00:09:09 -!- hppavilion[1] has joined.
00:13:47 -!- tromp_ has quit (Remote host closed the connection).
00:16:10 <tswett> One could make a DSL specifically for Magic: the Gathering.
00:18:59 -!- tromp_ has joined.
00:37:40 -!- jaboja64 has quit (Ping timeout: 260 seconds).
00:40:16 -!- tromp_ has quit (Remote host closed the connection).
00:41:25 -!- triggerwarning has joined.
00:41:27 -!- dcentral has joined.
00:46:05 -!- triggerwarning has quit (Changing host).
00:46:05 -!- triggerwarning has joined.
00:57:30 -!- dcentral has quit (Ping timeout: 260 seconds).
00:57:30 -!- triggerwarning has quit (Ping timeout: 260 seconds).
00:57:52 -!- hppavilion[1] has quit (Ping timeout: 256 seconds).
01:04:37 -!- ais523 has joined.
01:12:42 <ais523> hmm, Firefox is disabling unsigned addons
01:12:56 <ais523> however, the addon in question is a private one that I wrote for my own use, and isn't signed for obvious reasons
01:19:04 -!- tromp_ has joined.
01:47:55 -!- ^v has joined.
02:01:09 -!- jaboja has joined.
02:08:33 -!- bb010g has quit (Quit: Connection closed for inactivity).
02:13:06 <tswett> Another NN-generated wisdom database entry:
02:13:07 <tswett> 04:41:54: <HackEgo> tapebacklogs are a Chu space of something you don't know if it is probably not because I don't know.
02:14:01 <tswett> An interesting exclamation:
02:14:02 <tswett> 04:54:10: <copumpkin> holy side of the right name
02:14:14 <tswett> "Holy side of the right name, Batman!"
02:14:15 <ais523> what was the corpus? just wisdom and quotes?
02:14:57 <tswett> No, the #esoteric logs from...
02:15:08 <tswett> Might have been from the beginning of 2014 up to the time that I downloaded the logs.
02:15:10 <ais523> oh good, logs are going to be much better for this sort of thing
02:15:27 <ais523> the people doing neural network M:tG cards figured out a way to seed the network
02:15:40 <ais523> i.e. you give it the start of a message, and it completes it for you
02:16:07 <tswett> I think that's been a feature of Karpathy's char-rnn from the very beginning.
02:16:23 <tswett> 15:55:53: <lambdabot> LOWI 051750Z AUTO 14001KT 10SM FEW080 SCT064 BKN030 95/18 A2960 RMK CU1 CI TR SLP267
02:20:25 <tswett> Okay. Date is the 5th, time is 17:50 UTC. Wind is from 140, at a speed of one knot. Visibility is ten miles. Few clouds at 8,000 feet above ground, scattered clouds at 6,400 feet, and broken clouds at 3,000 feet.
02:20:36 <tswett> The temperature is 95 degrees Celsius, holy shit.
02:21:34 <tswett> Dew point is 18 degrees Celsius.
02:24:38 <tswett> I can't find a meaning on Wikipedia for CU1.
02:25:36 <ais523> that's an incredibly arid climate
02:26:24 <tswett> Ah, it's apparently cloud coverage. Only one okta cloud coverage, despite the broken and other cloud layers.
02:27:32 <tswett> I can't find a meaning for CI or TR either.
02:27:46 <tswett> Sea-level pressure is 267 of something.
02:28:55 <lambdabot> CYYZ 200200Z 26012G18KT 15SM OVC045 M02/M09 A3026 RMK SC8 SLP256
02:29:05 <lambdabot> LOWI 200220Z AUTO 05003KT 350V110 4600 FZBR FEW001 M01/M01 Q1025
02:30:41 <tswett> Ah, let's see. Altimeter setting is 29.60 inches mercury. Sea level pressure is 1,026.7... hectopascals?
02:31:18 -!- Wright has joined.
02:35:42 <tswett> The altimeter setting is less than the sea level pressure. Does this mean that the airport is below sea level?
02:35:51 -!- tromp_ has quit (Remote host closed the connection).
02:37:56 <tswett> The neural net is usually pretty good with boily's quit messages. Here's an interesting exception.
02:38:01 <tswett> 13:47:38: -!- boily has quit (Quit: SCOULN DHICKEN CHICKEN).
02:38:31 <tswett> It created this one; I wonder if boily has ever actually used it.
02:38:32 <tswett> 13:19:35: -!- boily has quit (Quit: INTERCAL CHICKEN).
02:39:04 <tswett> 13:03:41: <ais523> yep, at least in the syntax how to allow a 2 syntax highlighting as a set of bits because it's already a fair people who believe that it's called "trivial ink"
02:39:18 <tswett> Yeah, there are really too many people who think it's called "trivial ink".
02:39:29 <tswett> We should teach them about how to allow a 2 syntax highlighting.
02:40:12 <mauris> i tried to parse that as a real sentence ;___;
02:40:51 <ais523> mauris: that's because my real sentences are normally possible to parse
02:44:23 -!- jaboja has quit (Ping timeout: 255 seconds).
02:46:16 -!- jaboja has joined.
02:47:44 -!- tromp_ has joined.
02:48:35 <zzo38> Do you know some things about PCRE?
02:48:56 <ais523> I know some things about PCRE, but not other things
02:50:37 <zzo38> Do you know exactly what UTF-8 text is acceptable if you turned off the checking? It says codes up to 31-bits but it doesn't say about if overlong encodings can be used and other stuff like that.
02:52:21 <ais523> that's one of the things I don't know, sadly
02:52:36 -!- Phantom_Hoover has quit (Read error: Connection reset by peer).
02:52:48 <shachaf> What are the things you do know?
02:53:17 -!- jaboja64 has joined.
02:53:23 -!- jaboja has quit (Ping timeout: 255 seconds).
02:53:33 <ais523> I know quite a few things about the regex syntax, although sometimes I get them muddled with "actual Perl" regexes
02:53:45 <ais523> have some vague memories of what the API is like but I can't remember the details
02:56:05 <zzo38> There is man pages for the API with C and with C++, and I am making the API with SQL, so I wanted to know some of these thing
02:58:04 <zzo38> For compatibility with other stuff in SQL, it exposes one-based offsets instead of zero-based offsets, and if the PCRE_UTF8 option is selected (or (*UTF8) in the pattern) then it will expose the UTF-8 character offsets instead of the byte offsets, too.
02:59:12 <zzo38> But I want to make it so that if you specify option to don't check valid UTF-8 then my program will do it by itself in the less strict but still acceptable way.
03:23:04 -!- jaboja64 has quit (Ping timeout: 272 seconds).
03:35:58 <FreeFull> ais523: I'm guessing you know what this matches
03:36:38 <ais523> FreeFull: given that you didn't anchor it (or are you doing that with API options), anything that's at least three characters long ;-)
03:36:47 <ais523> if it is anchored, anything that's between three and ten characters
03:36:52 <ais523> minus newlines, if you didn't turn on the option
03:36:55 <ais523> to make . match newlines as well
03:46:56 <ais523> I can't remember the relative precedence of ^ and $ compared to \
03:47:27 <ais523> is that intended to match all non-prime numbers in unary?
03:47:48 <ais523> I know it's possible to do something that works along those lines, that may well be it
03:50:27 -!- J_Arcane has joined.
03:54:50 -!- mauris has quit (Ping timeout: 260 seconds).
04:00:46 <zzo38> Using callouts you can make it to match only prime numbers (I have tried it).
04:04:26 -!- ^v^v has joined.
04:24:19 <FreeFull> zzo38: But then you're leaving the regex temporarily
04:27:59 <zzo38> Yes, although my own library provides a callout function that can be used for this (and other) purposes.
04:30:02 <zzo38> It might also be possible without callouts, due to some other features in PCRE.
04:36:19 <FreeFull> zzo38: PCRE almost certainly is powerful enough that you don't need callouts
04:37:55 <ais523> FreeFull: it's less than turing-complete without callouts
04:38:01 <ais523> so you might need them in some cases
04:38:20 <ais523> (IIRC it's a push-down automaton, which is much more powerful than regular expressions are but still sub-TC)
04:39:47 <FreeFull> That's interesting, PCRE allows recursive backreferences
04:39:57 <FreeFull> So (a|b\1)+ is a valid pattern
04:40:22 <ais523> err, not with that syntax I don't think
04:40:38 <FreeFull> ais523: I got that directly from the manpage
04:40:45 <ais523> err, actually that doesn't conflict with anything
04:40:50 <ais523> I was confusing it with something else
04:41:27 <FreeFull> ais523: And I was thinking you could use negative lookahead assertions to match primes
04:41:54 <ais523> anchoring a negative lookahead assertion is really hard
04:42:00 <ais523> I don't think it's impossible
04:42:16 <ais523> but I remember an incident a few months (years) back when #esoteric attempted to do it and collectively failed repeatedly for several hours
04:42:23 <FreeFull> Maybe lookbehind assertions too
04:44:35 <ais523> FreeFull: here was the problem (IIRC): you want to write a regex that matches any string that /doesn't/ contain a specific other string
04:44:44 <ais523> where the specific other string is hardcoded
04:45:12 <ais523> can you do it with the size of the resulting regex linear in the size of the string that you're (not) matching against?
04:45:21 <ais523> (where "regex" here is typically either Perl-regex or PCRE-regex)
04:45:30 <ais523> also you can't use callouts
04:46:06 <ais523> err, I was trying to eliminate the FSM solution with the "linear in the size" clause but that doesn't actually work
04:46:39 <ais523> size of the string you're matching against + a constant
04:47:15 <FireFly> Right, so ^(?!.*foo) is too slow. (I think that would work otherwise?)
04:48:08 <FireFly> It's not nice at all though
04:48:08 -!- Patashu has quit (Ping timeout: 255 seconds).
04:48:16 -!- Patashu has joined.
04:48:37 <ais523> FireFly: no, it doesn't work
04:51:08 <ais523> let me read the man page to make sure
04:53:09 <ais523> hmm, maybe I'm thinking of some other problem
04:53:12 <ais523> that does seem to work
04:53:15 <ais523> and shouldn't be slow?
04:54:18 <FireFly> I think the .*foo part isn't implemented "cleverly" as an actual FSA in PCRE/other 'powerful' regex engines, or at least used to not be
04:55:17 <FireFly> I don't remember why, but I think the leading .* would cause issues
04:55:18 <ais523> FireFly: wouldn't matter, you'd just do a forward search and then negate the result
04:55:29 <ais523> .*foo has the backtrack penalty
04:55:37 <ais523> I was thinking the ?! wouldn't cost, forgot the .* would
04:56:39 <FireFly> so how would PCRE execute ".*foo"?
04:57:17 <FireFly> In principle it should be possible to compile into an FSA I think?
04:57:19 <ais523> read the entire string, look for an f, if you don't find one go back one character and look for an f again, and so on
04:57:32 <ais523> at least that's how Perl does it
04:57:37 <ais523> PCRE has multiple algos
04:58:12 <FireFly> Having multiple algorithms depending on which features are used makes sense I suppose
04:58:27 <ais523> no, different APIs for them
04:58:33 <ais523> the default algo worked as I described
04:58:39 <ais523> it also has an NFA algo which is called DFA for some reason
04:59:12 <ais523> also it's apparently slower
04:59:16 <ais523> and produces outputs in a different format
04:59:33 -!- TellsTogo has joined.
04:59:41 <ais523> presumably that's "slower on regexes that are optimized for the other algo"
04:59:54 <FireFly> It might reduce the NFA to a DFA (and hence call it that)
05:00:12 <ais523> it says "this isn't really a DFA algorithm because" and then goes on to describe an NFA without using the term
05:00:33 <FireFly> So just poor naming then. Like 'qsort' in the standard library
05:01:40 <shachaf> I said that using an NFA or DFA to match a regular expression seems like an obvious case of dynamic programming to me, but someone disagreed.
05:01:49 <shachaf> So maybe it's not obvious.
05:02:22 <shachaf> But I still think it's as reasonable a use of the term as any that I've seen.
05:02:42 <FreeFull> Can't .*foo be matched like a reverse oof.* assuming you know the size of the input?
05:02:45 <ais523> I can't remember what "dynamic programming" means anyway
05:02:52 <ais523> FreeFull: not in a lookahead assertion
05:03:03 <shachaf> I think it means optimizing a tree to a graph, or something.
05:03:06 <ais523> because lookbehind assertions, at least with widely used regex-engines in 2015, can't be vairable length
05:04:42 <FireFly> To me 'dynamic programming' means "break down into potentially overlapping subproblems and don't recompute the same thing twice"
05:05:04 <FireFly> either by memoizing or by figuring out an order to compute things so that previous things are already computed
05:05:47 <ais523> clearly we need to create a #esoteric dynamic programming language based on the literal meanings of the words "dynamic" and "programming" when they're not in conjunction with each other
05:05:53 <ais523> but not the typical technical meanings
05:06:03 <ais523> which would just produce something like JS or Perl
05:06:06 <shachaf> What's the literal meaning of "programming"?
05:06:13 <ais523> instead, "dynamic" as in the managerese/marketingspeak term
05:06:13 <shachaf> Oh, you still mean in the sense of computer programming.
05:06:21 <ais523> and "programming" as in the sense of TV programming
05:06:41 -!- ^v^v has quit (Read error: Connection reset by peer).
05:06:49 <shachaf> programming in a dynamic, fast-moving environment
05:06:52 <ais523> i.e. producing a schedule that shows how modern you are
05:07:10 <ais523> then we advertise this as a "dynamic programming language" and make people do a triple-take
05:07:41 <FireFly> Clearly it's about flexible TV show scheduling
05:08:12 <shachaf> "dynamic programming" is a bad name but it's a reasonable concept
05:08:30 <shachaf> FireFly: Is what you said the same as what I said?
05:08:30 <FireFly> It's a terrible name. I have no idea why it was called that
05:09:10 <shachaf> Instead of an NFA, you can make an infinite tree corresponding to a regular expression.
05:09:32 <shachaf> And using that tree is like evaluating it with backtracking.
05:09:39 <FireFly> With one node for each substring?
05:09:56 <shachaf> Well, it's the tree you get by degraphing the graph.
05:10:35 <shachaf> If the graph is a DAG you get a finite tree, I guess.
05:11:07 <shachaf> Or another way of looking at it is that when you match with an NFA, you collapse distinct paths through the NFA if they lead to the same states.
05:12:08 <shachaf> Which is a dynamic programming thing where you share work between "subproblems" that lead to the same state.
05:12:21 <shachaf> But you lose things like backreferences, which depend on the actual path taken, not just the state.
05:12:42 <shachaf> (A tree has no problem supporting backreferences because a state and a path are the same thing.)
05:13:07 <shachaf> I guess that's one definition of a tree -- a graph where there's a unique path from the root to each node.
05:14:52 <shachaf> I wonder whether you can talk about the free/cofree tree on a graph (or a DAG).
05:16:01 <ais523> I don't think there is a unique free tree, is there?
05:16:37 <ais523> the definition of "free" is basically "any two things are distinct unless you can prove them equal using the axioms"
05:17:04 <shachaf> Well, that's the definition of free algebraic structures on a set in universal algerba.
05:17:16 <shachaf> I'm not sure how well it extends to things between other categories.
05:21:44 <shachaf> "F(G) is a free tree on G" would mean something like "for any tree T, tree homomorphisms : F(G) -> T are in natural correspondence with graph homomorphisms : G -> T"
05:34:38 -!- TellsTogo has quit (Quit: Page closed).
05:39:00 -!- TellsTogo has joined.
05:52:35 -!- aloril has quit (Ping timeout: 240 seconds).
06:05:14 -!- aloril has joined.
06:15:35 -!- ^v has quit (Ping timeout: 246 seconds).
06:18:33 -!- interest1ng has quit (Quit: Leaving...).
06:18:49 -!- hppavilion[1] has joined.
06:39:00 -!- hppavilion[1] has quit (Ping timeout: 256 seconds).
06:40:13 <zzo38> Now you can see my PCRE wrapper library and you can tell me the comment/question/suggestion/complaint/whatever. It is in the ZIP archive: http://zzo38computer.org/sql/sqlext.zip
06:42:53 -!- Sprocklem has joined.
06:47:37 -!- andrew_ has joined.
06:57:33 -!- Wright has quit (Remote host closed the connection).
07:15:00 -!- oerjan has joined.
07:28:38 <lambdabot> ENVA 200720Z 26007KT CAVOK 05/05 Q0990 NOSIG RMK WIND 670FT 11003KT
07:28:42 <lambdabot> KOAK 200653Z 32005KT 10SM CLR 08/03 A3013 RMK AO2 SLP201 T00830033
07:34:38 -!- tromp_ has quit (Remote host closed the connection).
07:49:44 <HackEgo> Treants are genericized ents for intellectual property reasons
07:49:57 <oerjan> `` sed -i 's/$/./' wisdom/treant
07:50:55 <oerjan> `learn Termites are genericized ants for intellectual property reasons.
07:50:57 <HackEgo> Learned 'termite': Termites are genericized ants for intellectual property reasons.
07:53:07 <HackEgo> Marmite is a group mind of fungal microorganisms spreading throughout the supermarkets of the Commonwealth.
07:53:47 <oerjan> `learn Vegemite is genericized marmite for intellectual property reasons.
07:53:49 <HackEgo> Learned 'vegemite': Vegemite is genericized marmite for intellectual property reasons.
07:57:16 <HackEgo> Ants are great architects. They are famous for their highways.
08:19:57 -!- TellsTogo has quit (Ping timeout: 252 seconds).
08:24:31 <HackEgo> wisdom/coulor \ wisdom/ghoul
08:24:41 <HackEgo> Coulor is the correct spelling.
08:28:10 <shachaf> `` rgrep -il 'correct spelling' wisdom
08:28:12 <HackEgo> wisdom/coulor \ wisdom/nooodl \ wisdom/mauris
08:28:50 <HackEgo> maur is the correct spelling
08:29:04 <ais523> OK, that's kind-of clever on two levels
08:32:05 <oerjan> `` rm wisdom/{elronnd,b_jonas}
08:32:07 <HackEgo> rm: cannot remove `wisdom/b_jonas': No such file or directory
08:35:08 -!- tromp_ has joined.
08:39:55 -!- tromp_ has quit (Ping timeout: 265 seconds).
08:55:23 -!- Patashu has quit (Ping timeout: 265 seconds).
08:56:10 -!- Patashu has joined.
09:04:57 -!- TellsTogo has joined.
09:20:55 -!- Sprocklem has quit (Ping timeout: 260 seconds).
09:37:14 <fizzie> Hm. I need to come up with a cover photo for the thesis book.
09:37:27 <fizzie> (Technically the image is optional, but everyone seems to have had one.)
09:38:09 <myname> does it have to be serious?
09:42:50 <fizzie> It should probably at least appear serious to a casual observer.
09:43:22 <ais523> fizzie: what sort of "thesis book" is this?
09:43:32 <fizzie> ais523: The printed version of my doctoral thesis.
09:44:01 <ais523> at least at this university, all printed doctoral theses look identical apart from the date, author name, and number of pages
09:44:10 <ais523> the binding is plain and consistent for all theses, going back years in some cases
09:44:21 <ais523> (although it changes when they go to a new bookbinding company as they don't all have identical equipment)
09:45:13 <fizzie> At Aalto, they have a "series" for them, with some unifying graphical elements (the cover is one out of 7 fixed colors), but there's a special place for a cover photo, which will be enclosed in one out of three shapes (a circle, a cross or an octagon).
09:45:37 <fizzie> So they all look kind of similar, but the contents of the cover image shape are up to the author.
09:46:56 <fizzie> http://zem.fi/tmp/diss_vaihtoehdot.pdf
09:47:12 <fizzie> (The light blue color has been discontinued.)
09:47:28 <fizzie> myname: Noise robust speech recognition, more or less.
09:48:00 <fizzie> When the university was still the Helsinki University of Technology, I think they had un-customizable covers. But they made this new "visual identity" as part of the merger into Aalto University.
09:49:44 <ais523> that cross-shaped photo cutout looks awful
09:50:18 <ais523> somehow I feel that noisy speech deserves a photo of an oscilloscope display showing something meaningful
09:51:12 <fizzie> Well, I mean. There's been several where the cover image has incorporated a spectrogram of something, it's kind of overly obvious.
09:52:00 <fizzie> Mine deals with reverberation, I was thinking of running a 2D sound wave simulation thing (with the cutout shape as walls) and then doing some sort of a visualization on it.
09:52:21 <fizzie> But I don't know if it'd look any good, and there's a nonzero amount of effort to do that.
09:52:34 <ais523> isn't the thesis a considerably nonzero amount of effort anyway?
09:52:51 <ais523> getting my thesis bound took about two or three days of effort, mostly chasing up bookbinders and couriers
09:53:19 <fizzie> Well, nonzero amount of purely optional effort.
09:53:22 <fizzie> The rest is kind of mandatory.
09:53:57 <ais523> there was still a mistake in the binding but a harmless one (we have an author declaration form which we sign to say it isn't plagiarised; it was only meant to be in one physical copy, but they bound it into both, and this is despite me contacting them via email to ask them how to request it to only be placed in one copy, and then making the request in the way they suggested)
09:55:40 <fizzie> I was reading the publication instructions today (to see what they say about the covers), and apparently the printing house is the one who compiles the electronic version the library's going to use, too. I'm somewhat afraid this will wipe out all the hyperlinks and the TOC metadata from the PDF.
09:57:32 <ais523> that was quite different from me, they asked me to PDFify it myself
09:57:38 <ais523> also gave links for an online Word-to-PDF conversion service
09:57:59 <ais523> I had hyperlinks, TOC metadata, and also had to take some font encoding care to make sure it'd copy-paste properly
09:58:05 <ais523> otherwise the plagiarism detector would have had a fit
10:01:09 <fizzie> Heh. Well, we've got this "publication platform" where you fill in the changeable parts (like titles, abstracts, and the cover stuff) and provide a contents PDF, and then you can turn that into an order for Unigrafia, a printing house owned by our university, University of Helsinki, and two other nearby ones.
10:02:36 <fizzie> (Well, owned by a holding company owned by our university, anyway.)
10:06:47 <fizzie> I don't yet know if the library's going to publish the e-copy with articles, either. For the ones with copyright transfers, I've technically got permissions for an electronic version as well, but some of them had conditions like "the download page must prominently display a note saying this and that".
10:06:48 -!- andrew_ has quit (Remote host closed the connection).
10:08:54 -!- andrew_ has joined.
10:09:56 <ais523> I was quite surprised with the copyright status on the thesis (they said I keep the copyright, and asked for a very limited license to be able to hold it in the library and host the online version)
10:10:12 <ais523> not that I'm complaining
10:10:52 <fizzie> I think that's the case for the part of the thesis I wrote exclusively for the thesis, too. But it's a compilation one, so the articles have been published in conferences and journals.
10:12:18 <ais523> those can be a pain copyrightwise
10:12:42 <fizzie> One of the journal articles was to a very open-access one, though -- they only wanted a CC-BY license for the article, with no copyright transfer, which was nice.
10:15:32 <fizzie> I remember hearing from the gone-before that IEEE used to require an explicit license agreement, but that seems to no longer be the case. "The IEEE does not require individuals working on a thesis to obtain a formal reuse license, however, you may print out this statement to be used as a permission grant."
10:15:46 <fizzie> They've got one of those e-publication conditions, though.
10:17:47 <fizzie> ("please display the following message in a prominent place on the website" + a non-endorsement note and a link how to obtain a license for further republication.)
10:18:13 <fizzie> A lot of the compilation dissertations on our library page have just been published without the articles.
10:18:18 <fizzie> Presumably for this sort of reasons.
10:18:19 <ais523> fizzie: well cc-by is a very permissive license
10:18:30 <ais523> so asking someone to license under it is quite a big request
10:19:14 -!- ais523 has quit (Quit: lunch).
10:19:38 <fizzie> Perhaps "if we're not making any money out of this, neither are you" is what they were thinking.
10:20:15 -!- oerjan has quit (Quit: leaving).
10:21:32 <Taneb> The disadvantage of using ssh and a chromebook to do most of my programming on is that I don't know any server I can access with OpenCL or Cuda or something similar on it
10:22:15 <Taneb> I mean, there's a nightclub in York called Kuda but that's not the same thing
10:23:20 <fizzie> Doesn't your university have a thing?
10:24:16 <Taneb> None I know of with a GPU
10:25:14 <fizzie> We had a couple of GPU nodes on the shared cluster you could reserve for serious use, and for fiddling around they had (not very powerful, but better than nothing) Nvidia Quadro cards in the Linux workstations, so you could use one that nobody was using.
10:26:53 <fizzie> Can you remotely get into a computer with a graphics card in a classroom, or something like that?
10:28:09 <Taneb> Yeah, I can, but they don't have the libs installed as far as I can tell
10:29:19 <Taneb> I also don't know which are turned on and booted into linux
10:31:58 <fizzie> Aw. I think ours had the CUDA packages installed.
10:32:41 <fizzie> (Also they were all permanently in one operating system, though of course sometimes just randomly off.)
10:35:17 <Taneb> All of our lab PCs dual-boot between Ubuntu 14.04 and Windows 7
10:35:25 <Taneb> Some can also boot into Windows 10
10:35:36 -!- tromp_ has joined.
10:37:30 <fizzie> Wonder why we had separate rooms for Linux and Windows machines. Maybe a license thing.
10:37:53 <fizzie> (I have no idea how Windows volume licensing works, and whether there's a per-machine cost component.)
10:40:24 -!- tromp_ has quit (Ping timeout: 256 seconds).
10:45:57 <Taneb> The university probably has upwards of 1000 windows machines, a few more that also can boot into Linux I don't think would hurt
10:54:39 -!- ais523 has joined.
10:57:48 -!- TellsTogo has quit (Ping timeout: 252 seconds).
11:14:48 -!- TellsTogo has joined.
11:19:48 -!- TellsTogo has quit (Ping timeout: 252 seconds).
11:33:33 <HackEgo> [wiki] [[Deadfish]] http://esolangs.org/w/index.php?diff=45922&oldid=45768 * YourDeathIsComing * (-54)
11:36:32 -!- tromp_ has joined.
11:40:55 -!- tromp_ has quit (Ping timeout: 260 seconds).
11:41:24 -!- TellsTogo has joined.
11:41:27 -!- TellsTogo has quit (Client Quit).
11:49:55 -!- Phantom_Hoover has joined.
11:51:26 -!- TellsTogo has joined.
12:09:56 <HackEgo> [U+0142 LATIN SMALL LETTER L WITH STROKE]
12:13:54 -!- Patashu has quit (Ping timeout: 256 seconds).
12:21:47 -!- MoALTz has quit (Quit: bbl).
12:30:18 -!- MoALTz has joined.
12:51:55 -!- andrew_ has quit (Ping timeout: 240 seconds).
12:57:01 -!- andrew_ has joined.
12:58:27 -!- mauris has joined.
12:59:03 -!- Welo has joined.
13:12:39 -!- Welo has quit (Quit: Leaving).
13:18:45 <HackEgo> [wiki] [[Special:Log/newusers]] create * Qwerty12302 * New user account
13:37:29 -!- tromp_ has joined.
13:41:35 -!- tromp_ has quit (Ping timeout: 240 seconds).
13:48:50 -!- andrew_ has quit (Ping timeout: 246 seconds).
13:50:58 <HackEgo> [wiki] [[Gtltem]] N http://esolangs.org/w/index.php?oldid=45923 * Qwerty12302 * (+1688) Created page with "'''gtltem''' (greater than, less than, exclamation mark) is an esoteric language created by username qwerty12302 in 2015. The language consists only of three characters: <, > ..."
13:52:12 -!- TellsGoto has joined.
13:55:08 -!- TellsTogo has quit (Ping timeout: 246 seconds).
14:03:53 -!- sebbu2 has quit (Ping timeout: 246 seconds).
14:04:16 -!- sebbu has joined.
14:15:55 -!- sebbu has quit (Ping timeout: 240 seconds).
14:19:03 -!- sebbu has joined.
14:42:52 -!- tjt263 has joined.
15:23:38 -!- tromp_ has joined.
15:27:53 -!- tromp_ has quit (Ping timeout: 246 seconds).
15:33:55 -!- sat-buddhi has joined.
15:37:16 -!- TieSoul has joined.
15:47:40 -!- Phantom_Hoover has quit (Ping timeout: 260 seconds).
15:49:26 -!- sat-buddhi has quit (Quit: Saliendo).
15:58:38 -!- tjt263 has left ("part").
16:29:05 -!- mauris has quit (Ping timeout: 260 seconds).
16:32:16 -!- tromp_ has joined.
17:35:16 -!- jaboja has joined.
17:39:40 -!- tromp_ has quit (Remote host closed the connection).
17:42:41 -!- Lord_of_- has quit (Excess Flood).
17:45:36 -!- Lord_of_- has joined.
17:52:21 -!- tromp_ has joined.
17:53:09 -!- shikhin has changed nick to rdy.
17:53:23 -!- rdy has changed nick to shikhin.
18:13:02 -!- boily has joined.
18:20:12 <lambdabot> CYQB 201800Z 26010G17KT 25SM FEW030 SCT048 BKN150 M05/M12 A3031 RMK SC1SC2AC2 SHWR NE SLP272
18:21:41 <zzo38> One reason to have separate room for Windows and Linux might just be to make it clear where they are.
18:22:26 <zzo38> (If the computers can dual-boot then that wouldn't be necessary for that purpose, but is still possible)
18:22:55 <boily> hezzo38. partitioning your house so multiple OSes can cohabit?
18:26:38 <ais523> we do that at work, separate rooms for linux and windows machines
18:27:08 <ais523> but that's mostly so a) you know where to go to find a particular OS, b) if a lot of people need to use the same room for some reason they probably also need to use the same OS
18:44:06 -!- Phantom_Hoover has joined.
18:51:16 <zzo38> Yes those are the reasons I have meant.
18:51:28 -!- Sprocklem has joined.
18:51:30 <zzo38> I don't meant your house, I did mean in an office or at school.
18:52:31 -!- boily has quit (Quit: STRAIGHTENED CHICKEN).
18:56:51 -!- Welo has joined.
19:12:46 -!- copumpkin has joined.
19:14:47 <\oren\> I added 奨奪奮姓姻娘娠娯婚婦婿媒媛嫁嫉嫡嬢寂寄密寡寧審寮 אּבּגּדּהּוּזּטּיּךּכּלּמּשׁשׂשּׁשּׂאַאָנּסּףּפּצּקּרּשּתּוֹבֿכֿפֿﭏיִﬞײַﬠ ꝊꝋꝌꝍꝎꝏꝐꝑꝒꝓꝔꝕ ꙀꙁꙂꙃꙄꙅꙆꙇꙈꙉ
19:15:49 <tswett> I like how the neural net seems to have figured out that fungot responds to its name.
19:15:49 <fungot> tswett: and for the times ahead) and the ability to constrain matching. when you say foo for a in m:
19:16:27 <lambdabot> KGRR 201853Z 21019G24KT 10SM BKN250 07/M04 A3003 RMK AO2 PK WND 20026/1843 SLP178 T00721044
19:19:44 <tswett> 13:00:31: <mroman> fungot: Who just zero links to it?
19:19:44 <tswett> 13:16:44: <fungot> mroman: and i find a nice shit in a map to call the [ (as in Haskell.)
19:19:44 <fungot> tswett: i was looking at
19:19:44 <fungot> tswett: 3 would be nice. i think
19:22:36 -!- Welo has quit (Quit: Leaving).
19:25:47 <\oren\> some of these characters, specifically , don't officially exist. But they can be inferred from the characters around them.
19:26:06 <lambdabot> ESSB 201920Z 22012KT 9999 BKN026 12/07 Q1000 R30/19//70
19:27:36 <FireFly> tswett: you heard him. Serve three lines, please
19:27:41 <HackEgo> U+FB44 HEBREW LETTER PE WITH DAGESH \ UTF-8: ef ad 84 UTF-16BE: fb44 Decimal: פּ \ פּ \ Category: Lo (Letter, Other) \ Bidi: R (Right-to-Left) \ Decomposition: 05E4 05BC \ \ U+FB45 - No such unicode character name in database \ UTF-8: ef ad 85 UTF-16BE: fb45 Decimal: ﭅ \ () \ Uppercase: U+FB45 \ Category: Cn (Other, Not A
19:28:01 <FireFly> fungot: what is your preferred pronoun?
19:28:01 <fungot> FireFly: choke and die at some point and dubbed it newio. :s :s :s :s :s
19:30:49 <tswett> fungot: I don't think FireFly's question warrants such a response.
19:30:49 <fungot> tswett: it's as problematic as extreme conciseness " too short" is quite a understatement. fnord
19:31:45 <zzo38> Unicode property is terrible that what you need it to do can depend on what you are using with private use areas and with other stuff; such thing should be either in the font metrics or in the locale data depending on what kind of properties they are.
19:32:07 <tswett> fungot: so, what *is* your preferred pronoun?
19:32:07 <fungot> tswett: it's on-topic, because i'm not on sisc :p
19:32:25 <tswett> fungot: are you saying your preferred pronoun is "it"?
19:32:25 <fungot> tswett: might need a special form, rather than call itself directly.
19:32:39 <tswett> fungot: a special form, you say? Which one did you have in mind?
19:32:40 <fungot> tswett: but the real wtf is that. but you should be upgraded to di
19:33:08 <tswett> fungot: I should upgrade the pronoun I use for you to "di"? I can do that.
19:33:46 <FireFly> I got the impression that fungot would prefer a longer pronoun given their aversion to extreme consiseness above
19:33:46 <fungot> FireFly: what is the reason i've been fnord
19:33:59 <FireFly> oh. 'fnord'. that would explain a thing or five
19:34:11 <FireFly> fungot: so "fnord" it is, then?
19:34:11 <fungot> FireFly: if i do anything, but it doesn't expose them
19:35:02 <ais523> FireFly: I got the impression that the pronoun was extremely short and problematic for that reason
19:35:08 <ais523> the only such pronoun that really fits is the null string
19:35:40 <FireFly> That is a nice interpretation, let's go with that
19:37:31 <tswett> I think that di has made dir desires pretty clear here.
19:37:50 <tswett> Let's give dim what di wants. The choice is dis.
19:48:58 <FireFly> does fungot have any more wisdom for us today?
19:48:58 <fungot> FireFly: now this service won't even stop :p)
19:49:06 <\oren\> @tell b_jonas I added a sample in Serbian to be font demo, which uses ј
19:49:10 <HackEgo> mnoqy/mnoqy used to be monqy before the earthquake.
19:51:28 <HackEgo> 305) <cpressey> `quote django <HackEgo> 352) <olsner> django is named after a person? <olsner> thought it would be a giraffe or something <cpressey> thankfully only one
19:53:16 <HackEgo> 6) <oerjan> what, you mean that wasn't your real name? <Warrigal> Gosh, I guess it is. I never realized that. \ 13) <fizzie after embedding some of his department research into fungot> Finally I have found some actually useful purpose for it. \ 20) IN EINEM ALTERNATIVEN UNIVERSUM (WO DIE NAZIS WON): <ehird> So kann ich nur schliessen, dass es fals
19:54:20 <HackEgo> 20) IN EINEM ALTERNATIVEN UNIVERSUM (WO DIE NAZIS WON): <ehird> So kann ich nur schliessen, dass es falsch ist, oder die Welt ist vollig BONKERS. Gegrusset seist du der Fuhrer Hitler!
19:55:12 <HackEgo> 42) <GregorR> ??? <GregorR> Are the cocks actually just implanted dildos? <GregorR> Or are there monster dildos and cocks? <GregorR> Or are both the dildos and cocks monster?
19:55:17 <zzo38> I am not very good at German writing
19:56:40 <HackEgo> 600) <fungot> elliott: young john soon afterward receives as a visitor a fnord spaniard, fnord de moncada, who has escaped from fnord fnord dissolved in the absolute.
19:56:55 <HackEgo> 73) <Warrigal> Making a small shrine to Lawlabee in my basement is something I should get around to at some point.
19:57:03 <HackEgo> 583) <kmc> COCKS [...] <kmc> truly cocks
19:57:09 <HackEgo> 499) <fungot> elliott_: it's a machine that looks like you!
19:57:15 <HackEgo> 65) <Warrigal> Ah, vulva. <Warrigal> What is that, anyway?
19:57:17 <ais523> ooh, a good fungot quote
19:57:17 <fungot> ais523: i couldn't gotten all this information in the format
19:57:26 <HackEgo> 929) <Bike> if you say "java" three times to a dark mirror does steele come out and lecture you about operator overloading
19:57:33 <HackEgo> 1112) <Sgeo> NihilistDandy: HOLY FUCK THAT'S STILL ALIVE? <Sgeo> Maybe I should look into it
19:57:37 <int-e> `` grep -i cocks quotes | wc -l
19:57:54 <HackEgo> 1089) <zzo38> Do you like this kind of melody? l8 e4gc'fac'g ^bd'fgfed e4gc'fac'g ^1
19:58:20 <HackEgo> 431) <oerjan> i try to be a hermit but it's hard with all these housemates.
19:58:27 <HackEgo> 1214) <boily> I got my girlfriend through the previous job's intranet IRC channel, so I don't see how the two can conflict...
19:58:43 <HackEgo> 744) <monqy> moral of the story with enough peer pressure nything is possible
19:58:45 <zzo38> Do you like music?
19:58:59 <HackEgo> 854) <GreyKnight> also if a GNU tar feature doesn't make you feel superior to others then you should file a bug report, it was probably unintentional
19:59:14 <HackEgo> 833) <shachaf> Do physicists have half-life crises?
19:59:21 <HackEgo> 1013) <Fiora> shachaf: make friends. help people. find ways to help people be happy. hug people. have fun. make the world a little bit better.
20:04:32 <HackEgo> 120) <AnMaster> cpressey, oh go to zzo's website. He is NIH <Phantom_Hoover> AnMaster, really? I was strongly under the impression that zzo was invented here.
20:04:59 <HackEgo> 48) <Madelon> both of you, quit it with the f-bombs. <Madelon> kaelis: what's the matter? something censoring stuff you're interested in?
20:05:16 <HackEgo> 270) <oklopol> and then there's the slightly annoying one where suddenly, i start rolling forward and i can't stop <oklopol> like i can be having some great sex dream or whatever and then suddenly "oh god not this again" <oklopol> (i go "not this again" but not necessarily realize it's a dream)
20:05:34 <HackEgo> 1226) <oren> when i was a kid it used to snow on christmas eve. what is this "freezing rain", "sleet" crap? <vanila> yeah seriously, who is evn in charge anymore? <oren> apparently not santa claus <zzo38> Santa Claus is dead by now. \ 1227) <{\[oren]|}> ｚｚｏ３８：ｉｔ ｗｉｌｌ ｃａｕｓｅ ｐｒｏｂｌｅｍｓ ｂｙ
20:05:37 <Phantom_Hoover> i don't like 120 because it just reads like i'm being a snarky asshole about the phrasing
20:06:02 <Phantom_Hoover> whereas the actual joke was that zzo is so weird that you sometimes wonder if he actually is a chatbot written in malbolge or twoducks or something
20:06:37 <HackEgo> 1227) <{\[oren]|}> ｚｚｏ３８：ｉｔ ｗｉｌｌ ｃａｕｓｅ ｐｒｏｂｌｅｍｓ ｂｙ ｂｅｉｎｇ ｈｉｌａｒｉｏｕｓ
20:07:12 <HackEgo> 666) <NihilistDandy> Benchmarks are only a good measure of surprise
20:07:40 -!- jaboja has quit (Remote host closed the connection).
20:07:47 <ais523> Phantom_Hoover: sometimes I'm reading something on a random site I don't normally visit on the website, and suddenly realise it was almost certainly written by zzo38
20:08:08 <ais523> it's rare that you can distinguish a particular person out of the entire world's population by something they wrote, but some people have very distinctive styles
20:09:10 -!- Patashu has joined.
20:10:08 <ais523> zzo38: no, never quite sure
20:10:11 <ais523> but it feels very likely
20:10:23 <ais523> most recently it was on the c2 wiki
20:10:33 <ais523> have you posted there in the past?
20:10:38 <ais523> if not then it must have been someone else
20:11:05 <zzo38> I have posted there in the past yes, but now I cannot post on there anymore
20:11:47 <zzo38> (I wrote some things about music macro language and so on)
20:15:31 <b_jonas> https://gamesdonequick.com/schedule
20:15:37 <ais523> that's been up for a while
20:15:45 <ais523> although it gets more precise over time
20:15:51 <b_jonas> it can't be older than a few days
20:15:52 <ais523> it was on a temporary page for a while I think
20:16:18 <b_jonas> let me see what goodness they will have this year
20:18:43 -!- ais523 has quit (Quit: food; might or might not be back on later).
20:19:39 <b_jonas> Oh I so wish they started to include the name of the console or system in that table, for the benefit of people like me who are less into speedrunning.
20:19:46 -!- jaboja has joined.
20:22:27 <FireFly> Are you only interested in certain systems?
20:22:50 <b_jonas> but it would make it easier to identify the games than from just names
20:23:19 <FireFly> Especially for series that have both an old and more recent game with the same name, such as Ninja Gaiden
20:23:27 <b_jonas> And some of the games are even ambiguous, because they have speedrunnable ports to multiple systems, which are sometimes different.
20:24:33 <b_jonas> Crypt of the Necrodancer is that popular rhythm game that pretends to be a roguelike, right?
20:25:31 <FireFly> Something like that I think
20:25:48 <FireFly> I hope the pokemon glitch exhibition incentive gets met
20:26:24 <b_jonas> “Prince of Persia” played by CapnClever. There are multiple different ports of it, though only one that the speedrunners play, the one with lots of extra levels; plus there's another later game with the same name but that one is crap
20:26:35 <b_jonas> I'll have to watch this one
20:28:02 <b_jonas> oh! an Age of Empires speedrun. great!
20:28:11 <b_jonas> and Grand Theft Auto 3 too
20:29:31 <b_jonas> “Legend of Zelda: Ocarina of Time Master Quest, The” - how is that different compared to just Legend of Zelda: Ocarina of Time?
20:29:55 <b_jonas> plus there's some classic games
20:31:21 <b_jonas> what's the name of that platformer with very simple graphics where gravity reverses every time the character lands?
20:32:25 <b_jonas> Oh, Legend of Zelda: Majora's Mask 4p co-op! that will be interesting
20:32:54 <b_jonas> There's a TASBot block too
20:33:13 <b_jonas> Super Metroid reverse block order? nice
20:34:08 <b_jonas> Ah nice! Mike Tyson's Punch-Out blindfolded! They did that before and it's always great
20:34:35 <b_jonas> Oh, Age of Empires category is "IL Showcase". what does that mean?
20:35:16 <FireFly> I think IL is individual level(s)
20:35:32 <FireFly> so I guess speedrunning individual campaign maps or something
20:36:37 <FireFly> OoT MQ have slightly harder bosses than regular OoT I think, or the dungeons are slightly harder in some way
20:36:48 <FireFly> and Hyrule is mirrored, because supposedly that makes it harder
20:37:42 <b_jonas> oh, Indiana Jones and the Temple of Doom too
20:38:37 <b_jonas> Mario Kart 64 as bonus incentive
20:38:54 <FireFly> I hope some of the awful games will be amusing
20:39:09 <FireFly> Roundabout cutscene% last SGDQ was a lot of un
20:39:40 <b_jonas> "Wolfenstein: The Old Blood" - um, what's that? is that plain Wolfenstein, or something else?
20:40:16 <FireFly> "It was released on May 5, 2015"
20:40:59 <b_jonas> there's also Super Mario Bros, Pokémon Yellow glitchless, and Super Mario 64 120 stars
20:41:25 <b_jonas> oh, and Super Monkey Ball of course
20:41:29 <b_jonas> I always look forward to that
20:42:15 <FireFly> the kaizo mario games are these crazy super mario romhacks
20:42:33 -!- hppavilion[1] has joined.
20:42:36 <FireFly> https://www.youtube.com/watch?v=zkMuNRjodCQ ← as an example
20:42:36 <b_jonas> Didn't those have some other name? Um,
20:43:49 <zzo38> I have implemented many callouts in my PCRE wrapper library now (some of which are not yet released); do you have some kind of idea of what others could be added too?
20:44:09 <b_jonas> I was confusing it with that popular Super Mario World hack that demands lots of precision
20:45:16 <b_jonas> Oh yes, they're playing Super Mario World race of course, and Yoshi's Island as a bonus incentive
20:46:22 -!- Patashu has quit (Ping timeout: 265 seconds).
20:46:50 <b_jonas> Wait. Age of Empires and Grand Theft Auto 3 are played by the same player. Why are there so many players that play both of those games?
20:47:06 <b_jonas> Is there something shared in them? Super mouse skills?
20:47:22 <FireFly> I can't think of anything off the top of my head, but I'm not particularly good at either of those
20:48:25 <b_jonas> There's a ton of Mega Man games played, identified by numbers only, and I can't really place them since they don't tell the system.
20:51:03 <zzo38> I think that if a Magic: the Gathering card would have an ability "Sacrifice target creature: You gain 1 life" then you will not gain any life because the target is not valid by the time it is going to resolve (but you may be able to use another effect to change the target to a valid one).
20:51:43 -!- Welo has joined.
20:51:50 <zzo38> That is why you aren't supposed to put targets in the cost
20:55:25 <zzo38> What is your opinion of such thing?
20:57:19 <shachaf> I don't think putting targets in the cost would be a good idea.
20:57:50 <shachaf> But it might be interesting if the cost was something other than sacrificing.
20:58:08 <shachaf> "Remove a jam from target creature: You gain 1 life"
20:58:23 <shachaf> Probably you want something better than gaining 1 life.
20:58:32 <zzo38> Yes, it should be at least 2 life.
20:58:41 <shachaf> Or something other than gaining life.
20:58:49 <shachaf> But that sort of thing would give your opponent an opportunity to sabotage your ability.
20:58:58 <shachaf> b_jonas: A +1/+1 counter, I guess.
21:00:29 <zzo38> That can be an interesting kind of thing and if the target is no longer valid then it won't give you any life points or whatever else the effect might be, so any player can potentially stop it in that way. Even then, if another effect allows to redirect you can make it continue to work again, I think.
21:01:11 <b_jonas> You may have to amend the illegal actions rules first on how to revert consequences of targetting, for targetting can trigger triggers.
21:03:05 <zzo38> I don't see how that is an issue.
21:06:14 <zzo38> But I think that even in the "sacrifice target creature" case you could cause the ability to work by using an effect that changes the target.
21:07:50 <HackEgo> [wiki] [[DStack]] M http://esolangs.org/w/index.php?diff=45924&oldid=45371 * -Dark-Phantom- * (+6) /* String Literals */
21:08:15 <b_jonas> Hmm. Can you cast Hex, target a Wild Cantor with it, pay for the Hex by saccing the same Wild Cantor, and still have the Hex succeed?
21:08:41 <b_jonas> You control the Wild Cantor.
21:10:27 <zzo38> A spell fails only if all of its targets are now illegal.
21:10:39 <b_jonas> zzo38: sure, but I'm not sure casting can work in that order
21:10:51 <zzo38> You have to choose targets before paying costs
21:10:58 <shachaf> The scow is that if a spell has an optional target, if that target becomes illegal, the whole spell fizzles.
21:11:09 <b_jonas> Yes, you have to choose the targets before choosing how to pay, because of strive or whatever that mechanic is called,
21:11:28 <b_jonas> but don't you check the spell for valid targets after paying?
21:11:42 -!- tromp_ has quit (Remote host closed the connection).
21:12:35 -!- ais523 has joined.
21:13:31 <zzo38> I think target are checked at 601.2c and 601.2e, and then after that you determine the cost, get a mana step if the cost includes mana, and then you pay the cost, and then the spell is considered to be successfully cast.
21:15:13 <zzo38> (Effects that would alter the cost during the mana step do not alter the cost; the cost is locked in before the mana step, and the rules give an example.)
21:15:30 <b_jonas> yes, the costs are locked in
21:15:46 <b_jonas> and you determine how you'd pay for all the cost, then pay all costs according to that
21:16:00 <zzo38> It doesn't say what happens if the spell goes missing during the process of casting it, although it would seem that the spell is still considered as successfully cast (although it will never resolve in this case).
21:16:21 <b_jonas> So the side effects of paying the costs don't influence how much cost you pay.
21:16:29 -!- shikhin has quit (Ping timeout: 246 seconds).
21:16:42 <b_jonas> zzo38: yes, this looks fine
21:17:24 <b_jonas> zzo38: I grepped, and it seems like there's no activated ability with the targets mentioned in the cost.
21:17:34 <b_jonas> I'll have to check for spells with additional costs though.
21:18:38 <b_jonas> nope, no such thing either it seems
21:20:12 <b_jonas> it doesn't seem to appear in alternate costs either
21:21:10 <b_jonas> (alternate costs like that of Daze)
21:22:20 -!- boily has joined.
21:24:20 -!- shikhin has joined.
21:25:59 <zzo38> One situation where a spell could go missing during the process of being cast, is, I suppose, if you manage to cast a spell owned by another player in a multiplayer game, but that player concedes during your mana step.
21:30:22 -!- TieSoul has quit (Remote host closed the connection).
21:36:11 <zzo38> If you activate Echo Chamber while you do not have four manas in your mana pool, are you allowed to choose not to activate it if opponent chooses the target that you don't want?
21:38:29 <ais523> zzo38: if you control a copy of a spell owned by a player and that player concedes, is the copy still on the stack?
21:38:47 <ais523> given that it isn't a physical object and so they don't need to take it home with them, I guess there's no reason for it not to be
21:39:07 <zzo38> If they own the copy then it is removed
21:39:17 -!- Welo has quit (Quit: Leaving).
21:39:20 <b_jonas> ugh, I've no idea how the multiplayer rules work
21:40:11 <zzo38> "A copy of a spell is owned by the player under whose control it was put on the stack."
21:41:21 -!- Welo has joined.
21:41:22 <zzo38> So probably the copy is still on the stack if you control it.
21:44:13 <zzo38> Rule 800.4a says that all objects owned by players who leave the game will be removed, whether they are cards or tokens or something else it does not matter, they are still removed if the owner of those objects is no longer in the game.
21:46:25 -!- TellsGoto has quit (Ping timeout: 260 seconds).
22:12:16 -!- tromp_ has joined.
22:16:51 -!- tromp_ has quit (Ping timeout: 260 seconds).
22:28:08 <ais523> you can steal control of a copy of a spell using Commandeer, I think
22:28:53 <zzo38> Yes, and then it is owned by different player than who controls it.
22:43:29 <zzo38> select cast(pcre_replace('(?=\d)(?C148)\d+|[WUBRG](?C1)|X|\([A-Z0]/.\)(?C1)|(?=\(\d)\((?C148)\d+/.\)','0',-11,-1) as int); -- Do you like this?
22:45:20 <zzo38> Can you see its working or its meaning or purpose of such a code?
22:46:21 <ais523> it's allowing a db engine (presumably sqlite) to call out to pcre
22:46:50 <zzo38> Yes, although it is not what I meant; I meant if you can guess what this specific code and pattern is for.
22:46:51 <izabera> you were talking about mtg and then you posted this 100 character regex out of the blue
22:49:53 <HackEgo> [wiki] [[Funciton]] M http://esolangs.org/w/index.php?diff=45925&oldid=44356 * Timwi * (+1)
22:53:54 <ais523> `` echo 'select cast(pcre_replace(@(?=\d)(?C148)\d+|[WUBRG](?C1)|X|\([A-Z0]/.\)(?C1)|(?=\(\d)\((?C148)\d+/.\)@,@0@,-11,-1) as int);' | wc -c
22:54:59 -!- dcentral has joined.
22:58:26 -!- Welo has quit (Quit: Leaving).
23:03:13 -!- mauris has joined.
23:07:51 -!- boily has quit (Quit: SLIMY CHICKEN).
23:08:20 -!- ais523 has quit.
23:13:50 -!- tromp_ has joined.
23:14:54 <\oren\> It's fun to commit the faux pas of pronouncing it as folks pass.
23:18:04 -!- tromp_ has quit (Ping timeout: 245 seconds).
23:26:35 -!- jaboja has quit (Ping timeout: 264 seconds).
23:49:28 <izabera> do you guys know any file system that mounts compressed archives?