00:01:56 -!- conehead has joined.
00:11:59 <oerjan> <zzo38> If you have multiple copies of a picture or audio with different watermarks, can you then detect them and tamper with them? <-- i think it should be possible to design watermarks so that even after your tampering, they can still detect which originally watermarked copies you used
00:12:20 -!- Phantom_Hoover has joined.
00:12:54 <oerjan> so that if you get leaked copies from two sources, they can tell what the two leaks were
00:13:59 <zzo38> It may allow you to at least detect watermarks, I suppose, regardless if they can still detect it or not.
00:15:38 <HackEgo> [wiki] [[5command]] http://esolangs.org/w/index.php?diff=39697&oldid=39696 * Icepy * (+137)
00:16:07 <oerjan> as for detecting them, in a sense that's just diff'ing the copies.
00:17:05 -!- tswett has quit (Ping timeout: 276 seconds).
00:17:55 <MDude> If the watermarks don't cover the same area, and there are at least three, you could do a triple modular redundancy operation.
00:17:59 -!- tswett has joined.
00:18:41 <oerjan> right, but that would be a bad design if you wanted to detect collusion in leaking
00:19:11 <MDude> Though that'd both be just an easily automated subistitute for cropping and stitching the images, and also really bad practice on the part of the watermarker.
00:19:39 <MDude> I'm not sure what collusion in leaking even means here.
00:20:30 <oerjan> i mean, someone using two or more different copies in order to try to make a copy that doesn't tell where they got the originals from
00:21:29 -!- tswett_ has joined.
00:24:02 -!- tswett has quit (Ping timeout: 245 seconds).
00:24:02 -!- tswett_ has changed nick to tswett.
00:24:58 -!- M28_ has joined.
00:27:50 -!- M28 has quit (Ping timeout: 240 seconds).
00:28:05 -!- mhi^ has quit (Quit: Lost terminal).
00:28:32 -!- M28 has joined.
00:29:04 <zzo38> There is also various kind of compression which is usable; by itself it won't help but may be usable in combination with other things.
00:29:17 -!- M28_ has quit (Read error: No route to host).
00:29:48 -!- tswett_ has joined.
00:33:50 -!- tswett has quit (Ping timeout: 252 seconds).
00:33:50 -!- tswett_ has changed nick to tswett.
00:39:30 <oerjan> looking at several relevant wikipedia pages (Digital watermarking, Copy attack and Canary trap) none of them say anything about detecting this kind of multiple leak :/
00:40:13 <oerjan> of course that doesn't mean none of the systems listed have implemented it
00:40:35 -!- Phantom_Hoover has quit (Quit: Leaving).
00:43:55 -!- tswett has quit (Quit: tswett).
00:46:15 <oerjan> hm oh it's actually simple: just make n non-overlapping watermarks and for each recipient leave _out_ exactly one.
00:46:43 <oerjan> then your tampering cannot remove the ones you don't have the source excluding.
00:47:35 <oerjan> that's good, i was initially thinking it might need exponential amount of data
00:49:00 <oerjan> with that scheme, you can easily do MDude's trick if you have 3 copies, do get one with _all_ the watermarks.
00:50:02 <oerjan> maybe it does require exponential amount of watermark pieces after all.
00:50:34 <MDude> You could remove all the watermarks for which you have examples of images lacking them.
00:50:55 <MDude> But you'd need to be able to tell watermark from non-watermark.
00:51:01 <oerjan> MDude: yes but that would make things _worse_ for hiding your set of sources
00:51:37 <elliott> is it assumed that the watermarks are steganographic?
00:52:14 <oerjan> elliott: i'd think that would be necessary to prevent removing them?
00:52:34 <elliott> right. do such watermarks even survive, e.g. blurring?
00:53:07 <oerjan> there was something in those wikipedia articles about robust vs. fragile watermarks
00:53:12 <MDude> That would sort of make it a non-watermark, though?
00:53:27 <MDude> Since if it's hidden then no one can tell who marked it.
00:53:38 <MDude> Or that is is marked.
00:53:58 <oerjan> MDude: um hidden if you don't have the key to detect it, presumably
00:54:47 <MDude> I was thinking of watermarks on art, which are often there basically as a signature.
00:55:07 <oerjan> this is watermarking for source tracing / traitor tracing
00:55:12 <MDude> If the general audience can't see it, it can't tell them it's yours.
00:55:23 <Jafet> If each combination of watermarks is encrypted separately, there is no way to recover combinations from multiple versions
00:55:29 <oerjan> so basically the opposite of visible signature
00:58:23 <oerjan> Traitor tracing was another wikipedia page i looked at
01:07:01 <oerjan> Jafet: i am failing to interpret the meaning of your line, you may wish to state some assumptions?
01:09:47 <Jafet> I went with the assumptions that I assumed you were assuming
01:11:04 <Jafet> How does getting multiple copies let you mangle the watermark, anyway?
01:12:09 <elliott> e.g. averaging the pixels of two images?
01:12:46 <Jafet> Ideally, that would give you an image with both watermarks.
01:13:03 <zzo38> If you have a lot of copies, you can try to, instead of just averaging, use varying distributions at random.
01:13:44 <zzo38> You can also still apply other filters
01:14:15 <Jafet> You could pick k, then encode several watermarks such that any k images you distribute share at least one watermark. This mark can't be erased by averaging.
01:16:14 <zzo38> You could still try cropping, if you don't care about the entire picture. If you don't even care about the quality, just draw a picture. Some things are not allowed to be photographed, but you can still try to draw the picture. If it is spoken dialogue in a movie or play, you can try to use shorthand.
01:17:57 <oerjan> HackEgo: oh come on, there are plenty of h's in there already!
01:18:00 <MDude> One issue with drawing would be that it wouldn't neccisarily be considered reliable since you could jut draw whatever fake picture easily if you were a fake informant.
01:18:13 <zzo38> You could also add in your own watermarks
01:18:39 <MDude> Plus, your drawing might be detected as yours much like a signature.
01:18:47 <zzo38> MDude: Yes; you would have to assume that that doesn't matter, in addition to having a good quality copy of the picture also doesn't matter.
01:18:57 <Jafet> If you are watermarking classified information, you could replace some of the unimportant words using a language database, modulated by the watermarking signal.
01:19:26 <MDude> Edge detection might work.
01:19:46 <MDude> Only keep the lines that are important.
01:19:59 <zzo38> Jafet: You could do the same to tamper with it I suppose; how well a tamper work depend on the system.
01:20:22 <oerjan> Jafet: that's precisely one of the methods described in wikipedia
01:20:37 <oerjan> well, except for the modulation
01:21:19 <Jafet> I don't remember that, but wikipedia may have left a watermark in my mind
01:22:27 <zzo38> Esolang list of ideas has this: "Language with functions and classes that only draws/ describes pictures, and can be compressed into tight string. And can be used for image compression, so that if a compression program is presented with a picture, it tries to find the programs that draw at least some features on the picture. ... If compression is lossy enough, the result will look artistic in a strange way."
01:22:54 <zzo38> If such compression is possible, maybe it has effect on such thing?
01:23:50 <zzo38> In case of drawing a picture based on another picture, you can also have, many people are making a drawing, again from many different copies, and then many people will put it into the computer.
01:25:34 <oerjan> Jafet: https://en.wikipedia.org/wiki/Canary_trap
01:30:23 <zzo38> Yes, of course, internal conspiracies are another way.
01:30:58 <oerjan> zzo38: this somehow reminds me of once upon a time reading about a compression scheme using self-similar fractals
01:33:06 <oerjan> no later than the 90s, at most
01:33:27 <oerjan> (and no earlier than the 80s, for more obvious reasons)
01:38:26 <kmc> a compression scheme using self-transforming machine elves
01:38:36 <Sgeo> "Web browsers are a deliberately designed engine for remote code execution, a term which strikes fear into the hearts of information security professionals worldwide."
01:39:38 <zzo38> That is one reason to avoid it (although, such a "remote code execution" can be limited; it just happens to be way too complicated in web browsers).
01:40:02 <zzo38> Still, running the program on the server provides better compatibility anyways.
01:42:21 <MDude> I'd say a lot fo that comlpication is because browsers weren't made with that in mind, and in fact are more often than not designed with letting websites "control the user experience" as an unnervingly high priority.
01:42:33 <MDude> Much like the whole idea behind autorun.
01:42:53 <zzo38> Autorun is also bad.
01:43:18 <Sgeo> "tl;dr: use authenticated encryption. use authenticated encryption. use authenticated encryption. use authenticated encryption. use authenticated encryption. use authenticated encryption. use authenticated encryption. use authenticated encryption. use authenticated encryption. use authenticated encryption."
01:43:19 <zzo38> Not only that, but user experience may depend much on user's preferences and on suitability to devices and user interfaces.
01:44:46 <zzo38> Gopher menus are very well designed to be universal regardless of preferences and suitability to devices, however, this might not have necessarily been the intention. Nevertheless, it works well.
01:45:52 <zzo38> Autorun is a bad idea, however it may be useful to have a "autostart" option that can be turned on/off. Some game systems use this.
01:46:52 <zzo38> You can include functions in BIOS and hardware for protection of system when user doesn't want it tampered with by a program.
01:47:32 <Jafet> Use encryption authenticated by diginotar's iranian RAs
01:51:18 <Sgeo> I started watching a Planescape Torment LP but it seems like the person didn't finish
01:56:45 -!- nooodl_ has quit (Quit: Ik ga weg).
01:57:50 -!- MDude has quit (Ping timeout: 276 seconds).
02:00:40 <quintopia> zzo38: what if you got an android device as a gift
02:01:57 <zzo38> quintopia: I might try to reprogram it, if it has a keyboard; if not, I will just give it back
02:02:40 <quintopia> zzo38: so you wouldn't try to play android games?
02:03:27 <zzo38> I would want to reprogram it first, even though it still may run the Android system, I would want to make changes to it since the existing way has several problems.
02:05:29 -!- oerjan has set topic: Happy Towel Day! | PSA: fizzie is running the wiki now, contact him for any problems | brainfuck survey: https://www.surveymonkey.com/s/L82SNZV | https://dl.dropboxusercontent.com/u/2023808/wisdom.pdf http://codu.org/logs/_esoteric/ http://tunes.org/~nef/logs/esoteric/.
02:08:22 <elliott> you know I think that PSA can go by now.
02:09:06 -!- oerjan has set topic: Happy Towel Day! | PFFT! | brainfuck survey: https://www.surveymonkey.com/s/L82SNZV | https://dl.dropboxusercontent.com/u/2023808/wisdom.pdf http://codu.org/logs/_esoteric/ http://tunes.org/~nef/logs/esoteric/.
02:11:17 <fowl> how can person run a wiki
02:11:56 <oerjan> fowl: very stressfully
02:13:20 <tromp_> what does text mode mean in bf survey?
02:13:52 <Sgeo> Here's my BF opinion: Cells should expand to the right, leaving the tape to the left should wrap around to the right-most cell.
02:14:05 <Bike> we already did that joke Sgeo
02:14:11 <Bike> it ended in surreal numbers. do you really want that???
02:14:22 <Bike> i dunno, like.... *waves backwards in time* that way
02:14:50 <oerjan> we'd know when, if Gregor made the log search work again.
02:14:50 <Sgeo> I could stand to learn about surreal numbers
02:14:57 -!- hk3380 has quit (Ping timeout: 252 seconds).
02:15:06 <oerjan> Sgeo: wikipedia is thataway
02:15:45 <zzo38> Survey monkey not so good. Rewrite the survey in SQL.
02:15:50 -!- MDude has joined.
02:16:06 <Sgeo> Why does that statement remind me of Dr. Klaun?
02:16:12 <Bike> imo don't get the book "surreal numbers", it's too thin
02:16:19 <zzo38> I have written some (not released yet) extension which includes some statistics stuff and some other things too
02:16:38 <Sgeo> <zzo38> Survey monkey not so good. Rewrite the survey in SQL.
02:17:23 <fowl> surreal numbers
02:18:08 <oerjan> @google surreal numbers
02:18:08 <lambdabot> http://en.wikipedia.org/wiki/Surreal_number
02:18:09 <lambdabot> Title: Surreal number - Wikipedia, the free encyclopedia
02:18:11 <fowl> yea the wiki page doesnt clear anything up
02:18:20 <fowl> are you all mathematicians or something
02:18:29 * Sgeo wishes he was a mathematician
02:18:45 <Sgeo> Being a pseudomathematician as a kid doesn't count
02:18:49 <Bike> i'm a bikeologist
02:19:07 <Sgeo> Jeez, I was into a lot of pseudo stuff as a kid
02:19:26 <Sgeo> I remember buying a Deepak Chopra book once
02:19:26 <zzo38> I am not really a mathematician but I still do study and interest of mathematics.
02:19:38 <fowl> i was into pseudoephedrine as a kid
02:19:42 <fowl> does that count
02:19:51 <Bike> sgeo was super into pseudo-dionysius.
02:19:57 <zzo38> I have never purchased a book by Deepak Chopra, although I have read some. I have no intention to purchase such book.
02:20:11 <zzo38> (Howveer, I do have intention to read it, sometimes)
02:21:06 <MDude> Did you study pseudopods?
02:21:59 <HackEgo> /hackenv/bin/log: 2: cd: can't cd to /var/irclogs/_esoteric \ grep: ????-??-??.txt: No such file or directory
02:22:35 <zzo38> oerjan: When would they fix that, you expect?
02:22:54 <oerjan> zzo38: i am not sure he is intending to fix it.
02:24:03 <Bike> even a stopped clock
02:24:21 <oerjan> because he hasn't done so yet, and it's been a while.
02:25:15 <oerjan> and because he moved HackEgo to a different server, might mean he doesn't want to use that much resources on a single one.
02:27:19 <zzo38> What if he instead copy the file per day, or something like that?
02:27:53 <Bike> so it would always be ten years behind? how census
02:29:09 <elliott> zzo38: because people don't put work into something they don't care about unless they get paid for it.
02:29:12 <zzo38> No I mean copy yesterday's file to today.
02:31:09 <zzo38> Altneriatively since now SQLite is installed in it, you could even make it into one SQL database file (containing for all channel, date, etc), with one "IRC" table having the columns such as: timestamp, sender, command, channel, message, receiver. (The "receiver" can be applicable to such things as a KICK command.)
02:31:29 <elliott> he won't do that because he doesn't care about it.
02:32:18 <zzo38> How much does it cost?
02:33:29 <elliott> you'd have to ask Gregor for his contracting rate if he's willing to be paid for it at all, but I'm going to guess not significantly under $100/hr.
02:34:33 <coppro> elliott: is FlexibleContexts evil?
02:35:48 <oerjan> my secret conspiracy theory is that Gregor's not making the logs searchable because Aftran has threatened to sue him if he makes it possible to find his messages.
02:36:10 <oerjan> or possibly blackmailing
02:37:04 <oerjan> coppro: FlexibleContext is like the least evil extension which is only nonstandard because they cannot agree how to standardize it, or something,
02:37:44 <zzo38> Anyone who make the copy of the logs will make it searchable, though.
02:38:25 <Gregor> HackEgo and glogbot are no longer run on the same system.
02:38:34 <Gregor> That is the one and only reason why the logs aren't searchable.
02:38:48 <Gregor> They were searchable before because it was trivial to just link them into the right location.
02:39:00 <zzo38> Gregor: That is perfectly understood; the reason is why you didn't put it on the same system.
02:39:41 <Gregor> Because that system was constantly running out of hard disk space so I wanted to move one if its chroots to another system.
02:41:38 <zzo38> But, maybe is useful to have a copy, somehow
02:46:10 <Gregor> And that brings us to elliott's point :)
02:46:50 <zzo38> What if someone else install a copy, though?
02:52:29 -!- hogeyui has quit (Ping timeout: 264 seconds).
02:53:52 -!- hogeyui has joined.
03:10:16 <zzo38> This issue of 2600 contains the De Bruijn sequence for four-digit numbers.
03:10:30 -!- hk3380 has joined.
03:11:25 <zzo38> It also contains a letter that simply says, "Can you see it?" (there is no attachment or anything else like that)
03:12:27 <zzo38> There is also a letter about the most durable telephone number. Is it Hotel Pennsylvania's "PENnsylvania-5000"?
03:14:50 <kmc> why is it durable
03:15:08 <zzo38> They had the same telephone number since 1930, and still do.
03:15:42 -!- pikhq has quit (Quit: Reconnecting).
03:15:55 -!- pikhq has joined.
03:16:31 <shachaf> de bruijn sequences are the best
03:16:50 -!- pikhq has quit (Client Quit).
03:17:58 -!- pikhq has joined.
03:22:56 <zzo38> Since the name of the hotel matches the name of the telephone exchange, it can work very well!
03:32:39 <M28> my brainfuck JIT can run some benchmarks that take ~6 seconds in interpreters, in about 200 ms
03:33:37 <M28> also depends on the benchmark, some have many orders of magnitute in speed increase
03:41:37 <zzo38> I prefer tab separated values rather than comma separated values.
03:44:09 <pikhq> zzo38: Yes, as do I.
03:44:22 <pikhq> The CSV format isn't *that* bad, but TSV is significantly cleaner.
03:47:44 <elliott> M28: how does it compare to esotope-bfc?
03:48:17 -!- MoALTz_ has joined.
03:48:27 <M28> esotope is in python...
03:48:37 <M28> it generates C
03:48:48 <M28> elliott, do you have it there?
03:49:14 <elliott> http://mearie.org/projects/esotope/ oh, maybe it was rewritten
03:50:14 -!- MoALTz__ has joined.
03:50:16 <M28> I don't have a python env here, could you compile this for me to C? http://puu.sh/9027N/3c9c70e514.txt
03:51:10 <zzo38> One thing I don't like about indentation sensitive syntax is that, often in a C program I will add some lines for debugging purpose and those ones are at the left part of the line rather than being in the indented position. Also, there are many kind of ways to indent based on other things, so sometimes it isn't quite indented only by blocks anyways.
03:51:19 -!- MoALTz has quit (Ping timeout: 265 seconds).
03:52:05 <M28> it doesn't do as many optimizations as esotope, though, so we'll see
03:52:41 <M28> the problem is that esotope seems to just run the bf program if it doesn't have any i/o
03:52:51 -!- MoALTz_ has quit (Ping timeout: 252 seconds).
03:52:56 <elliott> because that can fail to terminate
03:53:05 <fowl> zzo38, indenting for something thats not a new scope? :(
03:53:06 <M28> yeah, but you got the idea
03:53:21 <pikhq> It's just able to optimize things quite a lot.
03:53:23 <elliott> "just running" a program but always halting is sort of what optimisation is.
03:53:27 <elliott> M28: anyway, http://sprunge.us/gSeC
03:53:32 -!- MoALTz has joined.
03:53:37 <M28> kk let me see
03:54:10 <zzo38> fowl: Yes, sometimes. One is case statements.
03:54:24 <zzo38> And then there is unindenting too like I described.
03:54:42 <elliott> with clang -O3 -march=native
03:54:58 -!- MoALTz_ has joined.
03:55:13 <elliott> (not that there's all that much genius optimisation you can apply to that C code, probably)
03:55:40 <M28> msvc just removes the whole block
03:55:41 <elliott> heh, though 0.028s with gcc
03:55:44 <M28> and replaces it with the output
03:56:02 <M28> takes 0 ms
03:56:07 <M28> let me open it with IDA
03:56:08 <elliott> I have not heard MSVC being very good at optimisation, so that surprises me.
03:56:33 <M28> it is very good at executing pure functions at compile time
03:56:47 -!- MoALTz__ has quit (Ping timeout: 276 seconds).
03:57:04 <M28> seems that it didn't remove the whole thing
03:57:04 -!- MoALTz__ has joined.
03:58:00 <M28> the compiled code is still somewhat similar to the C code
03:58:29 -!- MoALTz has quit (Ping timeout: 264 seconds).
03:59:23 <M28> right now, my JIT generates the code in 1 pass
03:59:27 -!- MoALTz_ has quit (Ping timeout: 245 seconds).
03:59:37 <M28> so I'll have to switch to generating an AST and all that crap to beat esotope
04:00:06 <M28> (it'll probably be just as fast as esotope, though)
04:00:14 <M28> unless I find some optimization that it doesn't do
04:01:23 -!- MoALTz has joined.
04:03:19 -!- shikhout has changed nick to shikhin.
04:03:53 -!- MoALTz__ has quit (Ping timeout: 265 seconds).
04:09:54 <zzo38> Now I made up a virtual table module of SQLite for making a "horizontal union". For example: CREATE VIRTUAL TABLE VT USING HORIZONTAL_UNION(X ORDER BY X ASC,Y ORDER BY Y DESC);
04:11:43 <zzo38> If X contains (1),(2),(3),(4),(5) and Y contains (1),(10),(100),(1000),(10000) then the result will be (1,10000),(2,1000),(3,100),(4,10),(5,1). Maybe it can be useful for some kinds of statistical calculations, or just if you want arbitrary matching.
04:12:28 <zzo38> Do you know much about statistics?
04:13:47 <Bike> i was a teenage statistic
04:21:31 <zzo38> I would want to have SQL extension to plot various graphics, such as statistical graphics and other things (such as geography and horoscopes).
04:21:32 -!- hk3380 has quit (Ping timeout: 252 seconds).
04:24:19 <coppro> oerjan: is TupleSections not less evil?
04:27:04 -!- Sprocklem has quit (Quit: brb).
04:27:29 -!- Sprocklem has joined.
04:29:14 -!- edwardk has quit (Quit: Computer has gone to sleep.).
04:29:15 -!- MDude has changed nick to MDream.
04:37:06 -!- edwardk has joined.
04:38:30 <oerjan> coppro: also harmless.
04:39:23 <oerjan> well mostly. actually TupleSections could get confusing if you forget an argument in a tuple.
04:42:16 <oerjan> so it's evil for the same reason as the default ghc Monad and Functor instance for functions.
04:43:07 <oerjan> and like making Num instances for functions etc.
04:44:18 <oerjan> i.e. it's cute for pointless golfing but shouldn't be on by default.
04:44:38 <elliott> that argument applies to all operator sections.
04:44:53 <elliott> forgetting a tuple element isn't really a very plausible error
04:45:14 <elliott> and it'll pretty much always give a reasonable type error, I think, since functions and tuples rarely mix
04:45:47 <elliott> you know what's evil about sections? (- 1)
04:46:34 <lambdabot> Could not deduce (GHC.Num.Num (a0 -> t))
04:46:35 <lambdabot> arising from the ambiguity check for ‘e_113’
04:46:35 <lambdabot> from the context (GHC.Num.Num (a -> t), GHC.Num.Num a)
04:46:35 <lambdabot> bound by the inferred type for ‘e_113’:
04:46:35 <lambdabot> (GHC.Num.Num (a -> t), GHC.Num.Num a) => t
04:46:47 <Bike> where did my life go so wrong
04:47:20 <oerjan> it's only a section of your life, don't panic
04:47:51 <oerjan> (especially as it's towel day)
04:51:05 <oerjan> kmc: well possibly not in SF yet
04:54:31 -!- Sprocklem has quit (Quit: bbs).
04:56:59 -!- Sprocklem has joined.
05:03:19 -!- edwardk has quit (Quit: Computer has gone to sleep.).
05:07:08 -!- BeingUntoDeath has joined.
05:09:02 -!- edwardk has joined.
05:09:34 -!- shikhout has joined.
05:10:04 -!- Sprocklem has quit (Quit: brb).
05:12:22 -!- shikhin has quit (Ping timeout: 245 seconds).
05:13:19 -!- oerjan has quit (Quit: Nite).
05:14:13 -!- Sprocklem has joined.
05:22:49 -!- Slereah has joined.
05:23:50 -!- Slereah_ has quit (Ping timeout: 240 seconds).
05:29:48 <zzo38> I have used the unofficial RLA instruction of 6502 to do three things at once: Do a bank switch, clear the carry flag, and clear the accumulator.
05:30:48 <zzo38> The program also uses ANC, ALR, LAX, AXS, and DCP.
05:32:42 <zzo38> ANC saves a byte and two cycles over using AND followed by CLC.
05:34:45 <fowl> zzo38, sometimes i think you're having conversations with people i can't see. this scares me.
05:36:06 <zzo38> Sometimes *I* think I am having conversations with people I can't see, but I generally doubt it.
05:43:41 <HackEgo> ittlycoin ariolameriacoin rumcoin topcoin beckoutcoin selcoin pathcoin ischersetcoin prefcoin explocoin mumcoin recurcoin rhizacoin circumcoin dispathisincoin iabcoin tricoin clccoin chasecoin sartcragcoin
05:44:44 <Bike> conversationswithpeoplecoin
05:48:46 -!- MoALTz has quit (Ping timeout: 265 seconds).
06:21:51 -!- BeingUntoDeath has quit.
06:24:50 <kmc> http://vanbever.eu/pdfs/vanbever_turing_icnp_2013.pdf "BGP has the same computing power as a Turing Machine"
06:25:14 <kmc> are you having a nice time?
06:25:52 <kmc> smoke BGP everyday
06:26:05 <fizzie> A bit tired time, I don't sleep so well in planes. It probably shall pass.
06:31:24 -!- drdanmaku has quit (Quit: Connection closed for inactivity).
06:32:22 <kmc> how long was your flight?
06:32:54 <fowl> didnt know you could use (T[]){t1,t2,t3} in c ._.
06:33:18 <kmc> is that standard or an extension
06:33:56 -!- glogbackup has quit (Remote host closed the connection).
06:34:09 <kmc> I guess it is standard C99, although GCC lets you use it in some additional places
06:34:13 <kmc> http://gcc.gnu.org/onlinedocs/gcc/Compound-Literals.html
06:34:21 <kmc> C is a wacky language
06:40:21 <fizzie> 9 hours, or something like that.
06:40:41 <fizzie> And a 6-hour timezone change.
06:41:35 <fizzie> Finland is +3 (in summer), I guess here must be +9 unless I miscalculated.
06:59:02 -!- Patashu_ has quit (Ping timeout: 240 seconds).
07:04:18 <kmc> did you learn yet about who is the boss of them all since 1992?
07:06:29 -!- password2 has joined.
07:07:26 -!- MoALTz has joined.
07:12:49 -!- conehead has quit (Quit: Computer has gone to sleep).
07:17:11 * kmc discovers the secret Rust attribute that lets you use private stuff from another module
07:25:42 -!- MindlessDrone has joined.
07:42:55 -!- Sorella has quit (Quit: It is tiem!).
07:54:45 -!- Slereah_ has joined.
07:55:04 <kmc> fungot: when all you have is a brain, everything looks like a brain
07:55:05 <fungot> kmc: mr president, commissioner, this report is very clear what is happening in that region of the european union should concern itself with all seas, particularly those not in paid employment. moreover, the commission sets up programmes to ensure that the objectives of the commission and the council to this question to the house.
07:55:46 <kmc> fungots fall on fungot falls
07:55:46 <fungot> kmc: we still take it for granted that there would not be our top priority. when species become extinct, they are not looking for scapegoats but are instead indicting the system.
07:56:19 -!- Slereah has quit (Ping timeout: 240 seconds).
08:00:32 -!- ter2 has quit (Ping timeout: 276 seconds).
08:31:47 -!- Slereah has joined.
08:35:05 -!- Slereah_ has quit (Ping timeout: 252 seconds).
08:55:36 -!- nooodl has joined.
09:21:52 -!- Slereah has quit (Read error: Connection reset by peer).
09:22:07 -!- Slereah_ has joined.
09:26:20 -!- Sprocklem has quit (Ping timeout: 276 seconds).
09:27:56 -!- edwardk has quit (Ping timeout: 258 seconds).
10:03:15 -!- edwardk has joined.
10:17:41 -!- Slereah has joined.
10:19:03 -!- Slereah_ has quit (Ping timeout: 252 seconds).
10:20:17 -!- password2 has quit (Ping timeout: 276 seconds).
10:23:12 -!- Patashu has joined.
10:23:36 -!- nooodl_ has joined.
10:27:17 -!- nooodl has quit (Ping timeout: 252 seconds).
10:36:39 -!- Phantom_Hoover has joined.
10:50:18 -!- Patashu_ has joined.
10:50:18 -!- Patashu has quit (Disconnected by services).
11:06:00 -!- xk002 has joined.
11:07:24 -!- newsham has quit (Quit: leaving).
11:09:01 -!- shikhin has joined.
11:12:23 -!- shikhout has quit (Ping timeout: 252 seconds).
11:24:03 -!- hk3380 has joined.
12:13:23 -!- xk002 has quit (Ping timeout: 276 seconds).
12:16:45 -!- MindlessDrone has quit (Quit: MindlessDrone).
12:30:53 -!- mhi^ has joined.
12:32:10 -!- yorick has joined.
13:11:00 -!- boily has joined.
13:11:18 <Taneb> Is it possible to fake <marquee> in CSS3?
13:16:22 <Taneb> boily, can you give me a pointer without too much spoilers?
13:16:30 -!- Sorella has joined.
13:20:14 <boily> Taneb: you have to think like a video animator, with @keyframes.
13:20:51 <boily> a nice demo, with transition effects → http://media02.hongkiat.com/marquee-css3-animation//demo/index.html
13:20:55 <Taneb> Oh yeah, bouncing marquees are a thing
13:26:26 -!- Patashu_ has quit (Ping timeout: 252 seconds).
13:27:44 -!- edwardk has quit (Quit: Computer has gone to sleep.).
14:02:55 -!- Sgeo has quit (Read error: Connection reset by peer).
14:13:16 <int-e> is there anybody else here who feels that adding animations to CSS is wrong?
14:13:31 -!- hk3380 has quit (Ping timeout: 258 seconds).
14:14:28 <int-e> ("only" in the form of interpolation, but still.)
14:21:11 <mroman> two people just recently submitted bug reports for Burlesque
14:27:19 <mroman> int-e: I rather prefer CSS animations than Flash
14:31:10 -!- xk002 has joined.
14:32:12 <int-e> mroman: The entire notion that web sites (which are not games or clever data browsers that use the time dimension to display data, or actual videos) need animations is wrong in my eyyes (literally more often than not)
14:33:13 <int-e> (I know that I'm not going to win any fight against this trend that has already gone most of the way, but I can still complain.)
14:34:13 <int-e> (I found that I even find Firefox' "smooth scrolling" feature annoying. I can switch that off though.)
14:34:25 <boily> smooth scrolling? bletch!
14:35:14 <int-e> it's just another kind of smooth transition that developers seem to believe users like so much.
14:36:09 <nortti> it is actually quite amazing how much stops working if you disable js
14:36:32 <elliott> it's amazing how much stops working if you disable http
14:36:51 <int-e> (coming from phones and tablet where input lacks precision, so you have to extrapolate what you are going to achieve using visual feedback; and suddenly, smooth scrolling becomes a boon.)
14:37:18 -!- ter2 has joined.
14:37:31 <nortti> I mean things that have no right to contain js, such as a download link that launches a script that parses a json embedded in thw page and the redirects to the file
14:37:39 <int-e> elliott: Ah but you can still browse "the web": there are still a couple of gopher sites out there.
14:39:34 <nortti> and then there are the infinite-scrolling webpages http://xkcd.com/1309
14:43:10 <int-e> Oh I'll try disabling layout.css.prefixes.{transitions,animations} in FF.
14:47:48 <mroman> int-e: well... the "web" has become a plattform for desktop applications too
14:48:07 <mroman> which kinda steers to being able to write os-independent applications for "the web"
14:48:41 <mroman> I start to think of browsers as virtual machines like .NET/JVM
14:49:25 <mroman> It's just not just hypertext anymore.
14:49:33 * int-e mentally moves mroman from the "undecided" into the "problem" bin ;-)
14:49:58 <nortti> unfortunatelly, some people tend to not realize what is an application and what is a hypertext page
14:55:17 -!- shikhin has quit (Ping timeout: 245 seconds).
15:01:10 <int-e> Another late realization is that tracking URL shorteners are perceived as a feature. E.g. "But all things being equal, it makes sense to select [an URL shortener] that provides tracking."
15:03:59 <int-e> source: http://searchengineland.com/analysis-which-url-shortening-service-should-you-use-17204 (note the domain name, it goes with the territory)
15:04:54 <int-e> This - research? - was prompted by a visit to the bit.ly front page yesterday.
15:07:10 <nortti> and this is why I like dy.fi; no tracking that I know of and forwards will go away after month on disuse (I think)
15:11:34 -!- oerjan has joined.
15:13:24 -!- hk3380 has joined.
15:25:38 -!- tswett has joined.
15:26:41 -!- M28 has quit (Ping timeout: 264 seconds).
15:28:32 -!- M28 has joined.
15:40:18 -!- edwardk has joined.
15:42:03 -!- conehead has joined.
15:52:38 -!- MDream has changed nick to MDude.
16:14:02 -!- M28 has quit (Ping timeout: 245 seconds).
16:15:03 -!- oerjan has quit (Quit: leaving).
16:15:58 -!- M28 has joined.
16:20:25 -!- realz has quit (Ping timeout: 255 seconds).
16:21:10 -!- realz has joined.
16:25:16 -!- tswett_ has joined.
16:27:49 -!- tswett has quit (Ping timeout: 252 seconds).
16:27:49 -!- tswett_ has changed nick to tswett.
16:35:03 -!- tromp__ has joined.
16:37:32 -!- clog_ has joined.
16:38:16 -!- edwardk has quit (Quit: Computer has gone to sleep.).
16:39:38 -!- erdic_ has joined.
16:44:04 -!- clog has quit (*.net *.split).
16:44:04 -!- erdic has quit (*.net *.split).
16:44:04 -!- tromp_ has quit (*.net *.split).
16:44:14 -!- erdic_ has changed nick to erdic.
16:45:20 -!- blitter64 has joined.
16:46:19 <mroman> damn. no I pushed one of my own warriors from the hill :(
16:48:37 <boily> you pushed one of your now warriors?
16:57:11 -!- edwardk has joined.
16:59:30 <kmc> mroman: it's a virtual machine and an operating system
17:00:10 <kmc> browsers have their own schedulers, memory managers, network stacks, graphics engines, UI toolkits, compilers, dev tools
17:12:09 <mroman> I wouldn't be surprised if Browsers had more source code than operating systems
17:12:50 <kmc> defining "operating system" is a tricky business
17:13:06 <kmc> the non-modular part of the Linux kernel is like 50,000 lines
17:13:13 <kmc> the set of all packages in Debian is probably hundreds of millions
17:13:17 <scoofy> linux kernel 3.6 has 15.9 million lines of code
17:13:18 <kmc> if not billians
17:13:20 <kmc> billions too
17:13:36 <kmc> i usually take "operating system" to mean something much broader than just a kernel
17:13:40 <kmc> but it's fuzzy
17:13:44 <scoofy> windows server 2003 is 50 million lines of code
17:15:09 <scoofy> firefox is about 15 million sloc
17:16:37 -!- metasepia has joined.
17:16:48 <kmc> Servo is only 100,000 SLoC but it doesn't do most of the browsery things yet :)
17:17:54 -!- tswett_ has joined.
17:18:07 <kmc> I'm writing the new HTML parser, which is about 4,000 non-blank lines and not done yet
17:18:19 <kmc> but I think it will end up substantially smaller than any other browser-quality HTML parser
17:18:26 <kmc> at least the tokenizer part is like 6x smaller than others
17:18:41 <Bike> how long is it after macroexpansion, tho :P
17:18:53 <mroman> hm. IT's actually better not to throw spl 0 bombs o_O
17:21:44 -!- tswett has quit (Ping timeout: 258 seconds).
17:22:59 <Jafet> Web browsers are operating systems
17:23:19 <int-e> firefox is the new emacs?!
17:23:40 <Bike> at least javascript has lexical scope
17:23:43 * int-e looks for a coffeemaker plugin
17:23:44 <Melvar> That seems like actually a pretty nice analogy.
17:24:11 <Bike> kmc: ensure that servo has M-x psychoanalyze-pinhead
17:24:13 -!- MindlessDrone has joined.
17:24:27 <boily> int-e: I have pentadactyl installed on my firefox. no emacs for me! :D
17:24:43 <int-e> there actually is one called "Morning Coffee", but it's about opening routine websites in tabs
17:24:51 <int-e> (I thought firefox had tab group bookmarks?)
17:24:55 <Melvar> Dangit, I was going to observe that there is a counterpart to evil-mode and the like at least.
17:25:08 <nortti> int-e: there is a htcpcp plugin"
17:25:23 -!- tswett has joined.
17:25:49 -!- tswett_ has quit (Ping timeout: 240 seconds).
17:25:57 <int-e> (I only looked for "official" addons)
17:27:00 <int-e> htcpcp is a good joke at least. "The implementation of this should be via a plugin API for beverages in general, so that the Download Manager can brew a nice cup of tea [...]"
17:27:56 -!- nooodl__ has joined.
17:30:43 -!- nooodl_ has quit (Ping timeout: 240 seconds).
17:31:27 -!- tswett has quit (Remote host closed the connection).
17:31:29 -!- clog_ has quit (Quit: ^C).
17:31:43 -!- clog has joined.
17:38:11 -!- tswett has joined.
17:41:14 -!- newsham has joined.
17:41:21 <kmc> shachaf: I'm having a lot of fun hacking on rustc so far
17:41:33 <kmc> it's a lot less scary than GHC, although I'm having trouble putting my finger on why
17:41:48 <kmc> maybe it's just that I know a bunch of the core rust developers and they're usually on IRC
17:48:46 <int-e> kmc: do the rustc sources contain the term "zonk"? (grep -r zonk . | wc -l in ghc's compiler subdirectory returns 907... and I still don't know what SPJ means when he talks about zonking.)
17:49:01 <int-e> (But I never asked.)
17:50:38 <olsner> it's something that ghc does to types, iirc
17:50:54 <kmc> int-e: nope
17:51:03 <kmc> they contain plenty of other weird shit though
17:51:10 <kmc> some of which has been documented by https://twitter.com/horse_rust
17:53:59 -!- xk002 has quit (Ping timeout: 252 seconds).
17:56:02 -!- conehead has quit (Quit: Textual IRC Client: www.textualapp.com).
17:56:41 <olsner> could be something with <> characters that got stripped by twitter?
17:57:06 <Bike> http://charlie.su/screen_shot_2014_04_20_at_12.01.43_pm-e79630c4f193a5.png No
18:03:37 -!- hk3380 has quit (Ping timeout: 245 seconds).
18:08:02 -!- shikhin has joined.
18:11:35 -!- tswett_ has joined.
18:14:14 -!- tswett has quit (Ping timeout: 252 seconds).
18:14:14 -!- tswett_ has changed nick to tswett.
18:16:04 -!- hk3380 has joined.
18:25:53 <zzo38> Is there pinball simulation which you would write the programming for what happen when various triggers are hit and displaying score and so on are in 6502 machine codes?
18:26:31 <Bike> yes, smoothed particle hydrodynamics should be adequate to represent this
18:43:37 -!- M28 has quit (Ping timeout: 245 seconds).
18:45:35 -!- M28 has joined.
18:48:15 -!- edwardk has quit (Ping timeout: 252 seconds).
18:48:58 -!- Sprocklem has joined.
18:54:22 -!- nooodl__ has changed nick to nooodl.
19:02:37 -!- conehead has joined.
19:11:45 -!- augur has quit (Ping timeout: 258 seconds).
19:15:35 <boily> playing terrafirmacraft. I just set fire to the whole thing...
19:18:53 -!- xk002 has joined.
19:22:44 -!- Patashu has joined.
19:22:55 <zzo38> I want to have some SQL extension for internet connection, both inbound and outbound connection. I don't know very well internet-base programming with C though it looks confusing to me. How would I do it?
19:39:19 -!- augur has joined.
19:39:51 -!- MindlessDrone has quit (Quit: MindlessDrone).
19:42:47 <mroman> zzo38: you wanna extend SQL with sockets?
19:52:08 -!- TodPunk has quit (Quit: This is me, signing off. Probably rebooting or something.).
20:03:04 -!- ais523 has joined.
20:03:36 -!- nortti has changed nick to hvidiecat.
20:03:52 -!- hvidiecat has changed nick to nortti.
20:04:29 -!- nortti has changed nick to hvidie.
20:05:22 -!- hvidie has changed nick to nortti.
20:13:46 -!- tswett has quit (Quit: tswett).
20:18:41 -!- drdanmaku has joined.
20:24:12 -!- hk3380 has quit (Ping timeout: 258 seconds).
20:25:28 -!- TodPunk has joined.
20:26:41 -!- stuntaneous has joined.
20:27:31 -!- hk3380 has joined.
20:30:35 -!- glogbackup has joined.
20:30:46 -!- mhi^ has quit (Quit: Lost terminal).
20:34:43 -!- glogbackup has quit (Ping timeout: 240 seconds).
20:35:06 -!- mhi^ has joined.
20:36:38 -!- Sgeo has joined.
20:42:41 -!- MoALTz_ has joined.
20:46:04 -!- MoALTz has quit (Ping timeout: 252 seconds).
20:49:56 -!- conehead has quit (Quit: Computer has gone to sleep).
21:04:15 -!- oerjan has joined.
21:23:32 <Taneb> It occurs to me that after next month I'll be living in a house with its own IRC channel
21:24:02 -!- Patashu has quit (Ping timeout: 245 seconds).
21:29:57 -!- sebbu has quit (Ping timeout: 252 seconds).
21:35:33 <Taneb> I enjoyed it, possibly due to my position as someone who doesn't really care about X-Men that much
21:37:13 <Taneb> However, it did feel a lot "Here's a bad thing, here's how we're going to fix the bad thing, we're fixing the bad thing, the bad thing is fixed"
21:37:50 <Taneb> That may put you at a disadvantage.
21:38:00 <Phantom_Hoover> (and even then i could tell the continuity was fucked)
21:38:08 -!- blitter64 has quit (Ping timeout: 252 seconds).
21:38:14 <Taneb> Oh god yeah, but that's true for the entire franchise.
21:39:47 -!- blitter64 has joined.
21:43:54 <Phantom_Hoover> i don't mean it was fucked wrt to the other films, i mean it was fucked in its own reference frame
21:45:01 <Taneb> Other than the time travel mechanic being ridiculous
21:46:16 <Phantom_Hoover> well i mean, future-wolverine remembers a bunch of interesting x-men stuff happening in his past
21:46:40 <Phantom_Hoover> but in his past peter dinklage built a robot in the 70s that killed everyone so how does he remember anything else
21:47:09 <Taneb> The robots didn't get going until the 2020s
21:48:10 <Taneb> On another note, I'm watching the elections coverage and it is making me sad
21:49:03 <ais523> oh, are the results coming out now?
21:49:13 <Taneb> North-Easts are out, at least
21:49:24 <Taneb> (2 labour, 1 ukip)
21:49:35 <ais523> hmm, not much of a change yet
21:49:40 <ais523> which is probably good
21:49:48 <Taneb> ais523, sharp rise in UKIP
21:49:57 <ais523> but only 4 results have been called
21:50:07 <Bike> something called "National Front" won a bunch of elections in France.
21:50:18 <Phantom_Hoover> on the bright side nick griffin fucked up the hashtag on his tweet after he lost: https://twitter.com/nickgriffinmep/statuses/470672756724748288
21:51:27 <Taneb> Well, until this election, the candidate I voted for lost outright
21:51:57 <ais523> I'm hoping that UKIP get lib-dem-ized
21:52:21 <ais523> i.e. a fringe party that starts looking serious, inadvertently ends up in power and gets completely humiliated as a result
21:52:42 <Phantom_Hoover> ukip are doing well because people with no political views beyond 'i'm vaguely annoyed at the current system' will vote for anyone promising a new alternative
21:53:17 <ais523> I actually liked the lib dems though :-(
21:53:21 <Phantom_Hoover> the european elections won't do that because nobody actually knows what their MEP does, and ukip sure as hell aren't going to tell them
21:53:39 <elliott> Phantom_Hoover: are you seriously suggesting people switched from lib dem to ukip?
21:53:48 <ais523> Phantom_Hoover: I do, I got a bunch of election propaganda because I've written to my MEPs in the past
21:53:56 <elliott> I rather suspect ukip is getting their votes from ex-tories instead
21:53:57 <ais523> elliott: I think he is, and even though it makes no political sense, I can believe it
21:54:19 <Phantom_Hoover> elliott, ukip has gained a lot of ground from labour too, from what i've read
21:54:39 <ais523> well the newspapers were spinning this as ukip vs. labour
21:54:50 <ais523> are there any scotland-specific EU parties, btw?
21:55:02 <ais523> like, do the SNP run in the EU elections
21:55:16 <elliott> Phantom_Hoover: labour I can believe too
21:55:20 <elliott> but lib dem seems like an entirely different crowd
21:56:37 <ais523> elliott: well the political situation in Birmingham University is mostly lib dem versus labour
21:56:57 <ais523> although I supported the Conservatives (up until the last election), I was a huge anomaly
21:57:42 <Taneb> I try to avoid the people who capital-c Care about politics, but in my social bubble it's seemed Lib Dems vs Green
21:57:54 <Taneb> That's probably more representative of my social bubble than of York
21:57:59 <ais523> Taneb: that says a lot about how the maintream parties are getting on, really
21:58:15 -!- nooodl_ has joined.
21:59:14 <ais523> although, the UK really needs three serious parties to avoid becoming like the US
21:59:19 <ais523> and atm, I'm not sure it has /any/ serious parties
22:01:52 <Phantom_Hoover> the US has two parties because its political system is a joke, not the other way round
22:01:58 -!- nooodl has quit (Ping timeout: 252 seconds).
22:02:15 <elliott> Phantom_Hoover: more like it's a feedback loop between the two
22:02:30 <Bike> i was going to say, there's that whole thing about the voting system causing two party lockin
22:02:57 <elliott> Bike: and two parties having strong incentive not to fix the system in turn
22:03:08 <Phantom_Hoover> a friend summarised up the US as "a democracy designed by people who were terrified of democracy"
22:03:33 <Phantom_Hoover> hence the stalemate-ridden legislature, the 3 cogs of government each smoothly locking the others in place, etc.
22:03:47 <Bike> i looked up the procedure for running a third party in my state once, basically you have to pay ten thousand dollars and get a bazillion signatures. dems and republicans don't have to obv
22:05:01 -!- sebbu has joined.
22:05:06 <Taneb> I didn't realise you were in the US, Bike
22:05:35 -!- sebbu has quit (Changing host).
22:05:35 -!- sebbu has joined.
22:05:40 <Bike> i moved here from luxembourg, shouldn't everyone know this by now
22:06:03 <MDude> Well with the cogs of government, only the legislature is particularly democratic.
22:06:34 -!- nooodl_ has changed nick to nooodl.
22:06:38 <Bike> the legislature with 90% encumbency
22:06:40 <MDude> The administraation has the president, who's elected, but most of it consists of the cabinet.
22:06:49 <ais523> Bike: in the UK, if you want to run as a candidate, you have to pay quite a bit of money but it's refunded if you get a reasonable percentage of the vote
22:07:04 <Taneb> £500 and 5% for Parliament
22:07:08 <Bike> do you have to do that even if you're the tories
22:07:15 <MDude> As oppossed to the supreme court, which is unelected and set for life.
22:07:22 <Taneb> Bike, I believe you do
22:07:30 <Bike> well there you go then.
22:07:33 <ais523> yeah but the tories always get the refund
22:07:48 <ais523> occasionally they don't in a particular constituency, but it makes the news
22:07:49 <Taneb> It's not like they can't afford it
22:08:03 <ais523> because it's so unusual
22:08:03 <MDude> I would think primaries would get democrats and republicans past any hurdles with fees.
22:08:22 -!- ais523 has quit.
22:08:31 <MDude> Just decide whoever you want and then you have a huge party putting cash behind whatever canidate it is.
22:08:54 <Phantom_Hoover> i remember an interview with some ukip guy saying how they were going to break into scotland, and the interviewer was choking back laughter for the whole thing because ukip to date haven't even come close to recouping their £500 here
22:10:59 <pikhq> It's honestly kinda weird that primaries are government-regulated...
22:11:14 <pikhq> At least in theory, primaries are a purely internal affair of the political parties.
22:11:25 <elliott> the party system seems to be weird everywhere.
22:11:28 <MDude> Yeah, I dunno what's up with that.
22:11:54 <Phantom_Hoover> pikhq, well because the two parties are the government, right?
22:12:01 <pikhq> But in general you become a member of a political party at the time you register to vote.
22:12:49 <pikhq> elliott: True facts.
22:13:33 <Taneb> It seems that re. parties, US has the opposite problem to Australia
22:14:09 <pikhq> But then the US has numerous other really weird electoral problems.
22:14:43 <pikhq> Such as the fact that the actual election of a president is by a handful of electoral college members.
22:14:46 <Taneb> Australia's system seems to be quite favourable to small parties in general, while still being harsh on individual small parties
22:18:39 <Taneb> East Midlands results being announced
22:19:04 <Taneb> UKIP 2, Con 2, Lab 1
22:19:32 <oerjan> > indexed (backwards each) . indices odd %~ (*2) $ [1,2,3]
22:19:34 <lambdabot> Couldn't match type ‘(->) s0’ with ‘Data.Functor.Identity.Identity’
22:19:34 <lambdabot> Expected type: [f b1] -> Data.Functor.Identity.Identity (f t)
22:19:34 <lambdabot> Actual type: Control.Lens.Type.Optical (->) (->) f s0 t a0 b1Couldn't matc...
22:19:34 <lambdabot> -> Data.Functor.Identity.Identity a2)
22:20:31 <Taneb> oerjan, you want "indexing"
22:21:25 <oerjan> > indexing (backwards each) . indices odd %~ (*2) $ [1,2,3]
22:21:27 <lambdabot> -> Data.Functor.Identity.Identity a1)
22:21:27 <lambdabot> -> p0 t0 (Data.Functor.Identity.Identity b)’
22:21:27 <lambdabot> with actual type ‘[i0]’Couldn't match expected type ‘GHC.Arr.Arr...
22:21:27 <lambdabot> with actual type ‘a0 -> GHC.Types.Bool’
22:22:35 <Taneb> oerjan, traverse rather than each
22:22:54 <oerjan> hm maybe. why would that matter?
22:23:41 <oerjan> actually it works in ghci
22:24:31 <oerjan> > indexing (backwards traverse) . indices odd %~ (*2) $ [1,2,3,4]
22:24:33 <lambdabot> -> Data.Functor.Identity.Identity a1)
22:24:33 <lambdabot> -> p0 t0 (Data.Functor.Identity.Identity b)’
22:24:33 <lambdabot> with actual type ‘[i0]’Couldn't match expected type ‘GHC.Arr.Arr...
22:24:33 <lambdabot> with actual type ‘a0 -> GHC.Types.Bool’
22:24:55 <Taneb> > indexing (backwards traverse) . indices odd *~ 2 $ [1,2,3,4]
22:24:57 <lambdabot> -> Data.Functor.Identity.Identity a0)
22:24:57 <lambdabot> -> p0 t0 (Data.Functor.Identity.Identity b)’
22:24:57 <lambdabot> with actual type ‘[i0]’Couldn't match expected type ‘GHC.Arr.Arr...
22:24:57 <lambdabot> with actual type ‘a1 -> GHC.Types.Bool’
22:25:12 <oerjan> oh right that's even shorter
22:25:18 <oerjan> but still no deal with lambdabot
22:25:32 <oerjan> int-e: why doesn't lambdabot handle this :(
22:25:49 <Taneb> > (indexing (backwards traverse) . indices odd) *~ 2 $ [1,2,3,4]
22:25:50 <lambdabot> -> Data.Functor.Identity.Identity a0)
22:25:51 <lambdabot> -> p0 t0 (Data.Functor.Identity.Identity b)’
22:25:51 <lambdabot> with actual type ‘[i0]’Couldn't match expected type ‘GHC.Arr.Arr...
22:25:51 <lambdabot> with actual type ‘a1 -> GHC.Types.Bool’
22:26:17 <Taneb> > (indexing (backwards traverse) . indices odd :: Traversal' [Integer] Integer) *~ 2 $ [1,2,3,4]
22:26:18 <lambdabot> Couldn't match expected type ‘(GHC.Integer.Type.Integer
22:26:18 <lambdabot> -> p0 GHC.Integer.Type.Integer (f GHC.Integer....
22:26:18 <lambdabot> with actual type ‘[i0]’Couldn't match expected type ‘GHC.Arr.Arr...
22:26:18 <lambdabot> with actual type ‘a0 -> GHC.Types.Bool’
22:27:15 <oerjan> :t indexing (backwards traverse) . indices odd
22:27:16 <lambdabot> Couldn't match expected type ‘a -> p0 a1 (f b)’
22:27:16 <lambdabot> Possible cause: ‘indices’ is applied to too many arguments
22:28:16 <Taneb> > (indexing (backwards traverse) . Control.Lens.indices odd) *~ 2 $ [1,2,3,4]
22:28:17 <lambdabot> Not in scope: ‘Control.Lens.indices’
22:28:17 <lambdabot> ‘Control.Lens.indexed’ (imported from Control.Lens),
22:28:17 <lambdabot> ‘Control.Lens.inside’ (imported from Control.Lens),
22:28:17 <lambdabot> ‘Control.Lens.iuses’ (imported from Control.Lens)
22:30:18 <oerjan> @ask int-e Why doesn't lambdabot have Control.Lens.indices twh
22:35:26 -!- yorick has quit (Remote host closed the connection).
22:36:22 <zzo38> I have now released a SQLite extension library. If any of you use SQLite, maybe it can be useful for you in some cases.
22:36:36 <elliott> oerjan: because Array, I think
22:36:46 <zzo38> gopher://zzo38computer.org/1sqlite http://zzo38computer.org/sql/sqlite.zip http://zzo38computer.org/sql/sqlite.txt
22:36:47 <shachaf> oerjan: import Control.Lens as Lens
22:36:50 <lambdabot> (Indexable i p, Applicative f) => (i -> Bool) -> Optical' p (Indexed i) f a a
22:37:05 <elliott> oerjan: yeah, I hid indices because
22:37:13 <elliott> arguably the lens one is more useful
22:38:22 -!- M28 has quit (Ping timeout: 258 seconds).
22:38:40 <oerjan> anyway, i got it tested and my SO comment made
22:39:12 -!- M28 has joined.
22:39:15 <Taneb> You do realise GHCi exists?
22:40:41 <oerjan> Taneb: i am a haskell exhibitionist
22:41:01 <oerjan> (and i have winghci open in another window already)
22:54:14 <zzo38> Some people say that SQL is not a real programming language, but I don't agree with that!!!
23:06:23 <zzo38> Some pinball game seem to be designed under the assumption that a free game is worth a lot more than anything else. I disagree; I think that it is usually not worth much at all but that precisely how much a free game is worth depends on the situation.
23:08:36 -!- mhi^ has quit (Quit: Lost terminal).
23:09:50 -!- nooodl has quit (Quit: Ik ga weg).
23:10:59 <int-e> @type Control.Lens.indices
23:11:00 <lambdabot> (Indexable i p, Applicative f) => (i -> Bool) -> Optical' p (Indexed i) f a a
23:11:14 <int-e> oerjan: it's there, but array came first.
23:11:50 <oerjan> > Control.Lens.indices
23:11:51 <lambdabot> Not in scope: ‘Control.Lens.indices’
23:11:51 <lambdabot> ‘Control.Lens.indexed’ (imported from Control.Lens),
23:11:51 <lambdabot> ‘Control.Lens.inside’ (imported from Control.Lens),
23:11:51 <lambdabot> ‘Control.Lens.iuses’ (imported from Control.Lens)
23:12:34 <oerjan> having to guess module abbreviations is annoying, though.
23:12:56 <oerjan> > Control.Lens.indexed
23:12:57 <lambdabot> No instance for (Data.Typeable.Internal.Typeable b0)
23:12:57 <lambdabot> arising from a use of ‘M678612761240765152220706.show_M6786127612407651522...
23:12:57 <lambdabot> The type variable ‘b0’ is ambiguous
23:12:57 <lambdabot> Note: there are several potential instances:
23:12:57 <lambdabot> instance Data.Typeable.Internal.Typeable Data.Dynamic.Dynamic
23:13:01 <lambdabot> Perhaps you meant ‘chars’ (imported from Data.ByteString.Lens)
23:13:25 <oerjan> > Control.Lens.indexed
23:13:27 <lambdabot> No instance for (Data.Typeable.Internal.Typeable b0)
23:13:27 <lambdabot> arising from a use of ‘M21213410986786399420731.show_M21213410986786399420...
23:13:27 <lambdabot> The type variable ‘b0’ is ambiguous
23:13:27 <lambdabot> Note: there are several potential instances:
23:13:27 <lambdabot> instance Data.Typeable.Internal.Typeable Data.Dynamic.Dynamic
23:13:33 <int-e> so you have to know that it does import qualified Control.Lens as Lens
23:13:41 <lambdabot> No instance for (Data.Typeable.Internal.Typeable
23:13:41 <lambdabot> Control.Lens.Internal.Indexed.Indexed)
23:13:41 <lambdabot> arising from a use of ‘M462241670695368513520745.show_M4622416706953685135...
23:13:41 <lambdabot> The type variables ‘i0’, ‘p0’ are ambiguous
23:13:51 <lambdabot> What lambdabot has in scope is at <http://code.haskell.org/lambdabot/State/L.hs>
23:13:52 <oerjan> ...but Control.Lens.indexed _does_ work...
23:15:21 <int-e> > Control.Lens.indices
23:15:22 <lambdabot> No instance for (Data.Typeable.Internal.Typeable
23:15:22 <lambdabot> Control.Lens.Internal.Indexed.Indexed)
23:15:22 <lambdabot> arising from a use of ‘M550549437821123700020776.show_M5505494378211237000...
23:15:22 <lambdabot> The type variables ‘i0’, ‘p0’ are ambiguous
23:15:37 <int-e> (I did the same for the other uses of "hiding")
23:15:49 <int-e> Did you know that "R" is System.Random?
23:16:13 <elliott> I apologise for contributing to the mess of the qualification.
23:16:16 <oerjan> that certainly looks random hth
23:16:21 <int-e> elliott: oh my, that's outdate
23:17:32 <int-e> https://github.com/int-e/lambdabot/blob/config/lambdabot/State/Pristine.hs is closer to the truth.
23:24:47 <HackEgo> [U+039C GREEK CAPITAL LETTER MU]
23:27:28 <HackEgo> [wiki] [[ΜCurse]] http://esolangs.org/w/index.php?diff=39698&oldid=39653 * Oerjan * (-4) Lowercase template should work as it's a greek capital M. Also intro format.
23:28:04 <HackEgo> [wiki] [[ΜCurse]] M http://esolangs.org/w/index.php?diff=39699&oldid=39698 * Oerjan * (-1) less space
23:30:26 -!- aretecode has joined.
23:33:20 <oerjan> int-e: that FIXME looks like it has already been fixed?
23:33:27 <lambdabot> What lambdabot has in scope is at <http://code.haskell.org/lambdabot/State/L.hs>
23:33:57 <oerjan> @where+ l.hs What lambdabot has in scope is at < https://github.com/int-e/lambdabot/blob/config/lambdabot/State/Pristine.hs>
23:34:00 <lambdabot> What lambdabot has in scope is at < https://github.com/int-e/lambdabot/blob/config/lambdabot/State/Pristine.hs>
23:34:16 <oerjan> @where+ l.hs What lambdabot has in scope is at <https://github.com/int-e/lambdabot/blob/config/lambdabot/State/Pristine.hs>
23:34:25 -!- Phantom_Hoover has quit (Quit: Leaving).
23:34:57 <boily> quintopia: QUINTOPIAAAAAAAAAAAAAH!
23:39:53 -!- MDude has quit (Ping timeout: 264 seconds).
23:41:55 <HackEgo> [wiki] [[Language list]] M http://esolangs.org/w/index.php?diff=39700&oldid=39692 * Oerjan * (+0) /* Non-alphabetic */ Ordo Novo
23:46:25 <HackEgo> [wiki] [[Joke language list]] http://esolangs.org/w/index.php?diff=39701&oldid=39691 * Oerjan * (+1) wat.
23:51:29 -!- edwardk has joined.
23:52:36 -!- conehead has joined.
23:54:16 <HackEgo> [wiki] [[Brainfuck implementations]] http://esolangs.org/w/index.php?diff=39702&oldid=39609 * Oerjan * (+20) consistent punctuation
23:55:20 <int-e> oerjan: stuff got out of order; as far as I understand, the comment also applies to deriving instance Show (f (Mu f)) => Show (Mu f) ... anyway it wasn't added by myself :)