00:14:16 was here a holy war about whether ++ and -- (and specifically, both prefix and postfix kinds of these) are such a good thing in a language? 00:15:06 just putting this out there for people who are interested: I've made a draft of what a 10+halt-waterclock interpreter for a TC language in The Waterfall Model could look like: http://nethack4.org/pastebin/and-high-rise.txt 00:15:13 arseniiv: not really a hot holy war, but python and ruby silently don't have them 00:15:31 I haven't proved this TC, nor implemented the actual interpreter yet 00:15:42 just described the basic design and what all the waterclocks are used for 00:15:46 so it's quite possible there's a mistake 00:15:50 but I thought I'd post it anyway 00:16:08 it's almost certainly possible to do better than 10 but I thought I'd err on the side of producing something TC 00:16:42 b_jonas: yeah Python is amidst sane ones here :) (for what I think) 00:17:18 -!- MtGBusyBeaver has joined. 00:17:29 ais523: do you also have a suspected order of magnitude on the upper bound of the waterfall matrix elements, as a function of the program size in some other more well-known computational model? 00:17:32 ais523: The Waterfall Model is nice 00:17:44 I'd like to know how many exponentials you need 00:18:15 b_jonas: two of the matrix exponents are probably single-exponential in the size of a 2-tag system that emulates the program 00:18:30 I am also curious to how big the initial setup needs to be 00:18:34 err, actually 1, in this construction 00:18:41 MtGBusyBeaver: a double-arrow number is enough, a single-arrow number might not be 00:18:55 ezpz 00:19:10 I wasn't quite sure how powerful the pre-BB layers would be 00:19:34 if you can reconfigure the machine between iterations you can make do with small numbers by constructing a Waterfall Model program whose purpose is to output a large number rather than to be a UTM 00:19:35 they are less powerful for sure, but they are still layers 00:19:57 so they are roughly one arrow each 00:19:58 ais523: yes, but how about a 150***3 or 150***4 number? I'd like to know how many exponentials you need. a double arrow number like 150***150 is likely enough. 00:20:05 but I figured you'd probably find a way to create a double-arrow number 00:20:27 b_jonas: well, the issue is that you're encoding the UTM you want to run as the digits of a number 00:20:30 not many of even your esolanguages need more than a few levels of exponential to simulate something sane 00:20:32 so it's exponential in the size of the UTM 00:20:43 ais523: yes, there are no inputs, only program 00:20:52 both in the turing machine you run and the waterfall program 00:21:07 I believed I only used a double-exponential (2**(2**x)) once, and even in that situation it later turned out to be unnecessary 00:21:07 that's why I asked as a function of the program size 00:21:11 *I believe 00:21:11 we can probably make something like 5-7 arrows with just what we have 00:22:00 ais523: though I guess in this case I have to ask this restricted to less than a hundred waterclocks, regardless of input program size, to be more relevant for M:tG 00:22:10 b_jonas: this is a specific construction with 10 waterclocks 00:22:14 + a halt waterclock 00:22:25 that I haven't figured out the details of yet, but have a basic plan for 00:22:43 MtGBusyBeaver: I also guess that just a double arrow number like 150***150 should be enough for most programs 00:22:55 four arrows like you mentioned earlier should be more than enough 00:22:58 the idea is to implement a version of High Rise in base 3 where carrying is disallowed, sequences 0 and 1 are empty, and sequence 2 alternates between a geometric progression and 0 00:23:47 that High Rise language, I'm about 99% confident has a fairly direct implementation of 2-tag 00:23:54 well once we run one BB we have access to that number so iterations really grow very fast 00:24:23 via arranging the base of sequence 2's geometric progression so that the distances between pairs of 2s in the data string (with one intervening 2 between them) represent tag system elements 00:24:26 MtGBusyBeaver: yes, the question is just how larege you need to start the bootstrapping 00:24:47 . o O ( wait, only just now I noticed that zero arrows is the multiplication (and both can be denoted by an empty string), so arrow notation is even better than I thought! ) 00:24:52 it almost certainly won't be a problem in your M:tG construction, but I'm still curious for reasons other than that construction 00:24:55 and I'm also fairly confident that a The Waterfall Model program using the waterclock assignment I linked above can be made to implement it 00:25:09 checking the logs for it 00:25:42 http://nethack4.org/pastebin/and-high-rise.txt this link? 00:25:43 ais523: that should work. and how many exponentials do you need to encode *to* a 2-tag system 00:25:54 now, if you have a double-arrow startup, you could make do with a fixed program (and simply varying the initial waterclock sizes) by writing a UTM in 2-tag and hardcoding that 00:25:59 MtGBusyBeaver: right 00:26:12 MtGBusyBeaver: although it's just a thought dump rather than presented in a way that's easy to follow 00:26:17 https://esolangs.org/wiki/High_Rise might help for some context 00:26:48 I will probably end up writing out the explicit proofs and Waterfall Model machine at some point, but likely not in the next few days because I have a day job 00:27:14 b_jonas: I don't know offhand how complex UTMs in 2-tag are 00:27:15 nice 00:27:43 ais523: I'm not asking for an UTM there, rather how large the tag system program is as a function of the program you're compiling from 00:27:48 the problem with languages that are "well known" to be TC is that I often haven't proved them TC myself and thus don't have much of an intuitive idea of what the complexities are like 00:28:03 ais523: for the Waterfall model, you need an UTM so that the program always fits in 150 or so waterclocks, because the M:tG construction can't handle more 00:28:23 b_jonas: well, this is the usual trick when optimising universal machines 00:28:37 you want your machine to be small, so you want as little code as possible, so much of the complexity is in the data 00:28:47 yeah, I guess that's true 00:28:59 if you only care about program size, not runtime (and we don't care about runtime here), then an UTM probably helps 00:29:08 just look at, e.g., Three Star Programmer or the I/D machine 00:29:28 well we have to do shenanigans to get 124 clocks 00:29:43 the interpreters are ridiculously simple and small, but writing a program to do anything is very complex because you have to code all the basic operations you want into the program itself 00:29:52 sure 00:29:55 12 and under requires much less restriction on the setu 00:30:00 setup 00:30:32 Three Star Programmer requires a complex one-time setup just to be able to write to specific memory addresses, for example, then needs to neutralise the extra copies of that setup on every future iteration because it has no way to run code only once 00:30:57 and it looks like 10 is going to be enough for TC-ness 00:31:00 MtGBusyBeaver: that's why I was aiming for 12 as a target, it's nice to have a concrete number rather than continuously worry about if I'm wasting a waterclock somewhere 00:31:12 the 10 is almost certainly improvable, but I can't immediately see how 00:31:18 "no way to run code only once" => oh no... Countercall 00:31:36 b_jonas: I think oerjan proved Countercall sub-TC? although I can't remember 00:31:50 did he? 00:32:02 it'll be in the IRC logs from around when Countercall was created 00:32:15 also reading the logs, there are actually two ways for the MTG program to halt 00:32:22 either by killing them 00:32:36 or by killing all of the other arcbonded creatures 00:33:11 the second one can be delayed in a BB method and we can prevent them from dying 00:34:24 so the creatures with arcbond act as the halting clock 00:34:36 one nice thing about your construction is that I think it has no limit on "self-reset" values, right? 00:34:43 a waterclock that zeroes can set itself to any value 00:34:49 yes 00:35:15 with the Hungry Lynx construction, I used Elesh Norn partly for the static +2/+2, giving me a self-reset max of 4, because I was worried it might not be high enough 00:35:26 (although I think we subsequently discovered that a self-reset of 3 is enough) 00:35:50 but large self-resets might be needed for the 10-waterclock construction 00:36:37 the +1/+1 from dralnu's crusade means we can't reset ourself smaller than 2 00:36:37 ais523: I'm not entirely sure, but I think oerjan only conjectured that Countercall is not TC, and proved something much weaker 00:36:44 ah, OK 00:36:54 MtGBusyBeaver: there's no issue with low self-resets 00:36:54 at least that's what it seems from checking the logs again 00:36:58 but it could be hiding anywhere 00:37:02 you can always just increase all the other waterclocks to compensate 00:37:15 yeah, but it is worth noting 00:37:35 right, definitely worth being aware of all your limitations 00:38:36 also if you need any of the "growing"/"flooding" type clocks those don't count towards the 12 limit 00:39:26 right, I figured that out; it might be interesting to try to create a UTM with some number of those 00:39:41 you'd want to store data using the length of time the program had been running 00:40:00 even with the help of a few normal clocks, though, it'd be a pain to keep things under control 00:40:09 yeah it seems really chaotic 00:40:21 volatile 00:43:32 ais523: anyway, it's useful to know that you will probably be able to prove that 10 or 12 clocks are enough 00:43:55 for turing-completeness that is 00:44:07 well that 10 proof looks close so 12 is almost assured 00:45:11 oh, since we're talking about M:tG, I wanted to advertise the play-by-post M:tG variant game that we're running: https://forums.giantitp.com/showthread.php?601011 00:46:55 three card blind always interesting 00:47:12 each round, players secretly choose a 3-card deck, then they're revealed, then each deck plays two games against each other deck, you don't lose for drawing from an empty library, all game information that would normally be hidden is known, we figure out the result of each game assuming perfect play, 00:47:30 that sounds like 3-card blind so far 00:47:41 I assume luck always goes against the player who put the luck card in their deck? 00:47:45 my first forum name was to participate in that on the old WotC 00:47:48 forums 00:47:59 players get a point for a draw and three points for a match won, the player with the most points in that round wins the round and an internet point and right to boast. 00:48:16 IIRC luck was decided by the opponent 00:49:21 ais523: yes, it's 3-card blind, and yes, random choices go against the person who put that card in their deck 00:51:26 and it's vanishing, which means that the cards from the deck that won a round are banned from future rounds, except certain cards like basic lands and storage lands are immune to getting banned this way. 00:52:44 and you aren't allowed to mulligan, despite that with the latest mulligan rule it would actually help you if you could 00:53:23 but apparently the player base is conservative and not used to the new mulligan rule yet, so they don't want to upset the meta from that 00:55:38 I can't think of many situations where you'd prefer to have a card in your deck than in your hand 00:55:46 miracles 00:55:47 maybe if you're going second and the opponent is playing Thoughtseize or the like? 00:55:57 oh yes, miracles, didn't think of that one but it's obvious now you said it 00:56:02 -!- arseniiv has quit (Ping timeout: 256 seconds). 00:56:31 ais523: that's sort of why the situation didn't come up before the current mulligan rule 00:56:37 but read the new mulligan rule: 00:57:44 you put cards from hand on the bottom of your library, so you choose which one you put there 00:57:49 rather than it being random 00:57:50 exactly 00:58:09 so now it mostly protects you from Burning Inquiry and other discards 00:58:25 well, it would 00:58:31 not completely, but to a very large amount 00:58:45 only we don't allow mulliganing, as a house rule, so nope 00:59:11 well, you have a house rule to not lose by outdraw 00:59:20 but that can happen during mulligans in regular MTG 00:59:24 so it's a bit of a weird interaction 00:59:44 (admittedly, it requires Shahrazad to be possible in a black-border game) 01:00:22 nevertheless, we did manage to get strange metas from newly released broken cards, most notably the non-black-bordered small print run cards Impatient Iguana and Mirror Lotus 01:00:34 s/Mirror Lotus/Mirrored Lotus/ 01:01:51 `card-by-name mirrored lotus 01:01:52 No output. 01:02:09 oh, wait, is this from Unsanctioned? 01:02:09 *newly released* 01:02:12 no 01:02:28 oh, Mystery Booster Playtest 01:02:42 ais523: technically does not require Shahrazad anymore, because of Karn Liberated, but that won't come up in a real game, because you'll win before it'd happen 01:02:46 ais523: yes 01:02:50 that isn't a silver-bordered card, it has a black border and a "not for constructed play" footer 01:03:05 ais523: I didn't say silver-bordered, I only said non-black-bordered 01:03:14 also not gold bordered 01:03:24 I'm not sure if it technically counts as having any border 01:03:34 the border is black, though, with a few intrusions 01:03:53 oh the playtest cards 01:03:54 but isn't that just the border of the card that the playtest sticker is sticked on? 01:04:10 the properties of that card shouldn't matter 01:04:19 even if you can somehow see them 01:04:40 b_jonas: the underlying card is part of the card, as printed, though 01:04:45 the sticker is also part of the card 01:04:51 they're on the same print sheet and printed together 01:05:03 that said, there probably won't be a judgement on this, because there are only three un-cards that care about whether a card is black-bordered 01:05:12 so the underlying card is part of the card; most of its properties don't matter because there's a sticker over it, but I'd argue it determines the border color 01:05:22 ais523: yes, but the underlying card shouldn't matter to determine characteristics 01:05:30 well I don't know 01:06:21 the sticker doesn't cover the border 01:06:24 maybe they are black-bordered then 01:06:38 it's a bit like host/augment, parts of the card are covered by the augment but those parts that aren't still count 01:06:48 or Curse of the Fire Penguin 01:06:53 `card-by-name curse of the fire penguin 01:06:54 Curse of the Fire Penguin \ 4RR \ Enchantment -- Aura \ Enchant creature \ Curse of the Fire Penguin consumes and confuses enchanted creature. \ UNH-R \ \ Curse of the Fire Penguin Creature \ 4RR \ Creature -- Penguin \ 6/5 \ Trample \ When this creature dies, return Curse of the Fire Penguin to the battlefield. \ UNH-R 01:07:17 was wondering how the bot would represent that thing 01:07:20 I guess that's a decent effort 01:07:38 ais523: I don't think playtest stickers are like that. if they were, I'd make an Ulamog proxy such that the sticker happens to not cover the mana cost {W} 01:07:51 b_jonas: the cards in question are not playtest stickers 01:07:58 they're cards whose design /looks/ like a playtest sticker 01:08:22 sure 01:08:25 but the sticker and underlying card are part of the same printed card, and in particular the underlying card is always the same 01:08:34 and admittedly they made a good job where the playtest sticker covers almost everything 01:09:16 I think I'd take border color and frame color from the underlying card, but pinline color from the sticker 01:09:28 ok, then they're black-bordered cards that aren't valid for tournaments 01:09:33 that said, I'm not sure if M:tG will ever have a card that cares about pinline color 01:09:34 Form of the mulldrifter is clearly on a mulldrifter though 01:09:44 even silver-bordered 01:10:02 that category already exists I think, because I think some very low print run cards were printed as black bordered with square corners rather than gold bordered 01:10:29 pinline color would be tricky because pre-Mirrodin cards don't have it 01:10:53 MtGBusyBeaver: heh, indeed 01:12:40 MtGBusyBeaver: and One with Death is on One with Nothing 01:13:07 One with Death is actually a fairly good card in constructed, IIRC 01:13:26 because there are a few combos that force your opponent to resolve one of your spells, and that's about the strongest possible payoff for one of those 01:15:03 ais523: that sounds a bit hard to do. isn't the normal route just donating an Illusions of Grandeur or a Phage? 01:16:26 Hive Mind + pact of negation 01:16:57 I think Hive Mind is the most common core of the combo 01:16:59 `card-by-name Hive Mind 01:17:00 Hive Mind \ 5U \ Enchantment \ Whenever a player casts an instant or sorcery spell, each other player copies that spell. Each of those players may choose new targets for their copy. \ M10-R 01:17:14 you don't even need anything else when using One With Death because the opponent's copy resolves first 01:17:48 the normal kill in tournaments uses pacts in the hope that the opponent doesn't have the right color of mana to pay for them 01:18:20 I see 01:19:38 I played a Phage deck twice in the three card blind, got a second place from it once, and that one doesn't try to get the opponent to resolve Phage, just puts it to play under the opponent's control (as one path to win; the other is to just play it and attack with it) 01:22:35 -!- MtGBusyBeaver has quit (Remote host closed the connection). 01:23:54 -!- MTGBusyBeaver has joined. 01:38:35 -!- ais523 has quit (Quit: quit). 02:11:49 -!- MTGBusyBeaver has quit (Remote host closed the connection). 02:20:26 -!- sprocklem has quit (Ping timeout: 240 seconds). 02:21:32 -!- sprocklem has joined. 02:31:25 -!- user24 has joined. 02:38:26 -!- user24 has quit (Quit: Leaving). 02:40:58 -!- Anas10808 has joined. 02:41:39 Hi 02:42:08 -!- Anas10808 has quit (Remote host closed the connection). 04:16:51 [[User:Quadril-Is]] https://esolangs.org/w/index.php?diff=70245&oldid=69894 * Quadril-Is * (+23) 04:17:21 -!- PkmnQ has joined. 04:27:28 -!- PkmnQ has quit (Remote host closed the connection). 07:23:13 -!- imode has quit (Ping timeout: 255 seconds). 07:34:02 -!- tromp_ has joined. 07:34:13 -!- tromp has quit (Read error: Connection reset by peer). 07:52:36 -!- Lord_of_Life_ has joined. 07:54:40 -!- Lord_of_Life has quit (Ping timeout: 265 seconds). 07:54:40 -!- Lord_of_Life_ has changed nick to Lord_of_Life. 08:13:06 -!- xelxebar has quit (Remote host closed the connection). 08:16:48 -!- xelxebar has joined. 08:27:23 -!- b_jonas has quit (Quit: leaving). 08:49:23 -!- xelxebar has quit (Ping timeout: 240 seconds). 08:52:22 -!- MDead has joined. 08:55:34 -!- MDude has quit (Ping timeout: 265 seconds). 08:55:37 -!- MDead has changed nick to MDude. 08:57:50 -!- xelxebar has joined. 10:07:08 -!- wib_jonas has joined. 10:08:36 fizzie: the log bot seems to have stopped yesterday 11:48:50 -!- kritixilithos has joined. 12:23:45 -!- arseniiv has joined. 12:38:08 -!- sprocklem has quit (Ping timeout: 256 seconds). 13:00:05 -!- LKoen has joined. 13:27:21 -!- LKoen has quit (Remote host closed the connection). 13:27:41 -!- LKoen has joined. 13:30:28 -!- kspalaiologos has joined. 13:33:31 -!- LKoen has quit (Remote host closed the connection). 13:36:06 -!- LKoen has joined. 13:42:46 -!- LKoen has quit (Remote host closed the connection). 13:44:13 -!- LKoen has joined. 13:49:38 -!- heroux has quit (Ping timeout: 240 seconds). 13:53:38 oh nice! 13:54:11 I created a stub at https://esolangs.org/wiki/Orca , and a few months later the person who created this esolang came along and edited more details onto the wiki 13:55:17 -!- atslash has quit (Read error: Connection reset by peer). 13:55:20 including a scan of a print magazine article reviewing the language (or its implementation) 13:55:57 nice 13:57:29 I don't know if they came to the wiki because I wrote the stub, or on the other hand the stub was useless and they'd have written the wiki article anyway, though 13:57:49 there is much more to explain, though 13:58:10 explanations of examples would be lovely 13:58:45 myname: there's a link saying "Tutorial Video on Youtube" 13:59:33 yeah, if i don't forget i will watch that later 14:03:01 ah, the log bot is working now. and the previous one might have been false alarm. 14:08:57 -!- 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.”). 14:26:44 -!- heroux has joined. 14:39:43 -!- FreeFull has quit (Ping timeout: 255 seconds). 14:49:25 -!- FreeFull has joined. 15:26:03 -!- xelxebar has quit (Ping timeout: 240 seconds). 15:28:06 -!- xelxebar has joined. 15:49:12 -!- kspalaiologos has quit (Read error: Connection reset by peer). 15:49:40 -!- kspalaiologos has joined. 15:50:53 `" 15:50:55 1/1:378) elliott: You have become the very thing you fought for! \ 1254) actually a small trebuchet onto the balcony might work 15:51:04 -!- kspalaiologos has quit (Read error: Connection reset by peer). 16:13:50 -!- wib_jonas has quit (Remote host closed the connection). 17:04:01 [[Infinite Goto]] N https://esolangs.org/w/index.php?oldid=70246 * PythonshellDebugwindow * (+181) Created page with "''Infinite Goto'' is a language by [[User:PythonshellDebugwindow]]. Programs in it will never end. ==Syntax== All lines have a number on them. Numbers less than 0 are treated..." 17:04:18 [[User:PythonshellDebugwindow]] M https://esolangs.org/w/index.php?diff=70247&oldid=70190 * PythonshellDebugwindow * (+20) /* Languages */ 17:19:46 -!- imode has joined. 17:27:03 the youtube video is interesting 17:27:10 now i want to install it 18:09:46 myname: there is a browser version also, the article says. Maybe it’s functional enough not to install locally (haven’t tried, but you’ve intrigued me into watching tutorial, maybe a bit later though) 18:13:49 i downloaded it, but i have no idea on how to create a midi device 18:24:13 I think I had a thing somewhere that creates a virtual midi device but that’s a windows one 18:31:49 -!- LKoen has joined. 18:34:23 -!- atslash has joined. 18:55:01 [[Infinite Goto]] M https://esolangs.org/w/index.php?diff=70248&oldid=70246 * PythonshellDebugwindow * (+965) 19:14:03 -!- kritixilithos has quit (Ping timeout: 240 seconds). 19:53:13 -!- Lord_of_Life_ has joined. 19:55:14 -!- Lord_of_Life has quit (Ping timeout: 240 seconds). 19:57:53 -!- Lord_of_Life has joined. 19:58:19 -!- Lord_of_Life_ has quit (Ping timeout: 255 seconds). 20:04:42 that thing is interesting, though I’m not comfortable with that kind of music environments 20:10:12 -!- LKoen has quit (Remote host closed the connection). 20:10:37 -!- LKoen has joined. 20:21:18 [[Infinite Goto]] M https://esolangs.org/w/index.php?diff=70249&oldid=70248 * PythonshellDebugwindow * (+509) /* Syntax */ 20:21:54 -!- b_jonas has joined. 20:22:01 [[Infinite Goto]] https://esolangs.org/w/index.php?diff=70250&oldid=70249 * PythonshellDebugwindow * (+83) /* Storage */ 20:38:20 -!- LKoen has quit (Remote host closed the connection). 20:38:41 -!- LKoen has joined. 20:39:50 -!- Phantom_Hoover has joined. 20:39:59 kmc, what level of smug prepper are you on right now 20:40:01 -!- Phantom_Hoover has changed nick to Guest69050. 20:40:08 lol 20:40:09 hi there 20:41:57 -!- Guest69050 has quit (Changing host). 20:41:57 -!- Guest69050 has joined. 20:42:08 -!- Guest69050 has changed nick to Phantom_Hoover. 21:07:06 [[Infinite Goto]] M https://esolangs.org/w/index.php?diff=70251&oldid=70250 * PythonshellDebugwindow * (+555) /* Syntax */ and /* Examples */ 22:31:48 -!- j4cbo has quit (Remote host closed the connection). 22:31:49 -!- dog_star has quit (Remote host closed the connection). 22:33:50 -!- j4cbo has joined. 22:36:06 -!- dog_star has joined. 22:36:51 -!- ^[ has quit (Remote host closed the connection). 22:36:51 -!- dnm has quit (Remote host closed the connection). 22:39:31 -!- ^[ has joined. 22:42:28 -!- dnm has joined. 22:46:17 -!- pikhq has quit (Remote host closed the connection). 22:49:57 -!- pikhq has joined. 22:50:17 -!- lynn has quit (Remote host closed the connection). 22:50:17 -!- ProofTechnique has quit (Remote host closed the connection). 22:53:10 -!- ProofTechnique has joined. 22:56:12 -!- lynn_ has joined. 23:19:44 -!- 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.”). 23:34:19 -!- Phantom_Hoover has quit (Ping timeout: 255 seconds). 23:38:07 [[Language list]] M https://esolangs.org/w/index.php?diff=70252&oldid=70242 * PythonshellDebugwindow * (+15) /* M */ 23:43:22 -!- oerjan has joined. 23:43:48 [[Language list]] M https://esolangs.org/w/index.php?diff=70253&oldid=70252 * PythonshellDebugwindow * (+20) /* I */