00:05:16 i've designed turing-complete language with only 5 instructions 00:05:32 do tell. 00:05:40 wait.. i can reduce it by one 00:06:30 it's name is XUML 00:06:45 that's a start. 00:07:08 it has 4 instructions: X = flip bit value 00:07:27 U = User interact (takes argument) 00:08:01 hmm wait 00:08:40 L = Loop (takes argument) 00:08:46 M = Move (takes argument) 00:08:48 Yay, a language starting with X. at last :D 00:09:15 i just noticed my syntax definition is wrong 00:09:22 lament: yeah, something like that... there are still a few books that looked promising at the library that i didn't check out; i'll keep looking... 00:13:07 i won't be at all surprised if nobody ever researched this seriously 00:13:53 mainstream CS seems to look at these things from a very different perspective 00:14:24 ignoring interactive IO and such 00:20:19 it seems the basic concepts CS operates with (in this regards) are not exactly the basic concepts relevant to esolangers 00:20:24 *regad 00:20:26 *regard :) 00:24:34 new language, hooray, fanfare etc: http://www.oceanbase.org/graue/qdeql/ 00:25:50 that's turing complete? 00:26:45 urgh i don't want to write XUML programms 00:26:59 (and did you forget about reenqueuing the 0 after a \ ?) 00:27:04 lament, i haven't proved it, but it intuitively seems so 00:27:12 hm? 00:27:24 if the byte is zero during a \ it just gets lost 00:27:25 is there any way to extend the queue size ? 00:27:33 yeah, it's dynamic 00:27:45 starts out empty, then you add to it 00:28:24 \ lets you enlarge and shrink the queue :) 00:28:31 what does - do on an empty queue 00:28:40 enqueues 255 00:28:44 okay 00:28:45 the boolfuck code: [[+]+] is LLLXXLXX in XUML 00:28:54 like it says, "dequeueing produces 0 if the queue is empty," so that's how it works 00:29:21 which is Loop(Flip Flip Loop(Flip) Flip) 00:29:28 jix, i like what i'm hearing 00:29:31 keep up the good work 00:30:25 you have to double flip (XX) because you cant place a subloop at the start of the parent loop 00:31:14 that is insane 00:31:34 i didn't wanted to add control characters 00:32:00 well, i meant insane in a good way 00:32:05 hehe 00:34:33 UXXUUXXXUXXUUXX prints a newline 00:37:45 but only if it's the last outputting command 00:46:53 auto converted (BF=>BOOF=>XUML) programs are going to be so fucked long 00:50:03 > is MMLXX and < is MLX 00:50:45 ... 00:55:07 I want to add a category for languages where the source is not a text file, like Piet. But what to name it? non-text based languages? doesn't sound too good in my ears... 00:55:23 non-ascii ? 00:55:38 that would exclude some text-based langs 00:55:45 include I mean 00:55:46 oh right 00:57:33 languages whose input is not in the form of a character stream is perhaps a bit verbose ;) 01:07:10 non-textual? 01:08:38 yeah, that's better :) 01:11:22 about the categories: do we really need categories for deterministic or one-dimentional? 01:11:40 shouldn't we just assume they are, unless otherwise categorized? 01:12:08 ++++++++++[>++++>++++++++++>+++++++<<<-]>>>++.<+.+++++++..+++.<++++.<+++[>----<-]>.>++++++++.--------.+++.------.--------.<<+++[>++++<-]>++.<++++++++++. 01:12:16 is autoconverted 14kb 01:13:03 but it outputs the same thing as the boolfuck=>xuml ULXULXULXXULXXULXULXXULXXULXXULXXULXXULXXULXULXXULXULXXULXULXULXXULXULXXULXXULXULXXULXULXULXXULXULXXULXXULXULXXULXXULXULXULXULXXULXXULXULXXULXULXULXXULXULXXULXXULXXULXULXULXULXULXULXULXXULXXULXULXXULXULXULXXULXXULXULXULXXULXXULXULXULXULXXULXXULXULXXULXULXXULXXULXULXXULXULXULXXULXULXULXXULXULXXULXXULXULXXULXULXULXXULXXULXULXXULXULXXULXXULXXULXULXULXULXXULXXULXULXULXXULXXULXXULX 01:13:29 the xuml handwritten version is even shorter 01:14:29 oh.. made some.. mistake in the autoconverter 01:14:34 kipple: Deterministic should indeed be the default, but one-dimensional is already the default 01:15:22 note that in http://www.esolangs.org/wiki/Esolang:Categorization there's not a category, meaning it's default 01:15:24 ah. that's what you've meant by not adding links for things like one-dimensional and Implemented? 01:15:34 yes 01:15:43 good 01:16:14 imperative and deterministic may be default too 01:16:21 btw, i don't think 2D langs should be a subcategory og multi-D langs 01:16:26 i think Implemented should be a category, though 01:16:37 who's going to want to search for Unimplemented languages? they're no fun 01:16:43 and there are a lot of them 01:16:54 and why not, kipple? makes sense to me 01:17:03 somebody looking for somehting to implement perhaps? 01:17:11 I don't see the benefit 01:17:18 maybe, but Implemented are generally more interesting, i think 01:17:24 agreed 01:17:45 do you see the drawback? i don't 01:18:07 multi-D-but-not-2D languages are special and can be easily found by looking in the base multi-D category 01:18:11 what's the problem with that? 01:18:12 I separated 2D and Multi = (3+)D, for the record 01:18:21 separated how? 01:18:23 just adds clutter to the multi-D page 01:18:36 no it doesn't, it only adds one category to that page 01:19:21 well, it's fine with me that way 01:19:24 well, true. lets keep it that way then 01:19:24 stuffing all the 2D langs on there would add more clutter 01:21:08 ok. I'm changing deterministic to default (removing category) 01:21:45 anybody disagree? 01:22:40 er 01:22:44 I was editing 01:23:05 ok. let me know when you're done then 01:24:42 done 01:26:36 btw, why is there an extra colon in the categories on that page? ( [[:Category:Nondeterministic]]) 01:27:02 because otherwise that page would also be added to the category 01:39:56 hmm, what counts as Implemented? 01:40:13 at least an implementation exists 01:40:15 what if only some features work, or the newest version of the spec isn't implemented, or the implementation is really buggy? 01:40:45 like, suppose the only Brainfuck interpreter supported +-<>., and [] support was "coming soon", implemented or no? 01:40:56 that's worth some notes in the text 01:41:12 I'd say implemented 01:41:29 me too, plus explain the situation in the text 01:41:33 fuzzy logic! 01:41:35 yes 01:45:36 do we need both Usability unknown and Unknown computational class? 01:46:02 well, Malbolge is Usability unknown and Finite state automata 01:47:15 hey! bitxtreme is a non-textual language, right? 01:47:41 isn't the source code binary? 01:48:08 true 01:48:52 well, now that category has at two languages! :D 01:49:00 Choon belongs there 01:49:07 oh wait, no it doesn't 01:49:13 source code only, right? 01:49:22 yeah 01:50:01 though it is a bit special 01:51:11 actually it's ascii 01:51:27 er, iso8859-1 01:51:30 I know. I was talking about the output 01:52:05 I'm talking about bitxtreme 01:52:07 there's no zero-dimensional category for NULL to go in? 01:52:28 I didn't think one single language would qualify for a category 01:52:50 I do! 01:53:00 then go ahead :) 01:53:33 hmm. so bitxtreme is text based after all? darn 01:54:05 someone needs to make a new zero-dimensional language somehow, i guess 01:54:09 How should we categorize language level? is low / high enough? 01:54:14 but it's not like such a thing is possible 01:54:24 Homespring and ZOMBIE would be "very high" 01:55:33 a scale then: turing tarpit - low - medium - high - high as a kite ;) 01:56:01 hm 01:56:15 well, i'm going to sleep, you have fun categorizing articles 01:56:26 and maybe you can add Qdeql to the wiki if you feel like it 01:56:30 good night 01:56:33 nite 01:56:34 -!- graue has quit ("Leaving"). 01:57:52 I'm not for categorizing language level 01:57:59 why not? 01:58:32 hm, on second thought maybe it helps looking for "powerful" languages 01:59:17 I think high/low should be enough 01:59:23 oh, var'aQ is not still there 01:59:34 or was it var'aq ? 01:59:50 * pgimeno adds a stub 02:00:13 I also think we should have a category for strongly metaphored languages (Chef, Shakespeare, ZOMBIE) 02:00:26 ugh 02:00:40 ZOMBIE looks pretty regular in my opinion 02:00:50 but yes, there are a few more 02:00:53 ok, perhaps not ZOMBIE 02:01:09 or maybe it should be called Themed languages 02:01:19 bf2xuml conversion is so baaad 02:01:32 boolfuck2xuml is ok 02:01:52 bad as in slow, or bad as in does not work? 02:02:10 bad as in output is very long 02:02:48 [-] auto-converted is a 500 byte XUML code 02:03:08 yuck 02:03:15 you need an optimizing compiler 02:03:28 yes 02:03:46 but the bad part is the bf2boolfuck 02:04:34 [-] is in boolfuck (not autoconverted) [+]>[+]>[+]>[+]>[+]>[+]>[+]>[+]> and autoconft to XUML LXMLXLXMLXLXMLXLXMLXLXMLXLXMLXLXMLXLXMLX 02:17:06 g'nite 02:19:58 -!- jix has quit ("Banned from network"). 02:25:38 -!- kipple has quit (Read error: 60 (Operation timed out)). 07:59:59 -!- clog has quit (ended). 08:00:00 -!- clog has joined. 09:20:44 -!- kipple has joined. 10:08:24 -!- puzzlet has quit (Read error: 104 (Connection reset by peer)). 10:27:39 -!- puzzlet has joined. 10:29:53 -!- puzzlet has quit (Client Quit). 10:29:57 -!- puzzlet has joined. 10:31:03 -!- puzzlet has quit (Client Quit). 10:31:06 -!- puzzlet has joined. 10:32:57 -!- CXI has quit ("If you're reading this, it's probably x-chat's fault."). 10:37:01 what client CXI was using? 10:37:54 -!- CXI has joined. 10:39:30 Xchat on WinDOS. 10:48:55 -!- sp3tt has joined. 11:01:10 -!- sp3tt_ has joined. 11:09:42 -!- sp3tt has quit (Read error: 60 (Operation timed out)). 11:29:46 -!- jix has joined. 11:34:26 moin 11:58:30 XUML parser done 12:33:58 XUML? 12:34:07 -!- J|x has joined. 12:34:21 -!- jix has quit (Nick collision from services.). 12:34:45 -!- J|x has changed nick to jix. 12:35:32 XUML interptreter done.. 12:38:41 What is XUML? 12:40:03 my tc language with only 4 instructions 12:40:08 X U M and L 12:40:48 and no other syntax elements (no () no [] no white-spaces no {}...) 12:43:44 moin 13:44:31 * pgimeno pages GregorR 13:50:53 wow autoconverted bf => XUML hello world is 12 kb 13:55:50 XLLLLMXXULXMMLXX is the first handwritten XUML program 13:56:04 an endless loop printing U:s 13:57:13 U = 01010101b 13:57:46 hehe 13:58:31 but xuml is little endian (because i want boolfuck => xuml conversion) so i need to start with a 1 13:58:33 handling bits one by one is awkward 13:58:50 but i don't need + and - just X 13:59:52 I'm only justifying the 12 Kb 14:03:31 not autoconverted it is much shorter 14:07:10 XLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLMXULXULXULXXUXUXXULXXUXXULXXUXXULXXUXXULXXUXXULXXUXXULXXUXXULXXUXUXXULXULXXUXXULXULXULXXUXUXXULXULXXUXXULXULXXUXXULXXUXXULXULXULXULXMMLXX 14:07:19 prints XUML\nXUML\nXUML\n.... 14:09:32 i'm in france next week. no internet.. no #esoteric :'( ;) 14:10:28 cu 14:10:31 -!- jix has quit ("Banned from network"). 14:34:48 -!- graue has joined. 14:36:21 hi graue 14:37:53 any clue of what kind of storage does [] use? 14:47:18 I mean [] a.k.a. Brackets 14:57:08 doesn't it say on the website? 14:59:13 it's too brief 15:02:42 does it have an interpreter or spec or anything? 15:03:40 yes 15:03:49 then that should tell you 15:03:50 interpreter 15:03:53 no spec 15:03:53 someone needs to add clunk: http://esoteric.sange.fi/essie2/download/clunk/ 15:11:20 interesting 15:11:24 reminds me of shrdlu 15:20:35 blagh, need to actually write this newsreader 15:23:26 woo, dodgy html parsing is a winner 15:28:31 Newsreader written in what? Brainfuck? 15:29:13 haha 15:29:18 worse, VB 15:32:33 graue: how do you rename a page? does it have to be deleted and the new one created? 15:40:09 http://esolangs.org/wiki/Special:Movepage 15:41:25 thanks 15:42:28 hum, then what? 15:43:26 "You have not specified a target page or user to perform this function on." 15:43:48 hmm 15:43:53 &target=pagename 15:44:57 it's a little silly - in theory there should be a link somewhere in the interface to do that 15:45:25 found it 15:45:54 it was not worth creating a redirection etc. 15:46:18 but now I don't know how to delete a page 15:49:28 you have to be an admin first 15:50:12 I'll leave that up to graue then (page: [[Category:Language]]) 15:51:25 CXI: you may be interested in this sed adder: http://www.formauri.es/personal/pgimeno/temp/addsed-r.txt 15:52:18 does your language work by replacing strings via regexps? 15:54:04 basically, yeah 15:54:59 nice, then it should be straightforward to adapt the adder to your language 15:55:35 well, it depends... :D 15:56:01 see, it actually stores the regexes in a list, goes through the list and rewinds every time it makes a replacement 15:56:46 oops, gtg, ttyl 15:56:52 alrighty, seeya 16:48:13 re 16:49:12 * CXI waves 16:49:15 does your lang admit \1, \2 etc.? 16:49:30 yeah... though it uses $1 $2 instead at the moment 16:49:37 I'll change it when I get around to cleaning up the code 16:49:45 k 16:50:23 right now I'm bashing away at this newsreader 16:51:43 when you say it rewinds, do you mean that it starts by the first RE of the list after each replacement? 16:51:53 yeah 16:52:32 think of it like a functional language with only one function :P 16:52:37 I don't think that's important then (for my adder) 16:53:53 what's the string's initial state? 16:54:04 is it user-given? 16:54:10 yeah 16:54:38 though incidentally you'd get stuck in a loop between s/2/11/ and s/11/2/ 16:55:01 what's the stop condition? 16:55:13 no patterns match :P 16:55:29 did I mention there's no /g flag? :P 16:55:38 yuck :) 16:56:18 still, it's easy to fix to adapt to these needs 17:04:47 G does what? 17:05:58 in a "s/regexp/replacement/g" sed statement, it replaces all occurences (as opposed to the first one when scanned from left to right) 17:07:13 Ah. 17:07:25 I want to read the 2P specs. 17:08:05 alright, I'll write one up quick-like 17:10:00 I was sorta putting it off because I wanted to know whether it could actually be useful 17:10:11 Thanks. 17:11:25 though keep in mind the interpreter won't actually match the spec until I fix it :P 17:13:02 useful as in "turing complete" or useful as in "easy to use"? I can tell you in advance that it's TC 17:13:13 ah, well then that's cool 17:13:16 -!- sp3tt_ has quit (Read error: 104 (Connection reset by peer)). 17:13:17 easy to use, hell no :P 17:13:29 hehe 17:14:24 you'd better specify which REs are admitted 17:14:36 if possible, don't restrict them to be Perl REs, please 17:14:44 I won't 17:15:08 *phew* 17:15:16 I'm thinking maybe gnu regex 17:15:28 is perl regex that different? 17:15:35 Posix REs would be good 17:15:50 POSIX EREs that is 17:16:01 well, there are quite a few extensions 17:16:27 the problem is supporting them in non-Perl interpreters 17:19:51 hmm 17:21:28 what are \1 \2 \3 called anyway? back-replacements? 17:22:37 back references IIRC 17:23:00 but only when used within a RE 17:23:04 yeah 17:23:20 i.e. not in the RHS 17:23:24 ah... 17:23:41 maybe sed has another name 17:23:55 eh, I'll just define them in the spec as back-references :P 17:25:29 "The replacement may contain the special character & to refer to that portion of the patter space which matched, and the special escapes \1 through \9 to refer to the corresponding matching sub-expressions in the regexp." (from the sed man page) 17:29:00 s/patter/pattern/ 17:29:38 -!- ChanServ has quit (Shutting Down). 17:31:37 -!- ChanServ has joined. 17:31:37 -!- irc.freenode.net has set channel mode: +o ChanServ. 17:42:42 heh, actually 17:42:52 it might not be possible to write a fair few things in it 17:43:04 if only because there's no way to distinguish what the user initially entered and what you're returning 17:43:35 as in, writing something to add the letter s to any inputted string 17:44:09 and /(.*)/(.*)s/ would just loop forever 17:44:12 er 17:44:22 I mean /(.*)/\1s/ 17:44:44 * CXI considers making the interpreter add '>' to the front of the input 17:49:57 hmm, but then you could never have a program that outputs a > 17:50:05 at the start of the output, anyway 17:51:28 you can escape the input string somehow at the start 17:52:53 e.g. s/\\/\\\\/g then s/^/\\i/ 17:55:54 but the problem is if the program was meant to output a string with \i at the start 17:56:53 then that'd be matched by the expression you used to check for ^\i and the program would loop infinitely 17:56:56 unescape it at the end; the regexps should then write \\i at the start instead of \i 17:57:24 -!- sp3tt has joined. 17:57:36 ah... hmm, yeah, that works 17:57:51 w00t, badge with brainfuck program on ordered :D 17:59:14 And also badges with a gnu, one that says "Proud filesharer", and of course one that says "How about a nice cup of stfu?". XD 17:59:28 heh, classy 17:59:33 what's the brainfuck program? 17:59:49 It prints "Your brain is fucked!" 18:00:00 that sounds like it'd be pretty long 18:00:40 Not with some clever multiplication... 18:00:42 ++++++++++[>+++++++++<-]>-.>++[<+++++++++++>-]<.++++++.---.>++++++++[<---------->-]<--.>++++++++[<++++++++>-]<++.>++[<++++++++>-]<.>++[<-------->-]<-.++++++++.+++++.>++++++++[<---------->-]<++.>++++++++[<+++++++++>-]<+.>++[<+++++>-]<.>++++++++[<---------->-]<---.>+++++++[<++++++++++>-]<.>+++[<+++++>-]<.>++[<--------->-]<.++++++++.------.-.>++++++++[<-------->-]<---. 18:00:57 Looks like this: http://www.knapp.nu/ShowBadge.aspx?id=2549463 18:01:32 yeah, not so bad, I guess 18:02:04 Should be here within 14 days :) 18:02:21 Ordered one with the firefox logo on too. 18:12:19 how about writing that in qdeql? 18:12:59 -!- graue has quit ("brb"). 18:17:15 qdegl? 18:26:26 -!- graue has joined. 18:27:59 puzzlet: around? 18:28:10 behind you 18:28:29 heh 18:29:15 just a note that in http://puzzlet.org/puzzlet/%EC%95%84%ED%9D%AC~Aheui in the links section there's a link redirecting to an "obsolete" page 18:29:29 ah, i forgot 18:29:45 even forgot to update those in Korean pages 18:31:10 * pgimeno wonders what each button will do in the JS interpreter 18:31:55 * puzzlet plans to make an extra frontend in English 18:32:41 that would be very nice for non-korean speaking people :) 18:34:39 to me it could say "press here to download and install Windows" without me noticing 18:35:48 which one? 18:35:59 any of the buttons 18:40:02 ah now i get it 18:41:42 * cpressey fears the wiki has gone category-crazy :) 18:42:04 what's the difference between {{Category:whatever}} and [[Category:whatever]] ? 18:42:10 the wiki goes wikipediastic 18:42:53 i'm not sure that {{Category:whatever}} is a valid syntax 18:42:56 {{}} is a template 18:42:59 [[]] is a link 18:43:04 oh, ok 18:43:07 thanks 18:43:07 so {{}} brings in the text from another article 18:43:12 np 18:43:18 ah, it will work that way 18:45:57 yeah, i think a lot of the categories are really bad ideas 18:46:31 in particular, low-level and high-level (how do you define that with respect to so many different programming styles?), and almost all of the other categories should be subcategories of "languages" 18:47:44 how does a subcategory help with respect to a category? 18:47:59 any there's going to need something like [[Category:Languages by storage types]] and so on 18:48:07 it means we don't have to put [[category:languages]] on practically every single page 18:48:38 graue: because the list will do the complete list of languages 18:48:52 wp has "article which should be a category" category... that way a topic can start life as an article, then eventually become a category when it's clear that it needs to be... maybe that model would work better than trying to categorize everything immediately 18:49:24 that's a good idea 18:51:19 I think that categories are a good way of getting an idea of what a language is like to start with 18:52:17 is it imperative? is it non-deterministic? etc. 18:53:14 of course that could belong to the description but using categories helps in making a cross-reference list at the same time 18:55:10 about the inclusion of most or all categories as subcategories of the Languages one, I don't know how to list e.g. all languages that way 18:57:11 -!- CXI has changed nick to coredumpage. 18:57:32 -!- coredumpage has changed nick to CXI. 19:02:02 for example, currently Turing tarpits is a subcategory of Turing complete, but you can't examine a list of all Turing-complete languages that includes Turing tarpits without entering each subcategory 19:02:39 is there a way around that? 19:03:24 [[List of Turing tarpits]]? 19:04:15 that would require manual editing of just another list 19:05:28 adding a language to a subcategory does not automatically add it to its parent category 19:11:04 so you can't examine a list of all Turing-complete languages that includes Turing tarpits without entering the subcategory 19:11:11 is it that hard to enter a subcategory? 19:11:30 that's not the problem 19:12:19 if all the current categories are subcategories of the Languages category, you can't have a list of Languages without entering every category 19:12:32 yes you can, by going to esolangs.org/wiki/Language_list 19:13:27 what's the use of that page which can't be done with a Languages category? 19:13:54 it doesn't require global modifications 19:14:03 for a category, you have to edit every page in the category 19:15:09 so what's a good non-esoteric high-level language i should learn? objective-c? java? ruby? erlang? 19:17:19 er... define "good" 19:18:07 you don't think any of those are good? 19:18:59 "good for what" is the question 19:19:16 java is quite more popular than the rest 19:19:44 but I don't know if popularity is meaningful for you 19:20:49 well, good for programming in 19:21:05 i don't care if it's buzzword-compliant or not, if that's what you mean 19:21:45 i don't want to do BOP in a strongly-hyped language 19:23:10 good speed-wise, ease-wise, self-explanatory-wise...? do you require it to be scriptable? have good string handling? etc. 19:25:20 what languages do you prefer? 19:25:23 graue: depends why you want to learn it (sort of like pgimeno said)... i can only say which non-eso languages i personally admire 19:25:26 erlang and lua 19:26:32 what's cool about erlang? 19:26:42 I quite like ruby 19:26:54 receiving messages based on patterns 19:27:09 at least, imo 19:27:34 is it (can it be) relatively fast? (not necessarily C-speed, but usable for some real-time stuff) 19:27:55 graue: it's billed as a "soft real-time" language (fwiw) 19:28:44 speed is, hmm, ok, in my experience... certainly acceptable for the things i use it for 19:28:53 it doesn't do so well on most shootouts though 19:29:02 although i'm not sure how much i trust the shootouts anyway 19:29:23 have you ever used Icon? 19:29:58 briefly. not for anything serious. i decided to go with lua instead, which has some similar features 19:30:28 icon was way ahead of its time... 19:32:16 i guess i'll study erlang in some more depth 19:32:44 *pitches in* give ruby a look too :P 19:33:18 what is the downside of ruby? 19:33:26 speed 19:33:34 it's so popular there must be a group of people who hate it 19:33:52 i'd like to hear from those people before spending much time on ruby, but it is interesting 19:34:14 heh, I'm not sure how easy it'd be to find any of those people 19:34:14 i'm not so big a fan of ruby, but i don't have any particular thing against it 19:34:46 I'm pretty quick to condemn a language, but I haven't been able to find much wrong with ruby at all 19:34:53 the syntax was a little strange at first 19:37:03 oh, erlang can load code into running systems, that's cool 19:48:01 graue: btw i suspect qdeql needs 2 queues 19:48:05 for TC 19:48:32 you have the problem (i think) of the length of the queue being unknown at any given point 19:49:04 so how do you (e.g.) know you've cycled through the entire thing if you e.g. want to get at a cell in the very middle 19:49:11 that's just a guess though 20:02:28 isn't that equivalent to the problem of the position of the tape pointer in brainfuck? 20:05:15 it seems to me that it should be possible to keep track of the length of the queue in a byte (or two, or three, or an unbounded number of bytes) that is kept accessible at all times 20:18:38 hmmm 20:19:17 i'm just not sure how you keep it accessible at all times 20:19:34 the tape in bf doesn't wrap around; you don't need to know the current length of it to navigate it 20:19:39 it seems like in a queue, you would 20:19:51 maybe if it was a deque 20:20:20 (then you could go "left" and "right" like in bf or a TM) 20:25:08 oh 20:25:09 hmm 20:25:11