00:32:03 -!- Corun has quit (Read error: 110 (Connection timed out)). 01:08:28 http://www.haskell.org/sitewiki/images/0/0a/TMR-Issue10.pdf Haskell in Haskell. 01:08:31 Intermediate language? 01:08:32 Unlambda. 01:14:01 closer to Lazy K, i should think 01:14:20 i was just quoting the reddit headline 01:14:30 ah 01:19:06 -!- tusho has quit. 01:26:14 Hahah. :) 01:27:07 THIS IS NO LAUGHING MATTER 02:02:24 -!- erus` has joined. 02:18:34 Idea: You're a wolf. You communicate by howling or not howling. At any given time, you know how many wolves are howling. Try to figure out what your friend is saying. 02:28:51 -!- bsmntbombdood has joined. 02:56:00 * Dewi looks around the channel for people called 'Idea' 02:57:17 probably just a horrible misspelling. actually, "Dewi" looks sort of close. 03:10:29 oerjan: Can't be me. I communicate by howling, but I'm not a wolf. 03:11:16 well that's what _you_ think 03:12:06 -!- sebbu has quit ("reboot"). 03:24:10 ihope: Unless the wolves use CDMA, I refuse. 03:24:27 There, I just violated an Agoran contract. 03:24:38 Which contract? 03:25:01 The pledge to ascend my current game of NetHack. 03:25:08 Ah. 03:25:18 Fool. 03:25:25 Where'd you die? 03:26:20 On dungeon level 3, of course. 03:26:36 That death was not nearly embarrassing enough. 03:29:36 * pikhq is curious 03:29:58 (what can I say? I'm curious about how people die in nethack) 03:32:18 Essentially, I starved to death. 03:32:49 Lessons learned from that game: Don't eat poisonous things, thereby decreasing your strength, then kick *three* sinks, thereby releasing the black puddings within, and cause them to divide by hitting them accidentally, then get most of your armor and both your weapons stolen by nymphs and become a wererat and summon so many pets that you have to kill some of them to move around. 03:32:59 As told to #nethack a few minutes ago. 03:33:29 Another lesson learned from that game: Don't pledge to win a game you've never won before. 03:33:37 Do pledge to get the Bell if you've done that before. 03:37:31 Also, if you're hungry, kill your pet. :p 03:37:51 Tasty. 03:40:04 -!- Sgeo has joined. 03:41:37 I killed my pet because he was in the way; I wasn't hungry. Then I starved to death anyway. 03:44:41 I just started a new game and already my armor class is below what it was in my previous game. 03:57:43 -!- Sgeo has quit (Read error: 104 (Connection reset by peer)). 03:59:42 -!- Sgeo has joined. 04:09:45 -!- pikhq has quit ("leaving"). 04:25:39 -!- oerjan has quit ("Good night"). 04:33:06 ihope: wait, you can *win* nethack? 04:34:02 nethack is one giant puddle of wtf 05:33:44 -!- GreaseMonkey has joined. 06:32:10 -!- Slereah_ has joined. 06:32:17 THEN WHO WAS PHONE? 06:43:21 huh? 06:50:09 -!- olsner has quit ("Leaving"). 07:02:27 -!- Sgeo has quit (Read error: 104 (Connection reset by peer)). 07:57:49 -!- CakeProphet has quit (Read error: 110 (Connection timed out)). 07:59:59 -!- clog has quit (ended). 08:00:00 -!- clog has joined. 08:25:06 -!- sebbu has joined. 09:34:02 * Mony is back (gone 09:36:42) 09:34:05 hi 09:35:18 Hi. 10:05:01 -!- kar8nga has joined. 10:10:01 -!- Slereah- has joined. 10:10:01 -!- Slereah_ has quit (Read error: 104 (Connection reset by peer)). 10:44:01 -!- kar8nga has left (?). 10:54:58 -!- GreaseMonkey has quit ("You're still in denial. It's easier to lie by saying that I'm lying than to face truth."). 11:06:17 -!- tusho has joined. 11:07:06 hello 11:08:50 Deewiant, hey I found an possible issue with TRDS, how does it handle the ? instruction 11:08:58 it need to give same result on the next run 11:09:45 hi tusho 11:29:38 AnMaster: heh, ask Mike 11:29:56 Deewiant, does ccbi do the right thing there? 11:30:06 no :-P 11:30:26 Deewiant, bet you don't want to need to change that ;P 11:30:29 "it should. Make a new release." 11:30:34 would be a simple change though 11:30:48 Deewiant, recording any ? yeah 12:01:38 -!- jix has joined. 12:03:18 Incorrect plurals. Discuss. 12:03:19 Databii. 12:05:04 * Mony miam 12:11:52 -!- erus` has quit (Read error: 104 (Connection reset by peer)). 12:13:19 -!- ErkiDerLoony has joined. 12:42:25 -!- kar8nga has joined. 12:50:40 -!- RedDak has joined. 13:03:05 -!- kar8nga has quit ("Leaving."). 13:16:26 * Mony is away: I will be back ! 13:19:23 -!- timotiis has joined. 14:12:24 -!- ihope has quit (Read error: 110 (Connection timed out)). 14:14:34 Hmm. 14:14:41 I type "publicly", not "publically". 14:24:59 -!- pikhq has joined. 14:40:51 -!- RedDak has quit (Remote closed the connection). 14:41:50 -!- CakeProphet has joined. 14:47:29 -!- Tritonio_ has quit (Remote closed the connection). 15:01:40 -!- ais523 has joined. 15:02:08 hi tusho 15:02:12 .................... 15:02:17 I JUST UNFOCUSED MY IRC CLIENT 15:02:18 THIS SECOND 15:02:27 * tusho stabs ais523's eyes out 15:02:39 * ais523 defends emself with a towel 15:02:42 useful things, towels 15:02:50 Aha! But... 15:02:54 * tusho brings out NUCLEAR TOWEL 15:03:02 It's nuclear, so yours dissolves along with your eyes. 15:03:18 well, I defeat that with a cluebat 15:03:24 now can we get back to talking about esolangs? 15:03:37 I went and implemented Shove 15:03:47 I couldn't think of an efficient way to do it so I did it a really inefficient way 15:03:56 heh 15:04:11 at least the way I did it makes it clear what's going on 15:04:41 what language? 15:04:44 don't say shove. 15:04:45 Perl 15:07:03 http://code.eso-std.org/shove/shove.pl 15:09:46 ok, I've put a few example programs there too now 15:09:58 modelling the Underload stack-manipulation primitives 15:10:00 -!- kar8nga has joined. 15:10:04 I have dup, concat, swap and enclose there 15:10:13 pop, eval and quote are trivial 15:10:34 so all I need now is a way to combine two programs into one and I'll have an Underload -> Shove compiler 15:11:00 combining two programs into one isn't trivial in Shove but I've thought of a way to do it, now I'm writing a precompiler that I can use to generate the compiler 15:12:20 ais523: shouldn't it say "by ais523"? 15:12:50 looks cool though 15:12:58 i think you'll need a stage-2 shove though 15:13:00 quite possibly, but I tend to sign realname on things that have copyright 15:13:00 that lets you define commands 15:13:03 because "swap" is long 15:13:23 tusho: yes, that would be important for practical use, atm I'm trying to work around the restrictions of the language though 15:13:32 ais523: yes 15:13:37 programming Shove feels a bit like programming a cross between Underload and SMITH, only 10 times more confusing 15:13:43 also, nowadays I generally sign copyrighted things tusho 15:13:50 i think it's sufficiently formal 15:13:59 how many things have been published under pseudonyms? loads 15:14:09 i don't see why it wouldn't stand up legally 15:17:13 -!- ihope has joined. 15:17:25 Tzi-ga-na-tood. 15:17:32 ihope: ? 15:17:38 Ello. 15:17:43 also, I put my new Shove interp online 15:17:47 http://code.eso-std.org/shove/shove.pl 15:23:14 hopefully it's clear enough for people to be able to deduce the spec from the interp 15:23:14 -!- ErkiDerLoony has quit (Read error: 104 (Connection reset by peer)). 15:23:25 oh, and ! S and n aren't part of the language 15:23:31 they're all NOPs, just with side effects 15:23:44 which make it possible to produce output and to run the program in a reasonable length of time 15:34:01 "NOPs with side effects" sounds like an oxymoron to me ;-) 15:35:32 Deewiant: they're NOPs from the language's point of view 15:35:37 I have a no I/O tarpit 15:35:43 but certain nops happen to cause output-like side-effects 15:36:01 yeah, I know what you meant, it just sounded a bit funny until I read the context :-) 15:36:20 another interesting point about Shove: in Befunge, if you go into stringmode from right to left, you get the string as you wrote it; in Shove, if you do that, the string comes out upside-down 15:36:30 or at least with v and ^ swapped and likewise for the other characters 15:36:47 this is a consequence of the way I wrote the interp, but I deliberately wrote the interp like that so that that would happen 15:37:00 ¿ʇɹoddns ǝpoɔıun ǝʌɐɥ ʇı sǝop os 15:37:21 unfortunately no, I wanted to do that but it couldn't write messages sideways very well 15:37:30 heh 15:37:32 so I just restricted myself to command characters and left other characters alone 15:38:46 -!- timotiis has quit (Read error: 110 (Connection timed out)). 15:44:50 oh that's a nice css effect 15:44:57 the background of blockquotes was fixed to all of them 15:45:00 so it was like they were cut holes in the page 15:45:05 to a fixed position background 15:45:13 that is clever 15:46:00 it's a bit too subtle though 15:46:02 the shape is white on beige 15:46:08 explains why I haven't noticed it before 15:49:15 -!- MikeRiley has joined. 15:52:21 you about AnMaster???\ 15:52:28 hm? 15:52:30 what? 15:52:40 AnMaster: ask him about ? 15:53:03 got a couple questions on funge108.... 15:53:13 been working on adding a mode for that in Rc/Funge-98... 15:53:21 Deewiant: what's your ? question? 15:53:22 MikeRiley, well ? in TRDS, shouldn't it mean that when you jump back in time the old ip (not jumped) will make the same random choice at ? 15:53:25 or? 15:53:31 ais523: that 15:53:32 oh, TRDS 15:53:35 would not make the same random choice... 15:53:50 MikeRiley, well fire away the questions 15:53:53 when an ip jumps back, the future no longer exists... 15:53:53 I /thought/ ? was one of the better-specified commands 15:54:01 I have been extremely busy with non-esoteric stuff the last week 15:54:06 so future events need not recur... 15:54:08 and will probably be that the next week too 15:54:12 so I may be hard to reach 15:54:17 fire away the questions 15:54:30 ok,,,first,,,on your new fingerprint method,,,,what format are you expecting URIs to be??? can you give me an example?? 15:54:57 MikeRiley, iirc I refer to an RFC and even give some examples 15:55:08 do not remember seeing any examples.... 15:55:15 http://example.com/my-finger-print.html 15:55:17 or whatever 15:55:19 just URIs 15:55:31 just a though....web addresses are a bad idea.... 15:55:36 doesn't even have to be URLs, though I guess in practice they will be 15:55:59 well, in that case I may end up calling a fingerprint apt://nethack to annoy people 15:56:09 for resource types, i was looking at f108: for the interpreter handprint, and fngr: for fingerprints... 15:56:24 ais523, is it a valid protocol defined in a RFC? 15:56:26 ;P 15:56:37 using a web address is problematic,,,, 15:56:37 AnMaster: no idea, but I see the links around every now and then 15:56:38 MikeRiley, no, makes no sense at all IMO 15:56:40 MikeRiley: use URIs 15:56:45 MikeRiley: there's a spec for them 15:56:45 MikeRiley, URI != URL 15:56:48 (note: not URL) 15:57:00 and what should the defined format be??? 15:57:05 MikeRiley: here is the rfc: 15:57:12 MikeRiley: http://tools.ietf.org/html/rfc3986 15:57:18 it's long; have fun 15:57:33 there are uri manipulation libs for just about every language 15:57:35 must be one for c 15:58:43 -!- ais523 has quit (Remote closed the connection). 15:59:01 -!- ais523 has joined. 15:59:02 asking an implementer to read through that rfc is not good, better to provide reasonable examples within your spec... 15:59:29 wb ais523 15:59:33 MikeRiley: so use a library! 15:59:48 anyway all URIs are reasonable 15:59:50 no reason to restrict them 16:00:04 so,,,i can just invent anything i want for them??? 16:00:10 no. 16:00:14 anything that obeys http://tools.ietf.org/html/rfc3986 16:00:20 then, layout a spec,,,, 16:00:24 MikeRiley: uhh 16:00:26 you mean 16:00:28 the rfc covers them in a general form... 16:00:28 duplicate the existing spec? 16:00:31 no,,,, 16:00:39 http://tools.ietf.org/html/rfc3986 is a specification for URIs 16:00:44 anything that obeys that is a valid URI 16:00:50 just like http: is a dfined subset....define a subset for this... 16:00:56 no 16:00:57 that's not the point 16:01:09 the point is that it SHOULD use http://, gopher://, ftp://, whatever 16:01:15 that's the point of using URIs 16:01:20 understood.... 16:01:31 but anything before the :// ??? i can define whatever i want??? 16:01:40 i think the resource type should be speced,,,, 16:01:48 MikeRiley: http://tools.ietf.org/html/rfc3986 16:01:55 i don't know how many times I need to paste that link.. 16:02:11 i do not intend to read through that document....too long 16:02:23 MikeRiley: so use a pre-existing library for manipulating URIs in C 16:02:30 http://en.wikipedia.org/wiki/Uniform_Resource_Identifier more info 16:02:31 ah, I finally found the thing before the :// 16:02:33 it's at http://www.iana.org/assignments/uri-schemes.html 16:02:44 they didn't define it in the RFC itself so they could change it later 16:02:51 of course 16:02:53 they just mandated IANA to keep a list, and that's where it is 16:02:54 no reason to restrict it though 16:02:58 just allow any valid URI 16:03:00 the rfc just says the a resource type has to be there, does not define them 16:03:02 and then you won't have to update it 16:03:06 srsly 16:03:09 no reason keeping track of that list 16:03:15 just allow anything that is a syntactically valid URI 16:03:23 that is how URIs are used all the time and how they should be 16:03:33 of course....so my point,,,i can define whatever resource type i want???? so like fngr://??? 16:03:47 so could do something like fngr://rcfunge98.trds 16:03:49 ??? 16:03:50 MikeRiley: don't, but it would be syntactically valid 16:03:50 well, no, because it isn't a URI if you don't use a registered scheme, I don't think 16:03:56 ah yes, ais523 is right 16:03:59 ok, you need to keep track of the list 16:04:00 not too hard 16:04:03 it would be well-formed just not valid 16:04:08 it's not like it has changed for what, years? 16:04:16 MikeRiley: in practice the URI will probably be 16:04:29 http://your.site/funge/fingerprints/name 16:04:37 and have that link to documentation 16:04:43 what about an irc:// URI to a fingerprint? 16:04:45 that would be great 16:04:48 ais523: hah, yes 16:04:50 or mailto 16:04:52 your.site is a bad idea.... 16:04:56 MikeRiley: no, it's not 16:05:01 yes it is.... 16:05:03 URIs are used as universal identificaiton and have been for quite a file 16:05:07 actually irc:// is not on the list 16:05:09 it is the Right Thing 16:05:09 which surprises me 16:05:11 example#1 catseye which defined some specs no longer exist 16:05:17 MikeRiley: yes it does 16:05:18 catseye.tc 16:05:24 anyway, URIs by definition aern't meant to change 16:05:24 different address tho 16:05:25 but I agree with MikeRiley to some extent, especially as not everyone has a website 16:05:31 ais523: that's why ESO is offering them 16:05:32 remember? 16:05:34 tusho: that doesn't prevent them changing in practice 16:05:38 example#2, the fingerprints that i have defined are on a different address... 16:05:41 http://fingerprints.eso-std.org/... 16:05:47 you change ESO's naming scheme every week afaict 16:05:53 no 16:05:55 next week it'll be befunge.eso-std.org/fingerprints 16:06:00 no, it won't 16:06:04 you don't understand what i'm saying 16:06:13 yes I do, I'm just disagreeing 16:06:19 websites are NOT as constant, and in my opinion should not be part of URIs for fingerprints.... 16:06:23 -!- Hiato has joined. 16:06:30 MikeRiley: URIs are URIs whether you'd like them to be URIs are not. 16:06:42 If you don't support a certain type of URIs you don't support URIs at all. 16:06:52 well in that case it shouldn't be "URI" but some other sort of identifier 16:06:58 no it shouldn't 16:06:59 back 16:07:01 you'd end up reinventing URIs 16:07:05 which is pointless 16:07:14 no, would still be in the format of uris,,,,just different information... 16:07:23 MikeRiley: the format of URIs includes the protocols. 16:07:27 If you disallow a protocol you do not support URIs. 16:07:28 i agree... 16:07:39 I highly reccomend you just stick with URIs. 16:07:40 i think the protocol should be there,,,and be defined for handprints and fingerprints 16:07:52 That's anti-ethical to the whole idea of URIs! 16:08:00 They're meant to design a resource protocol-agnostically! 16:08:07 er, not resource 16:08:09 you know what I mean 16:08:24 I will stick to plain URIs 16:08:36 because I want to be able to use gopher or whatever if I want 16:08:45 tusho: the problem is URIs aren't actually very good at their job 16:08:45 AnMaster: good :) 16:08:54 ais523: URIs are fine at their job. People aren't. 16:08:55 makes no sense to me to name fingerprints this way... 16:08:56 because people don't control Internet locations indefinitely in real life 16:08:59 anyway the risk of collision is *much* higher currently with 4 letters than with a URL 16:09:01 URI* 16:09:02 [[[ What makes a cool URI? 16:09:03 A cool URI is one which does not change. 16:09:03 What sorts of URI change? 16:09:03 URIs don't change: people change them.]]] - TBL 16:09:15 i agree about the 4 letters,,,,that was definitely a problem... 16:09:17 tusho: or their webhost changes them, or their financial situation 16:09:21 ais523: purl.net 16:09:40 tusho: what if it goes down tomorrow 16:09:44 I mean, seriously? 16:09:49 it probably won't happen but it might 16:09:49 ais523: it won't. It's got a huge amount of backing 16:09:49 not there have been any collisions yet,,,very few people have defined fingerprints..... 16:09:56 it's not just Yet Another Service 16:10:18 tusho: that attitude strikes me as being overconfident 16:10:23 no matter what website it's with respect to 16:10:51 ais523: purl has got a huge, huge amount of backing and effort & research to stay totally persistent put in it 16:10:51 Actually, a URI is meant to just *identify* the resource... Not necessarily point out where it is. urn:isbn:1234567890123, for example, is a valid URI... Well, except that that's not by any means a valid ISBN. :p 16:10:56 pikhq: Yes. 16:11:02 pikhq: yes 16:11:08 It seems like at least one person did not understand that. 16:11:15 Though I might just be missing something. 16:11:22 MikeRiley, I think 16:11:25 even so, I think it should probably have a date involved if you're going to base a URI on an URL 16:11:33 because the URL might end up owned by someone else eventually 16:11:51 URLs are a subset of URIs 16:11:57 "basing a URI on a URL" is a silly concept 16:12:23 which is back to my point, a web address is not relevent in a URI... 16:12:23 -!- Hiato1 has joined. 16:12:39 MikeRiley: of course it is 16:12:39 unless it is the URI is for a web based service... 16:12:55 fingerprints are not web-based services.... 16:13:01 they are features in an interprter,,,, 16:13:05 that's not how it works 16:13:11 well, I think it should be a data:// URI 16:13:20 that contains all the information needed to implement the fingerprint 16:13:28 e.g. the code for the fingerprint encoded in base64 16:13:29 data:// is reasonable.... 16:13:37 and i can put anything i want after that??? 16:13:42 MikeRiley: yes, but my suggestion is unreasonable on other grounds 16:13:46 -!- kar8nga has quit (Connection timed out). 16:13:49 which are??? 16:13:52 MikeRiley: there's a defined format for data:// 16:13:57 MikeRiley: you must support all URIs 16:14:02 otherwise you do not support URIs at all 16:14:04 ok,,,,so we are back to where i started... 16:14:07 MikeRiley: if you're going to have the entire code of your fingerprint in your program, then why bother with fingerprinting at all? 16:14:08 and therefore will not conform to befunge-108 16:14:36 all my fingerprint code is in my interpreter,,,,but it still requires the fingerprint mechanism to overload A-Z.... 16:15:01 i am trying to figure out how these should appear??? 16:15:05 yes, in the interpreter makes sense, in the program doesn't 16:15:23 when i use (,,,,,what should appear in the 0gnirts to load TRDS for example??? 16:15:38 MikeRiley: http://your.site/befunge/fingerprints/trds 16:15:39 or whatever 16:15:42 the problem is that funge-108 fingerprints will likely have names which are far too long for practical use 16:15:53 mailto:your.emai+trds@your.email.domain 16:15:54 your.site is unacceptable to me,....my site may change in the future.... 16:15:57 just look at any OpenOffice.org macro language source and you'll see what I mean 16:16:00 MikeRiley: will your email? 16:16:04 probably 16:16:06 mine definitely will 16:16:13 MikeRiley: ok, what won't? 16:16:18 anything related to the web is not a constant,,,, 16:16:34 i think fingerprints should have names that can be relied upon to be constant.... 16:16:48 a URI changing is a bug, a flaw, an error 16:16:50 in practice, yes, they do 16:16:52 so pick one that won't 16:16:59 pick a protocol that doesn't rely on something mallable 16:17:05 ais linked to a list of them 16:17:07 to pick one that will not,,,you cannot use any web address within the uri.... 16:17:12 correct 16:17:24 http://www.iana.org/assignments/uri-schemes.html 16:17:26 pick one 16:17:51 MikeRiley: in particular 16:17:54 urn: 16:17:57 is probably relevant to this 16:18:06 http://en.wikipedia.org/wiki/Uniform_Resource_Name 16:19:02 so something like: urn://fingerprint:rcfunge98:trds ????? 16:19:11 MikeRiley: why do you keep making things up? 16:19:16 why not read the definition of URNs... 16:19:17 i am not.... 16:19:20 just did.... 16:19:25 yes, you just made up urn://fingerprint:rcfunge98:tds 16:19:40 MikeRiley: http://www.ietf.org/rfc/rfc2141.txt 16:19:41 it's short. 16:19:45 a lot shorter than the URI definition 16:19:51 a few pages 16:19:58 ok, show me an example of how YOU would specify the trds fingerprint.... 16:20:07 http://www.iana.org/assignments/urn-namespaces 16:20:20 MikeRiley: why? 16:20:25 it's not my responsibility... 16:20:27 i want to see what you think it should be.... 16:20:32 you choose what kind of URI you want to use, e.g. URN 16:20:36 then you make one of those... 16:20:38 pretty simple 16:21:00 ok,,,,in that case, i will invent my own for my fingerprints.... 16:21:17 -!- Hiato has quit (Connection reset by peer). 16:21:39 MikeRiley: uh, as long as it accords to the spec... 16:21:46 and when somebody comes up with uris for fingerprints made by somebody else,,,,i will implement them then... 16:21:50 if it doesn't, you can't use it, because it's not an URN (or not a URI or whatever) 16:21:54 and... 16:21:57 you don't need to do anything special 16:22:01 just look up fingerprints by their uri 16:22:02 instead of name 16:22:06 it's trivial! 16:22:18 you have a database of installed fingerprints, instead of name, the key is the URI 16:22:24 which means i can do something like: fngr://rcfunge98/trds..... 16:22:29 no 16:22:29 that follows the rules of a uri.... 16:22:33 it isn't a valid URI 16:22:38 because the protocol is not one of the registered protocols 16:22:41 does not matter that fngr is not already defined by somebody else... 16:22:44 yes 16:22:45 it does 16:22:52 does the uri spec say that it must be pre-registered??? 16:23:02 that is what a URI is. 16:23:06 the protocol must be a valid URI protocol. 16:23:11 IANA maintains the list. 16:23:14 the little of it that i read just sad that you had to specify a resource type...and not what the types are... 16:23:37 "the little of it that I read" precisely 16:23:39 the little of it. 16:23:52 i accept that....too long for me to waste my time reading it.... 16:23:58 i would prefer to see reasonable examples.... 16:24:06 or else i will make my own way of doing it.... 16:24:20 MikeRiley: go ahead, as long as you don't care about funge-108 compliance. 16:24:38 i guess at this point,,,i do not....since nobody is willing to define what it should look like.... 16:24:52 MikeRiley: I can copy and paste the rfc if you'd like? 16:24:56 i will use my method,,,which will work in my interpeter... 16:25:00 i've told you exactly how it will look like 16:25:03 i am not going to read the rfc.... 16:25:04 by linking you to the definitive source 16:25:07 you just refuse to read it. 16:25:15