09 <fizzie> Well, that's curious.
08:45:01 <Jafet> ssh over icmp
08:45:03 <lexande> fungot: did you say something about a train?
08:45:03 <fungot> lexande: think of how far did you go" talk a lot, but they're always a good thing! stupid problems, but if they're sentient, and you shouldn't think about religion that hard
08:45:23 <lexande> i did go pretty far
08:46:09 <Jafet> I don't think religion is ready for sentient trains
08:46:13 <lexande> pretty much as far as i could, on a train, but i never reached scarytowne
08:59:46 -!- HackEgo has joined.
09:01:13 -!- impomatic has joined.
09:08:35 -!- MindlessDrone has joined.
10:09:50 -!- boily has joined.
10:18:39 -!- conehead has quit (Quit: Computer has gone to sleep).
10:33:49 <Taneb> Mornin'
10:40:51 <boily> 'matin
10:43:58 <Taneb> int-e, ping
10:51:30 -!- boily has quit (Quit: Poulet!).
11:28:17 <Taneb> @ask How awkward it be to make lambdabot not respond to > in #cs-york? That's the only real controversy about the bot in the channel
11:28:18 <lambdabot> Consider it noted.
11:28:22 <Taneb> Wait
11:28:25 <Taneb> @unask
11:28:25 <lambdabot> Who should I ask?
11:28:28 <Taneb> :(
11:28:33 <Taneb> @ask int-e How awkward it be to make lambdabot not respond to > in #cs-york? That's the only real controversy about the bot in the channel
11:28:33 <lambdabot> Consider it noted.
11:29:05 -!- ais523 has joined.
11:29:56 -!- How has joined.
11:30:04 <How> @messages-loud
11:30:04 <lambdabot> Taneb asked 1m 46s ago: awkward it be to make lambdabot not respond to > in #cs-york? That's the only real controversy about the bot in the channel
11:30:11 -!- How has quit (Client Quit).
11:30:35 <fizzie> Doing our part to keep things tidy.
11:33:37 <Jafet> @ask d41d8cd98f00b204e9800998ecf8427e hi
11:33:37 <lambdabot> Consider it noted.
11:34:40 <fizzie> That's too long to even change nick to. :(
11:36:08 <HackEgo> [wiki] testing, testing
11:40:25 <HackEgo> [wiki] still working?
11:40:30 <fizzie> I guess so.
11:41:47 <ais523> huh, Knuth did a few bugfixes to TeX again; apparently he said in 2008 that he would next look at TeX starting on 31 December 2013, and kept to his word
11:42:24 <slereah> Maybe he thought that the world would end in 2012
11:42:45 <ais523> no, he follows a geometric series for bugfixing TeX
11:42:56 <ais523> each review of bugs takes one more year than the last
11:43:03 <ais523> he fixed exactly one bug this time
11:43:24 <ais523> in TeX, that is
11:43:26 <ais523> plus one in Metafont
11:43:57 <ais523> both of which are incredibly minor (a missing space in a corner case that probably never comes up in practice, and a memory leak that leaks 7 words of memory)
11:44:47 <ais523> also apparently the computer he tested the memory leak on has "3 million words" of memory, which for any sensible word size, puts it in the low tens of megabytes or perhaps even lower
11:46:13 -!- nucular has joined.
11:46:13 -!- nucular has quit (Changing host).
11:46:13 -!- nucular has joined.
11:46:28 <ais523> <Knuth> These files contain almost exactly 7 megabytes altogether; and the new changes have altered about 3500 of those bytes.
11:46:31 <b_jonas> ais523: not geometric. I think it's quadratic only.
11:46:41 <ais523> err, yes
11:46:49 <ais523> sorry, not really thinking today
11:47:48 <Jafet> If it's truly a series, Knuth will be around to fix TeX bugs forever
11:48:00 <Jafet> I don't know whether that is supposed to be a comforting thought
11:48:26 <fizzie> Does this mean the version number gained a digit?
11:48:52 <b_jonas> Jafet: he has a specific statement: "My last will and testament for TeX and METAFONT is that their version numbers ultimately become $\pi$ and $e$, respectively. At that point they will be completely error-free by definition."
11:49:05 <ais523> fizzie: yes
11:49:41 <ais523> also, apparently the only regret Knuth has about TeX's original design is that the default rule thickness is 0.399993896484375 points rather than 0.4 points
11:49:49 <ais523> due to the use of binary rather than decimal fractions
11:50:11 <b_jonas> no way, where does he say that?
11:50:45 <ais523> near the end of the document
11:50:52 <Jafet> > 1 - toRational (0.4 :: Float) / (4 % 10) :: CReal
11:50:53 <lambdabot> Couldn't match type ‘GHC.Real.Ratio GHC.Integer.Type.Integer’
11:50:53 <lambdabot> with ‘Data.Number.CReal.CReal’
11:50:53 <lambdabot> Expected type: Data.Number.CReal.CReal
11:50:53 <lambdabot> Actual type: GHC.Real.RationalCouldn't match expected type ‘Data.Number.CR...
11:50:53 <lambdabot> with actual type ‘GHC.Real.Ratio a0’
11:50:56 <b_jonas> near the end of what?
11:51:03 <Jafet> > fromRational $ 1 - toRational (0.4 :: Float) / (4 % 10) :: CReal
11:51:05 <lambdabot> -0.00000001490116119384765625
11:51:37 <fizzie> > toRational 0.399993896484375
11:51:38 <lambdabot> 13107 % 32768
11:52:03 <ais523> b_jonas: his report on the bugfixes
11:52:10 <ais523> http://tug.org/TUGboat/tb35-1/tb109knut.pdf
11:53:22 <b_jonas> ais523: 2014? but http://www-cs-faculty.stanford.edu/~uno/abcde.html says the TeX updates "happened most recently in 1992, 1993, 1995, 1998, 2002, 2007, and 2013; following this pattern, I intend to check on purported bugs again in the years 2020, 2028, 2037, etc."
11:53:34 <b_jonas> how is that possible?
11:53:40 <ais523> b_jonas: 31 December 2013 is when he started working on it
11:53:44 <ais523> he didn't get it finished in a day
11:53:49 <b_jonas> ah
11:56:27 <ais523> <Knuth, on 14 Mar 1978> (Some day we will have personal computers and will live more normally.)
11:58:30 <Jafet> Perfectly normal.
11:58:58 <ais523> the "abnormal living" was sleeping at odd times of day in order to use the timeshare system when it was less loaded
11:59:19 <ais523> I live nowadays, and have a personal computer, but still sleep at odd times of day
11:59:49 <Jafet> Ah. I thought they had multiuser terminals already in 1978.
12:00:08 <ais523> I think probably they did, but everything was running on one central server
12:00:31 <fizzie> ais523: In the document, he also mentions "my home computer (a 3.6 GHz Xeon with 10 MB cache)" in the part where the build time of the TeXbook is discussed.
12:00:41 <ais523> fizzie: hmm, right
12:00:46 <fizzie> ais523: I'm not sure how this works with the three million words of memory.
12:00:58 <ais523> it might have been running on a VM or something?
12:01:15 <Jafet> Maybe his code runs completely in cache.
12:01:22 <fizzie> Possibly. And of course "home computer" and "home system" might refer to different things.
12:01:27 <Jafet> I would not be that surprised
12:02:00 <ais523> also, I was vaguely surprised that TeX gets an extra digit on the version number, rather than increasing the last digit by 1 or adding another digit if that would exceed pi
12:02:17 <ais523> like, I was expecting 3.1, 3.11, 3.12, 3.13, 3.14, 3.141, 3.1411, 3.1412, etc.
12:02:21 <ais523> that also converges to pi, just more slowly
12:02:39 <Jafet> "the systems [TeX and Metafont] have been converging to an error-free state."
12:03:16 <fizzie> And if you get right down to it, "it did take quite awhile to overflow 3 million words of memory on my home system" doesn't necessarily imply the system has only 3 million words, just that three million words is some limit that can be overflowed.
12:03:31 <ais523> indeed
12:03:55 <ais523> actually, the other thing that surprised me is that there's a logic bomb in the manual to prevent people compiling it
12:03:58 <Melvar> … π and ℯ
12:04:25 <ais523> I guess because it's TeX not LaTeX, that makes sense
12:04:45 <ais523> (in LaTeX, some people might find it easier to read in a different document class, so allowing them to recompile rather than using a precompiled version would be reasonable)
12:06:41 <Jafet> Maybe they don't want people to compile it.
12:06:47 <Jafet> "The source file texbook.tex for The TeXbook begins with the following lines: [...] % The file is distributed only for people to see its examples of TeX input, not for use in the preparation of books like The TeXbook."
12:07:20 <ais523> he didn't want people to make other books based on the formatting in that book
12:07:38 <ais523> that's different from compiling the book itself from source, though
12:07:39 <ais523> hmm
12:08:14 <Melvar> I would like to note that the English Wikipedia has articles named “Pi” and “e (mathematical constant)”, while the corresponding German ones are named “Kreiszahl” and “Eulersche Zahl”.
12:08:26 <Jafet> Well, it is also a book you can buy from Addison-Wesley.
12:09:12 <Jafet> Kreiszahl sounds like an article about number rings
12:09:35 <fizzie> "Pii (vakio)", "Neperin luku".
12:10:00 <Melvar> Jafet: It is, however, about the Circle-Number.
12:11:03 <fizzie> I like how Finnish name for e is "Neperin luku" lit. "Neper's number", named after John Napier except with the last name misspelled.
12:11:27 <ais523> and the first name absent?
12:11:38 <fizzie> Yes.
12:11:54 <fizzie> But that's not Finnish-specific; "Napier's constant" seems to be an English name.
12:13:15 <fizzie> (Also arbitrary: it's apparently either "Euler's number" or "Napier's constant", but not "Napier's number" like in Finnish.)
12:14:09 <Melvar> Jafet: “number ring” redirects to “Ring of integers” which corresponds to “Ganzheitsring”.
12:15:36 <fizzie> "John Napier of Merchiston (/ˈneɪpɪər/; 1550 – 4 April 1617) – also signed as Neper, Nepair – named Marvellous Merchiston, was a Scottish landowner known as a mathematician, physicist, and astronomer." Guess it's not *wrong* per se, just an alternative spelling.
12:16:05 <Jafet> https://el.wikipedia.org/wiki/Αριθμός_π uses latin letters for all terms except π (and an english-labelled diagram).
12:19:41 <Jafet> Ganzheitsring sounds like an article about middle-earth
12:20:28 -!- Sorella_ has joined.
12:21:25 -!- augur has quit (Quit: Leaving...).
12:22:09 -!- yorick has joined.
12:24:09 <b_jonas> I see
12:24:51 <Melvar> Jafet: …
12:32:23 -!- callforjudgement has joined.
12:36:24 -!- ais523 has quit (Ping timeout: 240 seconds).
12:36:29 -!- callforjudgement has changed nick to ais523.
12:45:39 -!- Sorella_ has changed nick to Sorella.
12:53:36 -!- idris-ircslave has quit (Quit: Terminated).
12:53:52 -!- idris-ircslave has joined.
12:57:21 -!- idris-ircslave has quit (Client Quit).
12:57:39 -!- idris-ircslave has joined.
13:04:23 -!- idris-ircslave has quit (Quit: Terminated).
13:04:41 -!- idris-ircslave has joined.
13:09:15 <Taneb> ^style europarl
13:09:15 <fungot> Selected style: europarl (European Parliament speeches during approx. 1996-2006)
13:09:25 <Taneb> fungot, there's a big debate happening tonight!"
13:09:26 <fungot> Taneb: madam president, on a point of order and by way of preparation in the committee on industry, external trade, research and energy is an extensive and complicated directive which will affect every single citizen, thereby violating the principle of equal treatment which puts the present commission, because the commission adopted a communication entitled: ' eeurope 2005: an information society is characterized by telecommuni
13:21:06 <fizzie> fungot: What do you think about the EU sanctions on Russia?
13:21:07 <fungot> fizzie: mr president, i believe that we must accept and address. one criticism that has been conducted. issues like this serve to underpin the peace process.
13:21:42 <fizzie> Typical politician, always ready with a seemingly coherent reply that still doesn't really say anything.
13:22:09 <Jafet> Hmm, what exactly are the sanctions not sanctioning?
13:24:18 <fizzie> They include at least some asset freezes. Could you summarize them for us, mr fungot?
13:24:18 <fungot> fizzie: mr poettering, the law. it is very important because it will determine our final vote.
13:24:29 <fizzie> Ah, I see.
13:25:25 <Jafet> I agree mr fungot, it's a good idea to sanction the law
13:25:26 <fungot> Jafet: i listened yesterday to mr schulz, mr poettering, chairman of the committee on foreign affairs, human rights, and the unions too, are sorry that we have begun to change little by little. mr kaklamanis, you ask yourself, can they contemplate the future with confidence.
13:28:04 -!- nycs has changed nick to `^_^v.
13:28:33 <fizzie> Today I learned (from a different channel) that strace gets it wrong if a 64-bit-code thread on Linux/x86-64 accesses the 32-bit syscall emulation layer via int 0x80.
13:28:36 <fizzie> http://sprunge.us/TdRe
13:30:19 <fizzie> It's a (nonimportant) write through the "syscall" interface (syscall number 1 in that table) followed by an exit(42) through the int 0x80 interface (syscall number 1 in *that* table), and strace thinks the latter is a write, too. (And looks for arguments where they would be for the "syscall" stuff.)
13:32:34 -!- tertu3 has joined.
13:58:18 -!- franklin8 has joined.
13:58:23 -!- franklin8 has quit.
14:01:51 <int-e> Taneb: at the momemnt the prefixes are global and there is no infrastructure for per channel settings at all.
14:03:58 <slereah> ahah, in some neural network book
14:04:07 <slereah> "Now forget everything about training sets and learning"
14:04:08 <slereah> DONE
14:04:10 <slereah> :D
14:04:17 <slereah> I don't know what I'm doing
14:09:14 <Jafet> Now learn it again in a different order while reading the text upside down
14:13:22 <Jafet> "Each month, the European Parliament moves back and forth to meet the EU obligation to hold meetings also in France."
14:16:10 -!- Patashu_ has quit (Ping timeout: 252 seconds).
14:17:15 -!- ais523 has quit (Write error: Connection reset by peer).
14:17:32 -!- ais523 has joined.
14:27:18 -!- Sprocklem has joined.
14:29:58 <Jafet> Apparently, people are using the Enron emails as a corpus http://www.cs.cmu.edu/~enron
14:37:52 -!- ais523 has quit.
14:42:26 -!- tertu3 has quit (Ping timeout: 252 seconds).
14:45:29 <fowl> ^style
14:45:29 <fungot> Available: agora alice c64 ct darwin discworld enron europarl* ff7 fisher fungot homestuck ic irc iwcs jargon lovecraft nethack oots pa qwantz sms speeches ss wp youtube
14:45:38 <fowl> ^style lovecraft
14:45:39 <fungot> Selected style: lovecraft (H. P. Lovecraft's writings)
14:45:43 <fowl> fungot, speak
14:45:44 <fungot> fowl: then they had sealed the entrances and left them without courage, words, or know why you shouldn't. maybe you ought to hear, though, had gilman been there; and when told of the frantic magician. his fright, turning to leave his post at the rope, as the gold of the sky ever since there was any village to watch his taciturn dwelling from the plain to the north in grotesque panic.
14:48:32 <Jafet> ^style enron
14:48:32 <fungot> Selected style: enron (subset of the Enron email dataset)
14:48:52 <Jafet> It's time for a bit of professionalism around here, fungot.
14:48:53 <fungot> Jafet: questions about the market let alone an inter-company transfer the money." guarantee of the corporate parent of the work of the thought of the last few years that
14:56:05 * int-e wonders what fungot's point is.
14:56:06 <fungot> int-e: in a. thus the next to the above the offer of a the best. sent: tuesday, may 1, you will not have the right to do." california is facing the process that will, and get you to resign.
14:56:46 -!- tswett has joined.
14:56:47 -!- tswett has quit (Changing host).
14:56:47 -!- tswett has joined.
15:01:46 -!- drdanmaku has joined.
15:05:42 -!- zzo38 has quit (Remote host closed the connection).
15:10:34 -!- edwardk has quit (Quit: Computer has gone to sleep.).
15:12:56 -!- tertu3 has joined.
15:27:00 -!- MindlessDrone has quit (Quit: MindlessDrone).
15:33:28 -!- mhi^ has joined.
15:58:42 -!- tertu3 has quit (Ping timeout: 276 seconds).
16:01:55 -!- slereah has quit (Quit: Leaving).
16:25:33 <Melvar> I have added to idris-ircslave obedience to configuration (per-channel) of line limit. What should the limit be in this channel?
16:26:29 <int-e> ask fungot?
16:26:29 <fungot> int-e: have not moved. on the financial officer of the financial, trade and market renewable in 2001 that will complete. jones;wes colwell;mark haedicke;vince kaminski;phillip allen;tim belden;chris calger;joe deffner;tim detmering;dave duran;joe kishkell;rob milnthorp;brian redmond;gary hickerson;george mcclellan;sally beck;brian stanley;keith dodson; cc, 8, and main lines " a" and " that advance post of our the eel ( primaril
16:27:39 -!- edwardk has joined.
16:28:05 <int-e> no, fungot, 2001 is too much for IRC
16:28:05 <fungot> int-e: the our files that we
16:31:18 <Taneb> Hello
16:38:29 <Melvar> The default is 5.
16:41:07 <mroman> hm.
16:41:15 <mroman> How do I detect x^n?
16:41:16 -!- FreeFull has joined.
16:41:34 <mroman> Given some array I want to know if it's x^n
16:41:39 <mroman> like
16:41:50 <mroman> 0,1,8,27,64,125
16:42:54 <mroman> ah well
16:42:59 <mroman> log probably
16:45:32 <mroman> ah
16:45:33 <mroman> wow
16:45:33 <mroman> ok
16:45:43 <mroman> 3.0000000000000004
16:45:47 <mroman> no wonder this is not working
16:47:23 -!- password2 has joined.
16:47:46 -!- password2 has quit (Max SendQ exceeded).
16:48:13 -!- Sprocklem has quit (Quit: Battery died).
16:48:22 -!- password2 has joined.
16:48:53 -!- password2 has quit (Max SendQ exceeded).
16:49:29 -!- password2 has joined.
16:49:36 <Jafet> > map (^3) [0..]
16:49:38 <lambdabot> [0,1,8,27,64,125,216,343,512,729,1000,1331,1728,2197,2744,3375,4096,4913,583...
16:49:41 <Jafet> > zipWith (-) <*> tail $ map (^3) [0..]
16:49:42 <lambdabot> [-1,-7,-19,-37,-61,-91,-127,-169,-217,-271,-331,-397,-469,-547,-631,-721,-81...
16:49:51 <Jafet> > zipWith subtract <*> tail $ map (^3) [0..]
16:49:52 <lambdabot> [1,7,19,37,61,91,127,169,217,271,331,397,469,547,631,721,817,919,1027,1141,1...
16:49:59 -!- password2 has quit (Max SendQ exceeded).
16:50:00 <Jafet> > zipWith subtract <*> tail $ zipWith subtract <*> tail $ map (^3) [0..]
16:50:01 <lambdabot> [6,12,18,24,30,36,42,48,54,60,66,72,78,84,90,96,102,108,114,120,126,132,138,...
16:50:06 <Jafet> > zipWith subtract <*> tail $ zipWith subtract <*> tail $ zipWith subtract <*> tail $ map (^3) [0..]
16:50:07 <lambdabot> [6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6...
16:50:14 <mroman> Yeah.
16:50:28 <mroman> You can derive until it only contains zeros
16:50:36 <mroman> that's essentially what I'm doing
16:50:52 <mroman> A derive, make some checks, derive again, make the checks, derive again, make the checks
16:51:43 <mroman> And currently it checks for a.) all zeroes, b.) x^n, c.) n^x, d.) n*x
16:51:58 <Jafet> @oeis 0,1,8,27
16:52:00 <lambdabot> The cubes: a(n) = n^3.[0,1,8,27,64,125,216,343,512,729,1000,1331,1728,2197,2...
16:52:48 <Taneb> @oeis 1,7,19,54
16:52:48 <lambdabot> Sequence not found.
16:54:56 <mroman> @oeis 9,1,7,7,0,3,1,5
16:54:57 <lambdabot> Sequence not found.
16:55:01 <FreeFull> @oeis 1,0,-1,0,1,0,-1,0
16:55:13 <lambdabot> Numerator of Bernoulli number B_n.[1,1,1,0,1,0,1,0,1,0,5,0,691,0,7,0,3617,0,...
16:55:31 <mroman> I'm abusing this stuff to judge how random some numbers are
16:56:11 <FreeFull> mroman: try it for n!
16:56:15 <mroman> (9,1,7,7,0,3,1,5 has random level 5, whatever the hell that's supposed to mean)
16:57:48 <FreeFull> :t scanl'
16:57:49 <lambdabot> Not in scope: ‘scanl'’
16:57:49 <lambdabot> Perhaps you meant one of these:
16:57:49 <lambdabot> ‘scanl’ (imported from Data.List),
16:58:00 <FreeFull> A strict scanl would be nice
16:58:03 -!- oerjan has joined.
16:58:07 <myname> what the hell @ bernoulli number to that input
16:58:09 <FreeFull> > scanl (*) 1 [1..]
16:58:11 <lambdabot> [1,1,2,6,24,120,720,5040,40320,362880,3628800,39916800,479001600,6227020800,...
16:58:14 <Jafet> > [ 321^n `rem` 10^6 `div` 10^5 | n <- [0..] ]
16:58:15 <lambdabot> [0,0,1,0,4,7,4,8,2,1,7,5,3,4,0,1,1,5,5,0,9,3,5,6,2,1,0,3,6,6,6,3,8,6,1,4,2,1...
16:58:44 <oerjan> int-e: just thought you should know about http://stackoverflow.com/questions/23337798/cant-build-lambdabot-with-ghc-7-8-haskell-src-exts
16:58:49 -!- oerjan has quit (Client Quit).
16:59:42 <Jafet> scanl doesn't need to be strict.
17:00:11 <FreeFull> I have found a need for a strict scanl before
17:00:20 <FreeFull> I don't remember what for anymore
17:01:25 <Jafet> > show =<< [ 321^(n+100) `rem` 10^6 `div` 10^5 | n <- [0..] ]
17:01:27 <mroman> Faculty has a random factor of 0.8
17:01:27 <lambdabot> "530514009063721311532538764312390304954886869678617557428179164663146490244...
17:02:30 <FreeFull> > let generate08 = 0.8 in generate08
17:02:31 <lambdabot> 0.8
17:03:26 <int-e> call it factorial, or you'll confuse the native speakers :)
17:04:05 <FreeFull> How did I read factor as generator?
17:04:32 <mroman> Math.sqrt(n+1) is also pretty random apparentely
17:05:01 <FreeFull> > scanl (/2) 0.4 []
17:05:03 <lambdabot> Occurs check: cannot construct the infinite type: b ~ a0 -> b
17:05:03 <lambdabot> Expected type: b -> a0 -> b
17:05:03 <lambdabot> Actual type: b -> b
17:05:16 <FreeFull> Nope
17:05:28 <FreeFull> I think I need an unfold
17:05:51 <int-e> @tell oerjan must be a windows thingy; I've had no trouble with haskell-src-exts 1.15.0.1 under linux.
17:05:51 <lambdabot> Consider it noted.
17:08:47 -!- password2 has joined.
17:10:57 -!- password2 has quit (Max SendQ exceeded).
17:12:29 <Jafet> sqrt is not random at all. (You could take the digits of one particular square root, which have a good chance of being iid)
17:24:11 <mroman> I know that it's not random
17:24:15 <mroman> But my algorithm doesn't ;)
17:25:13 <mroman> It detects Math.sqrt
17:25:17 <mroman> but not Math.sqrt(n+1)
17:28:03 -!- ^v has joined.
17:28:09 -!- password2 has joined.
17:29:53 -!- ^v has quit (Read error: Connection reset by peer).
17:32:05 -!- Barby1 has joined.
17:33:03 <Barby1> Holaaaaaaa<3
17:33:48 -!- nooodl has joined.
17:34:29 <Barby1> :-D
17:35:57 -!- Phantom_Hoover has joined.
17:36:29 <Taneb> Him Barby1
17:36:44 <Barby1> hello!!
17:37:05 <Taneb> *hi
17:37:32 <Barby1> como estas?
17:38:08 <elliott> Barby1: please stop coming to this channel.
17:38:20 <elliott> Barby1: por favor deje de visitar este canal.
17:38:34 <Barby1> porque?
17:39:00 <elliott> does anyone know spanish better than google here.
17:39:41 -!- Barby1 has quit (Quit: Leaving.).
17:39:52 <Taneb> elliott, does Barby1 have a history?
17:40:05 <elliott> Taneb: *!~canaima@* has a history
17:40:14 <Taneb> OK
17:40:18 <elliott> a history of joining and only speaking spanish for a few minutes before leaving
17:40:34 <elliott> like, 90% of the time the spanish welcome has been used, it's to them
17:40:38 <elliott> they just change nicks all the time
17:40:53 <elliott> maybe I should just set a ban? but they haven't done anything /wrong/, I guess, it's just... a waste of time for both parties
17:41:10 <Taneb> I am not very good at nethack
17:41:19 <Taneb> That was a bit of a non-sequitur, I know
17:41:27 <Taneb> But I was going to say it before I spotted Barby1
17:42:18 <Phantom_Hoover> try ftl, you'll be just as not good at it
17:42:40 <Taneb> Does ftl cost money
17:43:22 <Phantom_Hoover> yes
17:44:02 <Taneb> I am afraid I do not have money right now
17:50:13 <mroman> Any ideas how I can detect stuff like sqrt(n+1)?
17:50:24 <mroman> sqrt(n+x) in generall.
17:51:33 <mroman> funny
17:51:38 <mroman> It detects (n+x)^p
17:51:43 <mroman> but not (n+x)^(1/p)
17:52:54 -!- tertu3 has joined.
17:53:45 -!- password2 has quit (Ping timeout: 276 seconds).
17:54:45 -!- MoALTz has joined.
17:59:45 -!- MoALTz has quit (Quit: bbl).
18:04:44 -!- MoALTz has joined.
18:08:00 <Phantom_Hoover> http://www.reddit.com/r/Bitcoin/comments/1zpyba/100_secure_paper_wallet_cold_storage_tutorial_out/
18:08:04 <Phantom_Hoover> man this is brilliant
18:08:28 <Phantom_Hoover> "Use the linux command "echo -n my_dice_rolls | sha256sum" to generate a private key" your private key will be in your command history, for easy access!!
18:08:57 <myname> lol
18:09:28 <Phantom_Hoover> and all you need is 33 dice lying around!
18:13:27 <Bike> "Note: All source code produced by this campaign will always be publicly available under open source license. However building a WormSim from source is a complex operation that requires advanced engineering skills." advanced engineering skills to make
18:14:01 <myname> what is a wormsim
18:15:10 <Bike> a c elegans simulation.
18:16:48 <myname> what
18:17:19 <Bike> c elegans. a roundworm. nematode.
18:17:36 <Jafet> Advanced software engineering skills
18:18:47 <fizzie> Advanced social engineering skills to worm (npi) the building instructions out of someone.
18:19:27 <Bike> they provide "Eclipse Juno distributions for the main OSs"
18:28:39 -!- Sprocklem has joined.
18:42:45 <Taneb> In this roguelike I am helping with, the strongest starting armour is a hawai'in t-shirt
18:42:49 <Taneb> Followed by a law suit
18:43:07 <Sgeo> If I want to use rsync to transfer files between computers, is there an obvious choice for which is server and which is client?
18:43:26 <Sgeo> I'm just wondering if it's primarily a download thing or does two-way sync
18:43:39 <b_jonas> Sgeo: I use it both ways
18:44:22 <b_jonas> there's no obvious choice, but sometimes stuff could restrict you if you don't bother to set up an ssh or rsync server on both sides
18:45:01 <b_jonas> heck, I even use rsync for a few local copies
18:46:59 <Sgeo> How sick should one be before considering taking a sick day?
18:47:09 <Sgeo> I took a sick day today, and do feel sick, but not -very- sick
18:47:33 <b_jonas> Sgeo: depends on how many sick days you have
18:48:13 <mroman> Sgeo: If you can walk you're not sick enough
18:48:21 <b_jonas> think of how many you will want to keep for the rest of the year
18:49:19 -!- tertu3 has quit (Ping timeout: 252 seconds).
18:49:32 <Sgeo> I think it accumulates during the year, and the site that shows me how much time I have doesn't show how much will accumulate
18:50:08 <b_jonas> Sgeo: definitely get to learn early and often about how many days of leave you have
18:50:41 <b_jonas> if your calculation differs from that of your workplace by more than half a day, then talk to them about who doesn't understand the rules or who has made a mistake
18:51:07 <b_jonas> do that early even if you have enough days of paid leave for now, so that later it can't cause a problem
18:51:11 <Sgeo> I know I have 3 personal days a year that don't accumulate, and have been told by a coworker and my dad to generally use that first
18:51:41 <Sgeo> But I think that's separate from sick time, I don't think using personal days for sick time makes sense, maybe for vacation time
18:51:57 <b_jonas> the rules differ by country and occupation a lot of course, and depend on stuff
18:52:10 <b_jonas> and can be complicated
18:53:36 -!- tertu3 has joined.
18:58:00 -!- Phantom_Hoover has quit (Ping timeout: 240 seconds).
19:11:13 -!- Phantom_Hoover has joined.
19:22:24 -!- hexagon has quit (Quit: Here we are, going far to save all that we love - If we give all we've got, we will make it through - Here we are, like a star shining bright on your world - Today, make evil go away!).
19:22:35 -!- ^v has joined.
19:22:47 -!- hexagon has joined.
19:23:45 -!- hexagon has quit (Client Quit).
19:24:15 -!- hexagon has joined.
19:25:43 -!- nucular has quit (Quit: Switching to phone in a few minutes).
19:39:09 -!- ^v has quit (Quit: http://i.imgur.com/DrFFzea.png).
19:46:48 -!- Phantom_Hoover has quit (Read error: Connection reset by peer).
19:46:51 -!- conehead has joined.
19:47:05 -!- Phantom_Hoover has joined.
20:04:29 -!- tertu3 has quit (Ping timeout: 252 seconds).
20:04:40 -!- nisstyre has joined.
20:10:21 <Taneb> Aaaaah what can a CS society spend money on
20:11:48 <fizzie> A set of 3D-printed miniature figurines of famous computer science researchers.
20:12:13 <Slereah_> A quantum computer to play quantum DOOM
20:13:13 <kmc> a large collection of synthetic hallucinogens
20:15:27 <Bike> and booze
20:15:41 <Slereah_> Just a whole bunch of candy
20:15:51 <Slereah_> The world's most expensive dildo
20:15:55 <Slereah_> A piece of the true cross
20:16:10 <Slereah_> White slaves
20:16:24 <Slereah_> Ten kilograms of antimatter
20:16:36 <Slereah_> A cat skeleton soaked in kerosene
20:16:59 <Slereah_> All the tea in China
20:17:06 <Slereah_> Like fifty million bees
20:17:08 <Slereah_> a goat
20:17:10 <Slereah_> two goats
20:17:58 <Slereah_> The corpse of John the Baptist
20:18:07 <Slereah_> The pyramid of Giza
20:18:12 <Slereah_> a castle in Spain
20:18:16 <Slereah_> the sky
20:18:34 <fizzie> fungot: Suggest more.
20:18:34 <fungot> fizzie: please let me know the best is yet.?
20:19:02 <fizzie> fungot: Well, it's hard to say which one of those was best. Maybe you could suggest something clearly better than any so far.
20:19:02 <fungot> fizzie: we don't i mail this other project, i thus expect the following and let
20:20:14 -!- MoALTz has quit (Quit: bbl).
20:21:07 <Slereah_> An Indian graveyard
20:21:11 <nooodl> i wonder which of the things Slereah_ mentioned are purchasable, and then, which is most expensive
20:21:17 <nooodl> i guess you can buy all the tea in china?? "it's hard"
20:21:32 <Slereah_> A castle in spain is probably the most realistic and expensive
20:22:02 <Slereah_> unless you buy really a lot of candy
20:22:05 <Slereah_> Or white slaves
20:24:19 <elliott> nooodl: the world's most expensive dildo is pretty buyable, presumably
20:24:22 <elliott> also goats
20:24:27 <elliott> also candy
20:25:36 <Slereah_> Goats are not very expensive though
20:25:59 <Slereah_> Also I assume that the world's most expensive dildo is in a private collection
20:26:03 <Slereah_> it might not be for sale
20:26:46 <Taneb> In other news, we have a program which compiles fine on Ubuntu but very not fine on Arch
20:27:07 <nortti> how does one compile not fine?
20:29:28 <Phantom_Hoover> <elliott> nooodl: the world's most expensive dildo is pretty buyable, presumably
20:29:40 <Phantom_Hoover> it's probably made out of solid gold or something though
20:29:52 <Phantom_Hoover> not very affordable for a CS society
20:29:53 <Slereah_> My guess would be
20:29:53 <Taneb> nortti, NCurses fails
20:29:55 <Slereah_> Art dildo
20:30:02 <Slereah_> Made by a fancy artist
20:30:09 <Taneb> Phantom_Hoover, not particularly relevant either
20:30:12 <nortti> Taneb: compile or link?
20:30:14 <fizzie> Phantom_Hoover: Or perhaps it's just valuable because of its illustrious history.
20:30:15 <Slereah_> Phantom_Hoover : what about the sky
20:30:17 <elliott> Phantom_Hoover: cheaper than all the tea in china
20:30:51 <Taneb> nortti, runtime, in the clang memory sanitizer apparently
20:30:53 -!- barrucadu has joined.
20:30:56 <Taneb> But only there
20:31:04 <nortti> O_o
20:31:07 <Phantom_Hoover> Slereah_, well that's governed by national airspace treaties
20:31:12 <Taneb> Ask barrucadu
20:31:27 <Slereah_> Can you buy it
20:31:28 <barrucadu> We're trying to debug a memory error, and so used clang's memory sanitizer, but that throws out errors on Arch but not on Ubuntu - in libncursesw
20:31:32 <barrucadu> For the same version of ncurses
20:32:02 <barrucadu> This is the error: http://pastebin.com/SBCPRqqU
20:32:22 <barrucadu> Line 29 of main.c is this: https://github.com/HackSoc/LudumDare29/blob/master/main.c#L29
20:32:49 <Slereah_> Oh I see the problem
20:32:50 <Slereah_> #include <curses.h>
20:32:53 <Slereah_> IT IS CURSED
20:33:17 <barrucadu> Don't worry, we have (fallen) angels elsewhere in the code, they'll take care of any curses
20:33:31 <fizzie> See, the problem is that "binary packages encourage inconsistency and incompatibility, whearas source encourages unified development frameworks and integration."
20:33:47 <fizzie> (I just felt like quoting funroll-loops.info.)
20:34:03 <nortti> funroll
20:34:11 <nortti> very fun
20:34:47 <shachaf> @quote funsafe
20:34:48 <lambdabot> shachaf says: You can't spell "-funsafe" without "fun" and "safe".
20:35:49 <fizzie> That joke works p. well for the GCC option -funsafe-math-optimizations.
20:36:16 <fizzie> Also you can't spell -malign-double without maligning some perfectly respectable doubles.
20:36:56 <Slereah_> I keep a little script on my werk computer
20:37:00 <Slereah_> I found it on the internet
20:37:03 <Slereah_> it is like
20:37:07 <Slereah_> All the compiler warning
20:37:17 <Slereah_> I called it MAXIMUM OVERWARNING
20:37:29 <Slereah_> That way I have things to do in case I don't know what to do
20:37:42 <fizzie> Clang has a single flag for that, fortunately.
20:37:49 <fizzie> (It's called -Weverything.)
20:38:19 <Slereah_> Strangely enough -Wall doesn't do that
20:38:43 <Taneb> -Wall-no-really-when-I-say-all-I-mean-all
20:38:43 <FreeFull> > take 10 primes
20:38:44 <lambdabot> Not in scope: ‘primes’
20:38:47 <FreeFull> Damn
20:38:59 <Slereah_> For some reason gcc doesn't seem to like string literals?
20:39:01 <FreeFull> I don't think lambdabot has the primes package
20:39:02 <fizzie> -Wall doesn't, but -Weverything does.
20:39:03 <Slereah_> which I find weird
20:39:11 <Slereah_> They're pretty easy to compile, no?
20:39:19 <FreeFull> @let import Data.Numbers.Primes
20:39:19 <lambdabot> .L.hs:95:1:
20:39:20 <lambdabot> Failed to load interface for ‘Data.Numbers.Primes’
20:39:20 <lambdabot> Perhaps you meant Data.Number.Fixed (from numbers-3000.2.0.1)
20:39:20 <lambdabot> Use -v to see a list of the files searched for.
20:39:39 -!- nisstyre has quit (Quit: WeeChat 0.4.3).
20:39:53 <elliott> it did when I had it, I think
20:40:01 <fizzie> shachaf: Also you can't spell -funsafe without "NSA" it is a conspiracy
20:40:28 <Slereah_> And you can't spell -funsafe without enuf
20:40:46 <Taneb> So any ideas re: memory error?
20:41:38 <fizzie> Taneb: Install the package that comes with debugging info for the ncursesw lib, so that you get something more edifying than just offsets in hex. Then fix what's probably a ncurses bug.
20:42:57 <fizzie> If it's really the same (ncurses) code on both, and only trips a "use-of-uninitialized-value" on one, it could be different environment variables or terminfo databases or <any other environmental thing>. Though you never know about "same version" either.
20:43:20 <fizzie> I wonder if there are people who always compile with -ffinite-math-only out of some sort of philosophical principles.
20:43:45 <Slereah_> why would you want that
20:43:58 <fizzie> You know, people who don't believe in infinities.
20:44:00 <Slereah_> I am currently doing some log things so I really need my infinities
20:48:45 <fizzie> Taneb: FWIW, I get the same -fsanitize=memory error for libncursesw on my system, for a program that does only a initscr();.
20:49:15 <Slereah_> Are there C decompilers?
20:49:54 <Slereah_> Apparently so, yes!
20:50:10 <lexande> Phantom_Hoover, Slereah_: a friend who is an expert on such matters is aware of a $3500 gold-plated dildo and a $15k gold-plated vibrator, which may be the most expensive mass-produced examples, though yes presumably there are much more expensive ones sold as artwork at auction
20:50:12 <barrucadu> fizzie: Looking into it now
20:50:37 <barrucadu> Also, turns out the reason the error isn't coming up on Ubuntu is becasue the version of clang on the uni computers is too old to support the memory sanitizer
20:50:41 <barrucadu> *because
20:52:57 -!- Bike has quit (Read error: Connection reset by peer).
20:53:15 -!- Bike has joined.
20:59:14 <Taneb> ...we are now debugging ncurses
21:01:35 -!- Patashu has joined.
21:01:55 -!- Patashu_ has joined.
21:01:55 -!- Patashu has quit (Disconnected by services).
21:08:18 -!- augur has joined.
21:16:38 -!- Sprocklem has quit (Ping timeout: 240 seconds).
21:18:56 <fizzie> FWIW, my error comes with the call stack initscr -> newterm -> _nc_setupterm -> _nc_setup_tinfo -> _nc_read_entry -> _nc_first_db, and it's a strlen on the string "/home/fis/.terminfo" (values[dbdHome]) that seems just fine in gdb; not sure why the memory-sanitizer warns about uninitialized stuff in it.
21:19:01 <fizzie> But perhaps your error is different.
21:19:41 <Taneb> Our code now appears to work
21:20:07 <fizzie> (If I read the messages right, it doesn't like the null terminator at the end.)
21:20:45 <Taneb> Oh no it isn't
21:21:17 <fizzie> Well, the (ncurses) code does look incorrect, so maybe that's just fair.
21:21:47 -!- edwardk has quit (Ping timeout: 265 seconds).
21:24:53 -!- Patashu_ has quit (Ping timeout: 264 seconds).
21:28:01 <fizzie> Or, hm. I don't know. It looks okay on further inspection. Very kludgy, though. But the string's still the result of a regular snprintf, that should be suitably null-terminated. And actually the offset it's complaining about is one past the '\0', and it's not like the strlen should be looking there.
21:28:34 <fizzie> Unless it's some sort of "GCC knows the size of the buffer so it can use an optimized four-bytes-at-a-time strlen" kind of deal.
21:29:12 <fizzie> That actually sounds quite reasonable. In which case the use of uninitialized data would be just a false positive from the MemorySanitizer.
21:29:22 <fizzie> I'm going to call it that since then I won't have to keep looking at it.
21:30:30 <Taneb> Our program works only when memory is being watched
21:30:44 <kmc> quantum computing
21:30:52 <fizzie> (Possibly even glibc can assume things about page borders etc. and use a multi-byte strlen.)
21:32:30 <kmc> fizzie: I've been playing with SSE 4.2 string instructions and I think a lot of the example code out there doesn't handle page boundaries correctly
21:32:36 <kmc> I talk about that at the end of https://github.com/kmcallister/html5/blob/sse/src/tokenizer/buffer_queue.rs#L182-L252
21:34:20 <Taneb> `unidecode !
21:34:20 <HackEgo> ​[U+0021 EXCLAMATION MARK]
21:34:51 <shachaf> kmc: should be called terrifying_microöptimizations hth
21:34:56 <kmc> shachaf: yeah
21:35:15 <shachaf> and it should use a combining character in half the places
21:35:21 <shachaf> i hear rust will handle that correctly
21:35:26 <kmc> yeah
21:35:35 <kmc> i don't know if it normalizes --cfg arguments on the command line, though
21:36:15 <shachaf> --cƒg
21:36:33 <fizzie> kmc: I guess (I haven't looked if this is a thing that is done) one could just handle the *initial* not-16-byte-aligned part of a string specially, and then you can do movdqa and read in full 16-byte blocks and never worry about crossing a page boundary even if the read does go beyond the end of the string.
21:36:50 <kmc> imo it should also accept — instead of --
21:36:59 <kmc> fizzie: yeah
21:37:18 <kmc> otoh, that will often be slower
21:37:32 <FireFly> Is that an en or em dash?
21:37:37 <kmc> i think it's an em
21:37:41 <kmc> but i'm using a fixed width font
21:37:47 <kmc> I guess that means an em and an en are the same!
21:37:48 <shachaf> the difference is still visible
21:37:49 <shachaf>
21:37:51 <shachaf>
21:37:53 <kmc> praise jesus
21:38:05 <kmc> `coins
21:38:06 <HackEgo> sortcoin slash-01coin versedcoin univelacoin squeklycoin tclycoin brainftingcoin pooncoin loodbcoin purehancoin visiticoin 0x8coin excepiracoin returntzcoin owlcoin betticoin marcoin boskecoin mancecoin yellicoin
21:38:08 <FireFly> Oh, so it is
21:38:21 <kmc> if i'm gonna die for a word, my word is pooncoin
21:38:46 <shachaf> imo just map an extra page in at the end `j̀ust in casé´
21:38:56 <kmc> imo just handle SIGSEGV
21:40:49 <fizzie> kmc: glibc strlen seems to do more or less what I envisaged, except it also rounds *down* to do the unaligned part with a movdqa, cf. http://sprunge.us/hZDA
21:41:00 -!- boily has joined.
21:41:29 <fizzie> (I'm also going to go ahead and assume that's why the ncurses code fails with -fsanitize=memory, since it sounds plausible.)
21:43:17 <fizzie> (Er, "do the unaligned part with pcmpeqb to memory", but anyway.)
21:43:20 <Taneb> Yes, it goes beyond the end of the string
21:43:29 <fizzie> (I didn't really read the operands.)
21:44:27 <fizzie> Taneb: Well, that's perfectly legal, under some assumptions.
21:47:13 <kmc> if the president does it, it's not illegal
21:47:21 <shachaf> actually it should be µoptimizations
21:47:31 <kmc> shachaf: is that greek letter mu or mathematimu?
21:48:05 <shachaf> it's U+B5 MICRO SIGN
21:48:25 <Taneb> We've found the actual error we were looking for :)
21:48:28 <shachaf> which i get by pressing altgr+m
21:48:53 <fizzie> µoptimiꊔions.
21:49:00 <kmc> I do like it when code (usually kernel exploits) clears the WP bit in %cr0 so it can write to read-only pages
21:49:15 <kmc> much easier than walking the page tables
21:51:40 <boily> `? mathematimu
21:51:41 <HackEgo> mathematimu? ¯\(°​_o)/¯
21:52:22 <boily> `` echo 'A mathematimu is a quantum of mathematics. If you observe it, its codepoint can change.' >wisdom/mathematimu
21:52:24 <HackEgo> No output.
21:58:47 <boily> `ello fowl
21:58:48 <HackEgo> fellowl
21:59:01 <boily> fowl: you have been quoted. you exist in the PDF now.
21:59:33 <fizzie> (A fate worse than death.)
22:00:58 <fowl> eh its no worse than when i enlisted. now the govt has my fingerprints and dna on record :/
22:01:34 <kmc> which govt
22:01:39 <boily> which fingerprints?
22:01:43 <kmc> which dna
22:02:07 <fowl> us
22:02:09 <fowl> all of them
22:06:44 <Taneb> `quote fowl
22:06:45 <HackEgo> 1187) <fowl> one day we'll be able to put evil people inside mirrors and throw them into space like superman 2
22:06:56 <Taneb> `quote plastic spoon
22:06:57 <HackEgo> 400) <Taneb> Someone with that sort of grasp of logic shouldn't be allowed anything more computationally powerful than a plastic spoon
22:07:05 -!- oerjan has joined.
22:08:09 <oerjan> @messages-lost
22:08:09 <lambdabot> int-e said 5h 2m 17s ago: must be a windows thingy; I've had no trouble with haskell-src-exts 1.15.0.1 under linux.
22:13:18 <oerjan> <Jafet> Is fungot less volatile than btc <-- what isn't?
22:13:18 <fungot> oerjan: i moved to sp) regarding any of the street. 1717 main street, suite that enables the customers.
22:13:31 <kmc> fungotcoin
22:13:31 <fungot> kmc: you will get out the points of the information on the new more efficient.
22:14:26 <oerjan> @ask Jafet <Jafet> Is fungot less volatile than btc <-- what isn't?
22:14:26 <lambdabot> Consider it noted.
22:14:26 <fungot> oerjan: two years the big a book on the work of the business is mailing for individuals that dynegy to walk on the wharton. 1124.7) ( madison.
22:14:45 <boily> 1717 Berri is the address of Montréal's autocar terminus. fungot, what are you spying on from Montréal, eh?
22:14:46 <fungot> boily: new deal to keep the current and will find the time to get on the stack.
22:14:47 <oerjan> pesky @ask /whois race condition again
22:15:03 <boily> fungot: intelligence gathering, then.
22:15:03 <fungot> boily: could we schedule more to the issues on the new plants that not only survive, but not the one that the are two sections resulting in the lowest in three years that
22:15:07 -!- Phantom_Hoover has quit (Quit: Leaving).
22:16:16 <oerjan> boily: i hypothesize that there's an agency hidden at _every_ 1717 address hth
22:16:38 <boily> thausible.
22:17:17 <kmc> "autocar" means bus? confusing
22:17:23 <oerjan> or possibly a shared one, with clever portal doors
22:17:53 <oerjan> omnicar
22:18:47 <Jafet> @ask oerjan Crimea hth
22:18:47 <lambdabot> Consider it noted.
22:19:29 <boily> kmc: coach cars, I guess. intercity shuttle buses?
22:19:52 <boily> I like autocar myself. I find the word pleasing, like a zeppelin.
22:20:43 <fizzie> en:car = fi:auto.
22:20:48 <fizzie> A carcar.
22:21:04 <Bike> automatic car dispenser
22:21:28 <fizzie> http://upload.wikimedia.org/wikipedia/commons/d/d4/Smart_vending_machine_in_Shenyang.jpg
22:26:23 <nooodl> nl:auto too!
22:27:00 <oerjan> @messages-
22:27:00 <lambdabot> Jafet asked 8m 13s ago: Crimea hth
22:31:45 <kmc> fizzie: does it work?
22:33:06 <oerjan> @tell <ais523> each review of bugs takes one more year than the last <-- that's quadratic not geometric hth
22:33:07 <lambdabot> Consider it noted.
22:33:36 <elliott> oerjan: need a nick
22:33:44 <oerjan> argh
22:33:56 <oerjan> @tell ais523 <ais523> each review of bugs takes one more year than the last <-- that's quadratic not geometric hth
22:33:56 <lambdabot> Consider it noted.
22:41:56 <Taneb> oerjan, that was pointed out iirc
22:43:48 <oerjan> Taneb: another race condition :(
22:44:25 <oerjan> i need to switch to STM for my logreading.
22:49:59 -!- mhi^ has quit (Quit: Lost terminal).
22:54:25 <FreeFull> You should have used STM in the first place
22:54:33 <FreeFull> It's too nice to skip in multithreaded software
22:54:59 -!- edwardk has joined.
22:55:18 <boily> `ello edwardk
22:55:19 <HackEgo> Hello, edwardk !
22:55:46 -!- yorick has quit (Remote host closed the connection).
22:57:17 <edwardk> hya
22:58:42 <oerjan> wat
22:58:52 <oerjan> `ello wat
22:58:53 <HackEgo> watello
22:59:03 <oerjan> `cat bin/ello
22:59:03 <HackEgo> ​#!/usr/bin/env node \ // Generated by CoffeeScript 1.6.2 \ (function() { \ var consonant_then_o, ell_manglable, ends_with_consonant, ends_with_consonant_then_vowel, name, starts_with_o; \ \ name = process.argv[2]; \ \ if (!(name != null ? name.length : void 0)) { \ console.log('Usage: ello <name>'); \ process.exit(); \ } \ \
22:59:18 <oerjan> argh
22:59:19 <kmc> javascript? in my HackEgo?
23:00:07 <oerjan> `ello boily
23:00:08 <HackEgo> belloily
23:00:17 <oerjan> `ello edwardk
23:00:18 <HackEgo> Hello, edwardk !
23:00:21 <oerjan> `ello edwardk
23:00:23 <HackEgo> edwardkello
23:00:27 <boily> I think I'm now used to the classical “hoily.”
23:00:33 <Bike> `run cat $(which thanks)
23:00:34 <HackEgo> ​#!/usr/bin/perl \ $_ = (join " ", @ARGV) || `words`; s/^\s+|\s+$//g; print "Thanks, $_. "; if (/[aeiouyAEIOUY]/) { s/^[^aeiouyAEIOUY]*/Th/; } else { s/^./T/; } print "$_.";
23:00:46 <oerjan> hm it's sensitive to final space, but only sometimes...
23:03:54 -!- nooodl has quit (Remote host closed the connection).
23:04:15 -!- nooodl has joined.
23:07:34 <fizzie> kmc: I don't think so, in the sense that you could buy the car from it.
23:09:11 <kmc> sadness
23:09:19 <lexande> you could have a vending machine that sold car keys
23:09:25 <lexande> with the cars conveniently parked nearby
23:10:37 <oerjan> <FreeFull> A strict scanl would be nice <-- it's not that important for scanl unless you skip over elements in the list
23:10:38 <kmc> imagine the frustration after you put $15,000 in bills into the machine and then the keys don't quite fall off the corkscrew hook thing
23:11:03 <oerjan> because being strict is equivalent to evaluating the result elements in order
23:11:37 <FreeFull> oerjan: I had something that needed strict scanl but I don't remember why
23:11:37 <fizzie> There is also the http://spectrum.ieee.org/geek-life/tools-toys/vehicular-vending-machine which at least deals in cars.
23:11:44 <FreeFull> Something that would leak memory otherwise
23:14:04 -!- Sprocklem has joined.
23:17:57 <kmc> fizzie: nice
23:18:02 <kmc> zipcar is also like a car vending machine, in a sense
23:18:31 <oerjan> lenscar
23:19:17 <FreeFull> Use an ashtray as a lens, to magnify the car
23:23:03 <oerjan> <Taneb> Does ftl cost money <-- infinite money hth
23:24:11 <boily> what's the exchange rate in Canadian dollars?
23:25:20 <FreeFull> 1 USD = 1.10 CAD or something
23:26:10 <oerjan> ^google 1 USD in CAD
23:26:13 <oerjan> oops
23:26:17 <oerjan> @google 1 USD in CAD
23:26:17 <lambdabot> No Result Found.
23:26:17 <boily> so 1.1*infty.
23:26:36 <oerjan> how rude.
23:27:06 <boily> dammit. my stupid bot won't connect to freenode.
23:27:27 <oerjan> you were also going to duck it out?
23:27:54 <boily> me? what would make you believe that?
23:28:07 * boily mapoles his cuttlefish of a bot
23:28:23 <oerjan> a hunch.
23:28:30 -!- metasepia has joined.
23:28:31 <boily> ~duck 1 USD in CAD
23:28:32 <metasepia> --- No relevant information
23:28:37 <boily> IEAURGHGHGH!
23:28:38 <oerjan> also have you factored in that canada doesn't exist twh
23:28:59 <oerjan> `google 1 USD in CAD
23:29:00 <HackEgo> Failed to connect to socket 2. \ \ Looking up 127.0.0.1:3128 \ Making HTTP connection to 127.0.0.1:3128 \ Sending HTTP request. \ HTTP request sent; waiting for response. \ Alert!: Unexpected network read error; connection aborted. \ Can't Access `http://google.com/search?q=%31%20%55%53%44%20%69%6e%20%43%41%44' \ Alert!: Unable to access document.
23:29:21 <oerjan> fancy.
23:31:13 <boily> oh well. time to go drown my unexistant sorrows in some douteux stuff.
23:31:19 -!- boily has quit (Quit: DISQUIETING CHICKEN).
23:31:22 -!- metasepia has quit (Remote host closed the connection).
23:32:35 <oerjan> /wii Slereah_
23:32:38 <oerjan> oops
23:33:18 <oerjan> @ask Slereah_ <Slereah_> The corpse of John the Baptist <-- child or adult?
23:33:19 <lambdabot> Consider it noted.
23:53:44 <oerjan> @ask boily <boily> `` echo 'A mathematimu is a quantum of mathematics. If you observe it, its codepoint can change.' >wisdom/mathematimu <-- WHY DID I EVEN BOTHER TO MAKE `learn UNDERSTAND ARTICLES
23:53:44 <lambdabot> Consider it noted.
23:54:33 <Taneb> `? articles
23:54:34 <HackEgo> articles? ¯\(°​_o)/¯
23:54:43 -!- nooodl has quit (Quit: Leaving).
23:55:09 <Taneb> `run echo 'An article is something that `learn can understand.' > wisdom/article
23:55:10 <HackEgo> No output.
23:55:23 <shachaf> oerjan: for your own use?
23:55:24 <shachaf> hth
23:55:26 -!- ChanServ has set channel mode: +o oerjan.
23:55:32 -!- oerjan has kicked Taneb AAAAAAAAA.
23:55:37 -!- oerjan has set channel mode: -o oerjan.
23:56:06 <shachaf> Taneb++
23:56:19 -!- Taneb has joined.
23:56:42 <oerjan> ಠ_ಠ
23:56:42 <myndzi> ¯|¯⌠
23:56:43 <myndzi> /^\|
←2014-04-27 2014-04-28 2014-04-29→ ↑2014 ↑all