00:00:01 <int-e> impomatic: basically I decided that a population of 170 was too big to store it all as data
00:01:13 <oerjan> hm AMA should be beginning now
00:02:37 -!- lambda-11235 has quit (Read error: Connection reset by peer).
00:03:03 <oerjan> https://www.reddit.com/r/comics/comments/4dyros/im_david_morganmar_author_of_darths_droids_and/
00:06:57 <zzo38> How can I set up NNTP server and client on this Linux?
00:08:10 <oerjan> a1 ... am (b1 ... bn c) = fmap (a1 ... am) (b1 ... bn) c. so if you can handle fmap (a1 ... am) you can handle the former. but the latter is fmap^{\circ n-1} fmap a1 ... am .
00:09:30 <oerjan> and thus by induction you can handle any application of anything you can handle.
00:10:10 -!- hppavilion[1] has joined.
00:10:44 <hppavilion[1]> I'm formulating an idea for an Esolang that could actually be of use to someone
00:11:23 <hppavilion[1]> zzo38: A language based entirely on lazy-evaluated sequences
00:12:31 <hppavilion[1]> zzo38: Maybe sort of like Haskell, but the "Functions always return the same value for the same arguments" requirement would be dropped in favor of "Functions are lazily-evaluated sequences of values"
00:19:23 <hppavilion[1]> Here's a speculative example: http://pastebin.com/h3n9xdS6
00:20:33 <oerjan> shachaf: don't complain, it was just an act of chairity
00:21:17 <shachaf> I don't appreciate what you're doing and you should stop.
00:22:43 -!- lambda-11235 has joined.
00:23:54 <hppavilion[1]> lambda-11235: Given the latter part of your nick, I'm making a language that may be of interest to you
00:24:59 <hppavilion[1]> lambda-11235: It's based entirely on feeding lazy-evaluated sequences into iterators to make NEW lazy-evaluated sequences
00:27:02 <ais523> it's a language and a VM and a security model and a browser plugin
00:27:13 <ais523> some of these things work better than others
00:29:41 <lambda-11235> ais523: It's also a drink. I personally only use the VM and the library.
00:32:27 -!- bennyZ has joined.
00:32:35 <hppavilion[1]> Truth machine now available @ http://pastebin.com/h3n9xdS6
00:33:42 -!- bennyZ has left.
00:35:55 <oerjan> <ais523> at this point it's almost worth banning the username fragment, lots of Spanish-speakers using canaima seem to end up here when they shouldn't be for no obvious reason <-- i've been tempted.
00:36:26 <ais523> yes, I know you discussed it earlier
00:36:33 <ais523> the redirect-ban actually makes a lot of sense
00:36:41 <ais523> but you'd need a good description in the target channel
00:37:04 <Phantom_Hoover> <hppavilion[1]> lambda-11235: Given the latter part of your nick, I'm making a language that may be of interest to you
00:38:05 <lambda-11235> Phantom_Hoover: It can construct infinite fibonacci sequences.
00:39:34 <lambda-11235> hppavilion[1]: It reminds me a little bit of https://github.com/matz/streem.
00:40:28 <hppavilion[1]> lambda-11235: Really, it's just https://docs.python.org/2/library/itertools.html in the form of a programming language
00:41:38 <int-e> impomatic: darn, can't stop: http://sprunge.us/QVMB
00:41:48 <int-e> impomatic: using 32 threads... wee
00:45:02 <impomatic> int-e: have you still got a few ideas how to reduce it further?
00:46:14 -!- tromp_ has joined.
00:48:27 <int-e> I'd need to find a different approach...
00:53:15 <int-e> And I'm not very hopeful there... to manage to squeeze in more threads I think I'd have to compress the image data, but any decoding I can think of will need significantly more than 2 instructions per pixel (at least 2 extra instructions, I think)... so I'd need more than 50 threads, and that's optimistic?!
00:56:15 -!- tromp_ has quit (Remote host closed the connection).
01:12:25 <int-e> impomatic: apparently you got it done in 3.5 instructions per pixel... but that doesn't change the math significantly :)
01:13:56 -!- tromp_ has joined.
01:16:07 <impomatic> My code will display almost any picture, with any number of colours :-) (and I have code which is a few cycles slower which will display any picture).
01:17:44 <int-e> yeah I've just looked at the 0x2E cycle one
01:17:52 <impomatic> So I know whatever puzzle the author devises, if it has <=48 pixels it can be solved in <=7 cycles. Otherwise it can be solved in <=0x30 cycles
01:18:40 <int-e> > (0x14-5) * 31 / 170
01:19:23 <int-e> (that's instructions per visible pixel in my code)
01:20:09 <int-e> (there's some overdrawing and some pixels have to be erased... so it's worse than the optimal 2)
01:20:46 -!- hppavilion[1] has quit (Ping timeout: 252 seconds).
01:20:52 <int-e> rdococ: excellent points. I like the middle one best.
01:21:11 <shachaf> int-e: That reminds me of a joke told by Raymond Smullyan.
01:23:24 <shachaf> "232a" in https://archive.org/stream/WhatIsTheNameOfThisBook/What-is-the-Name-of-this-Book_djvu.txt
01:23:35 <shachaf> Hmm, that's not great OCR.
01:25:44 <int-e> at least there was no drum solo
01:28:01 -!- tromp_ has quit (Remote host closed the connection).
01:49:48 -!- XorSwap has joined.
02:01:30 -!- hppavilion[1] has joined.
02:12:28 -!- hppavilion[1] has quit (Ping timeout: 252 seconds).
02:17:06 <zzo38> I made some fix of Magic: the Puzzling: Codex but do you think it is more better now? Please tell me in case something is still wrong
02:18:29 -!- hppavilion[1] has joined.
02:20:57 <hppavilion[1]> I have decided that, for XML-like data storage, KML is the optimal Markup Language
02:22:11 <zzo38> I think that it can be different whether you want text marked or data marked, too.
02:23:39 -!- Phantom_Hoover has quit (Remote host closed the connection).
02:24:35 <hppavilion[1]> http://programmers.stackexchange.com/questions/314599/why-is-xml-called-a-language-exactly makes me sad.
02:25:17 <hppavilion[1]> Obviously, for mapping data and holding arrays of integers/strings, JSON is better
02:25:30 <zzo38> For some kind of data, INI or JSON or RDF might also help to form such a data. By having a reasonable kind of XML representation of such a data then you can use XSL with it too, although the XML would not be the ordinary representation of the data
02:26:25 <zzo38> TeX supports varying syntax
02:27:02 <zzo38> It is good if you want to process it with TeX, but only TeX can properly parse it
02:27:59 <hppavilion[1]> Because multiple bodies on a tag, plus more TeX in properties (not just a string). That's why.
02:28:45 -!- tromp_ has joined.
02:29:16 <zzo38> It does not even require that they are in square brackets; other delimiters are also possible, such as a number terminated by a full stop, or text terminated by a paragraph break.
02:29:52 <hppavilion[1]> The second choice- after XML- was MAGMA: Minimal Architecture for Generalized Markup Applications
02:33:54 -!- tromp_ has quit (Ping timeout: 276 seconds).
02:36:55 <zzo38> I have written some codes in TeX and with that it is possible to see some example of how it can be done.
02:37:10 <zzo38> It can even include chess notations.
02:40:16 <hppavilion[1]> JSON is for numbers, strings, bools, maps from those to JSON, and arrays of JSON
02:40:25 <hppavilion[1]> XML is for text that has certain properties applied to subsegments
02:41:33 <zzo38> Yes that is how JSON and XML are good for.
02:42:58 <hppavilion[1]> JSON is fairly limited in scope, but XML isn't good for storing any kind of data other than text, really (because it doesn't syntactically support anything /but/ text and tags)
02:43:29 <zzo38> For even simpler data than JSON there is INI, and for a more complicated data there is RDF which is even extensible
02:44:09 <tswett> I wouldn't say that that's what XML is for.
02:44:23 <zzo38> (Of course there is also tab-separated values is another format)
02:44:35 <tswett> I'd say XML is for representing data in a way that's both human-readable and machine-readable.
02:44:45 <tswett> So it *can* be used to represent pretty much any kind of data.
02:44:50 <hppavilion[1]> tswett: Yes, but the data it represents is only really good as text
02:45:02 <hppavilion[1]> tswett: You can represent non-textual data, but it isn't syntactically shown
02:45:13 <zzo38> tswett: Yes, although it isn't really very good for any purpose other than text markup I think
02:45:31 <tswett> hppavilion[1]: what's a type of non-textual data you're thinking of?
02:46:28 <tswett> I suppose. If you're applying XML to something related to integers, you'd probably just put the digits in there and call it good.
02:46:32 <zzo38> Yes, JSON does it more easily to write numbers (although it doesn't distinguish between integer and non-integer)
02:46:54 <tswett> <pixel red="0" green="32" blue="64"/>
02:47:18 <tswett> The problem there is that that XML doesn't indicate that the attributes there *have* to be integers.
02:47:19 <zzo38> In RDF there is representation for any type (and is extensible); the Turtle syntax has a reasonable way to represent both integer and fraction numbers
02:47:27 <tswett> That too. XML tends to be a lot more verbose than JSON.
02:48:36 <hppavilion[1]> JSON represents hashmaps and arrays, XML represents trees...
02:48:52 <tswett> hppavilion[1]: you know about YAML?
02:48:59 <zzo38> And, RDF represent directed graphs
02:49:25 <hppavilion[1]> Perhaps a language that represents Rings/Groups/Etc would be nice?
02:50:21 <tswett> Haskell-like syntax is usually perfectly fine for representing mathematical stuff.
02:50:51 <hppavilion[1]> tswett: What's a good data-y data structure not covered by those?
02:51:20 <tswett> Labeled directed graphs are sort of like the ultimate data structure.
02:51:36 <tswett> Though... I guess there are a lot of things that are "sort of like the ultimate data structure".
02:52:10 <hppavilion[1]> tswett: Perhaps a strongly-typed language of some sort? One that forces you to obey a format to prevent errors?
02:52:54 <tswett> Strongly-Typed Data Language.
02:53:09 <tswett> Except call it something punchy. Not STDL.
02:53:44 <hppavilion[1]> tswett: The parser would reject the data completely if it doesn't match a strict, dependently-typed format and return None
02:54:47 <hppavilion[1]> tswett: But it would obviously have dependent types
02:55:04 <hppavilion[1]> tswett: ...leading to a second language which expresses the typing
02:56:24 <hppavilion[1]> tswett: What data structures would it support by default? I'm thinking Maps, Objects (named tuples), Lists, Tuples, and Tagged Unions
02:56:30 <zzo38> Yes how are you going to do the ring/group/etc?
02:56:44 <tswett> Every data structure that is possible to define.
02:57:03 <tswett> Well, that's *one* data structure.
02:57:41 <hppavilion[1]> With Int, Real, Char, Bool, Top, and Void being the builtin scalar data types
02:57:52 <tswett> How about... the list.
02:58:03 <tswett> Everything else is just kind of a fancy list.
02:58:10 <tswett> As Lisp fans can tell you.
02:58:31 <tswett> A tuple is just a list with a specific length whose elements have specific types.
02:59:08 <tswett> A map is just a list whose elements are 2-tuples whose first elements are never duplicated, and whose order doesn't matter.
02:59:37 <tswett> What's your notation here?
03:00:07 <tswett> If $ denotes the end of the list, what is ] for?
03:00:38 <hppavilion[1]> tswett: $ force the list to end at a specific point
03:01:15 <tswett> Have a sort of property-based typing. Like...
03:01:20 <zzo38> With RDF it is possible to represent a looping list as well as an incomplete list
03:01:23 <tswett> The most basic type is, say, "thing".
03:01:41 <tswett> But you can make more complicated types. Like "list". And "list of integers".
03:01:53 <tswett> "list whose first element is equal to its third element"
03:02:13 <tswett> "list of integers, where each integer is greater than the last"
03:03:11 <tswett> Damn it, what have I gotten myself into? :D
03:03:25 <hppavilion[1]> tswett: OK, so the builtin type is "bits", which means "big hunk-o-bits" and the builtin structure is "list of these types"
03:04:14 <hppavilion[1]> tswett: Ignore my previous 3 messages, the idea sucked
03:04:22 <tswett> So here's part of my idea.
03:04:54 <tswett> The basic type is, I guess, "thing".
03:05:09 <tswett> Now, if you have a "thing", then you don't know anything about it. And since you don't know anything about it, you can't do anything with it.
03:05:25 <hppavilion[1]> listOfIncreasingValues :: [Int] (a->0>a->1) forall a
03:06:17 <zzo38> Can't it be similar to a set of integers too?
03:06:19 <tswett> If you have a "list of things", now you know that it's a list, so you can do list stuff with it, like getting its length.
03:06:38 <tswett> And you can rearrange it, and stuff. But you can't really do anything with the individual things in it.
03:06:48 <tswett> And... that's how it all goes.
03:06:59 <tswett> By the way, let me see if I can whip something up.
03:07:22 <zzo38> But even in Haskell you can do stuff if you do not know the type of something, but the only way can be done is to match the same type variable such as if it is (x -> x) then you can use the same input as output type.
03:09:32 <tswett> data IncreasingList where { Empty :: IncreasingList; Cons :: (x :: Int) -> (xs :: IncreasingList) -> (case xs of { Empty -> (); (y:_) -> x < y }) -> IncreasingList }
03:09:51 <tswett> So... Haskell doesn't actually let you do that.
03:10:05 <tswett> But you can do something really similar in Coq.
03:10:18 <tswett> But don't ask me to; I don't remember the syntax.
03:10:35 <tswett> Anyway, the idea here is...
03:10:57 <tswett> "Increasing list of integers" can just be defined as a list of integers, which satisfies the predicate "is increasing".
03:11:36 <zzo38> What is if you have define a set and then you will make the sorted list from such set
03:12:11 <tswett> But an alternative is to sort of weave it into the definition of your list, so to speak.
03:12:41 <tswett> hppavilion[1]: see you later.
03:15:43 -!- hppavilion[1] has quit (Ping timeout: 252 seconds).
03:37:39 -!- XorSwap has quit (Quit: Leaving).
03:44:42 <zzo38> I have implemented Black-Johansen in JavaScript now
03:46:05 <HackEgo> [U+130B8 EGYPTIAN HIEROGLYPH D052] [U+0020 SPACE] [U+130B9 EGYPTIAN HIEROGLYPH D052A] [U+0020 SPACE] [U+130BA EGYPTIAN HIEROGLYPH D053]
03:46:41 <zzo38> Do you know Egyptian writing?
03:46:59 <zzo38> shachaf: An algorithm for optimizing text encoding in Z-machine
03:47:29 -!- tromp_ has joined.
03:47:47 <zzo38> (The algorithm is made up by myself and by oerjan)
03:52:01 -!- tromp_ has quit (Ping timeout: 252 seconds).
03:52:10 <zzo38> The text is encoded as a series of 5-bit Z-characters, involving three shift states. Zero always represents a space; 1 and 2 and 3 are "frequent words"; 4 and 5 set the shift state, and 6 to 31 are emit the character. The initial shift state is 0, where 4 means temporary shift to state 1 and 5 means temporary shift to state 2. In state 1, 4 permanent shifts to 1 and 5 permanent shifts to 0. In state 2, 5 permanent shifts to 0 and 4 permanent shift
03:52:48 <zzo38> Each character has its own preferred shift state, for which it is part of that character set. In addition, Z-character 6 of state 2 means the next two Z-characters are interpreted as an ASCII code and it then emits an arbitrary ASCII character.
03:54:03 <zzo38> Infocom designed this but then presumably they did not figure out how to encode text in an efficient way using this, so they just omitted the use of permanent shifts and used only temporary shifts in their compiler.
03:57:34 -!- lambda-11235 has quit (Read error: Connection reset by peer).
03:58:15 <zzo38> Shift state 0 contains lowercase letters, shift state 1 contains uppercase letters, while shift state 2 contains the ASCII escape, line break, followed by 0123456789.,!?_#'"/\-:()
03:58:48 -!- lleu has quit (Quit: That's what she said).
03:59:02 -!- bb010g has quit (Quit: Connection closed for inactivity).
03:59:46 <zzo38> Possibly by this you can understand what strings [4,13,10,17,17,20,5,19,0,4,28,20,23,17,9,5,20] and [5,5,9,10,11,12,13,6,2,8,14,15,16,17,8] represent.
04:03:13 <lambdabot> Sum of proper divisors minus the number of proper divisors of n: a(n) = sigm...
04:03:32 <zzo38> They aren't any mathematical sequences; I described the encoding above
04:07:09 <zzo38> The second example above may be a bit silly but a better example of a text where Black-Johansen will encode more efficiently would be "Copyright (c) 1981, 1982, 1983, 1984, 1985, 1986 Infocom, Inc. All rights reserved." Using a permanent shift would encode the numbers much more efficiently than temporary shifts
04:10:18 -!- hppavilion[1] has joined.
04:11:51 <hppavilion[1]> tswett: Here's a piece of software that needs to be invented
04:13:37 <hppavilion[1]> tswett: Also, Excel is a good example, but PowerPoint is a better one
04:13:56 <tswett> Excel can definitely be scripted.
04:14:16 <tswett> I'm pretty sure Word can be scripted.
04:14:39 <tswett> Well, you need to be able to use the scripting language.
04:15:01 <tswett> And it's not like it's a particularly bad language or anything.
04:15:31 <hppavilion[1]> tswett: I'm thinking something like "Select word -> Right click -> Make Button", which opens an "Events" window
04:16:21 <tswett> Lots and lots of people have put lots and lots of effort into creating a programming system that can be used by people who don't already know programming.
04:17:31 <hppavilion[1]> tswett: Because I kind of want to make a Free Software Online office suite
04:17:38 -!- lambda-11235 has joined.
04:18:10 -!- tromp_ has joined.
04:19:31 <tswett> Fsoos. Nah, that's probably not a good name.
04:19:36 <tswett> hppavilion[1]: hey, let me ask you something.
04:19:43 <tswett> Do you want to start a project and see it through to completion?
04:20:00 <hppavilion[1]> tswett: I even wrote a short essay on it for school at some point
04:20:28 <tswett> You'll get better at it.
04:21:40 <hppavilion[1]> tswett: I'll go with Zodiac Office, because "Zodiac" is one of my stock names
04:22:12 <tswett> What sort of project do you think you might want to do?
04:22:20 <hppavilion[1]> I'll give the word processor the working name "VomitWordsOntoScreen"
04:23:10 <tswett> Many of the ideas you've described are the sort of thing a person may spend ten years on.
04:23:27 <tswett> Do you have any ideas that are, say, weekend-sized?
04:25:16 <hppavilion[1]> (on the topic of What If): Especially the diagram of wing shapes
04:27:38 <hppavilion[1]> I'll try making a tiny calculator with a Qt-based GUI
04:29:34 <tswett> That sounds like a good idea.
04:29:46 <hppavilion[1]> Prefix notation, because it's easier to parse AND nobody knows how to use it
04:30:49 <hppavilion[1]> All integers must be expressed in Dozenal, because it is clearly the superior system
04:45:03 * Sgeo has successfully written something similar to Haskell's reflection library in RUst:
04:45:04 <Sgeo> https://gist.github.com/Sgeo/3d7f8449ea5ba136038a9c677e744004
04:51:31 <zzo38> We will need the ability to manipulate the TV guide of the cable box by the use of SQL codes.
04:56:22 -!- iconmaster has quit (Ping timeout: 252 seconds).
05:01:15 -!- rdococ_ has joined.
05:01:33 -!- rdococ has quit (Ping timeout: 240 seconds).
05:02:58 <HackEgo> [wiki] [[Pyth]] https://esolangs.org/w/index.php?diff=46761&oldid=46231 * 86.180.125.40 * (-2) /* Quine */
05:04:00 -!- tromp_ has quit (Remote host closed the connection).
05:17:00 -!- centrinia has quit (Quit: Leaving).
05:33:58 <hppavilion[1]> tswett: Assuming that all querying must be done in the language, and data produced cannot be filtered any more beyond that, but that you want a technically-usable DB
05:35:49 -!- centrinia has joined.
05:43:15 -!- centrinia has quit (Ping timeout: 246 seconds).
05:47:59 -!- XorSwap has joined.
05:59:04 -!- oerjan has quit (Quit: To sleep, perchance to dream of, hm, bears maybe?).
06:06:20 -!- tromp_ has joined.
06:29:00 -!- tromp_ has quit (Remote host closed the connection).
06:49:15 -!- MoALTz has joined.
06:55:15 <\oren\> the minimal database is usually implemented as "a folder full of CSV files"
06:55:28 <\oren\> or sometimes "full of JSON files"
06:56:31 <\oren\> in C, it is often "a file containing arbitrary fwritten structs"
06:59:11 <ais523> minimal database is also often implemented via using the filename to store the value of one field
06:59:18 <ais523> thus using the filesystem as a hash table
06:59:51 <ais523> but really, if you have nonrelational data, you can't do much better than a hash table for your database anyway
07:00:17 <ais523> it's relational queries that databases are very good at, that and situations where you can't fit the whole thing in memory or you need to handle power loss without data loss
07:02:59 <hppavilion[1]> [X-POST/r/name-of-this-subreddit] flag must be in title when reposting OC from this sub.
07:05:08 <HackEgo> [wiki] [[Element]] https://esolangs.org/w/index.php?diff=46762&oldid=46440 * Kc kennylau * (+49) /* Interpreter */
07:07:41 <\oren\> AARGH why doesn't ksp have a "dump fuel" button
07:08:23 <\oren\> I need to fly in circles a bunch before I can land this damn thinf
07:08:56 <\oren\> either that or be really really good at flying
07:13:12 -!- XorSwap has quit (Quit: Leaving).
07:29:36 -!- tromp_ has joined.
07:30:29 -!- infinitymaster has joined.
07:33:52 -!- tromp_ has quit (Ping timeout: 244 seconds).
07:42:22 -!- lambda-11235 has quit (Quit: Bye).
07:59:49 <hppavilion[1]> I've invented a new buzzword-that-sounds-good-but-is-actually-horribly-awful
08:02:16 <hppavilion[1]> (It's in the same category as "Self-modifying code", "Java Applet", and "think outside the tesseract")
08:03:57 -!- ais523 has quit (Ping timeout: 248 seconds).
08:05:15 -!- infinitymaster has quit (Remote host closed the connection).
08:05:49 -!- infinitymaster has joined.
08:05:53 <hppavilion[1]> Ugh. Forbes is evil because they block you if you have an adblocker.
08:07:47 <hppavilion[1]> I think that we should send some legalbabble to them referencing nonexistant legislation so they'll get rid of it
08:10:03 -!- infinitymaster has quit (Ping timeout: 240 seconds).
08:17:15 -!- infinitymaster has joined.
08:20:35 <hppavilion[1]> I'm going to (attempt to) read Transgressing the Boundaries: Towards a Transformative Hermeneutics of Quantum Gravity
08:22:03 -!- infinitymaster has quit (Ping timeout: 264 seconds).
08:35:38 -!- infinitymaster has joined.
08:40:01 -!- infinitymaster has quit (Ping timeout: 244 seconds).
09:30:11 -!- tromp_ has joined.
09:34:18 -!- AnotherTest has joined.
09:34:42 -!- tromp_ has quit (Ping timeout: 260 seconds).
09:35:48 -!- me2 has quit (Ping timeout: 244 seconds).
09:49:44 -!- rdococ_ has quit (Ping timeout: 244 seconds).
09:50:54 -!- rdococ has joined.
09:54:51 -!- ais523 has joined.
10:03:02 -!- me2 has joined.
10:25:16 -!- hppavilion[1] has quit (Ping timeout: 252 seconds).
10:30:52 -!- tromp_ has joined.
10:34:19 -!- Phantom_Hoover has joined.
10:35:10 -!- tromp_ has quit (Ping timeout: 252 seconds).
10:55:20 -!- S1 has joined.
11:00:14 -!- S1 has quit (Read error: Connection reset by peer).
11:39:13 -!- S1 has joined.
11:58:53 <Taneb> Is brainfuck's overflow behaviour undefined in the C sense? That is, can a compiler optimize a program on the assumption that the program never over or underflows?
12:02:23 <int-e> I think it's implementation-defined, if you want to follow the C standard's terminology.
12:03:43 <int-e> Many implementations just use unsigned char cells. So +[+] will terminate. https://esolangs.org/wiki/Brainfuck_constants distinguished between "wrapping" and "non-wrapping"...
12:03:53 -!- Reece` has joined.
12:05:55 <int-e> The closest we come to actual undefined behavior is the case of stepping off the tape to the left. (but the brainfuck interpreter on anagol has the source code immediately to the left of the data and many programs submitted there rely on that fact...)
12:07:41 -!- gremlins has joined.
12:07:47 <int-e> And I believe that there are some implementations that have a two-sided unbounded tape.
12:08:57 -!- Reece` has quit (Ping timeout: 246 seconds).
12:18:39 -!- Reece` has joined.
12:20:02 -!- gremlins has quit (Ping timeout: 268 seconds).
12:27:59 -!- ais523 has quit.
12:29:17 -!- Reece` has quit (Ping timeout: 268 seconds).
13:36:14 -!- lleu has joined.
14:19:09 -!- S1 has quit (Quit: S1).
14:21:36 -!- boily has joined.
14:28:40 <lambdabot> CYQB 091300Z 31008KT 290V350 30SM SCT038 BKN076 M04/M11 A2972 RMK SC3AC2 SLP071
14:28:45 -!- Reece` has joined.
14:43:33 -!- iconmaster has joined.
14:54:58 -!- Elronnd has quit (Quit: Let's jump!).
14:59:04 <Taneb> int-e: I generally write implementations with a two-sided unbounded tape
14:59:17 -!- Elronnd has joined.
15:01:06 -!- lleu has quit (Quit: That's what she said).
15:16:44 -!- iconmaster has quit (Quit: Leaving).
15:23:03 -!- centrinia has joined.
15:34:48 <FireFly> int-e: so anagol brainfuck is in principle self-modifying?
15:40:00 -!- lambda-11235 has joined.
15:41:11 -!- Kaynato has joined.
15:42:01 -!- gremlins has joined.
15:44:09 -!- Reece` has quit (Ping timeout: 268 seconds).
15:44:13 <boily> there are quines. there are quine chains. are there pseudo-quine-chains that gradually evolve into an interpreter for the pseudo-quine-chain?
15:51:27 -!- centrinia has quit (Ping timeout: 264 seconds).
15:54:00 -!- tromp_ has joined.
16:04:02 <int-e> FireFly: yes it is
16:28:35 -!- boily has quit (Quit: GUILD CHICKEN).
16:48:12 -!- earendel has joined.
17:31:03 -!- yorick___ has changed nick to yorick.
18:08:27 -!- centrinia has joined.
18:09:45 -!- tromp_ has quit (Remote host closed the connection).
18:32:00 -!- earendel has quit (Ping timeout: 276 seconds).
18:54:42 -!- carado has quit (Ping timeout: 268 seconds).
18:55:21 -!- carado has joined.
18:56:15 -!- acertain has joined.
19:10:16 -!- tromp_ has joined.
19:13:07 -!- Froox has quit (Quit: *bubbles away*).
19:14:26 -!- tromp_ has quit (Ping timeout: 244 seconds).
19:26:43 -!- bb010g has joined.
19:52:51 -!- centrinia has quit (Ping timeout: 248 seconds).
19:56:21 <fizzie> Heathrow's (post-security) restaurant had the silliest knives I've ever seen.
19:56:54 <fizzie> They looked more like spoons than knives, the "pointy" end was entirely round.
19:57:07 <zzo38> What kind of food do they have?
19:57:09 <fizzie> Like a flat, narrow spoon.
19:57:34 <zzo38> Depending the kind of food it might be suitable, but it might not be suitable because in general that are no good you should need a proper metal knife
19:58:01 <fizzie> Just... British pub stuff. Burgers. Chicken breast. I think they might have something steak-like, but I'm not sure.
19:58:16 <fizzie> Could be they have real knives for special purposes.
20:01:52 <zzo38> I have been in a few British pub in Victoria, BC
20:06:10 <shachaf> I've been to a butterfly garden in Victoria, BC.
20:09:49 <Taneb> I've been to two British pubs in the past two days
20:11:45 -!- tromp_ has joined.
20:13:00 -!- earendel has joined.
20:15:30 -!- rdococ has quit (Quit: gtg).
20:16:39 -!- tromp_ has quit (Ping timeout: 276 seconds).
20:18:28 <b_jonas> Wow. In M:tG, the last ability of Skullbriar, the Walking Grave looks REALLY strange to me.
20:29:03 -!- p34k has joined.
20:30:32 <zzo38> Yes it does seem like strange, but can be worked.
20:35:02 <b_jonas> zzo38: I wonder if it could matter for that rule whether counters of the same name are distinguishable, but it doesn't seem so.
20:35:53 <zzo38> I think they are not distinguishable just multiples of the same name or different name
20:43:25 <b_jonas> I wonder why they didn't just make the ability such that it keeps the counters in the graveyard and stack but not in exile.
20:44:24 <b_jonas> Or maybe so taht it keeps the counters in the gravyard and stack and command zone, but not in exile.
20:45:37 <zzo38> If it loses that ability then it will not keep the counters
20:47:34 -!- Sgeo has quit (Ping timeout: 240 seconds).
20:50:34 <b_jonas> What was the way again to put a card into a library as a consequence of mana payment during a spell, that is, not at trigger speed?
20:51:32 <b_jonas> Not just reorder your library that is, but put something to a library from another zone.
20:52:28 <b_jonas> Ah yes, Darksteel Colossus or similar abilities.
20:59:12 <b_jonas> If I play a Jayemdae Tome and during playing, pay for it by sacrificing a Darksteel Colossus on Ashnod's Altar, and get unlucky with shuffling, can Jayemdae Tome cause me to try to draw the Colossus, which then doesn't do anything because it can't move?
21:02:22 -!- hppavilion[1] has joined.
21:02:44 <hppavilion[1]> Have you guys heard that the US Congress is trying to ban citizens from using impossible-to-break encryption?
21:03:26 <zzo38> I have not heard of that, but would not be surprised of such thing
21:04:16 <hppavilion[1]> zzo38: I have a feeling it'll lead to a new field of cryptography
21:04:35 <hppavilion[1]> "Impossible-to-break, but doesn't see like it on the surface. Also, make a billion of these"
21:06:51 <Phantom_Hoover> in the uk it's already a crime to refuse to give an encryption key to the police
21:06:54 -!- acertain has quit (Ping timeout: 246 seconds).
21:07:26 <hppavilion[1]> Phantom_Hoover: What if you don't have the key because of the way things were set up?
21:07:48 <b_jonas> This guideline M:tG is using that any real-world mammal needs to have a narrow creature type approperiate for its species is crazy. They have tons of creature types for mammals that are used only on a few cards.
21:08:10 <Phantom_Hoover> but if you have some actual means of accessing the data then you're compelled to give that up
21:09:11 <b_jonas> Mole is the latest one. They can't just add a "rodent" type because there had been rats and since ancient times.
21:15:29 <hppavilion[1]> Is there a way to execute JS in-browser as a sandbox?
21:15:47 <hppavilion[1]> Essentially, to tell it "You may only use these functions as builtins"
21:16:12 <hppavilion[1]> Preferably (almost necessarily) where "these functions" can be a mix of builtins and custom functions?
21:17:02 <b_jonas> Ok, so now the rules don't specify any canonical way to determine which side of double-face card is the front face. This is horrible.
21:18:32 <b_jonas> And they're caught unaware by it too, just look at this: http://gatherer.wizards.com/Pages/Card/Details.aspx?multiverseid=410049 shows the back face on the left side, unlike for most double-faced cards. HAHAHAHAHAHAHAHAHHA HOIST BY THEIR OWN PETARD!
21:29:03 -!- bb010g has quit (Quit: Connection closed for inactivity).
21:31:50 <b_jonas> zzo38: if you try to download and save the Comprehensive rules from wizards.com , be VERY careful, because the url has the date typoed
21:32:05 <b_jonas> so after I downloaded, it didn't sort as if it were the latest version
21:33:03 -!- acertain has joined.
21:34:19 -!- Reece` has joined.
21:35:03 -!- gremlins has quit (Ping timeout: 240 seconds).
21:49:18 <zzo38> I am using the HTML version so don't need to download directly from Wizards of the Coast
21:50:59 <b_jonas> Correction. I was looking at the wrong version of the rules, because of this date thing, and the rules do specify which face is the front, by naming both styles of icons, but Westvale Abbey still seems buggy in Gatherer.
22:02:05 -!- AnotherTest has quit (Quit: ZNC - http://znc.in).
22:11:34 -!- tromp_ has joined.
22:16:15 -!- tromp_ has quit (Ping timeout: 276 seconds).
22:16:40 <zzo38> It is possible in Node.js to execute a sandbox context but it is necessary to be careful by enabling strict mode and removing the default prototype of the sandbox context object as well as doing other stuff
22:19:27 -!- Reece` has quit (Ping timeout: 244 seconds).
22:23:49 <zzo38> You should also define any function needed inside of the sandbox
22:24:12 <b_jonas> zzo38: and set memory limits and other stuff I guess
22:26:43 <zzo38> They recommend running untrusted code in a separate process, which would be needed in order to set memory limits and that stuff anyways
22:27:03 <zzo38> There is also vm.runInDebugContext although it does not document how the debug context is working
22:34:18 <zzo38> You can use Object.defineProperty in order to remove access to JavaScript built-ins, it looks like.
22:47:54 -!- oerjan has joined.
23:00:43 -!- p34k has quit.
23:08:08 <\oren\> if they ban strong encryption in the usa, then maybe the databases will move to canada or (trump forbid) mexico.
23:09:59 <HackEgo> 8ballreplies \ autowelcome_status \ awesome \ cat \ conscripts \ construct_grams.pl \ delvs-master \ dict-words \ esolangs.txt \ esolangs.txt.sorted \ hello \ hello2.c \ hello.c \ lua \ maze \ maze.c \ radio.php?out=inline&shuffle=1&limit=1&filter=*MitamineLab* \ UnicodeData.txt \ units.dat \ WordData
23:11:00 <oerjan> `` echo '<' | fueue src/brainfuck.fu
23:11:56 <HackEgo> http://codu.org/projects/hackbot/fshg/index.cgi/file/tip/bin/fueue
23:14:31 -!- MoALTz has quit (Quit: Leaving).
23:15:02 <oerjan> `` echo '<' | fueue $(cat src/brainfuck.fu)
23:15:04 <HackEgo> Error: fueue received too many arguments. The Hello world program \ Hello, world!
23:15:33 <int-e> `` echo '<' | fueue "$(cat src/brainfuck.fu)"
23:16:10 <oerjan> i'm just not sure if i remember how bin/fueue takes its program.
23:16:26 <int-e> `` echo '+[+.]' | fueue "$(cat src/brainfuck.fu)"
23:16:51 <shachaf> `` echo '+[+.]' | fueue "$(cat src/brainfuck.fu)"
23:17:00 <shachaf> ok, that was the source of the beep
23:17:10 <oerjan> `` echo '<+[+.]' | fueue "$(cat src/brainfuck.fu)"
23:17:13 -!- Sgeo has joined.
23:17:19 <shachaf> please don't \a in public twh
23:17:22 <int-e> oh it beeps too? that must be annoying :P
23:17:55 <oerjan> shachaf: please don't use shitty clients in public twh
23:17:57 <int-e> it may be messing up colors in irssi, too, now that I look at it
23:18:16 <shachaf> oerjan: what's wrong with a client that supports notifications
23:18:42 <oerjan> shachaf: i dunno but i hear no beeps hth
23:18:48 <int-e> by default irssi can be fooled to some extent with control sequences starting with ^D (\004)
23:18:58 -!- ^v has joined.
23:19:53 <oerjan> anyway, all this was just because i didn't remember what i did with brainfuck.fu going left. seems i made it double-ended.
23:21:26 -!- boily has joined.
23:21:33 <oerjan> i guess it was easier to just use two identical stacks for initial tape
23:22:45 <oerjan> <int-e> And I believe that there are some implementations that have a two-sided unbounded tape. <-- based on this
23:24:08 <oerjan> too bad ais523 isn't here.
23:25:42 <lambdabot> ENVA 092150Z 12007KT 9999 FEW040 01/M02 Q1016 RMK WIND 670FT 17003KT
23:25:43 <int-e> sshd[32006]: Connection closed by 5.103.65.96 [preauth] <--- fine, but what's the point of doing this once every minute?
23:26:18 -!- staffehn_ has quit (Read error: Connection reset by peer).
23:26:19 <oerjan> and here i nearly changed to my summer jacket the other day...
23:26:25 -!- staffehn has joined.
23:27:42 <boily> I don't think I ever vesperally porthelloed ais523. it'd make a fine one: bonsais523r.
23:27:47 <lambdabot> CYQB 092200Z 31011G20KT 30SM SCT044 M02/M11 A2988 RMK SC3 PRESRR SLP125
23:27:51 <int-e> Perhaps something is rotten in the state of Denmark. [IP is assigned to an DK ISP]
23:28:18 <oerjan> <boily> there are quines. there are quine chains. are there pseudo-quine-chains that gradually evolve into an interpreter for the pseudo-quine-chain? <-- wat
23:30:50 <boily> you know about recursive quine chains, where a program output source in another language that output source in another language, etc... and then you get the original source?
23:31:33 <boily> what about pseudo-quine-chains, which approximatively get you the original source after a full loop, that you loop and loop and loop up until you get an interpreter to bootstrap the original loop.
23:32:18 <int-e> hm, same thing on another server, two different IPs there... Kuala Lumpur, and the Netherlands
23:33:07 <boily> time to go play some Dominion...
23:33:17 -!- boily has quit (Quit: MASSIVE CHICKEN).
23:33:44 <int-e> (some botnet for sure, but it still seems odd to scan ports that often... that's what I suppose this is, I'm too lazy to do a packet dump)
23:35:04 -!- hppavilion[1] has quit (Ping timeout: 252 seconds).
23:50:53 -!- Lord_of_Life has quit (Excess Flood).
23:53:41 -!- Lord_of_Life has joined.
23:57:55 -!- centrinia has joined.