00:33:13 -!- ais523 has quit (Ping timeout: 256 seconds). 00:43:03 -!- ais523 has joined. 01:31:05 -!- amby has quit (Quit: so long suckers! i rev up my motorcylce and create a huge cloud of smoke. when the cloud dissipates im lying completely dead on the pavement). 02:02:44 [[User talk:None1]] https://esolangs.org/w/index.php?diff=131865&oldid=130080 * TwilightSparkle * (+229) /* About Scratch */ 02:29:57 I would have never guessed before trying to implement it that the most arcane part of the Conedy spec is: "For each uppercase letter, there must be a corresponding lowercase letter (for which both letters case-fold to the same codepoint), and vice versa" 02:32:00 `unicode ĸ 02:32:04 U+0138 LATIN SMALL LETTER KRA \ UTF-8: c4 b8 UTF-16BE: 0138 Decimal: ĸ \ ĸ \ Category: Ll (Letter, Lowercase) \ Bidi: L (Left-to-Right) 02:32:58 `run python3 -c "print('ĸ'.upper() == 'ĸ'.lower())" 02:33:00 True 02:39:59 this might be a Python3 bug, but it's not especially clear what the meaningful result should be, other than upper() ever returning a result in the Ll category seems wrong. 02:41:02 -!- chiselfuse has quit (Remote host closed the connection). 02:41:25 -!- chiselfuse has joined. 02:45:50 https://codepoints.net/U+0138?lang=en shows it is lowercase, but explicitly maps its uppercase to itself :shrug: Are there other examples of this? 02:47:36 Sharing because Unicode is a frequent topic of discussion / complaint here. HackEso's `unicode cmd is also very convenient 02:55:49 I would have never guessed before trying to implement it that the most arcane part of the Conedy spec is: "For each uppercase letter, there must be a corresponding lowercase letter (for which both letters case-fold to the same codepoint), and vice versa" ← I knew at the time it was difficult, and worded that sentence specifically to at least be unambiguous 02:56:57 some languages have a casefold function/method that folds any letter to either its uppercase or lowercase version, consistently for each uppercase/lowercase pair, in a way that makes "same letter with different case" comparisons work – neither uppercase nor lowercase does that consistently IIRC 02:58:06 if there's an "equals-ignoring-case" method, that would work too – some regex engines can do that 03:24:03 yes, python has str.casefold for that 03:59:42 I had made a character code which makes case-folding much more consistent regardless of the languages; no complicated tables are needed, and the I which is normal, with a dot, and without a dot, are three different characters, so it does not have that problem. 04:00:10 However, whether or not this is suitable depends on your application; different character codes are suitable for different applications, anyways. 04:01:31 I happened to star with lowercase Unicode, then tried to find the uppercase other-half of the pair. Starting with Uppercase and using casefold seems like it will have less issues. ty 04:02:52 I'm now thinking upper and lower case are very context dependent, and not always consistently meaningful 04:06:42 Unicode category Ll has 2,155 entries, Lu has 1,791, so that's the limit for pair, and I should have started with Lu! 04:07:34 I'm now thinking upper and lower case are very context dependent, and not always consistently meaningful ← indeed! 04:10:18 Yes, it is very context dependent and not always consistently meaningful, I agree with that. 04:10:50 A common example for this is the letter i in Turkish. Its uppercase is İ, and the lowercase of I is ı. 04:12:10 There are also some fun examples where you have two lowercase letters corresponding to one uppercase, e.g. σ and ς for Σ. 04:12:18 that's one example that Unicode libraries don't normally handle by default – you have to request Turkic casefolding with separate methods, if it's even supported 04:12:44 JAA: a good one is that the uppercase of ß is SS which is two letters rather than one 04:13:03 ais523: Except not anymore! There is ẞ now! 04:13:05 …and although ẞ exists it isn't generally used, but its lowercase version is definitely ß 04:13:09 :-) 04:13:28 `` perl -E 'say uc "ß"' 04:13:30 ​ß 04:13:42 hmm, need to turn on Unicode 04:13:51 Related to that, IDNA 2003 vs 2008. 04:14:05 `unicode ẞ 04:14:07 U+1E9E LATIN CAPITAL LETTER SHARP S \ UTF-8: e1 ba 9e UTF-16BE: 1e9e Decimal: ẞ \ ẞ (ß) \ Lowercase: U+00DF \ Category: Lu (Letter, Uppercase) \ Bidi: L (Left-to-Right) 04:14:32 `` perl -C63 -Mutf8 -E 'say uc "ß"' 04:14:34 SS 04:14:38 there we go 04:15:07 `` perl -C63 -Mutf8 -E 'say lc uc lc "ẞ"' 04:15:08 ss 04:15:23 IDNA 2003 normalises ß to ss, so no domain names with ß are possible in it. IDNA 2008 fixed that, but it leads to incompatibilities in some fun places. Python still only supports 2003, and trying to do TLS certificate validation on a domain containing an ß leads to surprise fireworks. 04:15:23 `` perl -C63 -Mutf8 -E 'say fc "ẞ"' 04:15:24 ss 04:15:35 huh, that's an interesting casefold normalization 04:15:39 but I guess it makes sense, considering 04:17:15 ah, that's how to get perl to do that, I was missing -Mutf8 and switched to python to get HackEso to respond correctly 04:17:19 Yes, those cases where there are multiple ways to case-fold. 04:17:20 But yeah, also a very good example for this kind of nonsense. :-) 04:17:38 salpynx: -C is for reading/writing files and argv, -Mutf8 is for the program itself 04:18:41 But I think that the working of international domain names is no good; I had idea how to make a better way. It would involve a code to select a language (a domain component has only one language), which has its own character set that does not distinguish uppercase/lowercase, instead of using Unicode. And then, TLS certificates and most other file formats would use only the ASCII format of the domain name. 04:19:01 (Since the ASCII format is most portable and will be consistent, so it won't cause the problem.) 04:20:13 And then, each of these small character sets will have mapping to Extended TRON Code, althogh they would also be possible to use as stand-alone character sets, so a program can use them directly if desired (and in most cases would use them directly) 04:21:47 JAA: Thanks for the Greek and Turkish examples, sigma should have been obvious when I was trying to think of one. Intuitively Σ -> σ, and ς is a context dependent version of σ. I hope casefold agrees. 04:24:54 `` perl -C63 -Mutf8 -E 'say fc "Σσς"' 04:24:56 ​σσσ 04:25:02 it does 04:29:14 -!- tromp has joined. 04:31:58 [[Chef]] M https://esolangs.org/w/index.php?diff=131866&oldid=127241 * Indigo * (+25) Blog seems to no longer be online 04:33:01 [[Chef]] M https://esolangs.org/w/index.php?diff=131867&oldid=131866 * Indigo * (+38) 04:40:31 My Conedy "Hello, World!" sketch uses 104 letter pairs (one for each output bit), that's 5% of the available pairs. 04:40:31 I thought re-using nets for output would be impractical because every net has only one beacon to head for afterwards. 04:40:31 But there might be up to 3 distinct exit paths depending on the angle of incidence? 04:40:42 [[*&&^]] https://esolangs.org/w/index.php?diff=131868&oldid=131758 * Ractangle * (+25) 04:41:08 it would be amazing if the outcome of this were "Conedy is almost TC, but there aren't enough letters in Unicode to make it work" 04:41:12 I'm not totally convinced by that output convention. The input form makes more sense though. Encoding data in decimal places of rationals is what makes the language interesting. 04:41:54 It'd expand as Unicode does, so it'lll be Unicode version dependent 04:49:05 A more satisfying Hello World would be something that aimed and exited the playing field at, say x or y coord 04:49:07 `` python -c "print('10.' + ''.join([str(ord(c)).zfill(3) for c in 'Hello, World!']))" 04:49:09 10.072101108108111044032087111114108100033 04:55:00 -!- tromp has quit (Quit: My iMac has gone to sleep. ZZZzzz…). 05:02:05 About the Turkish alphabets, that is why I put a separate code for them: 0x702149 for dotless uppercase I but the 0x702249 for the ordinary uppercase I whose lowercase does have a dot but uppercase doesn't. And then, 0x709149 will mean the uppercase I including a dot. (This is not always useful in all applications, but for some applications this will help significantly.) 05:04:01 New versions of a character set ought to not cause programs to change, anyways. 05:14:53 These are reasons why I write software to not use Unicode, as much as possible. Other people do not think about the consideration properly, and will say, O, we will put Unicode; that way you can write international text and now the program is international. Well, that won't work, and it might also just get in the way, depending on what exactly your program is doing (it is not same for all programs, though). 05:23:08 Now if I understand correctly, a Conedy source file could contain either a Ii pair OR a İi pair unambiguously, and the spec is clear. 05:23:16 I was generating a list of valid pairs to compose scripts from. It never occurred to me that some valid pairs would be mutually exclusive. 05:24:16 Σσ και Σς are other examples of mutually exclusive pairs. 05:25:13 Maybe, if they should make a non-Unicode variant which can be better. For example, each cell contains an integer; each positive must have a corresponding negative and vice-versa; nonzero integers must be unique within the program; and the sum of all of the numbers in the program must add up to zero. 05:26:57 that is a lot more clear cut. I think the problem (which I run into frequently with eso-langs) is you want a distinct 1 character symbol for each numeric value 05:28:46 .. an idea I am working on involves extending utf8 to enable infinite codepoints to work with an alphabet with one symbol for every natural number 05:29:20 There is extensions like that of UTF-8, I think. 05:30:04 Instead of positive / negative pairs, would adjacent odd/even codepoints work? They wouldn't neccesarily look related, which is the advantage case has 05:30:25 However, UTF-8 is not the only character encoding (nor is Unicode the only character set, although an extension of UTF-8 beyond U+10FFFF is already not Unicode anyways) 05:30:48 Yes, odd/even like that would work, and is sometimes the way of converting integers into natural numbers by a 1:1 correspondence. 05:31:01 I would be super interested in knowing of a UTF-8 extension like that, the encoding to extend it is pretty obvious, the question is why anyone would bother :) 05:33:26 I did look, but there's enough confusion around UTF-8, I don't know what terms to search for to find a hypothetical infinite alphabet solution. 05:33:33 I don't know, but I think it is unnecessary and not helpful in this case anyways. Anyways, even without such an extension, UTF-8 can encode any unsigned 31-bit number. (This technically allows encoding TRON characters (up to 128 volumes) in UTF-8 without ambiguity, but it isn't a very good way of doing it anyways.) 05:33:58 I have seen the infinite variant of UTF-8 called "UTF-8-infinity", although I could not find it recently. 05:35:08 it's possible it's something I wrote on the esolangs wiki, that would be funny. I have at least two works in progress of the idea (but I keep getting sidetracked) 05:38:47 recently I realised that not only do I need to write the codec, I need to rewrite 'String' to work with it too, which is actually the more complex part, and figure out some intermediate representations that don't clash with existing ones 05:42:06 Just looking at TRON now. I have always liked Unicode, but suspected that if I had more experience or requirement for Chinese I'd change my mind pretty quickly. It seems a weak point, and TBH I avoid looking into the details, as some sort of cognitive dissonance avoidance. 05:43:58 zz038: when you say "0x702149 for dotless uppercase " is that TRON, or a custom encoding? 05:44:58 It is an unofficial "Extended TRON Code". So, it can be used with TRON code (it is a different plane number, so it does not overlap the existing codes) 05:45:52 Does TRON have an upper limit? 05:45:54 (Officially, you can use 0x212349 for the JIS character English uppercase I, though, which still is available in the Extended TRON Code, too) 05:47:05 TRON does not have an upper limit, although some encodings may have limits, and computer programs may have practical limits. There is a limit of 10648000 characters per volume, although the number of volumes is unlimited. TRON-32 encoding supports up to 256 volumes; TRON-8 and TRON-16 are theoretically unlimited.) 05:48:21 (The names "TRON-8", "TRON-16", and "TRON-32" are my own names for the encodings; as far as I know there are no official names for them, although the Ruby programming language uses "TADTextBE" for what I call "TRON-16BE", and "stateless-TADTextBE" for what I call "TRON-32BE". As far as I can tell, usually "TRON code" refers to the TRON-16 encoding (like how in the past, "Unicode" often referred to UTF-16).) 05:49:32 I wrote a article on Just Solve The File Format Problem Wiki about TRON character code, although there is a lot of stuff I could not figure out, so I could not include. 05:51:03 (My use of the word "volume" here is also my own; I don't know what (if any) official terminology exists for the concept that I am refering to, since most of the information is in Japanese and even that is difficult to find.) 05:55:26 -!- Sgeo has quit (Read error: Connection reset by peer). 05:57:46 Thanks, this is interesting, I'll have to read more. I have not encountered TRON before, but it seems like it's been around for a while. Knowing there is Ruby support is good, and makes sense. 06:01:51 -!- ais523 has quit (Quit: quit). 06:03:02 I have made a bitmap font for the first TRON plane, which is based on combining several other fonts, and then I added the braille by myself. (Actually, the list of fonts in esolang wiki (the article called "Font") does list this font.) Later, I would hope that bitmap fonts for the other planes can also be made up, including the unofficial Extended planes. 06:06:23 (I had also found TrueType fonts for the second and third TRON planes; however, they are split across fourteen fonts, with Unicode mappings that have no relation to TRON code nor Unicode, with PDF files describing the mappings. It seems difficult to actually make them useful; I have tried to parse data from the PDF automatically but could not get it to work, so it would have to be entered manually.) 06:25:11 [[Category:Input only]] N https://esolangs.org/w/index.php?oldid=131869 * Yayimhere * (+0) Created blank page 06:43:22 [[Gate]] https://esolangs.org/w/index.php?diff=131870&oldid=131829 * Yayimhere * (+24) 06:43:33 [[Gate]] https://esolangs.org/w/index.php?diff=131871&oldid=131870 * Yayimhere * (+10) 06:43:42 [[Gate]] https://esolangs.org/w/index.php?diff=131872&oldid=131871 * Yayimhere * (-1) 06:44:58 [[Gate]] https://esolangs.org/w/index.php?diff=131873&oldid=131872 * Yayimhere * (+37) 06:46:49 [[Gate]] https://esolangs.org/w/index.php?diff=131874&oldid=131873 * Yayimhere * (+93) 07:05:03 [[Gate]] https://esolangs.org/w/index.php?diff=131875&oldid=131874 * Yayimhere * (+344) 07:05:19 [[Gate]] https://esolangs.org/w/index.php?diff=131876&oldid=131875 * Yayimhere * (+5) /* not */ 07:12:12 [[Gate]] https://esolangs.org/w/index.php?diff=131877&oldid=131876 * Yayimhere * (+82) 07:22:45 -!- Noisytoot has quit (Remote host closed the connection). 07:25:43 -!- Noisytoot has joined. 07:31:24 zzo38: yes, the complications of the turkish i, the vietnamese i, and the russian vs serbian i/p and š/t force tradeoffs on us, they mean there's no single way to deal with letter case differences that's the best for all applications, sadly. 07:32:32 "Unicode category Ll has 2,155 entries, Lu has 1,791" => that's mostly because of letters that are used only in various IPA-like phonetic transcriptions though 07:34:16 "trying to do TLS certificate validation on a domain containing an ß leads to surprise fireworks" => hehe 07:39:01 salpynx: "extension of UTF-8 beyond U+10FFFF" => https://ucsx.org/ 07:46:33 salpynx, ais523: if you're running low on letter for a Conedy program, you could use characters composed from multiple code points as a cell, like и́И́, and hope that the interpreter handles them properly 07:49:36 ``` datei # we need a new password, right? 07:49:38 2024-07-01 07:49:37.558 +0000 UTC July 1 Monday 2024-W27-1 07:56:54 also this years ICFP contest names its associated esoteric language ICFP, which will result in an even weirder entry on the esowiki than the languages called GCC and GHC did 08:06:46 i started icfp 08:06:52 but i didn't understand how strings are not terminated 08:08:11 riv: in the source code, every token is terminated by a space 08:09:34 riv: note that the source code character set differs from the character set in which you read the strings, so this is ASCII space terminator, while space in the string is represented by ascii right bracket or something like that 08:09:54 oh that explains it haha 08:12:02 -!- __monty__ has joined. 08:12:23 b_jonas: UCS-∞ is perfect, thank you! The encoding seems... more thoughtful than what I came up with, they have thought about practical performance issues for very large code points. 08:17:43 b_jonas: "use characters composed from multiple code points as a cell, like и́И́" Good idea, that does seem an intuitive extension. I'll consider adding it, it probably won't happen by default. 08:28:17 Heh, I totally overlooked "Yet another useful property of one-to-six-byte UTF-8 is that the leading byte indicates the length of a code." 08:30:40 I'm not saying that this is necessary, it's quite possible that the few thousand available upper-lower case pairs are sufficient for TCness 08:37:09 this is kind of like the M:tG creature types thing 08:38:51 I was about to agree, but limiting the pairs in Conedy is like limiting the length of the source code. I that same same role M:tG creatures play? 08:40:59 Like how the number of waterclocks in The Waterfall Model are not the same as number of registers in a Minsky machine 08:42:14 -!- lisbeths has joined. 08:43:27 salpynx: the creature types thing comes up when we examine the eso-question of how we can do universal computations in M:tG. (there are multiple variants of what question you want answered.) the constructions that we have use creature types as a resource, so we compile from some intermediate language and how much of something the source program uses turns into how many creature types there are. the 08:43:33 number of creature types is around 279, but this number increases at most rules changes, because the new expansion set has cards that are of new creature types or mention new creature types (usually because they have rules for creating a token of that type, but there are exceptions like Coward). except I think the number of creature types decreased at the Lorwyn update. 08:43:40 -!- Koen has joined. 08:44:59 I wonder if you could design a programming language that deliberately uses some such list as a resource, and such that the limit is near some computational power jump. like a language where the source code is a list of countries, and it's TC iff Kosovo is a country 08:45:52 or maybe it depends on the scrabble word list, and whether QI and ZA are words changes the power 08:48:07 lol, that could probably be contrived somehow, by forcing elements to have to work together. Aren't most TC limits 1,2, or 3 (2-reg Minsky) (3 cell bf) 08:48:32 -!- X-Scale has joined. 08:51:21 I see creatures are explicitly equivalent to waterclocks in the flooding waterfall model. They represent source code and data, Conedy letter pairs just represent source code... they seem similar though 08:52:52 an efghij-style language where you need smartphones with different brands of operating systems for loops, so whether you can find two non-Android non-Apple phones decides if you can be TC 08:54:09 if the basic construction could be TC (maybe it is, maybe it isn't) it probably has the same kind of leeway that future specs could have more letters (/ creatures) ... so it's potentially unlimited 08:56:10 -!- X-Scale has quit (Ping timeout: 250 seconds). 08:57:28 a language where parts of the source code are read from the internet in such a way that you need to have access to domain names under different top-level domains to be able to write a program 08:59:05 having it conditional on useful discoveries (drugs, planets, mathematical proofs), maybe a Minsky machine with registers addressable by the names of planetary bodies which have independently evolved life? 09:00:21 I'd like to try running an Earth, Enceladus pair to see if that compiles... 09:03:15 -!- X-Scale has joined. 09:07:33 -!- jix_ has quit (Quit: quit). 09:20:53 Could someone please produce a chart showing the number of M:tG creatures vs. number of Conedy compliant Unicode character pairs over time, to indicate likely future trends? <== isn't there an esolang where asking in IRC is valid code? (trying my luck) 09:24:16 https://esolangs.org/wiki/IRP is a thing, yes. 09:30:10 -!- Koen_ has joined. 09:33:45 -!- Koen has quit (Ping timeout: 268 seconds). 09:33:52 -!- X-Scale has quit (Ping timeout: 250 seconds). 09:41:44 -!- jix has joined. 09:46:49 -!- wib_jonas has joined. 09:51:34 an esolang where the source code is a person or team, and the program is ran by transforming this person's or teams achievements in some sport to source code for a simpler esolang. if you want the program whose source code you are to do something interesting, then you have to bat/throw/swim/climb specific sequences in multiple ranked matches. 10:06:52 [[ICFP]] N https://esolangs.org/w/index.php?oldid=131878 * B jonas * (+363) Created page with "'''IFCP''' is an esoteric programming language involved in the tasks for [[:Category:ICFP contest|]] 2024. (No description given here yet because the contest is still ongoing for a few hours.) == External links == * [https://icfpcontest2024.github.io/ ICFP 2024 contest 10:08:31 [[User:B jonas]] https://esolangs.org/w/index.php?diff=131879&oldid=128843 * B jonas * (+33) /* Todo */ 10:20:47 [[User:Gilbert189/Languages in concept]] https://esolangs.org/w/index.php?diff=131880&oldid=131195 * Gilbert189 * (+7951) It's an esolang! It's a conlang! No, it's... whatever this is. 10:21:04 -!- Reinhilde has left (Now, I'm the vixen with a knife, and I want blood.). 10:25:56 -!- amby has joined. 10:29:21 [[Gate]] https://esolangs.org/w/index.php?diff=131881&oldid=131877 * Yayimhere * (+41) /* examples */ 10:29:31 [[Gate]] https://esolangs.org/w/index.php?diff=131882&oldid=131881 * Yayimhere * (+0) /* nand */ 10:40:38 [[Gate]] https://esolangs.org/w/index.php?diff=131883&oldid=131882 * Yayimhere * (+1) 10:46:33 [[List of ideas]] https://esolangs.org/w/index.php?diff=131884&oldid=131515 * Gilbert189 * (+146) /* General Ideas */ 10:54:56 [[Gate]] https://esolangs.org/w/index.php?diff=131885&oldid=131883 * Yayimhere * (+83) 10:54:57 [[Esolang:Sandbox]] https://esolangs.org/w/index.php?diff=131886&oldid=131483 * Unicodes * (+75) added 4 words :/ 11:05:48 [[ICFP]] M https://esolangs.org/w/index.php?diff=131887&oldid=131878 * None1 * (+0) ICFP or IFCP? 11:11:38 -!- lisbeths has quit (Quit: Connection closed for inactivity). 11:19:44 [[Gate]] https://esolangs.org/w/index.php?diff=131888&oldid=131885 * Yayimhere * (+4) /* not */ 11:20:47 [[Gate]] https://esolangs.org/w/index.php?diff=131889&oldid=131888 * Yayimhere * (+9) /* examples */ 11:28:46 [[Gate]] https://esolangs.org/w/index.php?diff=131890&oldid=131889 * Yayimhere * (+269) 11:29:43 [[Gate]] https://esolangs.org/w/index.php?diff=131891&oldid=131890 * Yayimhere * (-74) /* examples */ 11:31:10 [[Gate]] https://esolangs.org/w/index.php?diff=131892&oldid=131891 * Yayimhere * (+2) /* A OR B */ 11:31:31 [[Gate]] https://esolangs.org/w/index.php?diff=131893&oldid=131892 * Yayimhere * (+0) /* nand */ 11:31:43 [[Gate]] https://esolangs.org/w/index.php?diff=131894&oldid=131893 * Yayimhere * (+4) /* nand */ 12:04:03 [[Special:Log/newusers]] create * DorIsch * New user account 12:05:46 [[ICFP]] https://esolangs.org/w/index.php?diff=131895&oldid=131887 * B jonas * (+119) 12:37:43 [[Ulsl]] https://esolangs.org/w/index.php?diff=131896&oldid=129535 * Tommyaweosme * (+14) 12:40:12 [[Esolang:Sandbox]] https://esolangs.org/w/index.php?diff=131897&oldid=131886 * Tommyaweosme * (+45) u p s i d e d o w n 12:41:40 [[Esolang:Sandbox]] M https://esolangs.org/w/index.php?diff=131898&oldid=131897 * Tommyaweosme * (+8) t o c 12:57:34 [[Galveston]] M https://esolangs.org/w/index.php?diff=131899&oldid=101437 * Salpynx * (+145) /* External resources */ Link to UCS- Specification 12:58:26 -!- amby has quit (Ping timeout: 256 seconds). 12:59:02 [[GHost CPU]] N https://esolangs.org/w/index.php?oldid=131900 * B jonas * (+2255) Created page with "'''GHost CPU''' (GHC) is one of the languages used in the [[:Category:ICFP contest|]] 2014. GHC is designed to resemble an 1980s 8-bit microcontroller. It has a read-only code memory made of 256 instructions, a read-write data memory made of 256 bytes, 8 general-p 12:59:26 [[GHC]] N https://esolangs.org/w/index.php?oldid=131901 * B jonas * (+23) Redirected page to [[GHost CPU]] 13:30:12 -!- salpynx has quit (Quit: Leaving). 13:43:30 -!- FreeFull has quit (Ping timeout: 268 seconds). 13:44:52 -!- FreeFull has joined. 13:45:18 so the C standard says that in struct tm, the normal range of the tm_mon field is 0 to 11 inclusive, and, in the docs of the asctime function, even specifies the abbreviated names of the 12 months in normative text. This means the standard library could use the Julian calendar, the Gregorian calendar, a combination of the two with the transition 13:45:19 date dependent on the local timezone, or even a different 12-month calendar with the same month names; but it can't use a lunisolar calendar, since that would need 13 different month indexes in some years. that's a pity. if we do a calendar reform, gmtime/localtime/mktime might have to return the broken down date in the old Gregorian calendar for 13:45:19 compatibility with the C standard. 13:52:25 [[Gate]] https://esolangs.org/w/index.php?diff=131902&oldid=131894 * Yayimhere * (+79) 13:56:11 [[Gate]] https://esolangs.org/w/index.php?diff=131903&oldid=131902 * Yayimhere * (+116) /* examples */ 13:58:23 -!- X-Scale has joined. 13:58:56 [[Gate]] https://esolangs.org/w/index.php?diff=131904&oldid=131903 * Yayimhere * (+111) 14:00:49 [[Gate]] https://esolangs.org/w/index.php?diff=131905&oldid=131904 * Yayimhere * (+41) 14:01:08 that's a land type, not a creature type. 14:08:15 [[Gate]] https://esolangs.org/w/index.php?diff=131906&oldid=131905 * Yayimhere * (+173) /* A OR B */ 14:08:30 [[Gate]] https://esolangs.org/w/index.php?diff=131907&oldid=131906 * Yayimhere * (+6) /* A NOR B */ 14:10:37 -!- X-Scale has quit (Quit: Client closed). 14:12:54 [[Gate]] https://esolangs.org/w/index.php?diff=131908&oldid=131907 * Yayimhere * (+127) /* examples */ 14:18:30 -!- ais523 has joined. 14:21:28 [[Gate]] https://esolangs.org/w/index.php?diff=131909&oldid=131908 * Yayimhere * (+77) /* A AND B */ 14:28:10 [[Gate]] https://esolangs.org/w/index.php?diff=131910&oldid=131909 * Yayimhere * (+295) /* examples */ 14:28:23 [[Gate]] https://esolangs.org/w/index.php?diff=131911&oldid=131910 * Yayimhere * (+2) /* A > B */ 14:51:51 -!- perlbot has quit (Quit: ZNC 1.8.2+deb3.1 - https://znc.in). 14:51:51 -!- simcop2387 has quit (Quit: ZNC 1.8.2+deb3.1 - https://znc.in). 15:05:18 -!- simcop2387 has joined. 15:06:47 -!- perlbot has joined. 15:48:17 -!- X-Scale has joined. 16:11:32 [[General Compute Coprocessor]] N https://esolangs.org/w/index.php?oldid=131912 * B jonas * (+1482) Created page with "'''General Compute Coprocessor''' (GCC) is one of the languages used in the [[:Category:ICFP contest|ICFP contest]] 2014. GCC is a virtual machine intended for running a lisp-like untyped functional language compiled to it. There are built-in ins 16:11:56 [[GCC]] N https://esolangs.org/w/index.php?oldid=131913 * B jonas * (+41) Redirected page to [[General Compute Coprocessor]] 16:14:14 [[User:B jonas]] https://esolangs.org/w/index.php?diff=131914&oldid=131879 * B jonas * (-45) /* Todo */ 16:24:07 b_jonas: re: C restricting its time functions to 12 months: Java actually has constants for 13 months in order to work around that problem (the constants are normatively named JANUARY, FEBRUARY, etc., with the 13th being UNDECIMBER) 16:30:51 -!- ais523 has quit (Remote host closed the connection). 16:32:06 -!- ais523 has joined. 16:35:46 -!- wib_jonas has quit (Quit: Client closed). 16:41:07 -!- ais523 has quit (Remote host closed the connection). 16:42:21 -!- ais523 has joined. 16:49:24 -!- ais523 has quit (Remote host closed the connection). 16:50:38 -!- ais523 has joined. 17:18:59 Looks like that's the case for the old java.util.Calendar, but not for the new java.time.Month enumeration, which just has the usual 12 constants. 17:21:00 -!- X-Scale has quit (Ping timeout: 250 seconds). 17:46:45 -!- rodgort has quit (Quit: Leaving). 17:59:14 In operating system designs I had considered the calendar also with i18n, as well as text and other stuff. A calendar does not even necessarily have months/weeks/etc; so strftime and that stuff is not used mainly but for international calendars you will have a different function, and won't use the locales like C and POSIX are using. 17:59:52 For general timekeeping, there are four kind of timestamps: SI relative, SI absolute, UTC relative, UTC absolute. These can be converted into whatever kind of calendars you want to use, but these four kind are independent of the calendars in use. 18:01:51 (Distinguishing SI and UTC is necessary in order to deal with leap seconds. Furthermore, they use the number of seconds and nanoseconds; UTC timestamps may have a number of nanoseconds exceeding one billion in some cases. Furthermore, arithmetic on such timestamps may sometimes result in missing fields, which is just something that you will have to deal with.) 18:03:42 -!- rodgort has joined. 18:04:01 This way would also avoid the confusion involving Julian vs Gregorian calendar, hopefully. 18:05:03 -!- ais523 has quit (Quit: quit). 18:12:04 Also, neither Unicode nor TRON code nor any other character set solves all of the problems. Some people expect to solve it by just putting in Unicode, but that won't help and neither will TRON code. TRON code has many advantages over Unicode in my opinion, but what a program or file format should need is not something that is solved merely by putting in some kind of universal character set. 18:12:36 -!- Lord_of_Life has quit (Read error: Connection reset by peer). 18:14:01 What I had consider with operating system design though, is it can handle "long character sets" and "short character sets". The long character set is Extended TRON Code. Short character sets always use 8-bit codse. (Individual programs can do what they need to do, although these are the general uses by what is handled and interchanged by most things in the operating system that need to deal with text.) 18:14:11 (Not everything does need to deal with text, of course.) 18:16:23 -!- Lord_of_Life has joined. 18:23:32 -!- tromp has joined. 18:24:03 UCS-G is something that would make sense for some programs to support, if they deal with Unicode-based encodings such as UTF-8 and UTF-32 (although perhaps not UTF-16, since that is more complicated and is not the simple obvious thing to do which can be done directly and easily). 18:33:53 -!- tromp has quit (Quit: My iMac has gone to sleep. ZZZzzz…). 18:48:10 [[Gate]] https://esolangs.org/w/index.php?diff=131915&oldid=131911 * Yayimhere * (+1) /* XOR */ 18:49:06 [[Gate]] https://esolangs.org/w/index.php?diff=131916&oldid=131915 * Yayimhere * (+0) /* A OR B */ 19:00:23 -!- tromp has joined. 19:03:08 -!- lisbeths has joined. 19:35:56 [[Shape-Machine]] https://esolangs.org/w/index.php?diff=131917&oldid=131864 * Ractangle * (+2) 19:36:47 [[Shape-Machine]] https://esolangs.org/w/index.php?diff=131918&oldid=131917 * Ractangle * (-29) /* Python3 */ 19:37:10 [[Shape-Machine]] https://esolangs.org/w/index.php?diff=131919&oldid=131918 * Ractangle * (-101) 19:45:36 [[Shape-Machine]] https://esolangs.org/w/index.php?diff=131920&oldid=131919 * Ractangle * (+120) /* Implementations */ 19:56:56 `learn The password of the month is making the garbage collector hard to implement with its unpredictable lifetime. 19:57:00 Relearned 'password': The password of the month is making the garbage collector hard to implement with its unpredictable lifetime. 20:02:27 [[Brainfuck]] https://esolangs.org/w/index.php?diff=131921&oldid=130126 * Ractangle * (+43) /* Looping counter */ 20:02:41 [[Brainfuck]] https://esolangs.org/w/index.php?diff=131922&oldid=131921 * Ractangle * (+43) /* Looping counter */ 20:35:57 [[SPIKE]] https://esolangs.org/w/index.php?diff=131923&oldid=131853 * Ractangle * (-82) /* Commands */ 20:38:23 [[Hello world program in esoteric languages (N-S)]] https://esolangs.org/w/index.php?diff=131924&oldid=129200 * Ractangle * (+39) /* Spin */ 20:39:01 [[Hello world program in esoteric languages (nonalphabetic and A)]] https://esolangs.org/w/index.php?diff=131925&oldid=130114 * Ractangle * (-217) /* */ 20:39:40 [[Hello world program in esoteric languages (nonalphabetic and A)]] https://esolangs.org/w/index.php?diff=131926&oldid=131925 * Ractangle * (-105) /* */ 20:40:03 [[Hello world program in esoteric languages (nonalphabetic and A)]] https://esolangs.org/w/index.php?diff=131927&oldid=131926 * Ractangle * (+104) /* */ 20:42:37 -!- X-Scale has joined. 20:50:05 -!- tromp has quit (Quit: My iMac has gone to sleep. ZZZzzz…). 21:08:21 -!- salpynx has joined. 21:14:29 -!- HackEso has quit (Remote host closed the connection). 21:16:16 -!- esolangs has joined. 21:16:16 -!- ChanServ has set channel mode: +v esolangs. 21:21:03 [[Shape-Machine]] https://esolangs.org/w/index.php?diff=131928&oldid=131920 * Salpynx * (+121) /* Implementations */ I don't quite understand what this is, but here's an interactive graph. 21:22:11 -!- X-Scale has quit (Quit: Client closed). 21:29:05 -!- X-Scale has joined. 21:29:56 -!- X-Scale has quit (Client Quit). 21:32:33 -!- lisbeths has quit (Quit: Connection closed for inactivity). 21:33:17 Bah, `apt full-upgrade` (from Debian 11 to 12) wants to install apache2 for some reason. 21:36:15 Can't really tell why. As it stands, `aptitude why apache2` just says certbot (installed) suggests python3-certbot-apache, which depends on apache2, but that's just a 'suggests', and it's not planning to install python3-certbot-apache, I think `aptitude why` just tries to explain why it might currently be needed, not what might eventually need it after the upgrade. 21:37:38 It is planning to install libapache2-mod-php8.2, which might be the reason why it's planning to install apache2, so maybe it's something to do with the PHP upgrade. 21:37:39 Yeah, it's a bit of a pain to debug that kind of thing. 21:37:55 You could do `apt-cache rdepends apache2` and then compare that to what is being installed. 21:38:45 Another thing to try would be to disable recommended packages, but that might not be desirable in all cases. 21:40:07 It is going to install php8.2, which depends on "libapache2-mod-php8.2 | php8.2-fpm | php8.2-cgi, php8.2-common", but it's installing php8.2-fpm so it shouldn't also need to install libapache2-mod-php8.2 too. But I'm still guessing it's a PHP thing. 21:41:05 It's also going to use "986 MB of additional disk space" which is _slightly_ concerning because the wiki-machine is already using 66% (17G) of the available 25G of disk. 21:42:26 `apt full-upgrade --no-install-recommends` doesn't want apache2, so I guess it is a recommended package somewhere indeed. Maybe I'll diff what it's planning, and if nothing important seems to be left out I'll go with that. 21:58:12 why is php installed in first... never mind, it's for MediaWiki obviously 21:58:59 -!- Sgeo has joined. 22:12:06 -!- __monty__ has quit (Quit: leaving). 22:18:52 I just went with the default full-upgrade and then uninstalled apache2 after the fact. 22:19:15 The nginx metrics powering my fancy dashboard were broken by the upgrade though. 22:22:22 Just one more reboot, hopefully. 22:24:06 -!- esolangs has joined. 22:24:06 -!- ChanServ has set channel mode: +v esolangs. 22:30:12 I really should have some sort of a staging environment set up for this thing. 22:30:32 [[Talk:Main Page]] https://esolangs.org/w/index.php?diff=131929&oldid=130462 * Rdococ * (+196) /* Fix the name */ new section 22:37:07 Things seem to be mostly working, except the HackEso container doesn't. 22:38:16 It's set up in a systemd-nspawn thing, in which the UML stuff runs, but now `machinectl shell ...` just immediately exits. 22:39:48 [[Deadfish/Implementations (M-Z)]] https://esolangs.org/w/index.php?diff=131930&oldid=131602 * Zzo38 * (+286) Super ZZ Zero (assembly) 22:43:09 Wonder what's wrong with that. `machinectl login ...` does give me a getty "in" it (well, the hostname looks right and so on), but I can't log in because there are no users with passwords in there. 22:45:52 [[Talk:One Time Cat]] N https://esolangs.org/w/index.php?oldid=131931 * EvyLah * (+212) /* How does the program actually work? */ new section 22:46:01 [[Talk:One Time Cat]] https://esolangs.org/w/index.php?diff=131932&oldid=131931 * EvyLah * (+90) forgot signature 22:46:38 [[Talk:One Time Cat]] https://esolangs.org/w/index.php?diff=131933&oldid=131932 * EvyLah * (+0) 22:57:20 huh 22:59:05 Hmm. `systemd-run -M ... -P` seems to work for running things in the container. I think it must be some sort of pty and/or permissions problem. 23:03:05 Managed to dig down the error for the bot, which seems to be something networking-related. 23:11:16 -!- esolangs has joined. 23:11:16 -!- ChanServ has set channel mode: +v esolangs. 23:12:04 -!- X-Scale has joined. 23:18:51 Would a computer architecture that performs addition using a lookup table in memory be considered esoteric? 23:18:52 [[General Compute Coprocessor]] https://esolangs.org/w/index.php?diff=131934&oldid=131912 * B jonas * (+301) 23:21:01 Sgeo: likely, and I think there's such a machine somewhere though I don't remember the name 23:21:23 "The 1620 was announced in 1959 and targetted at the small scientific market. Its internal code name CADET was jokingly said to mean "Can't Add, Doesn't Even Try", referring to the use of addition tables in memory rather than dedicated addition circuitry." 23:22:35 I haven't found documentation for how the tables are meant to be loaded in memory. One reference to maybe loading the tables being part of the programs? 23:22:36 Sgeo: https://esolangs.org/wiki/BytePusher 23:37:16 Oh god now I want to add audio to that JS implementation 23:41:53 -!- X-Scale has quit (Ping timeout: 250 seconds). 23:48:40 -!- Taneb has quit (Quit: I seem to have stopped.).