←2015-10-03 2015-10-04 2015-10-05→ ↑2015 ↑all
00:04:52 <Phantom_Hoover> i'm more interested in data about what people in scotland actually speak at this stage
00:12:32 <Phantom_Hoover> from just that lecture i don't know how contemporary his speech actually is
00:27:53 <HackEgo> [wiki] [[Stackstack]] http://esolangs.org/w/index.php?diff=44590&oldid=43036 * LegionMammal978 * (+0)
00:31:34 <boily> “MPlayer crashed. This shouldn't happen.”
00:33:04 <boily> fungot: what's a 23-skidoo?
00:33:05 <fungot> boily: when they are revealed you don't have to crowd on computers without drscheme or the scheme community
00:33:39 <boily> fungot: SRFI 23: Error reporting mechanism?
00:33:39 <fungot> boily: oleg's treap implementation is an inadequate reification and the invocation process are part of a syntax-rules macro, which is not valid punctuation. it is fnord)
00:34:08 <boily> fungot: oh, oleg's involved. I suspected type hackery right off the bat.
00:34:08 <fungot> boily: thanks. aagh, postscript!" and even if they pay i won't get any judging done :(.
00:34:27 <boily> fungot: postscript has a type system?
00:37:29 <shachaf> boily: https://en.wikipedia.org/wiki/23_skidoo_(phrase)
00:39:06 <shachaf> Do I need one in order to 23 skidoo?
00:39:38 -!- Phantom_Hoover has quit (Read error: Connection reset by peer).
00:39:45 <boily> thachaf. tdh.
00:42:03 <Sgeo> shachaf, https://www.youtube.com/watch?v=pdxucpPq6Lc around 5:33, but watching the whole thing is more interesting, the skidoo license bit is a throwaway joek
00:42:05 <Sgeo> joke
01:05:30 -!- sc00fy has quit (Ping timeout: 240 seconds).
01:24:20 <tswett> ? porthello
01:24:21 <HackEgo> porthello? ¯\(°​_o)/¯
01:24:41 <\oren\> like portmanto but with hello
01:24:47 <tswett> Ah, of course.
01:24:57 <tswett> Are you sure it shouldn't be porthellau?
01:25:18 <tswett> Or perhaps even portmantellau.
01:25:27 <tswett> Portmanthellau?
01:25:48 <tswett> Natalie Portman-Thellau.
01:34:09 <doesthiswork> sgeo: http://existentialcomics.com/comic/1 is quite funny
01:34:39 <Sgeo> I think I've seen it before
01:34:54 <doesthiswork> I figured
01:40:53 <doesthiswork> the animation feels like someone having an imaginary argument
01:42:38 <doesthiswork> where someone sets up a straw man and demolishes it
01:44:39 <boily> porte-mantelleault.
01:51:52 <coppro> boily: nyc was fun!
01:56:25 -!- MoALTz has quit (Read error: Connection reset by peer).
01:57:13 -!- MoALTz has joined.
01:58:38 <\oren\> the wikipedia list of joyo kanji says 呪 means "charm" but I'm pretty sure it means "curse"
02:01:01 -!- Wright has quit (Ping timeout: 246 seconds).
02:03:08 <\oren\> looks like nelson agrees with me
02:06:10 <\oren\> 呪文 incantation 呪術 sorcery
02:06:22 <coppro> jisho says spell; curse
02:06:35 <coppro> fix it?
02:07:04 <coppro> but 呪文 can mean charm as well
02:07:52 -!- Wright has joined.
02:08:36 <boily> cheloppro! you survived! did you make outrageous hands?
02:08:47 <\oren\> well it isn't the most common meaning
02:08:54 <\oren\> so I fixed it
02:10:37 <tswett> I'm trying to think of Magic cards that would just break the game by confusing the rules.
02:10:46 <\oren\> btw "nelson" is the author of "The Japanese-English Character Dictionary"
02:10:47 <coppro> boily: no, best was a haneman. but I was reffing, not playing
02:10:49 <tswett> This one seems to be pretty good:
02:10:56 <boily> \oren\: could you add 露 twh
02:11:10 <coppro> including casual play, there was one yakuman on saturday and five (!) on sunday, including two kokushi by the same person (!?!?!?!?!)
02:11:13 <boily> coppro: Ō_Ō!
02:11:29 <tswett> "Target word on target face-up card becomes target other word on that card."
02:11:44 <boily> 本当?嘘だ！
02:12:17 <tswett> See that sorcery on the stack which happens to contain the word "land" in its text box? Well, that sorcery is now a land.
02:13:14 <\oren\> do lands even go on the stack?
02:13:38 <\oren\> boily: I'll put them on the list
02:13:47 <tswett> No, normally a land cannot be on the stack.
02:14:38 <boily> \oren\: do you have a large kanji backlog?
02:14:39 <\oren\> Also, when I used to play in middle school the stack was literally a stack
02:15:10 <tswett> The rules don't say what happens when a land resolves. They say what happens when an instant spell, sorcery spell, or ability resolves, as well as what happens when a permanent spell resolves.
02:15:19 <\oren\> boily: not really
02:15:45 <\oren\> I'm about to realase another revisio though
02:16:54 <tswett> Now, under the current rules, whenever something resolves, it must leave the stack somehow, and if it's a card, it goes to the graveyard if it doesn't go somewhere else for some reason.
02:17:25 <tswett> So I think we could reasonably conclude that if a land on the stack were to resolve, it would just go directly to the graveyard.
02:17:29 * boily plans to use \oren\'s font for strategic sections of the PDF
02:17:57 <boily> there should be more pictures in the PDF. the fizziecoin is nice, but it's a little bit lonely imeo.
02:18:40 <coppro> boily: the best was the baiman that got ruined by atamahane though
02:18:51 <tswett> But wait, that's not right.
02:19:01 <tswett> The rules state: "A spell is a card on the stack."
02:19:12 <tswett> So if a land is on the stack, that land is a spell.
02:19:29 <tswett> So is it a permanent spell or not?
02:19:53 <tswett> On the one hand, the rules give a listing of the types that a permanent spell can be, and "land" isn't one of those.
02:19:54 <boily> coppro: ow.
02:20:00 <coppro> tswett: lands are not allowed on the stack
02:20:21 <coppro> the rules do not specify what happens if a land is on the stack for the simple reason that it is not possible
02:20:22 <tswett> coppro: is there a state-based action that happens when a land is on the stack?
02:21:40 <tswett> On the other hand, this land spell would be a spell that's a permanent card. So I think I'd rule that actually, a land on the stack is a permanent spell.
02:21:56 <boily> time for some horizontality. 'night all!
02:22:02 -!- boily has quit (Quit: HYPOCHLORIC CHICKEN).
02:22:05 <tswett> N'ight.
02:22:31 <coppro> night
02:23:03 <coppro> tswett: it's like how there's no rule about an instant or sorcery on the battlefield
02:23:40 <tswett> Well, the rules say that if a permanent loses all of its permanent types, it remains on the battlefield as a permanent.
02:24:16 <tswett> The rules also say what happens if you're supposed to turn a permanent face up, but it wouldn't be a permanent if it were face up.
02:24:37 <coppro> right
02:24:47 <coppro> both those situations could exist
02:25:17 <coppro> there is no state-based action moving an instant or sorcery on the battlefield to its owner's graveryard, though, because that situation is impossible
02:25:20 <coppro> it's like a land on the stack
02:25:47 <coppro> the CR are designed to handle Magic as the game that actually is played
02:25:55 <coppro> (although there are corner cases where they have issues)
02:26:22 <\oren\> http://www.orenwatson.be/fontdemo.htm updated
02:26:32 <tswett> All right. But let's consider what would happen if we wanted to follow the comprehensive rules as closely as possible, while playing with "hack cards" such as this one.
02:26:38 <tswett> Think like a Nomicker!
02:27:52 <coppro> tswett: then the people inventing such cards should specify what happens
02:27:59 <tswett> My revised opinion is that if a land were to be on the stack and then resolve, it would be put on the battlefield.
02:28:04 <tswett> coppro: I guess I'd better start specifying, then.
02:40:47 -!- Wright has quit (Ping timeout: 256 seconds).
02:41:24 <tswett> You could make an odd sort of Magic nomic with that hack card.
02:41:53 <tswett> Two changes.
02:45:43 <tswett> First, add in the following card: "Greater Magical Hack. 6UU. Sorcery. Name a card, and choose two individual words on that card. The first word permanently becomes the second word on all instances of that card."
02:47:00 <tswett> Second, each player has an emblem that says the following: "You cannot win or lose. If you would draw a card, but every player's library is empty, then each player shuffles his or her graveyard into his or her library, then you draw a card."
02:47:32 <shachaf> Are Magic: The Gathering cards ever ambiguous in a way that's resolved by rulings?
02:47:40 <coppro> yes
02:47:44 <tswett> Yeah, I'm pretty sure.
02:47:49 <tswett> Anyone got an example?
02:47:52 <shachaf> That's a bit scow.
02:48:09 <coppro> the previous rules manager was of the opinion that all situations should have a resolution derived directly from the rules, but this is no longer in the philosophy of the rules design
02:48:15 <shachaf> Do they later change the rules to resolve the ambiguity?
02:48:16 <shachaf> Ah.
02:48:17 <coppro> so in certain edge cases the ruling stands alone
02:48:22 <tswett> ? scow
02:48:23 <HackEgo> scow? ¯\(°​_o)/¯
02:48:45 <coppro> there is no rule that enables Oblivion Ring and similar cards to actually work
02:48:58 <coppro> they operate only by fiat
02:52:56 <tswett> I guess whoever does the first Greater Magical Hack could trivially make the game unwinnable by naming Greater Magical Hack and then changing, say, "becomes" to "admires".
02:53:28 <tswett> Wait, no, because "admires" isn't on that card.
02:54:51 <tswett> You could just modify it so it doesn't make any sense: "The first word permanently and the second word on all instances of that card."
02:54:55 <coppro> change "first" to "second"
02:55:07 <tswett> Yup, that'll do it.
02:56:19 <tswett> Come to think of it, I don't think this version of GMH is open-ended.
02:56:33 <tswett> No, sure it is.
02:57:41 <tswett> You can use GMH to alter other cards in such a way that they too become hacking cards. I guess.
02:58:40 <zzo38> That "Greater Magical Hack" is not mathematically correct, I think
02:59:26 <zzo38> Also, I would think a land on the stack would go into play if it resolves; however, the rules do not explicitly say
03:00:32 <tswett> Ooh, here we go.
03:00:37 <tswett> Swirl the Mists says:
03:00:46 <tswett> "As Swirl the Mists enters the battlefield, choose a color word. All instances of color words in the text of spells and permanents are changed to the chosen color word."
03:01:05 <tswett> You could use GMH to turn "choose a color word" into "choose a text word".
03:01:20 <tswett> Presumably pretty much all words count as "text words".
03:02:23 <zzo38> No, it means nothing.
03:02:44 <Sgeo> Colorful pictures are text, what a language
03:02:59 <zzo38> There is no "text words" meaning anything; the answer is not yes and no!
03:02:59 <Sgeo> I wonder if MtG symbols will ever make it into Unicode
03:03:38 <zzo38> Probably not, but that's OK you don't have to use Unicode, or even if you do you can use the private use area.
03:04:03 <zzo38> I would want METAFONT programs to make the MtG symbols
03:04:15 <tswett> I think most MtG symbols already have Unicode characters you could use for them.
03:04:38 -!- Wright has joined.
03:05:16 <zzo38> If you want to make substitute, then yes probably you can.
03:06:03 <tswett> The white mana symbol is U+2600 BLACK SUN WITH RAYS, the blue mana symbol is U+1F4A7 DROPLET...
03:08:02 <tswett> I don't see a Unicode character that's just a skull, but there's a skull and crossbones. There are at least two tree symbols, but I don't know if either one is appropriate.
03:09:06 <shachaf> At one point I found substitutes for each MTG symbol.
03:09:17 <shachaf> At least, ones that rendered well in Google Hangouts.
03:15:17 <tswett> That emblem I came up with gave me a stupid idea.
03:19:20 -!- Melvar has quit (Ping timeout: 244 seconds).
03:19:48 -!- Melvar has joined.
03:22:19 <tswett> Variant: Nobody can win and nobody can lose except via the following mechanism. Once 30 spells named Runeclaw Bear controlled by a player have resolved, that player wins.
03:24:24 <tswett> Actually, make that Omega Myr instead of Runeclaw Bear.
03:33:46 <Sgeo> Google didn't index the logs with "transconic" yet
03:36:02 <zzo38> I have once had the idea of the card it is a creature card but when it comes into play it becomes an instant in addition to its other types. It was also given phasing, which in the rules current at the time that was made would cause it to remain phased out forever but in the next version it can phase in normally.
03:48:15 -!- Wright has quit (Ping timeout: 255 seconds).
03:58:39 <zzo38> Let's see other kind of Magic: the Gathering card to make up
03:59:30 <zzo38> Banding ;; {3}: ~ comes into play with a +1/+1 counter on it. Use this ability only during your upkeep and only if ~ is in your graveyard with exactly 1 card above it.
04:00:06 <shachaf> The graveyard shouldn't be an ordered zone.
04:00:09 <shachaf> That's scow to keep track of.
04:01:53 <zzo38> I like it to be, though. (Some other cards already do anyways, and I have also made some other cards that do)
04:08:49 <tswett> In otherwords, that would be "Put ~ from your graveyard onto the battlefield with a +1/+1 counter on it"?
04:12:12 <zzo38> Yes
04:16:32 <zzo38> Panglacial Wurm can be cast while searching your library, which mean you can activate mana abilities during that time; I want to see what kind of stuff can therefore be done with this.
04:20:51 <\oren\> shachaf: how it it hard to keep track of, if you have physical cards?
04:21:11 <shachaf> It means you can't reorder them.
04:21:24 <\oren\> why would you want to?
04:21:34 <shachaf> You often look through the graveyard and so on.
04:22:04 <shachaf> Some cards can be used from the graveyard and most can't. So you might want to move those to the top.
04:22:38 <zzo38> You should make all card with effect in the graveyard (unless the effect works only if it is on top) to be placed in the pile to stick out a bit so that you can find them. It is important if there is a triggered ability on one of them, regardless of whether or not the order matters.
04:36:11 <doesthiswork> did you know Patrick Stein invented negative length lists?
04:36:53 <zzo38> Madness ability can have some uses even in the case that you are unable to pay the madness cost.
04:37:03 <zzo38> doesthiswork: No. How does this work?
04:37:47 <doesthiswork> http://nklein.com/2012/02/the-anti-cons/
04:38:16 <doesthiswork> each thing you cons onto the list removes one anti-cons from it
04:38:42 <shachaf> Sounds like a free group?
04:39:01 <shachaf> I guess you would have to cons the same thing that you anti-cons to get that.
04:44:53 -!- nisstyre has quit (Ping timeout: 256 seconds).
04:49:30 <doesthiswork> that would make it less polymorphic and less useful if it was a free group
04:50:23 <doesthiswork> although maybe could retain the usefulness by having anti-cons return a value that later resolves into the consed value
04:50:38 <shachaf> Less polymorphic why?
04:50:57 <shachaf> I guess in their version there's only one anti-cons, rather than being able to anti-cons anything onto the list?
04:51:00 <shachaf> That's pretty strange.
04:55:50 <doesthiswork> because instead of being able to dissolve any value consed onto the list, a matching anti-cons would only be able to dissolve a single value.
04:58:52 <shachaf> Yes, like a free group.
04:59:03 <shachaf> It seems that anti-cons should obviously be symmetric with cons.
05:07:29 <doesthiswork> but is also tempting to combine it with 'pop'
05:13:52 -!- nisstyre has joined.
05:22:01 <zzo38> I think there are cases in Magic: the Gathering where a mana ability will end up not producing any mana, and which can still be used during a mana step or if a spell with split-second is on the stack.
05:22:55 <shachaf> I don't like the way "mana ability" is defined.
05:22:57 <shachaf> It's too vague.
05:25:49 <zzo38> It is a bit vague, which is why I wanted a mathematical definition of the game as a literate computer program.
05:26:47 <shachaf> Is Reflecting Pool's ability a mana ability if you control no other lands?
05:27:04 <shachaf> I think the answer was yes, because it "could" produce mana in some sense. But I don't like that sense much.
05:28:45 <shachaf> Wait, I'm mixing up two different objections I have now, I think.
05:28:51 <zzo38> Yes, rule 605.2 means that, it seems, or at least is supposed to
05:29:30 <shachaf> Ah, right.
05:31:14 <zzo38> Still it might be useful for a computer code to support properties such as :activated-mana-ability and :triggered-mana-ability, or to use macros to implement it (or both).
05:41:10 -!- scoofy has quit (Ping timeout: 240 seconds).
06:01:53 <zzo38> Is there a "Dead Reckoning Magic: the Puzzling"? Unlike chess, Magic: the Gathering has incomplete information (past, present, and future), so may be more difficult to define. But, other kind of retro puzzles can be possible to do
06:05:57 -!- doesthiswork has quit (Quit: Leaving.).
06:07:54 <shachaf> zzo38: I saw an unusual Chess: The Puzzling the other day.
06:07:57 <shachaf> http://hebdenbridgechessclub.blogspot.com/2011/02/hardest-chess-problem-in-world.html
06:09:35 -!- bender has joined.
06:09:35 -!- bender has quit (Changing host).
06:09:35 -!- bender has joined.
06:10:23 <zzo38> I have seen that one; it is based on old rules. Knowing how the old rules differ from the current rules, I can easily figure it out.
06:10:47 <shachaf> Yes.
06:12:38 <shachaf> Do you know any good Haskell: The Puzzlings?
06:12:53 <zzo38> I don't know how to play "Haskell: The Puzzling"
06:13:15 <zzo38> But also one idea would be a Magic: the Puzzling where the solution is almost completely different in one version of the rules than with a different version (preferably the immediately previous version, if possible)
06:13:53 <shachaf> I don't know how to play Magic: The Puzzling either.
06:14:03 <shachaf> I think you need a book for that, and I don't have the book.
06:16:35 <zzo38> You do not necessarily need a book; you can also find some on computer.
06:20:13 <zzo38> Are there positions in Magic: the Gathering for which it is impossible to decide whether or not they are legal?
06:22:32 <myname> what is a "position"? and isn't the only way to be impossible to decide for it to need infinitely long to get there?
06:22:50 -!- ^v has quit (Remote host closed the connection).
06:24:01 <pikhq> zzo38: I suspect not. To my knowledge the set of game states in Magic is countable, as well as the set of legal actions from any given game state.
06:24:14 <pikhq> (and, additionally, the set of *initial* game states is countable)
06:24:23 <shachaf> "countable" isn't enough.
06:24:40 <shachaf> Presumably there are "garden of eden" positions?
06:25:41 <pikhq> Ah, right, the set of game actions from a given game state need be not just countable, but demonstrably knowable.
06:26:02 <pikhq> Which I *think* is the case, but is... not something I'd just boldly assert.
06:26:23 <shachaf> I'm not sure why actions are directly relevant here. The question is whether a position is legal in the first place.
06:26:30 <zzo38> I don't know if there is something really strange that causes it not to be knowable.
06:26:40 <shachaf> pikhq: A certain individual is quite sauzzled right now.
06:26:54 <shachaf> I'm getting a lot of messages.
06:26:56 <pikhq> A position is legal if you reached it via a legal action from another legal position.
06:27:06 <shachaf> Oh, is that what it means?
06:27:24 <shachaf> OK.
06:27:50 <pikhq> Near as I can tell -- I don't recall game states *themselves* having any formal legality in Magic, but the actions themselves do.
06:28:02 <pikhq> i.e. the rules govern the state transitions, not really the state.
06:28:45 <coppro> correct
06:29:02 <quintopia> mtg is pretty popular. i'm still astounded.
06:29:18 <shachaf> Should I play the new set?
06:29:22 <shachaf> BFZ
06:29:33 <shachaf> There may be a jam brewing.
06:29:41 <quintopia> Borel-Frankel-Zermelo
06:30:02 <quintopia> or probably Bernays
06:30:43 <shachaf> béarnaise
06:32:08 <myname> doesn't each turing complete system has its own halting problem?
06:32:14 <zzo38> Is it possible to construct a position that you can win if and only if the number of counters added to some particular object is a prime number?
06:32:28 <quintopia> myname: you mean even if you add halting oracles?
06:32:33 <pikhq> Hmm. I'm not sure.
06:33:02 <pikhq> zzo38: I assume you mean "without creating a new card for exactly that purpose" of course.
06:33:15 <quintopia> zzo38: you mean if and only if it is any prime number?
06:33:16 <pikhq> (it's otherwise an uninteresting question. :))
06:33:26 <myname> quintopia: that too, but if mtg is turing complete, shouldn't it have a halting problem.and therefore an undecidavle state?
06:33:46 <quintopia> of course it does
06:34:03 <zzo38> Yes, if any prime number
06:34:14 <myname> so there is such a state
06:34:25 <pikhq> myname: Only in terms of whether or not the execution of that state halts.
06:34:31 <zzo38> And I mean using existing non-Un cards and standard rules that are at least as recent as the most recent card used
06:34:50 <quintopia> mtg is tc because someone has devised a utm for it. one could construct a game state such that that utm is running a tm which it is impossible to decide whether it halts
06:34:51 <pikhq> The legality of said state is a fairly different question than whether or not said state halts.
06:35:50 <pikhq> (legality is a question of how the state was reached, not what can be reached from the state)
06:36:23 <quintopia> ah i just now saw zzo's question
06:36:37 <myname> how so? assume you get a state that decides wether it halts or not, how do you proove it's legal?
06:37:03 <coppro> there is no such state
06:37:27 <quintopia> my guess would be "yes", on account of the halting problem when one runs the game in reverse. (assuming that mtg is also tc in reverse.)
06:37:51 <pikhq> myname: Whether or not the state can be reached from some other legal state is what you need to prove. I'm not sure if this is actually possible for all arbitrary states.
06:38:21 <pikhq> But it is certainly a very different question than whether or not a state halts.
06:38:51 <myname> couldn't you declare the state you are innas halting?
06:39:21 <pikhq> myname: Then the question is "does there exist a TM that halts", not "does this TM halt".
06:39:44 <pikhq> I think.
06:40:03 <myname> no. assume a state that decides wether or not you can reach this exact state
06:41:13 <quintopia> pikhq: let's suppose that one can construct an NTM based on running the rules of mtg in reverse. like if a card says "play this card in blah situation" instead, you can take it into your hand in that situation. if this is tc, then the legality question has an undecidable state.
06:41:15 <pikhq> Okay, let's go with "does there exist an input tape such that this TM halts", which... is a rephrasing of the halting problem.
06:42:11 <pikhq> Hum, yes, there must exist some state wherein it is not decidable if it is reachable.
06:42:14 <newsham> the original was "does this TM halt for all inputs", no?
06:42:28 <pikhq> Ah, so it was.
06:42:43 <pikhq> "For some input" might actually be decidable? I doubt it though
06:42:49 <coppro> no
06:42:55 <shachaf> Nothing is decidable.
06:43:02 <pikhq> Ah, no, it's proven that the more specific statement is definitely not decidable.
06:43:09 <myname> which is a superset of "does this tm.halts for a given input"
06:43:14 <newsham> lots of trivial tings are decidable.. just the nontrivial stuff is all undecidable :)
06:44:07 <pikhq> For instance, it's entirely decidable to determine if a given C program in a given C environment halts.
06:44:17 <pikhq> Not very practical, but definitely decidable.
06:44:24 <myname> how so?
06:44:36 <quintopia> C is LBA
06:44:41 <myname> ah
06:46:06 <newsham> pikhq: it is?
06:46:09 -!- hppavilion[1] has joined.
06:46:10 <coppro> pikhq: let M be "given M', decide if M' halts on some input." Define N to be "Given N', let N'' be 'Run each possible input on N' in parallel and halt if any of them do' and then run M on N''". Then N decides the HP.
06:46:20 <coppro> newsham: yes, bounded memory
06:46:38 <pikhq> Yep. C defines a system with a large but very definitely bounded state.
06:46:44 <shachaf> Depends on whether you allow certain file I/O.
06:46:45 <newsham> oh, FSM
06:46:57 <pikhq> It doesn't strictly define what those bounds may be, but does define that those bounds exist.
06:47:12 <myname> the fuck
06:47:19 <pikhq> shachaf: You can only really get past it with sockets.
06:47:30 <shachaf> You can do relative seeks in files.
06:47:32 <pikhq> Without sockets, well, the filesystem is *also* bounded.
06:47:39 <shachaf> whoa whoa whoa
06:47:46 <shachaf> I don't think the C spec defines anything about that.
06:47:54 <coppro> newsham: no, it's far stronger than a FSM
06:47:54 <shachaf> The file API can go to all sorts of things.
06:48:02 <coppro> shachaf: finite filename length, finite file length
06:48:04 <pikhq> Fuck non-POSIX C.
06:48:09 <newsham> coppro: how so?
06:48:11 <shachaf> coppro: Finite file length?
06:48:26 <newsham> coppro: your entire system can be encoded into a large FSM
06:48:42 <coppro> newsham: oh, misread as DFA, sorry
06:48:53 <pikhq> However, ISO C does still require bounded file length: fgetpos and fsetpos are ISO C.
06:48:59 <coppro> ^
06:49:07 <newsham> DFA is just as powerful as an non-dterministic FSM
06:49:17 <pikhq> Which means any given file position must fit within memory.
06:49:23 <pikhq> And is thus bounded.
06:49:40 <newsham> pikhq: what is fgetpos on stdin?
06:49:41 <coppro> likewise for file names, they must fit into memory
06:50:36 <pikhq> Well okay, there's three unbounded things in C: stdin, stdout, and stderr.
06:50:44 <shachaf> At least POSIX allows it to fail because it doesn't fit in an fpos_t?
06:50:56 <newsham> at any rate, you're not going to compute halting property of a real C program on a real machine
06:51:02 <pikhq> shachaf: But POSIX requires it to fit in an off_t. :)
06:51:09 <shachaf> I don't think so.
06:51:09 <newsham> and there are probably some ultrafinitists that would put up fists over that assertion ;-)
06:51:11 <pikhq> If it doesn't fit in an off_t *the open fails*.
06:51:36 <shachaf> I think I looked it up once and POSIX C allows it to be too big.
06:51:43 <\oren\> all programs will eventually halt when the computer running them is sucked into a black hole
06:52:08 <shachaf> I don't remember the details.
06:52:14 <shachaf> Maybe http://pubs.opengroup.org/onlinepubs/009695399/functions/ftell.html
06:52:18 <pikhq> newsham: Well, sure you can. Run two parallel simulations. Run one of them at twice the speed of the other. If the two states ever equal each other, it doesn't halt. If it halts, it halts.
06:52:21 <shachaf> EOVERFLOW
06:53:03 <newsham> pikhq: that is not a terminating program that answers the halting problem :)
06:53:20 <pikhq> newsham: As the number of states is bounded, it will always terminate.
06:53:24 <coppro> pikhq: that could fail if the repeatition has even period
06:53:31 <\oren\> printf("it will halt");
06:53:43 <coppro> *repetition
06:53:47 <\oren\> that solves it because all real programs halt
06:54:01 <coppro> pikhq: count the size of the state space, run that many steps
06:54:07 <coppro> if you're not done, return "doesn't halt"
06:54:19 <\oren\> coppro: that also works
06:54:21 <coppro> the busy beaver trick
06:54:36 <pikhq> coppro: That'll work, too.
06:56:02 <newsham> ah
06:56:29 <\oren\> hmm there should be a variant of BF where all programs begin with a header giving the number of cells and the bounds of each cell
06:57:14 <\oren\> thus making it subturing
06:57:56 <\oren\> is there a programming language that guarantees that all programs halt?
06:58:10 <pikhq> Add in something to specify the IO rules and you'll have come up with a practical means of accepting all BF programs with only trivial modifications. :)
06:58:19 <myname> \oren\: hq9
06:59:54 <\oren\> #!bf -c100 -v0-255 -lf
07:00:26 <\oren\> something like that could work
07:01:10 <newsham> oren: agda? (do you count productive coinduction?)
07:02:36 <shachaf> coins
07:02:38 <HackEgo> adheidcoin gotomcoin gnatiocoin cellcoin icacoin etamcoin mammarkcoin m-codcoin midcoin barcoin smitafncoin sing-boomcoin pugoddbcoin bubandcoin diiccoin quylcoin codecoin brimcoin sivecoin hersetcoin
07:02:43 <newsham> are there any high perf bf jits that will recover high level integer operations?
07:04:52 -!- scoofy has joined.
07:16:43 <quintopia> bubandcoin sounds wonderful
07:19:05 -!- zgrep has quit (Quit: ZNC 1.6.1 - http://znc.in).
07:21:42 -!- zgrep has joined.
07:39:03 <\oren\> coins
07:39:05 <HackEgo> tre-calcoin procoin vortecoin andlcoin l00ftcoin nouslycoin alagehottermotiacoin dutchytacoin eplexarchencoin fuperpilancoin udccoin rwancoin footicotenditediuscoin bjoriecoin hipcoin crabllcoin mahocoin uncoptercoin whenemcoin buicoin
07:39:52 <\oren\> mahocoin! mahomahomahomaho
07:42:23 -!- gamemanj has joined.
08:06:06 -!- hppavilion[1] has quit (Ping timeout: 255 seconds).
08:08:01 -!- TieSoul has joined.
08:23:21 -!- bb010g has quit (Quit: Connection closed for inactivity).
08:57:11 -!- oerjan has joined.
09:12:06 -!- zadock has quit (Quit: Leaving).
09:25:19 -!- J_Arcane has quit (Ping timeout: 246 seconds).
09:26:08 -!- Phantom_Hoover has joined.
09:35:42 -!- J_Arcane has joined.
10:00:11 -!- jalumar has quit (Ping timeout: 264 seconds).
10:00:52 -!- jalumar has joined.
10:24:15 -!- AnotherTest has joined.
10:28:45 -!- AnotherTest has quit (Ping timeout: 255 seconds).
10:45:11 -!- jalumar has quit (Ping timeout: 264 seconds).
11:21:12 -!- JesseH has quit (Read error: Connection reset by peer).
11:22:43 -!- mauris has joined.
11:55:17 -!- AnotherTest has joined.
12:56:11 -!- gde33|2 has quit.
12:57:39 -!- AnotherTest has quit (Ping timeout: 240 seconds).
13:00:52 -!- doesthiswork has joined.
13:03:17 -!- Patashu has quit (Ping timeout: 268 seconds).
13:13:03 -!- conehead has quit (Quit: ZNC - http://znc.in).
13:15:31 -!- conehead has quit (Changing host).
13:17:38 -!- doesthiswork has quit (Quit: Leaving.).
13:19:24 -!- oerjan has quit (Quit: Later).
13:20:08 -!- boily has joined.
13:40:57 -!- scoofy has quit (Ping timeout: 256 seconds).
13:56:09 -!- doesthiswork has joined.
14:14:50 -!- ais523 has joined.
14:38:04 -!- bender has quit (Quit: [In order to stop people from openly defecating after drinking too much vodka, the Russian government decided to provide free poo-tins.]).
14:40:09 <tswett> cat bin/coins
14:40:10 <HackEgo> words ${1---eng-1M --esolangs 20} | sed -re 's/( |$)/coin\1/g' | rainwords
14:42:28 <tswett> run echo ${1---eng-1M --esolangs 20} 14:42:29 <HackEgo> ​--eng-1M --esolangs 20 14:42:51 <tswett> run words --eng-1M --esolangs 20 14:42:53 <HackEgo> memboja cli sot pain ofc rna iincalake recunt reved enever2 enta magist anazranm tam bam squ toelter revertrit trantum pine 14:43:04 <tswett> paincoin 14:43:21 <tswett> cat bin/words 14:43:21 <HackEgo> ​#!/usr/bin/perl \ use strict; use warnings; \ use v5.10; \ use open qw( :encoding(UTF-8) :std); \ use File::Basename 'dirname'; \ use Storable 'retrieve'; \ use List::Util qw(sum min); \ use Getopt::Long qw(:config gnu_getopt); \ BEGIN { \ eval { \ require Math::Random::MT::Perl; Math::Random::MT::Perl->import('rand'); \ }; \ #wa 14:45:33 -!- Wright has joined. 14:46:18 <boily> tswellott. 14:46:39 <boily> coins --french 14:46:41 <HackEgo> bayncoin 14:46:50 <boily> coins --french 20 14:46:51 <HackEgo> boucatcoin rgildeecoin chécoin iiecoin delhscoin ruitcoin grimity'scoin martiquecoin argétèrcoin mésotencoin boranifiecoin préciamcoin attelumescoin caiendcoin ajustetencoin greecoin regemcoin ocatissairecoin rangiquicoin nidcoin 14:46:56 <b_jonas> is that like bacon? 14:46:58 <b_jonas> bacoin 14:47:00 <b_jonas> hmm 14:47:13 <b_jonas> coins 14:47:15 <HackEgo> 1mprelainskingcoin axocoin sagringpuzzlecoin outcoin vestciendcoin pathcoin wheredcoin pucedcoin keecoin phabismcoin binarycoin noilogcoin ialcoin entcoin trudewacoin jentroncoin dzimmicoin purcoin bakcurrownicensilaincoin cuddlequecoin 14:47:31 <b_jonas> binarycoin! 14:47:38 <boily> bakcurrownicensilaincoin! 14:47:39 <b_jonas> entcoin 14:47:49 <boily> b_jhellonas. 14:47:58 <b_jonas> cuddlequicoin is nice 14:48:03 <b_jonas> cuddle-que-coin 14:49:34 <boily> le cuddle qui cointe. 14:52:53 <boily> wisdom coin 14:52:54 <HackEgo> find: wisdom/*coin*': No such file or directory \ /cat: : No such file or directory 14:53:00 <boily> ? coind 14:53:01 <HackEgo> coind? ¯\(°​_o)/¯ 14:53:05 <boily> ? coin 14:53:06 <HackEgo> coin? ¯\(°​_o)/¯ 15:15:40 -!- boily has quit (Quit: LASCIVIOUS CHICKEN). 15:18:32 -!- Wright has quit (Ping timeout: 272 seconds). 15:21:56 -!- Vorpal has joined. 15:29:23 <tswett> Given five distinct points on the plane, is there always exactly one conic section which passes through all of them? 15:30:21 <tswett> Hmm, the answer is no if they're collinear. 15:30:33 <tswett> All right, five distinct noncollinear points. 15:30:49 <int-e> . o O ( Coincidence is a contraction of coin-incidence, referring to the event of a coin hitting a flat surface with random result. ) 15:34:12 <ais523> tswett: a line is also a conic section 15:34:24 <ais523> you cut the cone along a tangent to one of the sloping sides 15:34:30 <ais523> (it's not normally thought of as one, but it is) 15:34:39 <tswett> Yeah, but there are multiple conic sections passing through five collinear points. 15:35:18 <tswett> A pair of lines is a conic section. 15:45:28 -!- doesthiswork has quit (Quit: Leaving.). 15:53:07 -!- glowcoil has joined. 15:54:03 -!- Wright has joined. 15:56:04 <ais523> presumably all the points have to be distinct, too? 15:56:22 -!- scoofy has joined. 15:57:04 <ais523> actually, I think you can find an ellipse and parabola that intersect in five points 15:57:24 <ais523> that aren't coanything (although would have a line of symmetry) 16:03:45 -!- doesthiswork has joined. 16:05:44 <b_jonas> ais523: no you can't. 16:08:04 <b_jonas> For five points in the (real or complex Euclidean) plane, no two coinciding and no four collinear, there's exactly one conic section incident on all five; if also no three of the points are collinear, then that conic section is not degenerate (not two lines, but a hyperbola or parabola or ellipse, counting a circle as an ellipse). 16:09:17 <ais523> aha, ofc if you have four colinear (and the fifth not on the line) 16:09:40 <ais523> you could take an arbitrary line through the fifth 16:09:47 <b_jonas> ais523: yes. 16:10:03 <b_jonas> and note that any line intersects a non-degenerate conic in at most two points. 16:11:43 -!- scoofy has quit (Ping timeout: 252 seconds). 16:12:16 <b_jonas> The situation is much more difficult if you're looking for conics incident on four points and touching a line, in which case in the generic case (and I won't tell what that means now) there can be exactly zero or exactly two conics matching them, but I don't know a book or article that explicitly claims this; 16:13:06 <b_jonas> so I should try to ask a three-parter question on MathOverflow asking for a reference about (a) one line and four points, general case, (b) one line and four points, all cases, (c) two lines and three points, in which case I don't even know what's the answer I expect. 16:13:21 <b_jonas> I might have to make it two separate questions, I'll have to think it over. 16:13:45 <b_jonas> However, the Hajós book does explain the five point in a plane case completely. 16:16:58 <\oren\> unidecode 露漫 16:16:58 <HackEgo> ​[U+9732 CJK UNIFIED IDEOGRAPH-9732] [U+6F2B CJK UNIFIED IDEOGRAPH-6F2B] 16:20:04 -!- contrapumpkin has joined. 16:23:26 -!- copumpkin has quit (Ping timeout: 240 seconds). 16:24:37 <\oren\> I have expressed my annoyance at the Unicode Foundation's method of naming chinese characters before 16:26:58 <b_jonas> \oren\: you shouldn't care about the _name_ of the character, because that's not the important part. some names are bad, and about three even have obvious typos that can't be changed now. 16:27:43 <\oren\> unicode 露 16:27:44 <HackEgo> U+9732 CJK UNIFIED IDEOGRAPH-9732 \ UTF-8: e9 9c b2 UTF-16BE: 9732 Decimal: &#38706; \ 露 (露) \ Uppercase: U+9732 \ Category: Lo (Letter, Other) \ Bidi: L (Left-to-Right) 16:28:10 <\oren\> RRgh. why does the word "dew" appear nowhere here?! 16:28:19 <b_jonas> \oren\: also, when China and Taiwan can be involved, then the naming could be a political issue. 16:28:30 <\oren\> that's what the damn character means 16:28:54 <b_jonas> \oren\: then you look that up in a dictionary. online or paper, whatever. 16:29:20 <b_jonas> oh, that reminds me 16:29:50 <b_jonas> what's the unicode encoding for the fixed-width space used after the en-dash at the start of a line when it indicates a dialog line, typically in languages other than English? 16:30:14 -!- nortti has quit (Remote host closed the connection). 16:30:53 <\oren\> yeah but they had no problem naming ㊧ 16:31:03 <\oren\> unidecode ㊧ 16:31:03 <b_jonas> unicode ㊧ 16:31:04 <HackEgo> ​[U+32A7 CIRCLED IDEOGRAPH LEFT] 16:31:05 <HackEgo> U+32A7 CIRCLED IDEOGRAPH LEFT \ UTF-8: e3 8a a7 UTF-16BE: 32a7 Decimal: &#12967; \ ㊧ \ Category: So (Symbol, Other) \ Bidi: L (Left-to-Right) \ Decomposition: <circle> 5DE6 16:31:29 <b_jonas> \oren\: or BELL or all the other crazy dingbats and smiley faces 16:32:03 <\oren\> so why isnt 左 CJK IDEOGRAPH LEFT!?!?!? 16:33:38 <b_jonas> \oren\: because then the next Chinese communist leader who happens to have that hanzi in their wife's name or whatever will complain that her wife isn't left and that this is a personal insult and that the unicode consortium must immediately pay him a zillion dollars or else he will take out his vengeance with atomic missiles 16:34:35 <\oren\> argh 16:35:17 <b_jonas> no really 16:35:19 <b_jonas> the names aren't important 16:35:28 <b_jonas> not important enough to get insulted over, at least 16:37:03 <b_jonas> no seriously, what's the unicode encoding of a fixed with space \hbox{ } ? surely there is such a character, I just don't know which one 16:37:26 -!- Wright has quit (Ping timeout: 240 seconds). 16:39:28 <\oren\> hmm I can't find any character like that 16:39:50 <ais523> b_jonas: it's non-breaking space, it's in latin-1 16:39:56 <ais523> I might even be able to type it 16:40:00 <b_jonas> ais523: no. non-breaking space isn't fixed with 16:40:04 <b_jonas> unidecode 16:40:05 <HackEgo> ​[U+00A0 NO-BREAK SPACE] 16:40:07 <ais523> it's the same width as a normal space 16:40:12 <ais523> surely 16:40:20 <b_jonas> ais523: exactly, but it also stretches and shrinks like a normal space 16:40:29 <b_jonas> non-breaking space is what ~ types in plain TeX 16:40:34 <ais523> oh, hmm 16:40:39 <b_jonas> what I want is \hbox{ } , a space that doesn't shrink or stretch 16:40:43 <ais523> I'm not sure if it makes sense for such a thing to be in Unicode 16:40:46 <b_jonas> you can call it \hbox{\ } too, doesn't matter 16:40:53 <ais523> because I'm not sure Unicode has specific behaviour for justified text 16:41:09 <\oren\> maybe the en quad or the em quad? 16:41:13 <b_jonas> ais523: why not? just because it is there doesn't mean everyone has to use it. it'd just be a character you may use to represent that behaviour if you want. 16:41:22 <b_jonas> \oren\: no, those are not the right width 16:41:43 <b_jonas> \oren\: this one is exactly the width of a normal space in the font 16:42:05 <b_jonas> the quads are also fixed with, but often larger 16:42:24 <ais523> whereas things like bigskip are proportional, right? 16:42:44 <\oren\> there's also 3 per em 4 per em and 6 per em spaces 16:42:48 <b_jonas> ais523: I don't know what bigskip is. 16:42:58 <ais523> b_jonas: it's a vertical spacing directive 16:43:14 <ais523> I'm not sure of the exact semantics 16:43:33 <\oren\> of course in my font these all have visible glyphs 16:44:04 <\oren\> because oren is an asshole to typesetters 16:47:27 <b_jonas> ais523: note for example that (the plain TeX) \, (thin math space) and \quad doesn't shrink or stretch, \ has the size and shrinks or stretches exactly like a normal space that's not after punctuation, \> is just like the math space around an infix operator and can shrink to almost nothing, and \; is like the space around an equation sign and is not much wider but doesn't shrink at all. 16:47:50 <b_jonas> (ok, not quite shrinks to nothing, \> actually shrinks to half its normal size) 16:50:36 <\oren\> 16:51:04 <\oren\> there are 10 different spaces on the previous line 16:51:17 <HackEgo> [wiki] [[LOLCODE]] http://esolangs.org/w/index.php?diff=44591&oldid=44360 * 73.51.84.54 * (-101) /* Keywords */ rm duplicate PRODUKT 16:51:41 <b_jonas> \oren\: don't forget 16:51:52 <\oren\> yeah there is that too 16:52:00 <b_jonas> and I think there's more 16:52:12 <b_jonas> I think there's a unicode property somewher 16:52:28 <b_jonas> I guess I should look among spaces marked by that property 16:52:29 <\oren\> if you use my font you can see and recognise them all 16:59:52 -!- ais523 has quit (Quit: dinner). 17:11:45 -!- Phantom__Hoover has joined. 17:12:59 <\oren\> phantom_helloover! 17:15:09 -!- Phantom_Hoover has quit (Ping timeout: 256 seconds). 17:19:50 -!- atrapado has joined. 17:29:47 <\oren\> I've uploaded a new otf. ⴀⴁⴂⴃⴄⴅⴆⴇⴈⴉⴊⴋⴌⴍⴎⴏⴐⴑⴒⴓⴔⴕⴖⴗⴘⴙⴚⴛⴜⴝⴞⴟⴠⴡⴢﬀﬁﬂﬃﬄﬅﬆ〄〆〇〈〉《》【】〒〓〔〕〖〗〘〙〚〛￠￡￢￣￤￥￦因団固園圧坂堂場塩士声売変夏夕外多夜夢好並享亭他仮仰仲露漫 17:33:31 <b_jonas> \oren\: what did I do wrong? it doesn't show up again 17:33:38 <\oren\> reload a few time 17:33:54 <\oren\> I think AWS is caching the .css again 17:34:09 <b_jonas> \oren\: I think it's that otf conversion thing again, because I don't even see the old font 17:34:12 <b_jonas> I see only some fallback font 17:34:15 <b_jonas> I dunno 17:34:32 <b_jonas> \oren\: the image version also seems to be very old 17:34:55 <\oren\> can you look at the javascript console for any css errors 17:35:09 <b_jonas> \oren\: even if it caches, when I shift-reload, the browser is sending those http header thingies that tell the cache to refresh it. 17:35:23 <b_jonas> or at least it's supposed to 17:35:56 <b_jonas> \oren\: it says "Failed to load resource: the server responded with a status of 403 (Forbidden)" 17:36:26 <b_jonas> \oren\: and it says neolettersK.otf gives the 403 17:36:27 <\oren\> oh 17:36:32 <\oren\> fuck hold on 17:37:14 * \oren\ chmod +r /var/www/html/* 17:37:22 <\oren\> there now it should work 17:37:47 <b_jonas> oh! can I load your secret passwords file too then? 17:37:53 <b_jonas> also, the font loads now 17:38:11 <b_jonas> the image version is still wrong 17:38:20 <\oren\> nah I don't keep passwords in the www directory 17:38:43 <b_jonas> \oren\: the first two lines show up wrong though, probably because the browser is making the lines bigger because of the underlines 17:38:52 <b_jonas> \oren\: could you perhaps change the CSS to remove the underlines from the links? 17:39:05 <b_jonas> you don't want underlines to this font anyway 17:39:36 <\oren\> how do I do that? 17:39:52 <\oren\> a {underline:none?} 17:39:59 <b_jonas> \oren\: no, 17:40:14 <b_jonas> I think it's :link, :visited { what you said above } 17:41:14 <b_jonas> http://www.w3.org/TR/CSS21/selector.html#link-pseudo-classes 17:41:31 <b_jonas> and I don't think it's underline:none either 17:41:40 <b_jonas> it's text-decoration: none 17:42:52 <\oren\> k, done 17:42:55 <b_jonas> \oren\: :link, :visited { text-decoration: none } /* I hope */ 17:43:07 <b_jonas> um, with a semicolon after none for good measure 17:43:23 <b_jonas> \oren\: hmm, it still doesn't look good 17:43:26 <b_jonas> dunno why though 17:43:33 <b_jonas> hmm wait 17:43:48 <b_jonas> yes, definitely doesn't look good 17:44:04 <b_jonas> stupid browser 17:44:40 <\oren\> i updated the image so you can see how it looks on my compy 17:44:48 <b_jonas> \oren\: anyway, now you have 〇 twice in the sample 17:44:57 <b_jonas> in the CJK punctuation and in the hanzi group 17:45:38 <b_jonas> \oren\: I think at this point it's probably a browser stupidity, but removing the underline is probably a good idea anyway if you want to leave that text in your font 17:45:53 <\oren\> 〇ops. 17:46:28 <b_jonas> I see you've added more kanji/hanzi 17:47:16 <\oren\> yeah 17:47:41 <b_jonas> hmm, there seems to be something wrong though 17:47:48 <b_jonas> there's a character that's not in your font, but in the sample 17:48:07 <b_jonas> 17:48:21 <b_jonas> or at least that's how it shows up in my browser 17:48:34 <b_jonas> can you check on that? it's in the last line of hanzi in the samples 17:48:44 <\oren\> Oh, yeah your right 17:50:37 <\oren\> Oh. I put the glyph on the wrong charater 17:57:49 -!- ais523 has joined. 18:05:20 -!- boily has joined. 18:06:27 <\oren\> hiboily 18:06:44 <boily> \helloren\! 18:06:48 <\oren\> iadded 露 and 漫 18:06:50 <boily> you added the characters! 18:06:56 <boily> that's what I was checking ^^ 18:09:17 <boily> next, could you please add, at your leisure, 羅針饂飩 and 檎? 18:09:30 <boily> (but take your time. these are becoming silly.) 18:10:14 <b_jonas> boily: um, how are you chosing these particular characters? 18:11:15 <boily> strokeful characters :D 18:11:37 <b_jonas> ah 18:11:46 <b_jonas> so you are just choosing characters that are hard to draw? 18:12:11 <boily> mainly kanji versions of common words that are usually written in kana. 18:12:29 <boily> (and yes, perverse admiration for high stroke counts.) 18:12:53 <b_jonas> shouldn't the font contain kanji that are actually commonl? 18:13:10 <\oren\> well, whatever 18:13:26 <b_jonas> \oren\: did you add any of the mirrored or vertical alternate glyphs for the variable characters? 18:15:37 <\oren\> b_jonas: At some point I'll figure out how to do alternate glyphs, either that or just make a script that produces a rotated version the font. 18:16:29 <b_jonas> \oren\: um, just rotating isn't enough, you may have to shift when you rotate or mirror, and you as a font designer have to decide the shift amount that looks good 18:16:39 <b_jonas> you might even want to change more pixels when you rotate 18:17:58 <boily> ♪ ro ro, rotate your font ♪ 18:21:20 <HackEgo> [wiki] [[LOLCODE]] http://esolangs.org/w/index.php?diff=44592&oldid=44591 * 73.51.84.54 * (+34) /* Keywords */ added <nowiki> tags around pipe characters in comment of loop keyword, to prevent them being interpreted as part of the wikitable and not displaying part of the text. 18:23:13 -!- oerjan has joined. 18:29:11 -!- MDude has joined. 18:31:05 <\oren\> hellỗếrjan 18:33:00 <oerjan> orevening i'm not even going to try those accents 18:40:25 -!- JesseH has joined. 19:06:27 <tswett> Why does Unicode have spaces of different widths? 19:06:40 <b_jonas> tswett: which ones in particular? 19:07:07 <b_jonas> it has them because they're useful 19:07:34 <b_jonas> like, I can write three-digit-grouped numbers with thin spaces like 3 000 000 19:07:58 <tswett> Oh, how about the three-per-em space. Why does it have that one? 19:08:09 <b_jonas> dunno, someone probably uses that for something 19:08:13 <b_jonas> I don't know about each of them 19:08:51 <b_jonas> I know the thin space and the quad (en space) and the qquad (em space) and the figure space is useful 19:09:15 <b_jonas> and I'd like the fixed-normal-width non-breaking space 19:09:37 <b_jonas> but I don't know if there's such a character 19:11:33 -!- nortti has joined. 19:26:01 -!- fizzie has joined. 19:26:57 -!- coppro_ has joined. 19:28:18 * oerjan gives up listening to that podcast about darths & droids for the second time 19:28:23 -!- AnotherTest has joined. 19:28:51 -!- mysanthrop has joined. 19:29:08 <oerjan> goddammit hosts, learn to focus 19:30:44 -!- Alcest has quit (Ping timeout: 260 seconds). 19:30:46 -!- APic has quit (Ping timeout: 260 seconds). 19:31:22 -!- APic has joined. 19:32:23 -!- fizzie has quit (Ping timeout: 240 seconds). 19:32:24 -!- coppro has quit (Ping timeout: 240 seconds). 19:32:30 -!- myname has quit (Ping timeout: 240 seconds). 19:32:31 -!- mtve has quit (Ping timeout: 240 seconds). 19:32:35 -!- mysanthrop has changed nick to myname. 19:33:38 -!- Alcest_ has joined. 19:35:14 -!- hppavilion[1] has joined. 19:35:34 <oerjan> ? porthello 19:35:35 <HackEgo> porthello? ¯\(°​_o)/¯ 19:35:44 -!- zadock has joined. 19:35:53 <oerjan> ? portmanteau 19:35:54 <HackEgo> portmanteau? ¯\(°​_o)/¯ 19:36:55 -!- atrapado has quit (Quit: Leaving). 19:40:51 <fizzie> thanks portmanteau 19:40:52 <HackEgo> Thanks, portmanteau. Thortmanteau. 19:42:25 <b_jonas> oh wow 19:44:20 <fizzie> Achievement unlocked: rode the Tramlink: https://googledrive.com/host/0B4J9OAzXNfZAOXVEeDV1cXRTUXM 19:44:46 <fizzie> It was more trainy than the trams I'm used to. 19:44:57 <fizzie> Didn't even run on normal streets. 19:45:00 -!- fizzie has changed nick to fizzie. 19:46:08 -!- TieSoul has quit (Remote host closed the connection). 19:48:42 -!- Alcest_ has changed nick to Alcest. 19:51:28 -!- zadock has quit (Quit: Leaving). 20:05:05 <izabera>  paste "$(type -p thanks)"
20:05:07 <HackEgo> http://codu.org/projects/hackbot/fshg/index.cgi/file/tip/bin/thanks
20:05:54 <fizzie> ^show thanks
20:05:54 <fungot> >2,[>,]+15[>+6>+7>+3>+2<4-]>-6.>-.-7.+13.-3.+8.>-.>+2.<5[<]>[.>]>3+2.>.<3.<2[<]>[[-<2+>+>]+<-97[-4[-4[-6[-6[-4[>-<[-]]]]]]]>[[>]>2-11.<3[<]<.>3[.>]>3.>5][-]>]<3[[<]>2[.>]>5.>2]
20:06:08 <fizzie> Time for a random accuracy spot-check.
20:06:12 <fizzie> ^thanks portmanteau
20:06:12 <fungot> Thanks, portmanteau. Thortmanteau.
20:06:16 <fizzie> Pass.
20:06:35 <izabera> what does ^show do?
20:06:45 <boily> ^thanks
20:06:45 <fungot> Thanks, . T
20:06:51 <boily> thanks
20:06:53 <HackEgo> Thanks, com. Thom.
20:06:57 <fizzie> boily: No fair, corner cases.
20:07:04 <boily> mouah ah ah.
20:07:04 <fizzie> izabera: Shows the source code of a fungot ^command, if it's not a built-in.
20:07:04 <fungot> fizzie: maybe a good idea. i never knew it happened some 40 seconds ago.
20:08:57 <Taneb> ^Thanks Thor
20:09:02 <Taneb> ^thanks Thor
20:09:02 <fungot> Thanks, Thor. Thor.
20:09:44 <Phantom__Hoover> ^thanks hanks
20:09:44 <fungot> Thanks, hanks. Thanks.
20:10:04 <fizzie> ^thanks thanks
20:10:05 <fungot> Thanks, thanks. Thanks.
20:10:12 <fizzie> fungot: Stop repeating yoursellf.
20:10:13 <fungot> fizzie: they are sorted in a list
20:10:46 <Phantom__Hoover> ^bless fungot
20:10:53 <Phantom__Hoover> :(
20:12:37 <Taneb> ^thanks fungot
20:12:37 <fungot> Thanks, fungot. Thungot.
20:13:47 <gamemanj> ^thanks ABCDEF
20:13:47 <fungot> Thanks, ABCDEF. TBCDEF.
20:14:38 <boily> ^botsnack
20:14:38 <fungot> Oh nom nom nom!
20:19:58 <fizzie> I think case-insensitivity is also one differentiating feature, I didn't bother adding it.
20:20:12 <fizzie> thanks ABCDEF
20:20:13 <HackEgo> Thanks, ABCDEF. ThABCDEF.
20:21:24 <boily> thanks àèìòù
20:21:25 <HackEgo> Thanks, àèìòù. Tèìòù.
20:21:35 <boily> ^thanks àèìòù
20:21:35 <fungot> Thanks, àèìòù. Tèìòù.
20:21:56 <boily> thanks aaaaa
20:21:57 <HackEgo> Thanks, aaaaa. Thaaaaa.
20:22:03 <boily> ^thanks aaaaa
20:22:04 <fungot> Thanks, aaaaa. Thaaaaa.
20:22:25 <boily> thanks bbbbb
20:22:26 <HackEgo> Thanks, bbbbb. Tbbbb.
20:22:32 <boily> ^thanks bbbbb
20:22:33 <fungot> Thanks, bbbbb. Tbbbb.
20:22:43 <boily> good enough.
20:22:46 * boily pats fungot
20:22:46 <fungot> boily: are you using
20:22:52 <boily> fungot: my right hand
20:22:53 <fungot> boily: question not directed directly to forcer, anyone can make it in
20:23:02 <boily> guys! anybody can pat fungot!
20:23:02 <fungot> boily: i'll modify the core to split up .po files at some point today. all three of them
20:24:29 -!- \oren\ has quit (Read error: Connection reset by peer).
20:25:36 <fizzie> fungot: You've been internationalized?
20:25:36 <fungot> fizzie: oh, sorry, i don't
20:26:08 <gamemanj> internationalizing fungot
20:26:09 <fungot> gamemanj: wouldn't want to buy something. it wasn't supposed to be optimized into a nop.
20:26:25 <gamemanj> Ok, what universe does that bot live in
20:26:53 <gamemanj> aanyway, internationalizing fungot sounds like a great way to go mad
20:27:07 -!- \oren\ has joined.
20:27:07 <fungot> gamemanj: ( ...although that first one more like non-real-time pacman) evaluate the lambda expressions are hard to reason about a program running between irc and the unix 10 convention, and mid and res due to it checking k just before
20:31:07 <boily> gamemanj: why? fungot's code is very simple and easy to follow.
20:31:08 <fungot> boily: watch out for hot surfaces, not implicit ones. fnord are quite nice. then my old transmeta laptop might be an i/ o bound to the resulting brainfuck code.
20:31:31 <gamemanj> simple and easy to follow befunge? Is that possible?
20:31:38 <boily> fungot: you know I like your hot surfaces.
20:31:38 <fungot> boily: is anyone here familiar with chef? xd a 2d fnord games somewhat easily
20:32:12 <boily> gamemanj: you start from the begining, until you reach the end, and then you stop.
20:39:02 <oerjan> <gamemanj> aanyway, internationalizing fungot sounds like a great way to go mad <-- i say we get b_jonas to make a corpus out of those hungarian poems he mentioned a while ago.
20:39:02 <fungot> oerjan: by me. he felt that he is a sleep specialist after all. :) :(. i hate e-mailing strangers out of the core macro system make it much easier
20:43:41 -!- gde33 has joined.
20:49:19 -!- gamemanj has quit (Ping timeout: 240 seconds).
20:56:20 <int-e> i hate e-mailing strangers out of the core macro system as well
20:56:22 -!- variable has joined.
20:58:03 <oerjan> well if you're trapped in the core macro system it may be your only hope
20:58:35 <boily> macros unnerve me.
20:59:15 <boily> b_jonas: I guess you were happily designated to be part of the Fungot Translation Vanguard. sok szerencsét!
20:59:39 <fizzie> I think fungot's code is mostly relatively clean, except for the babbling part.
20:59:39 <fungot> fizzie: brainfuck coding time for compress.bf: 15 minutes. i dare you. unfortunately i usually work for utility companies
20:59:57 <fizzie> That one looks more traditionally befungey.
21:00:04 -!- hppavilion[1] has quit (Ping timeout: 246 seconds).
21:00:36 <fizzie> I'm a bit unnerved at the thought of fungot working for utility companies.
21:00:36 <fungot> fizzie: why is perl 6?) was looking for a way to increment each one at each subsequent iteration, i realised that was a fatal flaw
21:03:33 <fizzie> TIL: my work arrival time correlates only very weakly (ρ=0.106) with the time I leave.
21:04:31 <boily> Fungot Power Inc.
21:09:49 <shachaf> ^style oots
21:09:50 <fungot> Selected style: oots (Order Of The Stick)
21:09:57 <shachaf> fungot power?
21:09:57 <fungot> shachaf: too much clean and safe places for us, " oh, i get it.
21:12:19 <boily> shachaf: Markov Chain Lightning delivered to your home!
21:12:35 <shachaf> fungot lightning rods?
21:12:35 <fungot> shachaf: the more open-minded, you know that no one leaves. point blank range. normally, i would work, you can i get you anything about xykon. we just sort and therefore, the six other pages at " nodwick" gift.
21:14:31 -!- AnotherTest has quit (Ping timeout: 244 seconds).
21:20:30 <int-e> fungot's irc style is scary
21:20:30 <fungot> int-e: your brother and, apparently, there, about that... the cleric skipped to the part about the oath" story is just that was the point, can anyone hear, oh firstborn child of mine, all mine and make my will known.
21:20:49 <int-e> fungot: loop?
21:20:50 <fungot> int-e: my name is " roy's archon" thing, sir, and i know that, i'm so good that the dwarf, " stabby" is.
21:21:35 <fizzie> fungot: Can that sword alone stop?
21:21:35 <fungot> fizzie: and, as it is with my people, i would get the message would actually put us in mortal danger is infinitesimally low.
21:27:32 -!- Phantom_Hoover has joined.
21:27:47 -!- mtve has joined.
21:29:54 -!- gde33 has left.
21:30:12 -!- Phantom__Hoover has quit (Ping timeout: 268 seconds).
21:36:24 -!- hppavilion[1] has joined.
21:45:58 <boily> https://dl.dropboxusercontent.com/u/2023808/path4161.png
21:53:57 -!- coppro_ has changed nick to coppro.
22:03:42 -!- variable has changed nick to trout.
22:04:44 -!- Wright has joined.
22:05:09 <hppavilion[1]> Hellu
22:12:21 <boily> helloppavelloiellon[1]!
22:12:44 <hppavilion[1]> OK then
22:13:01 <hppavilion[1]> I'm trying to get a chess board working in Python
22:13:19 <hppavilion[1]> I'm going to create a suite of esochesses that people on this channel can play if they want
22:14:24 -!- oerjan has quit (Quit: Nite).
22:17:22 <hppavilion[1]> (Or, I suppose, people on other channels xD)
22:17:31 <hppavilion[1]> (Or even people that don't even use IRC)
22:18:01 <HackEgo> [wiki] [[User:Zzo38/Programming languages with unusual features]] http://esolangs.org/w/index.php?diff=44593&oldid=44450 * Zzo38 * (+1993)
22:19:17 <zzo38> This computer has Python 2.7.3; will that one work?
22:20:41 <hppavilion[1]> zzo38: Probably. I'm currently using Anaconda 3, but I'll try not to use any non-backwards-compatible features if I can remember
22:21:12 <hppavilion[1]> (no @ operator for me :,(.)
22:21:44 <hppavilion[1]> (OK, to be fair, Anaconda 3 is only for Python 3.4 currently, so I can't use the @ operator anyway)
22:22:52 -!- ineiros has quit (Ping timeout: 246 seconds).
22:22:59 -!- aloril has quit (Ping timeout: 240 seconds).
22:23:07 <hppavilion[1]> I really have no clue what I'm doing ATM xD
22:23:27 <hppavilion[1]> I only have a reader's understanding of chess
22:23:40 <hppavilion[1]> (That is, I don't often play)
22:24:53 -!- ineiros has joined.
22:27:20 -!- atehwa has quit (Ping timeout: 265 seconds).
22:27:49 -!- atehwa has joined.
22:30:01 -!- aloril has joined.
22:31:43 <Phantom_Hoover> hey fizzie does fungot have an installation manual
22:31:43 <fungot> Phantom_Hoover: i get to go. welcome go puberty. ( does something with a stick, and you, the gods, my poor useless eyes! helping!
22:33:55 <boily> what's the @ operator?
22:36:48 -!- hppavilion[1] has quit (Ping timeout: 272 seconds).
22:36:56 <fungot> fizzie: for that, my family would like my mode of transit returned, do i not, lieutenant, do i not, lieutenant, do i not, lieutenant, do i not, lieutenant, do i not, lieutenant, do i not, lieutenant, do i not, lieutenant, do i not, lieutenant, do i not, lieutenant, do i not, lieutenant, do i not, lieutenant, do i not, lieutenant, do i not, lieutenant, do i not, lieutenant, do i not, lieutenant, do i not, lieutenant, do i not, l
22:37:12 <fizzie> Phantom_Hoover: It might be missing something, though; I think there was a bit of troubleshooting involved, the last time someone tried to run it.
22:42:04 <fizzie> boily: It's nothing in particular in Python (except for the decorator syntax), but from 3.5 onwards you can overload it as an infix binary operator.
22:42:49 <Phantom_Hoover> so it's just a spare operator, basically?
22:43:23 <fizzie> Yes. The motivation chapter says it's specifically meant so that you can have * and @ for elementwise and matrix multiplications.
22:43:56 <fizzie> That's why the magic method names for overloading @ (and @=) are __matmul__, __rmatmul__, __imatmul__.
22:45:40 <Phantom_Hoover> well that's a bit silly
22:46:46 <fizzie> I think they're courting all those numerical programming folks.
22:47:29 <Phantom_Hoover> oh right, it is actually defined as matrix multiplication
22:48:21 <izabera> fizzie: can you add the fi legature in zem.fi ?
22:48:22 <Phantom_Hoover> i thought you meant it was undefined in the stdlib but they'd made it possible to define it
22:49:19 <fizzie> Phantom_Hoover: That's what I mean.
22:49:41 <fizzie> Phantom_Hoover: "But isn't it weird to add an operator with no stdlib uses?" "It's certainly unusual --" http://legacy.python.org/dev/peps/pep-0465/
22:49:42 <Phantom_Hoover> well that's a bit silly
22:50:18 <fizzie> They do define "recommended semantics" for anyone using it.
22:54:42 <fizzie> "The swirly shape is reminiscent of the simultaneous sweeps over rows and columns that define matrix multiplication"
22:54:45 <FireFly> izabera: ligature characters in domain names is probably a bad idea
22:54:53 <izabera> it's not in the name
22:55:04 <izabera> that text is in the banner
22:55:09 <FireFly> ah
22:56:15 <FireFly> fizzie: that is very colourful
22:57:08 <fizzie> Re domain names, at least Chrome goes to zem.fi if I type in zem.ﬁ -- but I'm not sure what I was supposed to add and where.
23:00:08 -!- Patashu has joined.
23:03:37 -!- mauris has quit (Ping timeout: 246 seconds).
23:05:13 -!- trout has quit (Ping timeout: 246 seconds).
23:11:11 <FireFly> Perhaps to have the "zem.fi" heading use a hardcoded ligature character
23:12:37 <fizzie> I don't really want to do *that*. I should update the related CSS stylings, though; it's using the nonstandard -webkit-font-feature-settings while apparently there's nowadays a simpler "font-variant-ligatures" too.
23:13:26 <fizzie> Although I couldn't get the 'fi' ligature to show up even after twiddling those; but it's supposed to exist in the font, it's the same font I used for the image.
23:19:43 <fizzie> It renders right for me in the word "fit" at http://www.georgduffner.at/ebgaramond/design.html -- not sure what's up with that.
23:32:08 <fizzie> Must be something related to pulling it from Google's WebFonts thingie; the other site hosts it locally.
23:41:24 -!- lleu has quit (Quit: That's what she said).
←2015-10-03 2015-10-04 2015-10-05→ ↑2015 ↑all