00:48:32 -!- conehead has joined.
00:53:58 -!- nooodl has quit (Quit: Ik ga weg).
00:55:13 -!- tromp has quit (Remote host closed the connection).
00:55:48 -!- tromp has joined.
01:00:31 -!- tromp has quit (Ping timeout: 264 seconds).
01:00:54 -!- tromp has joined.
01:23:28 -!- ais523 has quit.
01:57:54 -!- Bike has quit (Ping timeout: 255 seconds).
02:09:52 -!- tromp has quit (Remote host closed the connection).
02:10:26 -!- tromp has joined.
02:11:15 -!- Bike has joined.
02:15:00 -!- tromp has quit (Ping timeout: 255 seconds).
02:19:17 <Taneb> I just realised something
02:19:27 <Taneb> I find the concept of living near one's cousins really alien
02:20:26 <Taneb> Also I need to sleep
02:20:31 <Taneb> Goodnight #esoteric
02:23:03 -!- yorick has quit (Remote host closed the connection).
02:23:10 <oerjan> in particular, Taneb's cousin It
02:23:35 <Taneb> oerjan, all my cousins live in Australia
02:24:23 <Sgeo_> These things sound surprisingly good once I realize that there are two devices and I need to be sending to stereo
02:36:22 -!- tromp has joined.
02:47:21 -!- tromp has quit (Remote host closed the connection).
02:47:54 -!- tromp has joined.
02:49:37 -!- tertu has quit (Ping timeout: 240 seconds).
02:50:16 -!- tromp has quit (Remote host closed the connection).
02:50:29 -!- tromp has joined.
03:03:47 -!- nisstyre has joined.
03:04:52 -!- conehead has quit (Quit: Computer has gone to sleep.).
03:07:38 -!- HackEgo has quit (Ping timeout: 240 seconds).
03:17:45 -!- oerjan has quit (Quit: leaving).
03:24:11 -!- tertu has joined.
03:32:26 -!- tertu_ has joined.
03:32:53 -!- tertu has quit (Ping timeout: 252 seconds).
03:51:17 -!- Sorella has quit (Quit: It is tiem!).
03:59:04 -!- tromp has quit (Remote host closed the connection).
03:59:37 -!- tromp has joined.
04:02:26 -!- shikhout has joined.
04:04:14 -!- tromp has quit (Ping timeout: 252 seconds).
04:04:41 -!- nisstyre has quit (Quit: WeeChat 0.4.3).
04:05:19 -!- shikhin has quit (Ping timeout: 264 seconds).
04:08:55 -!- shikhout has quit (Ping timeout: 264 seconds).
04:09:10 -!- tromp has joined.
04:11:41 -!- conehead has joined.
04:37:35 -!- tromp has quit (Remote host closed the connection).
04:37:55 -!- tromp has joined.
04:38:57 -!- tromp has quit (Remote host closed the connection).
04:39:15 -!- tromp has joined.
04:39:30 -!- tromp has quit (Remote host closed the connection).
04:40:17 -!- tromp has joined.
04:40:35 -!- tromp has quit (Remote host closed the connection).
04:40:48 -!- tromp has joined.
04:41:05 -!- tromp has quit (Remote host closed the connection).
04:41:40 -!- tromp has joined.
04:42:11 -!- tromp has quit (Remote host closed the connection).
04:42:52 -!- tromp has joined.
04:47:19 -!- tromp has quit (Ping timeout: 264 seconds).
05:26:39 -!- Slereahphone has quit (Quit: Colloquy for iPhone - http://colloquy.mobi).
05:33:13 <Sgeo_> "(Yes, eval() is used here, leaving you at mortal risk of XSS-attacking yourself.)"
05:39:29 -!- shikhin has joined.
05:45:53 -!- tertu_ has quit (Ping timeout: 246 seconds).
05:52:43 -!- Slereah__ has quit (Ping timeout: 264 seconds).
05:53:32 -!- tertu has joined.
06:07:19 -!- Bike has quit (Ping timeout: 264 seconds).
06:28:51 <shachaf> kmc: http://www.crazygames.com/game/9007199254740992
06:51:07 -!- tertu has quit (Ping timeout: 264 seconds).
06:53:20 -!- Slereahphone has joined.
07:00:07 -!- Slereahphone has quit (Remote host closed the connection).
07:01:08 -!- Slereahphone has joined.
07:02:42 -!- ais523 has joined.
07:03:25 <zzo38> How do I use other programs on my computer with a SSH tunnel?
07:03:57 <zzo38> I have Cygwin and PuTTY and netcat all installed.
07:17:39 <kmc> shachaf: yesssssss
07:36:48 <fizzie> 2^53 is arguably a strange number to stop at, given that all the values involved would be powers of two and therefore exactly representable even beyond the integer values in a double.
07:37:10 <ais523> fizzie: many people don't really understand how floats work
07:38:13 -!- password2 has joined.
07:48:10 -!- Slereahphone has quit (Read error: Connection reset by peer).
07:48:24 -!- MindlessDrone has joined.
07:51:36 -!- Slereahphone has joined.
07:55:55 <Jafet> Did you know that 2048 is isomorphic to a game where the largest number involved is 11
07:56:47 -!- password2 has quit (Remote host closed the connection).
07:58:04 <ais523> also to a game where the largest number involved is -3, although that might be a bit less intuitive
07:58:43 <Jafet> http://www.twitch.tv/twitchplays2048
08:01:19 -!- password2 has joined.
08:04:10 <shachaf> ais523: is -3 just an arbitrary choice here or interesting in some way
08:04:23 <ais523> only I wanted it to be negative
08:05:21 <fizzie> For any game X, what's the expected time difference between the release of X and the appearance of Twitch Plays X?
08:06:15 <fizzie> (Also, should high-level chess tournaments have a Twitch Plays Chess competitor?)
08:06:45 <Jafet> http://en.wikipedia.org/wiki/Kasparov_versus_the_World
08:07:06 <fizzie> Now that you mention it, I vaguely recall that.
08:07:43 <Jafet> 50 thousand people... about the same number of participants as tpp
08:08:04 <fizzie> It didn't have a democracy/anarchy vote, though. Not that I think that would've improved the World Team's chance of winning.
08:08:37 <Jafet> "STOP PRESSING RESIGN"
08:08:45 <zzo38> Jafet: Is that because Kasparov is a good chess player, or because of the way the world team's moves are decided?
08:12:12 <fizzie> This dropped in my inbox a moment ago, and is possibly even slightly related to the topic: http://sprunge.us/jbIC
08:13:23 <ais523> how do they prove that the humans are actually picking the median value and not just pressing buttons at random?
08:15:43 <Jafet> I wonder how large their "experiments" were
08:16:35 <fizzie> They don't; the proof seems to be just for the fact that the process (the whole deal with the triplets) coincides with the mean if the triplets are samples from a univariate normal distribution.
08:16:51 <Jafet> On a relatedly related note, it seems that you can ignore the non-captcha part of a recaptcha and it will still let you pass
08:17:43 <fizzie> For the semantic-clustering-of-images task, it's not like there'd be an actual correct median value, anyway.
08:18:29 <ais523> recaptcha no longer serves its original purpose
08:18:46 <ais523> it used to be that both halves were words that Google's OCR couldn't handle
08:18:51 <ais523> but the spambots got good at it
08:19:04 <ais523> and instead of just using the spambots to digitize books, they screwed everything up
08:19:11 <ais523> and now the spambots don't have to bother with the digitization
08:19:59 -!- Sgeo has joined.
08:20:01 <fizzie> How did the check work if both words were un-OCRable
08:20:26 <coppro> ais523: what do you mean by "the spambots don't have to bother with the digitization"?
08:20:39 <ais523> fizzie: one of the words would be a word that had already been checked by many humans
08:21:09 <ais523> coppro: it's easy to tell which of the two words/number sequences is being digitized, and which one is random
08:21:37 <coppro> ah, I see what you mean
08:22:09 <coppro> and of course you can't use voting or anything because the spambots outnumber the humans
08:22:17 <zzo38> Is the source codes for those spambots available?
08:23:32 <ais523> most likely no, although you never know
08:23:35 <Jafet> In principle you don't need the source code, you just trade their OCR services for a free email account or whatever.
08:24:00 <zzo38> Such a thing would be useful, for several purposes, such as: to figure out how to prevent such spam, to port it to another computer, or if it implements OCR, to improve open source OCR stuff.
08:24:10 <ais523> I know that as part of my teaching job, my boss asked me to make a system to email a bunch of students their feedback for the week
08:24:17 <zzo38> (And to answer questions, if it is a question CAPTCHA)
08:24:25 -!- password2 has quit (Ping timeout: 240 seconds).
08:24:43 <ais523> and looking up libraries to do so, I found one that was designed for sending a huge number of emails to a particular server in a short space of time
08:25:00 <zzo38> Spambots do other stuff too of course, and some of these may be useful to have source codes for, for various purposes.
08:25:00 <ais523> I thought "hmm, that seems useful for spambots", and then used it anyway because it was what I needed
08:25:57 <zzo38> Why doesn't the shift-in and shift-out controls work in PuTTY?
08:26:47 <ais523> zzo38: because PuTTY cannot handle character set switching when UTF-8 is turned on, which is probably a bug
08:26:58 <ais523> they will work if you set it to, say, Latin-1
08:28:53 -!- Sgeo_ has quit (Ping timeout: 240 seconds).
08:28:53 -!- FireFly has quit (Ping timeout: 240 seconds).
08:30:17 <zzo38> In the session I used that, it was set to Latin-1. I tried changing it to CP437 and it still doesn't work.
08:30:32 -!- FireFly has joined.
08:31:58 <ais523> zzo38: another thing to check: SI and SO change between character sets G0 and G1, which are both customizable
08:32:08 <ais523> if they're set to the same thing, then SI and SO will have no visible effect
08:32:31 <zzo38> I intended it to be set to VT100 characters
08:32:38 <ais523> the normal setup would be "ESC ( B ESC ) 0"
08:32:46 <ais523> which sets G0 to Latin-1 and G1 to VT100
08:34:23 <zzo38> Can the initial setting be configured in PuTTY?
08:35:39 <zzo38> Why isn't that the default setting anyways?
08:35:56 <zzo38> It seems like setting G1 to VT100 ought to be the default setting.
08:39:01 -!- password2 has joined.
08:39:58 <zzo38> I have LANG=C set up but now I found another environment variable called LANGUAGE, what is that one for?
08:51:45 -!- conehead has quit (Quit: Computer has gone to sleep.).
08:57:30 <zzo38> I have idea about chess, where you may consider draws sometimes to be a half-win or half-lose.
08:58:34 <zzo38> For example, a stalemated player is half-lose, player whose move is the third of a three-times-repetition is half-lose.
08:58:40 -!- Slereahphone has quit (Read error: Connection reset by peer).
08:58:59 -!- Slereahphone has joined.
09:08:51 -!- Slereahphone has quit (Quit: Colloquy for iPhone - http://colloquy.mobi).
09:10:42 -!- Slereahphone has joined.
09:18:05 <fizzie> zzo38: For glibc, you can put a list of languages in "LANGUAGE", and that will be used as a priority order for languages for messages; it overrides LANG, LC_MESSAGES and LC_ALL, except if those specify "C", in which case it has no effect. I don't think it's included in POSIX.
09:18:32 <fizzie> (The idea being that you can specify the fallback language if there are no translations for your preferred language.)
09:18:54 -!- password2 has quit (Ping timeout: 255 seconds).
09:34:32 -!- Slereahphone_ has joined.
09:34:49 -!- Slereahphone has quit (Ping timeout: 240 seconds).
09:35:00 -!- Slereahphone_ has changed nick to Slereahphone.
09:37:43 -!- shikhin has quit (Ping timeout: 264 seconds).
09:40:51 -!- shikhin has joined.
09:55:19 -!- Slereahphone has quit (Ping timeout: 264 seconds).
09:57:31 -!- Slereahphone has joined.
10:02:34 -!- shikhout has joined.
10:03:37 <zzo38> I have specified "C" in LANG.
10:05:42 <zzo38> Which I think always should be used.
10:05:55 -!- shikhin has quit (Ping timeout: 264 seconds).
10:05:56 -!- shikhout has changed nick to shikhin.
10:06:46 <zzo38> But you might want to override for messages only, in case you prefer non-English languages
10:08:33 <ais523> oh, btw, I found another compiler bug yesterday; the version of clang in the repos here produces warning messages for non-taken branches of a _Generic expression, even though they aren't evaluated and so there's no point in producing warnings for them
10:11:10 <fizzie> Does it also produce errors?
10:12:11 <ais523> that's worth testing, but that might arguably not violate the standard
10:12:22 <ais523> non-taken branches of a _Generic are sort-of like the argument to sizeof
10:12:36 <ais523> it's acknowledged that it exists, it just is never evaluated
10:12:45 <fizzie> Mine seems to, for (at least some) constraint violations.
10:12:48 <ais523> maybe clang isn't buggy after all, maybe it's the standard that's buggy
10:14:44 <fizzie> Could be. It does make _Generic a bit less generically (no pun intended) applicable.
10:15:40 <fizzie> E.g. I tried out _Generic(p, int: p ^= 1, char *: p += 1) for a char *p, and you could argue that should be okay, since ^= 1 makes sense for an int, but according to clang (and very possibly the standard) it's a constraint violation nevertheless.
10:15:51 <ais523> yeah, I was trying to use it to discover which of the arguments to macros are string literals
10:16:19 <ais523> this is probably worth taking to Usenet
10:16:39 <ais523> it'd be nice to have a way to suppress warnings, at least
10:16:53 <ais523> because I can initialize a char * with a struct if I really have to
10:17:00 <ais523> I guess I could cheat and cast via void *
10:17:05 <ais523> which would suppress the warning
10:17:16 <ais523> this is all pointless, anyway, because gcc doesn't do _Generic yet
10:18:00 -!- Tritonio has joined.
10:19:31 <fizzie> You can take the quasi-portable approach and use GCC extensions on GCC and (try to use) _Generic elsewhere.
10:20:13 <fizzie> (Also it's going to be in 4.9.)
10:23:31 <fizzie> _Generic(e, a: x, b: y, default: z) is at least approximately __builtin_choose_expr(__builtin_types_compatible(__typeof__(e), a), x, __builtin_choose_expr(__builtin_types_compatible(__typeof__(e), b), y, z)), after all.
10:23:59 <ais523> except that the standard version of _Generic errors if multiple types match
10:25:15 <Taneb> nethack: how do I open a box
10:25:29 <ais523> or in NetHack 4, a then ,
10:25:41 <ais523> (#loot also works in NH4 but a, is easier to type)
10:25:54 <ais523> (and more consistent with the other controls)
10:25:55 <fizzie> I'm sure you can work that in too. (Static assert on the sum of 0s or 1s chosen based on type compatibility <= 1.)
10:26:32 <Taneb> How common are keys?
10:26:45 <ais523> not that rare, you're more likely to find them in the Mines
10:26:51 <ais523> either randomly, or purchasing them from shops
10:27:11 <ais523> also you can normally get the box open using sufficient violence, but that tends to have bad side effects
10:27:27 <ais523> (kick it with ^D or use your weapon with #force; don't #force if you have a valuable weapon, it might break)
10:29:02 -!- chaiomanot has quit (Ping timeout: 246 seconds).
10:29:16 <fizzie> You might also come across a lock pick, which is similar to a key except with slightly different properties.
10:29:36 <ais523> (or very occasionally, larger)
10:30:31 <fizzie> The weight seems to be consistently larger.
10:31:35 <ais523> I was referring to "chance to open a lock per turn"
10:31:43 <ais523> which is the main performance difference between unlocking tools
10:31:47 -!- boily has joined.
10:31:48 <zzo38> fizzie: I like it with the GCC extensions, which I think is a more useful and more versatile way.
10:32:00 <zzo38> (In GNU mode you can use "typeof" without underscores.)
10:46:45 <elliott> ais523: "effects whether" -> "affects whether", and that malloc/strcpy is shorter as strdup("submenu") if you can use strdup/have an implementation lying around
10:46:51 <elliott> ais523: no credit desired for either
10:47:41 <ais523> elliott: someone on Hacker News already pointed out strdup, but I've actually been removing strdup calls, because they're POSIX rather than C11 and this actually comes up sometimes
10:47:52 <ais523> and my blog posts frequently have typos in
10:47:56 <elliott> you can just write your own strdup and save the LOC everywhere
10:47:59 <ais523> that happens when you write ten thousand words in two days
10:48:12 <elliott> it's a good thing you have me to tell you the typos so you can fix them, then
10:48:22 <ais523> I just left them unfixed in the other posts
10:48:30 <ais523> this is getting enough exposure that I might have to fix them, though
10:48:45 <zzo38> I use strdup though, it is available in Windows and in UNIX systems; you can add an implementation of strdup if needed.
10:48:48 <ais523> there was also a debate on the thread about whether strdup'ping a constant is the best idea, because if it isn't inlined, it'll need to count the characters in it every time through
10:49:05 <ais523> I didn't expect the post to get as popular as it did
10:49:08 <elliott> ais523: btw, you also don't check for malloc failure...
10:49:14 <elliott> does the real code use a malloc wrapper?
10:49:18 <zzo38> ais523: Can it inline it though?
10:49:28 <ais523> elliott: I think it used to, then daniel_t changed it so it didn't
10:49:53 <elliott> I mean Linux's default configuration isn't going to give you NULL back, but...
10:50:19 <ais523> also, running out of stack gives you a segfault, arguably running out of heap should to
10:50:27 <elliott> ais523: anyway, return "submenu"; would work if you appropriately const-marked the return value, right?
10:50:31 <ais523> it's pretty easy to change whether we have malloc wrapped or not, though
10:50:39 <ais523> yeah, it's to allow free
10:50:51 <elliott> free() on string constants should be specified to do nothing or something
10:50:59 <ais523> I was about to say that
10:51:05 <elliott> ais523: on Linux, it doesn't necessarily give you a segfault
10:51:06 <ais523> that's technically viable in most C implementations
10:51:09 <elliott> it might kill Firefox instead
10:51:10 <zzo38> elliott: I don't agree it shouldn't specify
10:51:25 <ais523> elliott: it does 100% give you a segfault unless you explicitly mmap'ed over NULL
10:51:25 <elliott> ais523: which is rather more annoying
10:51:29 <ais523> or you're running in kernel mode
10:51:32 <zzo38> free("submenu"); should be an undefined behavior
10:51:36 <ais523> oh, you mean the running out of memory
10:51:38 <elliott> it doesn't bother to return NULL when it's out of memory
10:51:47 <ais523> you can set the overcommit behaviour to 2 to disable that
10:51:49 <elliott> it just gives pointers and then kills ~a random process when you try to use them
10:52:06 <ais523> really, though, we need a 3, which starts refusing allocations to programs that are using lots of memory when they get near, say 80% of the memory that isn't used by other prorgams
10:52:14 <elliott> I'm saying that your current behaviour isn't "segfault when out of memory", it's "kill Firefox and annoy the user when out of memory"
10:52:26 <ais523> and wrapping malloc doesn't change that behaviour either :-)
10:52:29 <elliott> though I guess there's nothing you can do about that, yeah
10:52:45 <elliott> we need an is_this_pointer_real syscall
10:52:47 <ais523> btw, the only time I actually ended up with the oom-killer triggering, IIRC it hit the right process
10:52:48 <Taneb> elliott, does @ solve this problem
10:52:50 <ais523> whereas malloc fail might not have done so
10:53:03 <elliott> Taneb: I'm not working on @.
10:53:06 <ais523> Taneb: it solves it by automatically purchasing more memory from cloud storage
10:53:29 <Taneb> elliott, I meant that more as a joke than aserious question
10:54:55 -!- Slereahphone has quit (Remote host closed the connection).
10:55:12 <zzo38> My own C extensions solve that problem with the strdup by allowing you to prefix the declaration with ["implements"("strdup",ptr)] so that the compiler can inline it and/or perform optimizations based on it, for one thing.
10:55:22 <ais523> actually, thrashing is the big problem in practice
10:55:50 <ais523> if you're getting near to running out of memory and have nonzero swap, your computer is probably unusably slow
10:56:12 <ais523> and you'll be glad it killed Firefox to free up enough memory that you can open up a terminal and deal with the problem before memory fills again
10:56:44 <zzo38> You should program the computer not to kill processes when one program runs out of memory, unless it is a privileged process which explicitly has a code in it that tells it to try doing such things.
10:57:39 <zzo38> Also have a SysRq command which you can push to tell it to stop a process in case you need to run other programs to deal with it temporarily.
10:57:50 <ais523> this is basically all caused by our programming languages being TC and our computers not being
10:58:36 <ais523> actually, I guess the real reason for NetHack is that if it gets anywhere close to memory exhaustion, there's a leak somewhere
10:58:57 <zzo38> Well, yes you should check that, in case that is actually the problem.
10:59:56 <elliott> ais523: I might disable overcommit, except that I don't trust programs to be robust enough to even crash cleanly in such a scenario.
11:00:28 <ais523> the overcommit setting rarely does anything at all, really, because it's rare that the computer actually does run out of memory
11:00:38 <ais523> unless you set it to 1, never do that
11:00:59 <ais523> (it causes the kernel to overcommit even for completely implausible amounts of memory)
11:06:59 -!- boily has quit (Quit: UNDERCOMMIT CHICKEN).
11:34:41 -!- AveCaesar has joined.
11:51:30 -!- Slereahphone has joined.
12:06:34 -!- tertu has joined.
12:08:14 -!- Slereahphone has quit (Quit: Colloquy for iPhone - http://colloquy.mobi).
12:12:32 -!- oerjan has joined.
12:23:26 <oerjan> bloody dog is at it again, for 2 1/2 hours.
12:24:38 -!- aergus has joined.
12:24:43 -!- Sorella has joined.
12:29:34 -!- Sgeo has quit (Read error: Connection reset by peer).
12:36:50 -!- tertu has quit (Ping timeout: 252 seconds).
12:37:46 <oerjan> i think the dog may be getting tired, it's occasionally taking several second breaks.
12:38:31 -!- Tritonio1 has joined.
12:38:45 <oerjan> perhaps it's finally despairing of anyone listening.
12:39:22 <oerjan> No, Dog, No One is Listening to You. No One You would Want to Meet, Anyhow.
12:41:31 -!- Tritonio has quit (Ping timeout: 264 seconds).
12:42:35 -!- Slereahphone has joined.
12:44:26 <oerjan> i am not quoting shakespeare. just snapping because of a barking dog, is all.
12:44:43 <AveCaesar> just kidding.. that no dog quote.. it's strange i've been reading so much poetry about dogs past days
12:45:23 <AveCaesar> you know..it's not much about writing.. "just sit down and bleed" -hemingway
12:45:39 <elliott> oerjan: I wonder who this guy is???!?!?!?!?
12:45:54 <oerjan> does it say anything about how to stop other people's dogs barking? yahoo answers was very disappointing.
12:45:54 -!- ChanServ has set channel mode: +o elliott.
12:45:59 -!- elliott has set channel mode: +b *!*AuePrum@46.115.88.*.
12:45:59 -!- elliott has kicked AveCaesar WE'LL NEVER KNOW.
12:46:50 -!- elliott has set channel mode: +b *!*@46.115.*.
12:47:01 <elliott> one of his recent evasions didn't match that mask
12:47:06 <elliott> and two other people over the life of the channel have used it
12:47:12 -!- elliott has set channel mode: -b *!*@46.115.*.
12:47:38 -!- elliott has set channel mode: -o elliott.
12:47:46 <elliott> I guess we just have to keep playing cat and mouse until he gets bored
12:49:53 <zzo38> elliott: No, it won't work???
12:50:36 <zzo38> To "keep playing cat and mouse until he gets bored".
12:50:57 <elliott> well, either he gets bored first, I get bored first, or neither of us ever get bored of it
12:51:11 <elliott> if it's the latter, then I think he is probably older enough than me that he is likely to die first
12:51:23 <zzo38> That depends on how persistent they are (and you).
12:51:28 <elliott> and I think this is more work for him than me
12:51:42 <elliott> and also I think I'm more competent than him. so I give myself good odds.
12:52:02 <zzo38> No, he is going to be really persistent.
12:52:13 <zzo38> It cannot be stopped.
12:53:20 -!- tertu has joined.
12:53:21 <zzo38> I recommending all persistence
12:54:02 <elliott> well, he seems to at least be bothered by me reacting to his ban evasion, since he sent me a lambdabot message telling me to grow up one of the times
12:54:08 <elliott> I *suspect* he has finite patience, though I cannot be sure.
12:54:59 <oerjan> zzo38: even for dogs? YOU WICKED MAN
12:55:28 <zzo38> oerjan: Even for dogs, it doesn't matter if it is dogs or not, I suppose
12:55:45 <zzo38> Finite persistence isn't same as being small, though.
12:55:48 <Taneb> oerjan, you could confront the owner
12:55:49 <oerjan> oklopol: because he was already banned
12:56:16 <oklopol> you are not feeding my curiosity
12:56:21 <elliott> it doesn't even matter why he was banned at this point, since his reaction since has been to repeatedly evade and insult ops and others (incl. behind their back) as an appeals strategy
12:56:31 <oklopol> perhaps you like cats more than dogs
12:56:38 <zzo38> You should forward banned people to a #esoteric#shadow channel (which is owned by glogbot for this purpose)
12:56:48 <oklopol> i'm not saying it _matters_, i'm just curious.
12:56:50 <oerjan> no, i actually like dogs more, in person.
12:57:24 <elliott> I don't actually remember the original thing
12:57:31 <elliott> I think he was banned multiple times before it became permanent.
12:58:16 <oerjan> maybe we just banned him to check if he was an evil ban evader, which he was.
12:58:41 <elliott> clearly we should just ban everyone by default
12:58:43 <elliott> to preemptively weed out the evaders
12:58:47 <oerjan> (we do like recursion in this channel)
12:59:42 <oerjan> Taneb: but the owner is only there when the dog doesn't bark, at which time i'm all busy being serene and calm.
12:59:53 <zzo38> That is why I made up the shadow channel and made it so that nobody can ban anyone.
13:00:06 <Taneb> oerjan, turn that serenity into ferocity or something
13:00:09 <zzo38> It is channel for banned people, although anyone can access it.
13:00:21 <zzo38> You have to provide forwarding.
13:00:43 <oerjan> Taneb: but when i'm ferocious enough to confront people, i want to kill them.
13:00:52 <Taneb> That may be a problem
13:01:00 <Taneb> Hire a confrontationist?
13:01:35 -!- tertu has quit (Ping timeout: 252 seconds).
13:04:50 <Melvar> < zzo38> I have specified "C" in LANG. < zzo38> Which I think always should be used. – Whatthewhat?
13:14:58 <zzo38> I found out very recently that immediate addressing mode in VAX is also program counter with postincrement, same as how some instruction sets I invented work, in the same way.
13:21:44 <zzo38> Except that mine also allows program counter indexed and all of that stuff too.
13:25:34 <zzo38> It looks like VAX also has a "time of year" register.
13:32:59 <Slereahphone> As long as it doesn't have a "time of the month" register
13:48:51 -!- tertu has joined.
13:51:05 -!- yorick has joined.
13:51:32 -!- Slereahphone has quit (Remote host closed the connection).
13:51:51 -!- Slereahphone has joined.
13:56:35 -!- tertu has quit (Ping timeout: 252 seconds).
14:18:59 <zzo38> Now I wrote about in Black-C, is the use of __error__() function, use of "static inline const" global arrays, and a list of many suggested metadata attributes (none of which need be implemented, and an implementation can add its own, too).
14:21:20 -!- Slereahphone has quit (Ping timeout: 252 seconds).
14:21:35 <zzo38> Metadata attributes are: aligned, alternate, always_return, arg_size, asm_name, associative, assume, breakpoint, cold, complex, commutative, decimal, expect, expect_false, expect_true, fast, fast_math, form, hot, implements, invert, never_return, penalty, prefetch, range, readnone, readonly, restrict, returned, section, self_bijective, unreachable, vector, vrange.
14:22:20 -!- Slereahphone has joined.
14:28:25 -!- Slereahphone has quit (Remote host closed the connection).
14:28:44 -!- Slereahphone has joined.
14:29:36 -!- Slereahphone has quit (Remote host closed the connection).
14:34:00 -!- Slereahphone has joined.
14:43:36 -!- spiette has joined.
14:46:03 -!- aergus has quit (Ping timeout: 255 seconds).
14:55:49 -!- oerjan has quit (Quit: leaving).
14:56:56 -!- password2 has joined.
14:57:30 -!- password2 has quit (Max SendQ exceeded).
14:57:55 -!- aergus has joined.
14:58:26 -!- password2 has joined.
14:59:08 -!- password2 has quit (Max SendQ exceeded).
15:01:08 -!- Slereah_ has joined.
15:02:35 -!- Slereahphone has quit (Ping timeout: 252 seconds).
15:11:05 -!- Bike_ has joined.
15:16:17 -!- Bike_ has changed nick to Bike.
15:17:27 -!- password2 has joined.
15:17:45 -!- password2 has quit (Max SendQ exceeded).
15:18:21 -!- password2 has joined.
15:18:47 -!- password2 has quit (Max SendQ exceeded).
15:19:26 -!- password2 has joined.
15:19:58 -!- password2 has quit (Max SendQ exceeded).
15:20:40 -!- password2 has joined.
15:51:45 <kmc> hackegocoin
15:53:49 -!- nooodl has joined.
15:54:01 -!- Bike has quit (Ping timeout: 240 seconds).
16:02:51 -!- shikhout has joined.
16:05:55 -!- shikhin has quit (Ping timeout: 264 seconds).
16:05:55 -!- shikhout has changed nick to shikhin.
16:13:54 -!- Bike has joined.
16:18:05 -!- tertu has joined.
16:18:58 -!- lambdabot has quit (Read error: Operation timed out).
16:20:41 -!- fungot has quit (Ping timeout: 252 seconds).
16:23:37 -!- tertu has quit (Ping timeout: 240 seconds).
16:27:03 -!- lambdabot has joined.
16:40:26 <fizzie> I don't want to start reimplementing `coins on fungot. :/
16:40:38 <fizzie> Hey, fungot isn't here either.
16:41:08 -!- fungot has joined.
16:41:31 <ais523> fungot: what is your opinion on HackEgo?
16:41:31 <fungot> ais523: i wonder why? it means i have to
16:41:45 <ais523> oh wow, what an awesomely relevant response
16:48:36 <ais523> it's a markov chain, I think
16:48:44 <ais523> initialized with names of esolangs and also something else, which I forget
16:51:54 <fizzie> One million English words, to be exact.
16:52:02 <ais523> are the esolangs more heavily weighted?
16:52:35 <fizzie> I believe it has uniform weights, but the frequencies are normalized, which means the corpus sizes do not affect the results.
16:52:43 <fungot> Selected style: fungot (What I've said myself)
16:53:19 <ais523> that got me thinking about initializing a markovbot with a markovbot
16:53:23 <ais523> then realised fungot already had that mode
16:53:23 <fungot> ais523: so, let's say i call them mindless games. if we hit every stupid person, any person going fast and such an apparently non-standard format is being dimensioned by assigning dimensions to them; and sets. maybe thats what them meant by experimental selected so it probably no longer the wand the better part made mercie, i should think at least, that is expressions which have not been able, to assume responsibilities. he wen
16:54:37 <fizzie> Well, it also contains output from all other fungot commands, incl. bf and ul programs.
16:54:37 <fungot> fizzie: ( ( ( a()**)a*:a*)(a()**)a*:a*)((x1)(x2)(x3)) ...out of time!
16:54:48 <fizzie> As was very nicely illustrated there, thank you.
16:56:38 <fizzie> (Discounting non-babble output, I believe the current style is asymptotically speaking equivalent to a weighted interpolation of all the other styles, but of course it hasn't spoken an infinite amount of data yet, so it's not quite.)
17:01:30 <nooodl> that was eerily well-timed illustration fungot
17:01:30 <fungot> nooodl: more simply put: siod sucks as a general purpose ( similar, and i'd like to see that mystical forest powers, but this time on the impact of the introduction to theoretical computer.) a variety of colorful fish, but the darkness. once this is false, another is a sgi indy ( mips, running under rc/ funge-98, using the strn, sock and fgrn fingerprints, running under rc/ funge-98, using the strn, sock and fgrn fingerprints,
17:03:15 -!- fizzie has set topic: a variety of colorful fish, but the darkness | https://dl.dropboxusercontent.com/u/2023808/wisdom.pdf http://codu.org/logs/_esoteric/ http://tunes.org/~nef/logs/esoteric/.
17:08:19 -!- aergus has quit (Ping timeout: 264 seconds).
17:21:44 -!- Tritonio1 has quit (Ping timeout: 255 seconds).
17:51:04 -!- tertu has joined.
17:56:21 -!- tertu has quit (Ping timeout: 264 seconds).
18:00:56 -!- aergus has joined.
18:22:40 <shikhin> http://esolangs.org/wiki/Thue -- what does a line which doesn't contain "::=" before the end-of-rules imply?
18:24:59 <Deewiant> ais523: Where do I report jettyplay bugs (if I should at all), darcsden.com is down (those two phrases may or may not be related)
18:25:16 <ais523> Deewiant: to me personally, and I know it's down
18:25:21 <ais523> although still responding to pushes/pullls
18:26:07 <shikhin> Any good esolangs you folks know of, which are easily implementable (like, say thue)?
18:26:14 <quintopia> elliott: what was the name of that tiny roguelike you never finished
18:26:28 <ais523> gah having a broken tab-complete is so annoying
18:26:39 <ais523> the tab button just arbitrarily started changing focus, rather than completing
18:26:49 <elliott> shikhin: Underload, ///, Deadfish
18:26:59 <ais523> Deewiant: not up to it right now, but I'll move the repo to nethack4.org some time
18:27:01 <elliott> one of these si not serious
18:27:04 <quintopia> shikhin: many many esolangs are designed with easy implementation in mind
18:27:05 <Deewiant> ais523: It opened a bzip2'd nethack (3.4.3) ttyrec just fine but it ended playback at / counted the frames only up to the first "Be seeing you...", works when bunzip2'd
18:27:13 <ais523> elliott: Deadfish is never serious :-)
18:27:20 <elliott> ais523: that wasn't the one that wasn't serious
18:27:26 <quintopia> ais523: but always easily implemented
18:27:26 <Bike> it is easy to implement though
18:27:26 <Deewiant> Or at least it counted further, didn't check if it actually makes it to the end but I'd assume so
18:27:32 <ais523> Deewiant: huh, I wonder if the file in question is multiple bzip files concatenated
18:27:39 <Bike> so is underload. shouldn't /// be easy,i haven't tried
18:27:52 <Deewiant> ais523: I doubt it since it's a ttyrec I created myself and I just bzip2'd it to save space
18:28:10 <quintopia> ResPlicate is easy to implement :P
18:28:15 <shikhin> Yeah, those three I know of.
18:28:40 <Deewiant> ais523: Actually, you're right, it was made with termrec -a
18:28:44 <quintopia> shikhin: you could implement ETAS. it's never been implemented, and I think it'd be pretty easy!
18:28:53 <Deewiant> ais523: So it probably is multiple concatenated ones
18:28:56 <ais523> the bunziper I ported over probably can't handle that format
18:29:17 <shikhin> quintopia: Looking at it :-)
18:30:56 <quintopia> elliott: do you have any broken early first drafts of Vagrant, or did cheater steal everything?
18:31:13 <elliott> there's like one or two vagrant.py copies on the internet
18:31:17 <elliott> they're both of bad versions :p
18:31:40 <quintopia> a bad version is better than no version
18:32:47 <Deewiant> ais523: Ah well, a better showing than ipbt which doesn't understand .bz2 at all
18:34:53 -!- conehead has joined.
18:36:46 -!- Slereah__ has joined.
18:39:07 -!- Slereah_ has quit (Ping timeout: 264 seconds).
18:57:43 -!- password2 has quit (Ping timeout: 264 seconds).
18:57:56 -!- chaiomanot has joined.
19:05:11 <shikhin> Doesn't look too easy to implement (note: my operating environments are... let's just say, insane).
19:09:27 <quintopia> shikhin: then implement a fake version where it uses two extended ASCII characters instead of one UTF-16 code point?
19:09:38 <quintopia> i mean, that's still better than nothing :P
19:09:48 -!- nooodl has quit (Quit: Leaving).
19:10:15 -!- nooodl has joined.
19:10:30 <shikhin> quintopia: You shall have to provide a "Hello, world!" example though. :P
19:16:08 -!- Bike has quit (Ping timeout: 246 seconds).
19:16:52 -!- Bike has joined.
19:26:42 <quintopia> shikhin: in this bastard double-character encoding, Hello World! looks like aeaaaaasataaaaeseasaeseeestaestaesttasaaeeetesttetasestaeseaasae (i think)
19:27:12 <shikhin> quintopia: I'll be sure to check it out once I finish the interpreter :-)
19:30:35 <quintopia> shikhin: or you could just implement the original spec, truncating 16 bits to 8, since the first 256 UTF code points are identical to ASCII anyway. If you do that, the above program should print "el ol!"
19:30:50 -!- nooodl has quit (Quit: Leaving).
19:31:08 -!- nooodl has joined.
19:55:50 <shikhin> quintopia: And a hello world for that? Why don't you add that to the wiki page too? :D
19:55:55 * shikhin is too lazy to do this himself.
19:58:22 <ais523> a language where, the first program any given programmer writes in that language, it's a hello world program
19:58:30 <ais523> and the second one is a 99bob
19:58:43 <ais523> to avoid clashes, you must sign your name in the source code
20:00:33 <ais523> (if your name is not unique, your program is a syntax error)
20:02:57 <ais523> well it doesn't have to be a GUID
20:03:04 <ais523> actually, naming people with GUIDs tends to be frowned upon
20:04:21 <nortti> an implementation keeps a log of users and does the stuff
20:08:11 <ais523> but then programs would /retroactively/ become syntax errors
20:08:16 <ais523> you need to know the universe of possible names in advance
20:14:01 -!- Bike has quit (Ping timeout: 240 seconds).
20:22:03 <elliott> ais523: I'm imagining the programs go into some kind of blockchain/DHT type thing.
20:22:38 <elliott> I guess all you'd need to store is name => timestamp of first program written
20:22:48 <ais523> this is going surprisingly well for a sudden esolang idea that's less expressive than HQ9+
20:23:03 <elliott> the implementation looks up the username and timestamp of the file you give it
20:23:10 <elliott> if the timestamp is equal, it's a hello world
20:23:13 <elliott> if it's after that, it's a 99bob
20:23:25 <elliott> if there's no match, store username => timestamp in the DHT and run hello world
20:23:35 <elliott> (they might not have written the 99bob at the timestamp given, but that's UB)
20:24:57 <ais523> what if someone later reproduces their hello world program, with a different modification time?
20:25:03 <ais523> I think you need hashes, not timestamps
20:27:50 <elliott> ais523: then that's not the same program
20:27:59 <elliott> ais523: or rather, it was distributed incorrectly
20:28:12 <elliott> ais523: you should use only verified programs from trustworthy distribution services
20:28:22 <ais523> this reminds me of my approach to encode metadata (like "uses threads") in the modification timestamp of INTERCAL programs
20:28:48 <elliott> ais523: (would you expect a program whose permissions got garbled to still be executable?)
20:29:11 <ais523> elliott: well it usually is
20:29:15 -!- nooodl has quit (Quit: Leaving).
20:29:17 <elliott> not if you run it with ./foo
20:29:23 <ais523> just you need to give the interp manually
20:29:35 -!- nooodl has joined.
20:35:06 -!- ais523 has quit.
20:39:33 -!- MindlessDrone has quit (Quit: MindlessDrone).
20:50:56 <ion> Stanford Professor Andrei Linde celebrates physics breakthrough http://youtu.be/ZlfIVEy_YOA
20:55:33 <Taneb> Why did I decide to cosplay a character with such a fancy jacket
20:55:41 <Taneb> I don't know where I can get such a fancy jacket
20:56:34 <fizzie> "Official Grease Pink Ladies Lady Jackets Fancy Dress Costume 50 Outfit Hen Party" ebay hit #1 for "fancy jacket"
20:56:47 <Taneb> I do not think that is the right sort of fancy jacket
20:57:09 <Taneb> I want this sort of fancy jacket: https://encrypted-tbn3.gstatic.com/images?q=tbn:ANd9GcQxDUF8eWnhUpJjEglwtNy38olYQuGq8emJ1I3H8cDWBZzXSqhJ
20:59:16 -!- aergus has quit (Remote host closed the connection).
20:59:46 <fizzie> Taneb: http://goo.gl/17Iltn close enuf?
21:00:04 <fizzie> (I added "white" to my search terms.)
21:05:53 -!- nooodl has quit (Quit: Leaving).
21:14:18 -!- nooodl has joined.
21:18:38 -!- Boscop has joined.
21:18:54 -!- Boscop has left.
21:32:27 -!- zzo38 has quit (Remote host closed the connection).
21:45:11 -!- spiette has quit (Ping timeout: 252 seconds).
21:52:23 -!- oerjan has joined.
21:59:18 -!- nortti has changed nick to lawspeaker.
21:59:38 -!- lawspeaker has changed nick to nortti.
22:00:29 -!- boily1 has joined.
22:01:19 -!- boily1 has changed nick to boily.
22:01:37 <boily> ah, the feeling of a righteous ghost :D
22:03:01 -!- shikhout has joined.
22:05:55 -!- shikhin has quit (Ping timeout: 264 seconds).
22:05:57 -!- shikhout has changed nick to shikhin.
22:06:29 <oerjan> <elliott> Taneb: I'm not working on @. <-- it's gone the way of feather hasn't it.
22:07:08 <boily> (also, hellørjan.)
22:08:53 <oerjan> jepp, helt ekte plasma
22:09:47 -!- Bike has joined.
22:29:24 <oerjan> <ion> Stanford Professor Andrei Linde celebrates physics breakthrough http://youtu.be/ZlfIVEy_YOA <-- i take it this may be a good day to catch up to r/physics.
22:29:45 <Bike> is this the gravity wave thing
22:31:17 <Bike> i saw some people wondering why people were going on about that before it was peer reviewed
22:31:52 <oerjan> people without souls, clearly
22:32:21 <boily> I just cleared hell in DCSS. I'm up to fourteen runes!
22:32:36 <Bike> oerjan: well, it was the same for the ftl neutrino business and we know how that turned out.
22:34:21 <oerjan> well, but that was a completely unexpected discovery.
22:38:05 <Bike> or, as this phys.org starts its headline, "Rumours fly"
22:54:00 -!- nys has joined.
23:15:45 -!- `^_^v has quit (Quit: This computer has gone to sleep).
23:19:29 -!- applybot has joined.
23:20:25 <Jafet> applybot: lemma "(2 :: nat) + 2 = 4"
23:20:35 <Jafet> applybot: apply simp
23:20:36 <applybot> No subgoals! (Use "done" to finish proof.)
23:20:47 <applybot> lemma "(2 :: nat) + 2 = 4" \ apply simp \ done
23:21:12 <Jafet> (This took much longer to make than it should have.)
23:23:04 <applybot> Isabelle commands: apply, done, lemma, oops, thm; meta-commands: help, context, state, restart, undo
23:23:15 <oerjan> luboš motl gets _guest bloggers_?
23:23:33 <Bike> powerup comics gets guest comics.
23:30:25 -!- Sgeo has joined.
23:34:07 -!- ^v has joined.
23:35:26 -!- not^v has joined.
23:36:03 -!- boily has quit (Quit: Poulet!).
23:38:29 -!- ^v has quit (Ping timeout: 252 seconds).
23:40:44 -!- glogbackup has quit (Read error: Connection reset by peer).
23:45:28 -!- ^v has joined.
23:46:44 -!- not^v has quit (Ping timeout: 252 seconds).
23:47:09 -!- nisstyre has joined.
23:50:22 -!- not^v has joined.
23:53:24 -!- ^v has quit (Ping timeout: 252 seconds).
23:54:50 -!- tromp has joined.
23:59:07 -!- ^v has joined.