←2020-03-04 2020-03-05 2020-03-06→ ↑2020 ↑all
00:06:43 -!- xelxebar has quit (Ping timeout: 240 seconds).
00:14:11 -!- arseniiv has quit (Ping timeout: 258 seconds).
00:15:23 -!- LKoen has joined.
00:22:29 -!- imode has quit (Quit: WeeChat 2.7.1).
00:24:57 <MTGBusyBeaver> bad news for the MTG turing machine for the most damage challenge: Arcbond triggers can be interleaved with death triggers. This seems like a hard problem to fix.
00:41:39 -!- imode has joined.
00:46:39 <esowiki> [[User:PythonshellDebugwindow/(Unnamed language)]] M https://esolangs.org/w/index.php?diff=70161&oldid=70156 * PythonshellDebugwindow * (+652) /* Input/Output */
01:04:09 -!- LKoen has quit (Remote host closed the connection).
01:26:58 <zzo38> Try the Unusenet newsgroup un2.org.zzo38computer.magic.maths if you are interested in those kind of things with Magic: the Gathering; there may also be some Usenet newsgroups about such thing although I do not know which ones they are.
01:35:38 -!- imode has quit (Ping timeout: 240 seconds).
01:38:59 <esowiki> [[User:PythonshellDebugwindow/(Unnamed language)]] M https://esolangs.org/w/index.php?diff=70162&oldid=70161 * PythonshellDebugwindow * (+285) /* Cat for 1 char */ FINISH
01:47:47 <esowiki> [[User:PythonshellDebugwindow/(Unnamed language)]] M https://esolangs.org/w/index.php?diff=70163&oldid=70162 * PythonshellDebugwindow * (+69) /* Interpret PlusOrMinus 1 char at a time */ done
02:16:45 -!- imode has joined.
03:21:53 -!- xkapastel has quit (Quit: Connection closed for inactivity).
03:32:48 -!- longname has quit (Read error: Connection reset by peer).
03:35:54 -!- longname has joined.
04:46:10 -!- imode has quit (Ping timeout: 255 seconds).
04:57:59 -!- FreeFull has quit.
05:30:49 -!- studio has joined.
05:31:19 -!- studio has quit (Read error: Connection reset by peer).
05:59:18 <esowiki> [[ALIMBIHNN]] https://esolangs.org/w/index.php?diff=70164&oldid=70158 * JonoCode9374 * (+60)
06:18:24 <esowiki> [[ZFC++]] N https://esolangs.org/w/index.php?oldid=70165 * Hakerh400 * (+7714) +[[ZFC++]]
06:18:28 <esowiki> [[Language list]] https://esolangs.org/w/index.php?diff=70166&oldid=70152 * Hakerh400 * (+12) +[[ZFC++]]
06:18:29 <esowiki> [[User:Hakerh400]] https://esolangs.org/w/index.php?diff=70167&oldid=70087 * Hakerh400 * (+12) +[[ZFC++]]
06:35:26 <esowiki> [[ZFC++]] M https://esolangs.org/w/index.php?diff=70168&oldid=70165 * Hakerh400 * (-1)
07:45:12 -!- Lord_of_Life has quit (Ping timeout: 256 seconds).
07:46:37 -!- Lord_of_Life has joined.
08:41:50 -!- b_jonas has quit (Quit: leaving).
09:07:19 -!- ^[ has quit (*.net *.split).
09:07:22 -!- Cale has quit (*.net *.split).
09:07:22 -!- APic has quit (*.net *.split).
09:07:22 -!- Bowserinator has quit (*.net *.split).
09:19:56 -!- arseniiv has joined.
09:43:20 -!- ^[ has joined.
09:43:20 -!- Cale has joined.
09:43:20 -!- APic has joined.
09:43:20 -!- Bowserinator has joined.
09:43:58 -!- APic has quit (Max SendQ exceeded).
09:49:09 -!- APic has joined.
09:58:13 -!- cpressey has joined.
10:12:14 <cpressey> Good morning. I have a vague idea for an esolang. It's like a Turing machine, but instead of writing symbols on the tape, it writes continuations.
10:13:03 <cpressey> Now, if you could compare continuations, you could use them like symbols, and that would be boring, so ideally, what the machine does with these continuations written on its tape, is to continue them.
10:13:17 <cpressey> But I don't know how that would work.
10:15:43 <cpressey> I don't actually think it *would* work. One of the salient aspects of a Turing machine is that it looks at what symbols are on the tape, it compares them against something. To do something far removed from that would be very un-Turing-machine-like.
10:16:23 <cpressey> So maybe you can compare them, even though that's boring.
10:17:07 <cpressey> Or... well, I'll think about it.
10:21:08 -!- wib_jonas has joined.
10:32:17 <cpressey> Maybe tape cells contain pairs of continuations. But what determines which one is chosen?
10:47:22 <cpressey> In the absence of closures, continuations are basically jump labels. Maybe this would be more like OISC et al.
11:23:33 <cpressey> On a Turing machine, jump labels are state labels. So, in this, state labels and symbols are the same thing?
11:24:28 <cpressey> "If the state written on the tape is not S3, transition to state specified on the tape and move head left"
11:24:51 <cpressey> You have to use something like "not", obviously.
11:27:53 <cpressey> But then, you are introducing boolean expressions in these conditions. That's more complexity than a simple table, and it's less pleasing.
11:28:36 <cpressey> But maybe not much, maybe you could still keep it simple somehow.
11:30:13 -!- longname has quit (Read error: Connection reset by peer).
11:32:02 -!- longname has joined.
11:38:52 -!- MTGBusyBeaver has quit (Remote host closed the connection).
12:25:06 <cpressey> "If the state written on the tape is less than the current state, transition to it and write Sn on the tape and move the head L/R. Else if greater, transition to it and write Sm and move L/R. Else transition to Sz and maybe do other stuff."
12:25:23 <cpressey> Which assumes a total order on states but whatever
12:25:47 <cpressey> You can't make an omelette and whatever
12:26:59 <cpressey> I think we can drop "And maybe do other stuff". So every state's tuple is like (state, direction, state, direction, state).
12:27:52 <cpressey> S1=(S4, L, S5, L, S9), and so forth
12:28:05 <cpressey> I think this might work. I don't know that it's very interesting.
12:36:00 <cpressey> Wait wait. Does making two small holes in an egg in order to blow it out count as breaking it? If not, then it should be possible to make an omelette without breaking any eggs.
12:42:06 -!- xkapastel has joined.
13:07:48 -!- laerlingSAP has joined.
13:13:18 <esowiki> [[User:PythonshellDebugwindow/(Unnamed language)]] M https://esolangs.org/w/index.php?diff=70169&oldid=70163 * PythonshellDebugwindow * (+18) /* Interpret PlusOrMinus 1 char at a time */
13:44:52 -!- kspalaiologos has joined.
14:40:27 <arseniiv> cpressey: an interesting idea!
14:41:31 <arseniiv> what would this turn into if one uses delimited continuations as an inspiration? or (algebraic?) effects?
14:41:33 <longname> Can't you just make one on the bottom and let it all drip out?
14:42:59 <longname> Omletwise, that is, no comment on the esolang idea
14:54:30 <esowiki> [[Special:Log/upload]] upload * JulienDelplanque * uploaded "[[File:Langton ant.png]]"
14:58:35 <esowiki> [[BytePusher]] https://esolangs.org/w/index.php?diff=70171&oldid=70125 * JulienDelplanque * (+230) Added Langton ant BP software.
14:58:45 <arseniiv> what would this turn into if one uses delimited continuations as an inspiration? or (algebraic?) effects? => hm I think these aren’t compatible with the spirit that TC is “imperative”
14:59:33 <int-e> "Do this now!" "I'll do it later and get back to you."
15:00:03 <int-e> (delimited by end of working day? :P)
15:00:58 <int-e> (I still don't get delimited continuations.)
15:01:18 -!- heroux has quit (Ping timeout: 260 seconds).
15:01:24 -!- erdic has quit (Ping timeout: 248 seconds).
15:01:31 <cpressey> I don't pretend to understand what delimited continuations are or how they work
15:01:44 -!- heroux has joined.
15:01:48 -!- erdic has joined.
15:54:41 -!- kspalaiologos2 has joined.
15:55:32 -!- kspalaiologos has quit (Remote host closed the connection).
15:55:39 -!- kspalaiologos2 has changed nick to kspalaiologos.
16:00:34 <kspalaiologos> Any TI wizards out there? Having problems with TI-Connect, I essentially own a TI83+ and a silverlink, I'm able to transfer programs from the calculator to my PC, but not the other way round
16:01:10 <kspalaiologos> always getting an error on TI side saying "error in xmit"
16:01:28 <kspalaiologos> tried various USB ports, tried clearing ram, tried changing batteries
16:01:52 <kspalaiologos> I removed the drivers too
16:02:02 <kspalaiologos> but installed them soon afterwards again
16:03:09 <wib_jonas> kspalaiologos: I suspect that you're the most qualified TI calculator wizard here
16:03:27 <kspalaiologos> interesting
16:03:49 <kspalaiologos> I'm tired of BASIC and I'd like some of my assembly programs sent over. I've got them in a correct .8xp format
16:06:08 <wib_jonas> kspalaiologos: does it work if you transfer the same BASIC program back that you have sent from the calculator?
16:06:10 -!- kspalaiologos has quit (Quit: Leaving).
16:09:48 -!- kspalaiologos has joined.
16:10:06 <kspalaiologos> I did reboot
16:10:17 <kspalaiologos> and I'll check that
16:12:47 -!- imode has joined.
16:13:48 <kspalaiologos> works!
16:13:51 <kspalaiologos> how so
16:15:18 <arseniiv> I have a weird game for y’all: take four points randomly in some area, make a Bézier curve on them, then vary one of the points to make that curve have the least length possible, by hand :o
16:16:51 <arseniiv> <cpressey> I don't pretend to understand what delimited continuations are or how they work => sad :( I think I had some understanding one time, and that required to use a call stack and cut/paste segments of it
16:17:12 <arseniiv> int-e: ^ too
16:17:57 <arseniiv> though I think shift…reset, or was there something less confusing to express them, is nonetheless less clear than effects+handlers
16:19:37 <arseniiv> it’s like `shift` does the work which should be done in another place (cf. try…except for exceptions) and because of that delimited continuations are weird and unhandy. Call/cc is too by virtue of being a `shift` without `reset`
16:19:53 <kspalaiologos> I think I may have compiled my app for ION shell
16:20:01 <kspalaiologos> I changed it to an asm( executable
16:20:09 <kspalaiologos> doesn't work aswell though
16:20:20 <arseniiv> kspalaiologos: what’s ION?
16:20:37 <kspalaiologos> TI83+ shell
16:20:50 <arseniiv> ah
16:21:02 <kspalaiologos> I'll try Devpac8x
16:23:17 <kspalaiologos> xaxaxaxa it worked
16:24:06 <kspalaiologos> it doesn't seem to function though
16:24:12 <kspalaiologos> damn
16:25:08 <kspalaiologos> I'm so out of ideas, what could go wrong
16:28:02 <arseniiv> re. shift…reset: a call to `reset` marks a place in the call stack and then when we call `shift`, that marks the other end of the segment which is cut away and packaged into a delimited continuation which is fed, as a kind of closure, to the function passed to `shift`. If `shift` doesn’t happen, so be it, the `reset` mark would perish when we step out of the call to `reset`. Yes this is more or less easy but nonetheless I’d have an effectful pri
16:28:02 <arseniiv> mitives instead of `shift` and handlers instead of both `reset` points and passing a “handler” to `shift`. This is the way with exceptions (`throw` as a handlerless `shift`, `catch` as a handler, though we don’t get a continuation because the effect is the most boring one possible)
16:31:18 <arseniiv> more specifically, because `throw :: exc -(Throws)> Void`, so if we would get a continuation, it would be an uncallable one (a → Void), and so we usually aren’t given one. If I get it right
16:31:39 <arseniiv> I would need to consult a paper on this if someone would be interested in these details
16:32:03 <int-e> we had a link to a paper here a while ago that cast continuations as evaluation contexts
16:32:25 <arseniiv> int-e: delimited too?
16:32:28 -!- tromp has quit (Remote host closed the connection).
16:32:34 <int-e> yes, delimited
16:32:49 <arseniiv> nice
16:32:51 <int-e> A context being a term with a single hole... it has a root and the hole where you can plug stuff in.
16:33:33 <arseniiv> btw if there’s a formulation more like effectful primitives + handlers instead like shift…reset, I may love delimited continuations back again
16:33:57 <cpressey> OK, so `shift` gets a ... a "delimited continuation" that extends from the `shift` to the `reset`. What happens when you call this "delimited continuation"? I mean, what happens when it finishes?
16:34:14 <cpressey> Does it return to somewhere?
16:34:37 <int-e> cpressey: you plug the argument into the place where reset was invoked.
16:34:48 <cpressey> So it returns to the `reset`.
16:34:59 <cpressey> So it's not a continuation at all, it's a regular function.
16:35:02 <arseniiv> but I think it isn’t too possible as there can be many sorts of effects, and we’ll need dynamic typing in the worst case, static in the best, and then we’ll get algebraic effects
16:35:25 <int-e> It's not really true that 'shift' gets a delimited continuation as an argument either. you can perform actual computations inside shift before `reset` is performed. It's all rather messy.
16:35:26 <arseniiv> cpressey: `shift` effectively would return out of a call to `reset`
16:35:45 <arseniiv> (that is, if it will at all)
16:36:05 <int-e> cpressey: And that part (evaluation before reset) makes it much closer to a continuation
16:36:12 <int-e> s/makes/brings/
16:36:24 <kspalaiologos> verdict: z88dk is garbage, omit it if you'd like to keep your sanity, tigcc doesn't target TI83 at all, therefore there is no C compiler for TI83+
16:36:26 <cpressey> Just never mind, I've already given up again
16:36:31 <int-e> you just "fix" the part where continuations never return.
16:36:31 <arseniiv> int-e: yeah `shift` gets a handler which it applies to a d. continuation
16:37:07 <int-e> (because that part makes continuations highly non-modular)
16:37:25 <int-e> but I'm not convinced that delimited continuations are actually modular, so, meh.
16:37:47 <cpressey> But they have so many useful applications, like...
16:38:03 <cpressey> Uh...
16:38:16 <arseniiv> <cpressey> So it's not a continuation at all, it's a regular function. => it’s both; one of the points of delimited continuations is that they are functions and so are composable and easy to do something with
16:38:41 <int-e> cpressey: Writing papers.
16:39:30 <arseniiv> cpressey: it would be easier with a picture in mind, I think. I read a page somewhere, with simple examples, and now it seems I can succesfully rebuild the understanding from scratch
16:39:48 <arseniiv> I don’t remember what I’ve read though
16:40:49 <cpressey> Maybe I just associate continuations too closely with continuation-passing style.
16:40:49 <int-e> cpressey: I mean they're good for writing papers *about*, of course.
16:41:36 <arseniiv> <int-e> but I'm not convinced that delimited continuations are actually modular, so, meh. => don’t you think that’s because of `shift` taking a handler instead of `reset`? (Because I do)
16:41:41 <int-e> cpressey: Well CPS is special in that the continuations are never duplicated.
16:42:26 <cpressey> int-e: ? I thought CPS was special because *everything* is a continuation (and everything ends by calling another continuation).
16:43:01 <int-e> Sure, it's also special in that way.
16:43:29 <cpressey> Maybe in practice, you'd also never duplicate a continuation, buuut... maybe there's a counterexample for that too, dunno
16:43:55 <arseniiv> CPS is good to implement, yeah. Nonlinear use of continuations seems to be a big problem in implementing analogous things (like alg. effects), as they get too costly AFAIR
16:44:02 <int-e> You don't do callCC in CPS.
16:45:14 <int-e> (where a continuation escapes itself and then later on you might continue at that point in execution several times... which is just weird.)
16:45:20 <arseniiv> BTW is there a computational framework in which CPS translation is trivial (like, you write it in the first place)
16:45:35 <int-e> CPS is a very sand fragment of what continuations can do.
16:45:42 <int-e> *sane
16:46:10 <kspalaiologos> sdcc as always nails the task
16:46:32 <int-e> `? sdcc
16:46:34 <HackEso> sdcc? ¯\(°​_o)/¯
16:47:14 -!- sprocklem has quit (Ping timeout: 265 seconds).
16:50:35 <cpressey> arseniiv: I hesitate to even mention this, but there's a strong similarity between CPS and Haskell's "do" notation
16:51:34 <int-e> ...
16:51:44 <int-e> I think you mean the stuff that the do notation *hides*
16:51:57 <int-e> foo >>= continuation
16:52:01 <cpressey> Right, "what Haskell's 'do' notation desugars to", is what I meant
16:54:57 <cpressey> do { x1 <- action1 ; x2 <- action2 ; mk_action3 x1 x2 }
16:55:02 <cpressey> becomes: action1 >>= (\ x1 -> action2 >>= (\ x2 -> mk_action3 x1 x2 ))
16:56:10 <cpressey> There's obviously some relationship between monads and continuations, and I've even heard people whisper about it on occasion, but I'm not sure I've ever seen it in broad daylight
16:58:09 <cpressey> (I don't think it's Cont. Cont is for people who can't handle CPS without it being wrapped in a monad. Or something.)
17:04:51 <int-e> Cont is more or less CPS, except that it also allows some of the more insane stuff that leads to coroutines and callCC. And I've seen people say that Cont is really implementing delimited continuations rather than full ones. (Obviously there's no way to go beyond the initial runCont. But inside that it felt more like full continuations to me.)
17:06:22 -!- tromp has joined.
17:06:36 <int-e> :t lookup
17:06:37 <lambdabot> Eq a => a -> [(a, b)] -> Maybe b
17:06:40 <int-e> :t M.lookup
17:06:41 <lambdabot> Ord k => k -> M.Map k a -> Maybe a
17:07:42 <int-e> (I find that argument order counter-intuitve)
17:09:47 <int-e> Mostly because it violates the principle that the argument you're less likely to vary should go first, but also because it's the opposite order from M.! and I'd like to use `lookup` as an infix operator. Sigh.
17:11:04 -!- tromp has quit (Ping timeout: 256 seconds).
17:14:25 <cpressey> The answer https://stackoverflow.com/a/3323122 starts with "The first thing to realize about the continuation monad is that, fundamentally, it's not really doing anything at all." and a comment on it rephrases that as "So basically, bind is just CPS-transformed function application?" and I feel those are getting pretty close to the matter.
17:15:32 <cpressey> I still hate and fear monads, btw
17:16:19 <cpressey> Ruination, I tell you
17:16:32 -!- imode has quit (Ping timeout: 258 seconds).
17:17:12 <int-e> cpressey: https://wiki.haskell.org/wikiupload/1/1d/Monica_monad_falconnl.png
17:18:18 <int-e> "This is Monica Monad, and she's a Warm Fuzzy Thing. Just giving a face to SPJ's alternative name for monads :)" (from https://wiki.haskell.org/Haskell_logos/New_logo_ideas which is huge)
17:22:22 <arseniiv> <cpressey> Right, "what Haskell's 'do' notation desugars to", is what I meant => yeah, this is an interesting point! I didn’t connect these two before
17:29:05 <arseniiv> <int-e> (Obviously there's no way to go beyond the initial runCont. But inside that it felt more like full continuations to me.) => I think I’ve seen `reset` implemented using precisely `runCont` in there
17:30:01 <arseniiv> <int-e> Mostly because it violates the principle that the argument you're less likely to vary should go first, but also because it's the opposite order from M.! and I'd like to use `lookup` as an infix operator. Sigh. => yeah I think I was stunned by that too
17:30:53 <kspalaiologos> I did chess on TI83
17:31:05 <kspalaiologos> and it's only 2 kilobytes
17:32:15 <arseniiv> kspalaiologos: how often does it win you? (if you tested enough times)
17:32:24 <kspalaiologos> I just did the openings
17:32:33 <kspalaiologos> it has 6 ply search depth
17:32:46 <kspalaiologos> I beat it up with king's gambit but it's kinda unfair because the computer is materialistic
17:33:00 <kspalaiologos> and when it realises it should start defending instead of greedy taking my pieces it's already too late
17:33:02 <arseniiv> anyway I hadn’t written any chess at all so it should be cool!
17:33:31 <kspalaiologos> the C code size itself is around 800 bytes
17:33:42 <kspalaiologos> but it works, somehow
17:33:49 -!- tromp has joined.
17:34:11 <kspalaiologos> next I'll try maybe tetris
17:34:27 <kspalaiologos> I did already pong in BASIC, it was a real torture tho
17:34:36 <kspalaiologos> took me around two boring lectures to finish it
17:35:16 <kspalaiologos> there are some problems tho
17:41:51 -!- cpressey has quit (Quit: A la prochaine.).
17:41:53 <kspalaiologos> damn it crashed in the middle of the game
17:42:45 <int-e> resign!
17:43:21 <int-e> Well, "rage quit" would be more accurate :P.
17:43:35 <arseniiv> roflol
17:43:45 <arseniiv> rofoldl*
17:45:16 <arseniiv> some haskell designs there associate with something Hebrew, like that double lambda thing, maybe it resembles an aleph…
17:47:37 <arseniiv> > let { the [x] = x; the _ = error "the: there are many or no values" }
17:47:39 <lambdabot> <no location info>: error:
17:47:39 <lambdabot> not an expression: ‘let { the [x] = x; the _ = error "the: there are man...
17:48:15 <int-e> lambdabot's not a repl
17:48:18 <int-e> there's @let though
17:48:38 <int-e> and @undef to undo all the damage @let can do
17:49:29 <arseniiv> > the "a"
17:49:31 <lambdabot> 'a'
17:49:47 -!- FreeFull has joined.
17:49:56 <int-e> > the "ä"
17:49:58 <lambdabot> '\228'
17:50:06 <arseniiv> > the "ä"
17:50:09 <lambdabot> *Exception: the: there are many or no values
17:50:19 <arseniiv> HAHAHA
17:50:32 <arseniiv> UNICODE IS ILL-DEFINED
17:51:24 <arseniiv> (not that it’s news or that I really think so)
17:51:46 <int-e> I'm surprised.
17:51:54 <int-e> I really think so. :P
17:52:15 <kmc> @let 2 + 2 = 5
17:52:17 <lambdabot> Defined.
17:52:18 <arseniiv> it’s ill-defined but not so ill-defined as to say that for the sake of it
17:52:19 <kmc> > 2 + 2
17:52:21 <lambdabot> error:
17:52:21 <lambdabot> Ambiguous occurrence ‘+’
17:52:21 <lambdabot> It could refer to either ‘Prelude.+’,
17:52:32 <int-e> > 2 L.+ 2
17:52:34 <lambdabot> 5
17:52:37 <int-e> @undef
17:52:37 <lambdabot> Undefined.
17:52:44 <kmc> qualified infix operators always look Wrong
17:52:55 <int-e> I got used to them.
17:52:57 <arseniiv> int-e: that’s cru… ah, you’ve undone that already
17:53:07 <arseniiv> > :t the
17:53:08 <esowiki> [[XENBLN]] M https://esolangs.org/w/index.php?diff=70172&oldid=70110 * PythonshellDebugwindow * (+124) /* Datatypes */
17:53:09 <lambdabot> <hint>:1:1: error: parse error on input ‘:’
17:53:10 <kmc> > let 2 + 2 = 5 in 2 + 2
17:53:12 <lambdabot> 5
17:53:19 <kmc> > let 2 + 2 = 5 in 1 + 1
17:53:20 <arseniiv> my poor the
17:53:22 <lambdabot> *Exception: <interactive>:3:5-13: Non-exhaustive patterns in function +
17:53:24 <kmc> F
17:53:35 <int-e> kmc: I use M.! quite a bit.
17:53:55 <arseniiv> kood emcening
17:54:44 <int-e> > let the x | [x] <- x = x in the []
17:54:46 <lambdabot> *Exception: <interactive>:3:5-24: Non-exhaustive patterns in function the
17:54:50 <esowiki> [[Special:Log/move]] move * PythonshellDebugwindow * moved [[User:PythonshellDebugwindow/(Unnamed language)]] to [[User:DINAC]]: got a name
17:55:39 <int-e> Haskell: Oh my god, it's full of syntax!
17:58:16 <int-e> arseniiv: the disappointing thing about your ä is that I couldn't see the difference in the terminal... somebody canonicalized it properly along the way.
17:58:50 <int-e> (irssi, screen, xterm... I don't want to figure out which)
17:59:19 <int-e> arseniiv: But the logs gave you away, and confirmed what the error message already suggested.
18:00:33 <arseniiv> int-e: ah, I didn’t think it would be non-obvious
18:00:52 <kmc> was it actually canonicalized? or just rendered the same way
18:01:12 <kmc> if a font has a character for ä then it should use that for the combining-character version too
18:01:34 <kmc> you can usually tell when the combining sequences are rendered by combining glyphs
18:01:37 <kmc> because they look like ass
18:02:37 <arseniiv> kmc: yeah, frequently with fonts which aren’t up-to-date or whose creators think nobody uses non-accented non-latin
18:03:01 <arseniiv> or when font is okay but the renderer is not
18:03:58 <int-e> kmc: I pasted it into od -tx1
18:04:13 <kmc> ah
18:04:51 <arseniiv> the question is could an Unicode analogue be simpler? (IDK, but I know many people think it could be, but I’m not sure many of them aren’t overlooking things)
18:06:28 <kmc> there's a hot-button question if ever there was one
18:07:07 -!- batman_nair has joined.
18:07:12 <kmc> given all the design requirements I think it couldn't be made that much simpler
18:08:07 <esowiki> [[DINAC]] N https://esolangs.org/w/index.php?oldid=70175 * PythonshellDebugwindow * (+8303) Better
18:08:37 <esowiki> [[User:DINAC]] M https://esolangs.org/w/index.php?diff=70176&oldid=70173 * PythonshellDebugwindow * (-8459) Blanked the page
18:08:43 <kmc> for example the requirement of round-trip compatibility with all legacy encodings adds a lot of complexity
18:08:46 <kmc> or at least a lot of characters
18:09:00 <kmc> that's why accented characters exist in both combined and combining-sequence versions
18:10:25 <kmc> but round trip compatibility is a nice thing to have because then you can design systems (programming languages, databases, etc) where all text is canonically represented as Unicode and you convert to/from legacy encodings at the edges
18:12:35 <int-e> arseniiv: I wasn't confused about what you did, I just looked for confirmation. Also, it could've been a zero-width space :)
18:16:34 <int-e> > text $ reverse "äa"
18:16:36 <lambdabot> äa
18:16:53 <int-e> > text $ reverse "äa"
18:16:55 <lambdabot>
18:17:00 <int-e> arseniiv: ^^ :)
18:18:35 -!- LKoen has joined.
18:20:08 <arseniiv> int-e: nice!!
18:20:45 <arseniiv> <int-e> Also, it could've been a zero-width space :) => a good idea too
18:21:55 <arseniiv> <kmc> that's why accented characters exist in both combined and combining-sequence versions => oh, I didn’t think about the compatibility reason
18:22:39 <arseniiv> Unicode makes more sense now
18:25:38 <int-e> https://xkcd.com/927/
18:26:11 <int-e> Arguably we could get away without emoji. Or having a ton of different math fonts in there.
18:26:20 * int-e shrugs.
18:26:39 <int-e> Of course it's not that easy.
18:26:58 <arseniiv> yeah math fonts are a weird thing
18:27:07 <int-e> If Unicode didn't accept emoji, we'd have vendor-specific encodings for those, because clearly there's a desire to have them embedded in text.
18:27:26 <arseniiv> I wouldn’t say “a mistake” aloud but I’ll be thinking that secretly
18:27:33 <int-e> And I suspect the same kind of reasoning is behind the math fonts.
18:27:56 <arseniiv> but does anyone use them seriously?
18:28:05 <int-e> Otoho, why isn't there a full set of Shogi pieces?
18:28:37 <arseniiv> someone should propose
18:28:58 <int-e> chess: ♔♕♖♗♘♙♚♛♜♝♞♟ shogi: ⛉⛊☖☗
18:29:08 <int-e> seems a bit unfair.
18:30:27 <int-e> (Also the turned Shogi pieces look different from the normal ones to me. Fun.)
18:30:44 <int-e> (But that's a font issue, of course.)
18:31:42 <int-e> "Otoho", hmm. Some Japanese influence there? :P
18:42:09 -!- wib_jonas has quit (Remote host closed the connection).
18:51:33 -!- kspalaiologos has quit (Quit: Leaving).
18:51:37 -!- batman_nair has quit (Remote host closed the connection).
18:58:56 -!- tromp has quit (Remote host closed the connection).
19:02:30 -!- tromp has joined.
19:22:39 -!- tromp has quit (Remote host closed the connection).
19:37:57 <zzo38> For shogi, I think you can just use the kanji, sometimes upsidedown in case of a diagram
19:44:16 -!- imode has joined.
19:46:00 -!- Lord_of_Life_ has joined.
19:47:42 -!- Lord_of_Life has quit (Ping timeout: 256 seconds).
19:48:50 -!- Lord_of_Life_ has changed nick to Lord_of_Life.
19:54:38 -!- tromp has joined.
19:59:35 -!- tromp has quit (Ping timeout: 272 seconds).
20:05:22 -!- ArthurStrong has joined.
20:15:17 -!- b_jonas has joined.
20:18:22 <b_jonas> `olist 1194
20:18:23 <HackEso> olist 1194: shachaf oerjan Sgeo FireFly boily nortti b_jonas
20:37:49 <arseniiv> is there a distinction between consistency and soundness in logic? I think I missed that one but now I suspect the first is syntactic and the second is semantic
20:39:55 <spruit11> That should be on wikipedia.
20:40:48 <arseniiv> ah, seems it’s more complicated: a set of formulas is or isn’t consistent, but soundness is a property of a particular logic. Hm I think there still were other uses of these words
20:40:54 <arseniiv> spruit11: fair
20:42:04 <spruit11> The usual distinction is soundness vs completeness.
20:42:32 <spruit11> Consistency is of course also an issue but most systems you'll encounter are consistent.
20:42:58 <spruit11> Unless you're interested in para-consistent logics, the subject doesn't come up a lot.
20:50:09 <esowiki> [[User:PythonshellDebugwindow/(Unnamed language)]] M https://esolangs.org/w/index.php?diff=70177&oldid=70174 * PythonshellDebugwindow * (-24) Blanked the page
20:51:54 <arseniiv> ah, I think it clarified for me. Consistency of a set Γ is “not (Γ ⊦ ⊥)” and consistency of a logic is the consistency of ∅; and there is no “soundness of Γ” as it’s just not obvious how and for what reason would one define that
21:02:03 <esowiki> [[User:PythonshellDebugwindow]] M https://esolangs.org/w/index.php?diff=70178&oldid=70148 * PythonshellDebugwindow * (-12)
21:07:40 <esowiki> [[DINAC]] M https://esolangs.org/w/index.php?diff=70179&oldid=70175 * PythonshellDebugwindow * (+13) /* Subpages */
21:17:03 <spruit11> That doesn't seem right to me but it's a long time ago for me too.
21:17:40 <spruit11> Uhm, it should be possible for classical logics to state false theorems.. I guess.
21:18:09 <spruit11> I.e., 3=2 is simply false. That doesn't make a logic inconsistent.
21:18:14 <spruit11> Dunno.
21:19:26 <int-e> spruit11: it's okay; {3=2} is an inconsistent set of statements; it doesn't make the logic itself inconsistent
21:42:53 -!- tromp has joined.
21:46:04 <spruit11> Right. So the question is what Γ denoted in that statement.
21:46:22 <spruit11> But okay, too long ago. I refer to wikipedia.
21:47:36 -!- tromp has quit (Ping timeout: 256 seconds).
21:56:18 -!- sprocklem has joined.
22:09:36 <esowiki> [[DINAC/STDLIB]] N https://esolangs.org/w/index.php?oldid=70180 * PythonshellDebugwindow * (+1821) Created page with "DINAC '''STDLIB''' is a library of functions you can use in your code. They are grouped together for ease of access, and from there the individual functions are arranged alpha..."
22:09:54 <esowiki> [[DINAC/STDLIB]] M https://esolangs.org/w/index.php?diff=70181&oldid=70180 * PythonshellDebugwindow * (+4)
22:10:29 <esowiki> [[DINAC]] M https://esolangs.org/w/index.php?diff=70182&oldid=70179 * PythonshellDebugwindow * (+45) /* Comments */
22:27:14 -!- tromp has joined.
22:29:33 -!- tromp has quit (Remote host closed the connection).
22:29:59 -!- tromp has joined.
22:38:10 -!- atslash has quit (Quit: Leaving).
22:45:38 -!- imode has quit (Ping timeout: 256 seconds).
22:53:06 -!- imode has joined.
22:54:24 -!- atslash has joined.
23:23:14 -!- LKoen has quit (Quit: “It’s only logical. First you learn to talk, then you learn to think. Too bad it’s not the other way round.”).
23:31:37 -!- WillGibson has joined.
23:36:16 <WillGibson> helloooooo
23:41:43 <WillGibson> anyone in saudi
23:45:00 -!- WillGibson has left.
←2020-03-04 2020-03-05 2020-03-06→ ↑2020 ↑all