←2016-01-22 2016-01-23 2016-01-24→ ↑2016 ↑all
00:04:18 <\oren\_> just miniaturize and make faster versions of cpu designs that have expired patents
00:04:30 <\oren\_> voila: open source cpu
00:05:57 <zzo38> I would intend the entire computer and not only the CPU (although, CPU is also part of it)
00:06:35 <vanila> RISCV
00:13:29 <quintopia> helloily
00:26:51 <quintopia> what do tonight?
00:27:37 <hppavilion[1]> How about a non-isolating programming language?
00:30:07 <zzo38> Which means what?
00:30:50 <hppavilion[1]> zzo38: You know what an isolating language is? In linguistics?
00:31:39 <zzo38> No I do not know, but I can learn, I can try to look it up in Wikipedia
00:31:45 <hppavilion[1]> zzo38: OK
00:33:31 -!- bb010g has joined.
00:41:46 <boily> quintopia: on my last few minutes for the day. what do you tonight tonight?
00:44:34 <hppavilion[1]> What kind of language does the world need? Not necessarily for valid usage; more like a J-like language
00:44:48 <hppavilion[1]> I'm thinking either UsefulFuck or Fueue#
00:49:34 <boily> horizontime!
00:49:46 -!- boily has quit (Quit: THIN CHICKEN).
01:09:08 <FireFly> hppavilion[1]: are you implying J doesn't have uses?
01:09:27 <hppavilion[1]> FireFly: It certainly has uses
01:09:42 <hppavilion[1]> FireFly: But how often is it seriously used in business?
01:11:19 -!- lleu has quit (Quit: That's what she said).
01:12:35 <FireFly> Sure, it's pretty esoteric
01:12:58 <hppavilion[1]> FireFly: That's what I meant
01:13:19 <hppavilion[1]> FireFly: It's used for golfing more than for serious programming
01:13:35 <FireFly> Umm, not sure about that (though possibly)
01:13:57 <fizzie> How about serious business golfing?
01:14:22 <FireFly> I don't follow the jsoftware.com mailing lists, but I'm pretty sure it's used seriously in actual businesses
01:14:45 <hppavilion[1]> FireFly: No one does that anymore
01:14:50 <hppavilion[1]> I mean fizzie
01:14:56 <hppavilion[1]> fizzie: But we used to xD
01:15:19 <hppavilion[1]> Of course, it was more compiler golfing than code golfing, but that's not the point
01:19:57 <fizzie> http://code.jsoftware.com/wiki/Stories -- "success stories" from businesseses.
01:20:27 <fizzie> Wonder if there's any J users at work.
01:20:50 <fizzie> Getting the work laptop out at 1:20am doesn't appeal enough to look for any.
01:23:47 -!- p34k has quit.
01:25:36 -!- XorSwap has joined.
01:29:04 <mauris> I am definitely a J Convert.
01:29:18 <mauris> yup, apl/j/k still feels like a total cult when i read about it.
01:29:39 <mauris> some really strange ideologies about programming in those communities :/
01:30:19 <shachaf> mauris: hauris
01:30:34 <mauris> hi, shachaf!
01:30:38 <mauris> how arechaf
01:31:02 <shachaf> amchaf mauchaf
01:33:10 <quintopia> mauris: if you like J, you'll love Jelly!
01:34:46 <mauris> disclaimer: i think they're both really cool, just inelegant and not something i'd seriously use for code i want anyone to see
01:34:53 <mauris> (outside the context of code golf...)
01:35:11 <Sgeo> J is interesting but I can't imagine ever learning it well enough to be able to come back to code written some time ago and understand it without a refresher in J
01:35:16 <Sgeo> With the exception of simple stuff
01:35:40 <mauris> shachaf: is mauchaf the new popular #esoteric slash pairing, titwties hth hand
01:35:57 -!- vanila has left ("Leaving").
01:37:02 <mauris> Sgeo: i think part of the J mentality -- or at least the one that's popular in its community -- is just that you treat reading code very differently
01:37:15 <shachaf> mauris: that is the way that i extract sympathy?
01:37:23 <mauris> i asked a J user "how do you even piece apart all of the @: and & and [: stuff??"
01:37:28 <shachaf> i've never been part of a slash pairing before
01:37:43 <shachaf> other than the ever-present HackEgo/shachaf "hachaf", of course
01:38:00 <mauris> and they replied "i just throw the tails into the J interpreter and see what they do to the input"
01:38:33 <mauris> was going for: "that is the worst thing i've ever said"
01:39:43 <shachaf> i could go for a juicy maurjan piece
01:41:44 <mauris> sha523 is my OTP and also my favorite cryptographic hash function
01:43:14 <shachaf> you can't get a real OTP out of a cryptographic hash function hth
01:43:40 <mauris> is this an amazing crypto joke
01:43:50 <mauris> oh my god i just got it
01:46:32 <shachaf> Taneb: is combining your name with someone else's name Taneb-approved
01:46:51 -!- ais523 has joined.
01:46:56 -!- ais523 has quit (Changing host).
01:46:56 -!- ais523 has joined.
01:48:45 <\oren\_> I'm working on a new library to replace ncurses
01:50:43 <\oren\_> why? because ncurses is too complicated
01:51:22 <mauris> i've always been sorta disappointed in the lack of a middle ground when it comes to terminal graphics libraries
01:52:03 <mauris> it seems as if you either work directly with ncurses, or you work with something that has Widgets and Layers and Fields and whatever
01:52:13 <\oren\_> specifically, color support is way too complex for ncurses
01:52:32 <ais523> \oren\_: I wrote my own curses replacement a while back
01:52:41 <ais523> although it was meant to be very easy to port existing curses programs to it
01:53:08 <\oren\_> ah... yah that definitely won't be the case here
01:56:16 <\oren\_> basically, this library is much more directly attached to the ANSI escape codes that it outputs
02:00:18 <\oren\_> e.g. instead of writing "\e[1;32;44m" you write oiocolor(1,LIME,NAVY);
02:00:41 <\oren\_> where 1 is the file descriptor of the terminal
02:02:09 <hppavilion[1]> The Language of Syntactic Odds and Ends
02:02:23 <hppavilion[1]> For example, 15 different kinds of comment.
02:02:41 * Sgeo misses the Factory ABCDEF language
02:06:49 <hppavilion[1]> I am more than half tempted to make a valiant effort towards implementing Compute/IO
02:07:11 <hppavilion[1]> Oh my god
02:07:39 <hppavilion[1]> What happens if we train Fungot on the specifications for programming languages and ask her to make a programming language?
02:09:04 <Sgeo> Still better than PHP </low-effort-jokes>
02:09:32 <hppavilion[1]> Sgeo: Seriously, I think we should do that
02:09:36 <ais523> \oren\_: I wouldn't call something a curses replacement unless it does screen addressing
02:09:37 <hppavilion[1]> Make an Esolang-defining robot
02:10:24 <\oren\_> ais523: oh it does do that
02:10:49 <\oren\_> but it doesn't optimize your output with a refresh-cycle and double buffering
02:11:05 <Sgeo> I should learn ANSI escape codes
02:11:12 <ais523> character encoding handling?
02:11:17 <\oren\_> http://www.orenwatson.be/ansi.htm
02:11:19 <ais523> Sgeo: they're not hard, /but/ they are inconsistent between terminals
02:11:27 <hppavilion[1]> Sgeo: No, you should memorize all of unicode
02:11:28 <\oren\_> ais523: it only supports utf-8 right now
02:11:34 <ais523> I did a lot of experimentation to figure out which were the most stable
02:11:37 <ais523> \oren\_: what if the terminal doesn't support utf-8?
02:11:46 <ais523> are you at least sending codes to put the terminal into UTF-8 mode?
02:11:47 <Sgeo> ais523, but, um. crud. what does the world do then?
02:12:03 <ais523> Sgeo: typically handles things badly
02:12:04 <\oren\_> ais523: nope :P
02:12:27 <ais523> the original plan was to use a database called termcap/terminfo that listed the appropriate codes for each terminal
02:12:40 <ais523> this doesn't work in practice because almost every terminal claims to be xterm
02:13:03 <Sgeo> Why don't they claim to be xterm then implement exactly xterm's escapes?
02:13:40 <\oren\_> xterm has too many
02:14:12 <\oren\_> it's ludicrous. the ones on my page are afaict the lowest common denominator
02:15:04 <ais523> Sgeo: apparently the xterm devs bitch about this
02:15:16 <ais523> complaining that so many emulators don't implement all the codes they're supposed to
02:15:25 <ais523> most likely, it's a case of stopping development when things seem to be working
02:15:55 <Sgeo> Why don'
02:16:15 <Sgeo> Why don't they not claim to be xterm then? Or is it a Mozilla UA fiasco again where everything only works with "xterm"?
02:17:49 -!- hppavilion[1] has quit (Ping timeout: 265 seconds).
02:19:55 <fizzie> "Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/48.0.2564.82 Safari/537.36"
02:20:01 <fizzie> So much of that is complete nonsense.
02:20:54 <\oren\_> WTF make up your mind as to what you are?!
02:21:01 -!- EasyAsShlt has quit (Read error: Connection reset by peer).
02:22:02 <ais523> Sgeo: they dont want to deploy their own termcap databases
02:22:51 -!- vodkode has joined.
02:23:06 <\oren\_> who in their right mind would?!!?
02:23:16 <fizzie> \oren\_: Well, Microsoft Edge on desktop sends "Mozilla/5.0 (Windows NT 10.0; <64-bit tags>) AppleWebKit/<WebKit Rev> (KHTML, like Gecko) Chrome/<Chrome Rev> Safari/<WebKit Rev> Edge/<EdgeHTML Rev>.<Windows Build>".
02:23:28 <fizzie> That's the same lies as Chrome + Edge/X tagged on the end.
02:24:18 <fizzie> At least Chrome uses a (fork of) WebKit, unlike Edge.
02:25:37 <\oren\_> I see.
02:25:52 <\oren\_> hey wait why do I have an underscore
02:25:58 -!- \oren\_ has changed nick to \oren\.
02:26:22 <ais523> does anyone actually read user agents any more?
02:26:34 <\oren\> who in their right mind would?!!?
02:27:02 <fizzie> I think they still do, at least for old-IE workarounds.
02:27:03 <\oren\> they're nothing but a pack of lies anyway
02:27:11 <\oren\> oh, good point
02:27:15 <Sgeo> >.>
02:27:22 <\oren\> <.<
02:27:37 <\oren\> ^.^
02:27:41 <ais523> http://webaim.org/blog/user-agent-string-history/
02:27:59 <Sgeo> ais523, that's old, it's gotten worse
02:28:02 <fizzie> That could do with a paragraph about Edge.
02:28:08 <ais523> Sgeo: I know
02:28:13 <ais523> but I don't have a newer page about it
02:28:20 <ais523> and it's a funny and amusing enough summary as it is
02:28:49 <Sgeo> IE11 pretends not to be MSIE but can be instructed by Microsoft to act like IE9. It can also be instructed to send the IE9 UA but apparently this isn't guaranteed, at least in one case.
02:29:32 <fizzie> I wonder if there's some sort of nominal WebKit/Blink version correspondence mapping that Chrome uses to set the WebKit revision in the UA it advertises.
02:30:09 <ais523> what rendering engine does Opera use nowadays?
02:30:10 <fizzie> At least both my Chrome/47.0.2526.80 and Chrome/48.0.2564.82 say AppleWebKit/537.36.
02:30:22 <Sgeo> Yes, I have seen this happen.
02:30:55 <fizzie> I think Opera's now built on top of Chromium.
02:31:05 <fizzie> So, Blink.
02:32:08 <fizzie> "Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/28.0.1500.52 Safari/537.36 OPR/15.0.1147.100"
02:32:22 <fizzie> As seen there for Opera 15.
02:33:05 <fizzie> And of course crawler-bots tend to be "Mozilla/5.0 (compatible; <crawler version goes here>)".
02:33:39 <Sgeo> I think there's a Mozilla or Firefox reference in the HTML5 spec somewhere
02:33:57 <\oren\> anyway, I've now added oiomovexy, which is a macro for oiomoveyx
02:34:13 <\oren\> which is also called oiomove
02:34:25 <fizzie> My access.log has 4 visits with the user agent "null".
02:34:34 <ais523> \oren\: does it swap the arguments?
02:34:35 <Sgeo> https://developer.mozilla.org/en-US/docs/Web/API/NavigatorID/appCodeName
02:34:43 <\oren\> ais523: yeah
02:34:59 <Sgeo> https://developer.mozilla.org/en-US/docs/Web/API/NavigatorID/appName
02:35:23 <Sgeo> Oh, it's removed from the standards but still in every browser, I guess
02:36:14 <fizzie> ...and 24 requests where the user agent is a perl script that tries to download something that's almost certainly malware.
02:36:23 <\oren\> Note: Do not rely on this property to return a real product name. All browsers return "Mozilla" as the value of this property.
02:36:34 <\oren\> WTFWTFWTFWTFWTF
02:37:24 <fizzie> http://sprunge.us/dMGV <- nice user agent.
02:37:32 <ais523> occasionally I get browsers that try to load URLs that are on an entirely different site
02:37:40 <fizzie> (It's the part that starts "() ...".)
02:37:40 <ais523> presumably in the hope that my web server has been configured as a proxy for some reason
02:37:49 <ais523> they have http:// and everything
02:38:15 <ais523> also that reminds me a ton of the Shellshock exploit
02:38:41 <fizzie> Yes.
02:38:49 <ais523> presumably it's hoping to hit a CGI-generated page
02:38:54 <ais523> old-fashioned CGI that is
02:39:04 <ais523> and then have the resulting page run an old verson of bash
02:39:47 <fizzie> There's also one hit from "Cloud mapping experiment. Contact [email address in pdrlabs.net]".
02:41:13 <\oren\> that's why all my cgi is sed or C
02:41:23 <\oren\> (yes, sed)
02:41:26 <\oren\> lol
02:41:27 <ais523> that's probably one of the experiments that contacts every IPv4 IP to see what happens
02:41:38 <ais523> \oren\: it can still happen with C CGI, if the C happens to run system() (which goes via bash)
02:41:42 <fizzie> Yes, their web page says they're mapping the Internet.
02:42:06 <fizzie> Speaking of IPv4, our home ISP started offering native IPv6 the other day. Feels very futuristic.
02:42:32 -!- hppavilion[1] has joined.
02:43:46 <\oren\> nice
02:44:22 <Sgeo> fizzie, is the download dangerous or just executing the download?
02:44:28 <Sgeo> What's in the image
02:46:59 <fizzie> I wonder. I don't see anything obvious about trying to execute it or anything, just the download bit.
02:47:29 <fizzie> But possibly they wait for a "XSUCCESS!" reply and then send a second request that tries to run it.
02:48:30 -!- MDude has joined.
02:50:51 <hppavilion[1]> A programming language where the code is a matrix?
02:52:33 <hppavilion[1]> A nested matrix, no less
02:58:23 <HackEgo> [wiki] [[Quiney]] https://esolangs.org/w/index.php?diff=46254&oldid=39066 * * (-3) ..Why is there a reference to parenthesis in a sea of brackets?
03:00:47 <ais523> right, they're probably testing to see which systems are vulnerable, in the hope of exploiting them later
03:02:45 <\oren\> hmm those image url's don't seem to load
03:04:01 <\oren\> doesn't seem to exist
03:09:02 <\oren\> eh, whatever
03:26:18 -!- ais523 has quit.
03:28:38 -!- XorSwap has quit (Ping timeout: 250 seconds).
03:29:18 -!- XorSwap has joined.
03:29:40 -!- XorSwap has quit (Read error: Connection reset by peer).
03:34:12 -!- Phantom_Hoover has quit (Read error: Connection reset by peer).
03:55:25 -!- mauris has quit (Quit: Leaving).
04:18:36 -!- oerjan has joined.
04:19:55 -!- iformas has joined.
04:30:31 -!- iformas has quit (Quit: Page closed).
04:39:47 <zzo38> Do you know if efficient byte arrays can be implemented in a JavaScript code (with any of the common implementations)?
04:41:50 -!- vodkode has changed nick to EasyAsShlt.
04:44:24 <hppavilion[1]> Does anyone 'round these parts know what technique fungot uses to generate sentences?
04:44:24 <fungot> hppavilion[1]: ( if x x disjunct2)) disjunct1) consequent alternative) if condition, then going back and watching old classes
04:44:31 <hppavilion[1]> fungot: How /do/ you generate sentences?
04:44:31 <fungot> hppavilion[1]: what annoys me is that i can't use if as a function
04:44:55 <hppavilion[1]> So it doesn't use IF at all, at least not as a function...
04:45:20 <zzo38> The back cover of this issue of 2600 includes a picture of a Japanese food container labeled "unix ware" on top of a TV set labeled "Color Television TH-2600XE"
04:48:01 <oerjan> hppavilion[1]: fizzie knows hth (it's called "n-grams" with variable n, iirc, which is somewhat like markov chains but keeping track of n (-1?) previous words.
04:48:06 <oerjan> )
04:48:47 <hppavilion[1]> oerjan: What I want to do is take some of the rather well-written Esolangs articles and feed them to a random sentence generator xD
04:48:49 <oerjan> basically there are pregenerated tables of probabilities for each word in a style
04:48:50 <hppavilion[1]> See what happens
04:49:25 <oerjan> hppavilion[1]: oh. you need a rather large "corpus". you'd probably want the _whole_ wiki as one input.
04:49:29 <hppavilion[1]> I doubt it'd produce anything /usable/
04:49:31 <hppavilion[1]> oerjan: Ah
04:50:04 <oerjan> i think there are some pointers in fungot's github
04:50:04 <fungot> oerjan: ( how the fuck is " data" does? i'm not following with all these rxrs... isn't scheme fnord bloated?
04:50:07 <oerjan> ^source
04:50:07 <fungot> https://github.com/fis/fungot/blob/master/fungot.b98
04:50:26 <hppavilion[1]> oerjan: I'd also like to feed it some REAL language specs
04:50:34 <oerjan> (snip everything from blob, i guess)
04:51:06 <oerjan> hppavilion[1]: note, it doesn't actually works with specs. it works with a corpus of examples. and it doesn't split up words.
04:51:10 <oerjan> *-s
04:51:36 <hppavilion[1]> oerjan: What do you mean exactly? Why wouldn't it work with specs?
04:52:04 <oerjan> it's essentially a probabilistic FSA, so it cannot handle everything context-free (like, why fungot cannot match brackets)
04:52:04 <fungot> oerjan: pass the variable's name to the argument tuple ( list a)) ( define ( fnord conn nick) language? i didn't do
04:52:57 <oerjan> hppavilion[1]: well the model generator programs work with examples. i guess you could try to make a model in another way. but it would _still_ be FSA-ish.
04:53:25 <hppavilion[1]> oerjan: True, true
04:53:25 <oerjan> fungot: thank you for the demonstration
04:53:25 <fungot> oerjan: will follow the standard idiom won't work with any libc you want to do it. thanks
04:53:43 <hppavilion[1]> oerjan: I don't want to generate a formal language; I just want it to spit out commands xD
04:54:01 <oerjan> hppavilion[1]: it's basically just a bit more advanced than the "obvious" markov generators.
04:57:51 <oerjan> hppavilion[1]: btw the model is not itself in befunge, it's a separate file (or two?) that fungot does lookups in.
04:57:51 <fungot> oerjan: i mean, i think) is just ( define void ( if f f)) if i'm not using zsh, you can use
04:59:08 <oerjan> also, i think fizzie had a perl version of the lookup program, or something. one that also matched brackets, or something.
05:00:04 <oerjan> hppavilion[1]: for the wiki, there's also the problem of splitting things nicely into sentences for analysis. that's the reason fungot's wikipedia style uses talk pages instead of actual articles.
05:00:57 <oerjan> and that some other styles have abrupt endings
05:01:01 <hppavilion[1]> fungot: Tell me more about OR
05:01:01 <fungot> hppavilion[1]: would look nice anyways
05:01:33 <hppavilion[1]> fungot: Huh? How is that related to OR
05:01:33 <fungot> hppavilion[1]: various :) the one that comes to mind), then i removed the first to
05:02:08 <hppavilion[1]> Huh. fungot doesn't respond to private messages.
05:02:08 <fungot> hppavilion[1]: so generic-room serves no purpose other than to the prototype-based oo of ecma-script
05:02:11 <hppavilion[1]> Shame.
05:06:46 -!- clog has quit (Ping timeout: 240 seconds).
05:06:57 <oerjan> hppavilion[1]: fungot's commands do, but not the babble.
05:06:57 <fungot> oerjan: why do you keep asking me stuff like that... manifest meaning having been made much easier by the assertion), rather than just define it
05:07:16 <hppavilion[1]> oerjan: Ah
05:07:39 <hppavilion[1]> oerjan: I think I'll make a random generator for EF
05:07:44 <hppavilion[1]> (Formerly EBNF)
05:07:53 <oerjan> hppavilion[1]: oh hm no, i checked, it does respond.
05:08:06 <hppavilion[1]> oerjan: Seems like an easier place to start
05:08:36 <oerjan> you may, however, have hit fungot's babble limit.
05:08:36 <fungot> oerjan: obviously i had drunk too much coffee or 2) they're too conservative. :)
05:08:49 <hppavilion[1]> fungot: Yeah, generic-room is practically useless
05:08:50 <fungot> hppavilion[1]: i had fnord. the guile manual just mentions the idea of
05:11:44 <hppavilion[1]> 6_>2_+7:_(>3-*,
05:14:21 <\oren\> what's that?
05:15:07 -!- MDude has changed nick to MDream.
05:16:43 <quintopia> hppavilion[1]: you're always talking about making things...where do you keep all the code?
05:17:02 <hppavilion[1]> quintopia: On my computer, githubbing the good stuff
05:17:11 <hppavilion[1]> quintopia: I start too many projects, really
05:17:45 <hppavilion[1]> Nothing can ever keep my attention for too long; I've started hundreds of little and large projects.
05:17:54 <hppavilion[1]> It's unfortunate :/.
05:18:30 <oerjan> the latest xkcd doesn't seem to reach his usual standards of destruction
05:19:06 <oerjan> er
05:19:11 <oerjan> *xkcd what-if
05:24:56 <hppavilion[1]> I
05:24:57 <hppavilion[1]> AM
05:25:03 <hppavilion[1]> DISAPPOINT
05:25:07 <hppavilion[1]> >.<
05:26:24 <oerjan> yes, what does he mean by letting most of humanity survive like that
05:27:06 <oerjan> on the bright side, it's the second week in a row. maybe he's reducing standards to increas volume.
05:27:22 <oerjan> *+e
05:27:34 <hppavilion[1]> oerjan: True
05:27:38 * oerjan looks suspiciously at his keyboard.
05:27:49 <hppavilion[1]> oerjan: I submitted what I think was a pretty good question a while ago
05:28:00 <hppavilion[1]> Much better than cars reentering the atmosphere
05:28:15 <hppavilion[1]> I suppose it was a bit advanced though; not something one can just work out on a sheet of paper.
05:58:05 <quintopia> can someone explain to me why this sed command isn't matching anything in the file, even though regex tester confirms the regex matches at least one line?
05:58:24 <quintopia> sed -n '/^[a-zA-Z]+\s.*/p'
06:02:03 <pikhq> Huh, I'm at a loss.
06:02:36 <\oren\> hmm, I'm not sure what elese I should put in oio.h
06:03:23 <\oren\> well, ncurses has a lot of helpers for drawing various graphics, I guess I can add some of that
06:04:39 <\oren\> like using ▁▂▃▄▅▆▇█ to graph an array of integers
06:05:06 <\oren\> actually, that doesn't require any ncurses like stuff anyway
06:05:40 <\oren\> also, s/integers/numbers/
06:06:55 <oerjan> quintopia: wrong regexp style, you want at least \+
06:08:07 <pikhq> Oh right, sed takes *basic* regexes!
06:08:20 <pikhq> POSIX basic regexes don't include \+.
06:08:25 <pikhq> Or +
06:08:37 <pikhq> Do include \{1,\} though.
06:08:46 <oerjan> um \+ worked fine for me.
06:08:56 <oerjan> (i only tested that, not the rest)
06:09:12 <pikhq> oerjan: That's a GNU extension.
06:09:24 <oerjan> mayhaps.
06:09:28 <pikhq> Which POSIX *permits* but doesn't specify.
06:10:44 <pikhq> POSIX also doesn't specify \s, the POSIXly correct version would be: sed -n '/^[a-zA-Z]\{1,\}[[:space:]].*/p'
06:11:06 <pikhq> Or, sed -n '/^[[:alpha:]]\{1,\}[[:space:]].*/p'
06:13:17 <oerjan> quintopia: adding -r might help instead
06:13:45 * oerjan honestly has no clue about the variations among non-perlish regexps
06:13:45 <quintopia> what is -r
06:13:52 <oerjan> quintopia: extended regexps
06:14:04 <quintopia> posix is so verbose
06:14:28 <pikhq> And that's not POSIX. (sadly)
06:14:41 <pikhq> (I don't even know why, POSIX *specifies* extended regexps!)
06:14:52 <pikhq> oerjan: Basic is the weird one, basically.
06:14:58 <oerjan> OKAY
06:15:32 <pikhq> ACK
06:16:02 <quintopia> pikhq: sed understands \s at least. the \+ worked
06:16:14 <pikhq> quintopia: Both are GNU extensions, not POSIX. :)
06:16:44 <pikhq> Though, lots of stuff relies on them.
06:22:51 <hppavilion[1]> centrifᵫtal force
06:23:00 <hppavilion[1]> \oren\: ᵫ?
06:32:05 <oerjan> hppavilion[1]: i'm pretty sure that's unethical hth
06:32:35 <hppavilion[1]> oerjan: ?'
06:33:09 -!- oerjan has set topic: Go tromp! | The international hub for magic gathering and deployment. | Effi's finest fluffy waffles | https://dl.dropboxusercontent.com/u/2023808/wisdom.pdf http://codu.org/logs/_esoteric/ http://tunes.org/~nef/logs/esoteric/?C=M;O=D | https://esolangs.org/ | This part of the topic was mungled in 2016.
06:33:53 -!- EasyAsShlt has changed nick to Vodkode.
06:34:10 <oerjan> hppavilion[1]: slinging fᵫtuses around is not generally accepted behavior
06:34:39 <hppavilion[1]> Ah
06:34:48 <oerjan> i hear in parts of america it may get you killed hth
06:34:50 <hppavilion[1]> Back in my day...
06:35:01 <hppavilion[1]> oerjan: Especially if you're black.
06:36:14 -!- oerjan has set topic: Go tromp et al.! | The international hub for magic gathering and deployment. | Effi's finest fluffy waffles | https://dl.dropboxusercontent.com/u/2023808/wisdom.pdf http://codu.org/logs/_esoteric/ http://tunes.org/~nef/logs/esoteric/?C=M;O=D | https://esolangs.org/ | This part of the topic was mungled in 2016.
06:36:35 <oerjan> i guess it was a team effort
06:38:05 <hppavilion[1]> oerjan: What did tromp do exactly?
06:38:18 <oerjan> http://tromp.github.io/go/legal.html
06:39:11 <hppavilion[1]> oerjan: Oh right, that
06:40:11 <hppavilion[1]> oerjan: Wait, is tromp_'s last name actually tromp?
06:41:10 <lifthrasiir> make computer Go great again
06:41:10 <oerjan> yes hth
06:41:26 <zzo38> It says "John Tromp" if the WHOIS command is used
06:41:30 <lifthrasiir> yup
06:41:30 <zzo38> Therefore, I would think so
06:41:31 <hppavilion[1]> Awesome. That is a truly awesome last name.
06:41:42 <oerjan> it tromps most other names
06:41:47 <hppavilion[1]> ...
06:41:54 <hppavilion[1]> CSI screech.
06:42:24 <oerjan> i think i may have got a new favorite pun victim.
06:42:59 <hppavilion[1]> Yep
06:43:37 <zzo38> I would now want reviews for Magic: the Puzzling: Codex.
06:45:07 <hppavilion[1]> My god.
06:45:14 <hppavilion[1]> J has the whilst control structure.
06:47:26 <hppavilion[1]> Shutting down the channel now, we're through here.
06:51:54 -!- Moose has joined.
06:58:40 -!- clog has joined.
07:00:05 <\oren\> guess what I graphed!
07:00:07 <\oren\> █ ▅█▂█
07:00:07 <\oren\> ▂ ▅█ █▅ ▅████
07:00:08 <\oren\> █▂█▂██▅████████
07:00:30 <oerjan> something with strange square boxes in
07:00:34 <hppavilion[1]> \oren\: randrange(0, 9)?
07:00:52 <\oren\> nope, its the first 15 digits of pi!
07:01:32 <\oren\> if your font supports the block characters properly that is!
07:01:38 <oerjan> hmph, i would have recognized that if i could actually see all chars properly </snotty>
07:04:07 <myname> why only 15?
07:04:14 <oerjan> i think lambdabot had some graphing function but i have no idea how to find it.
07:04:30 <\oren\> http://ctrlv.in/701168
07:04:55 <oerjan> and it may have been something someone temporarily coded up, anyway.
07:05:15 <oerjan> @list-modules
07:05:15 <lambdabot> activity base bf check compose dice dict djinn dummy elite eval filter free fresh haddock help hoogle instances irc karma localtime metar more oeis offlineRC pl pointful poll pretty quote search slap source spell system tell ticker todo topic type undo unlambda unmtl version where
07:05:54 <oerjan> it's probably in L.hs somewhere. i just don't remember where to browse that.
07:06:08 <oerjan> @help pretty
07:06:08 <lambdabot> pretty <expr>. Display haskell code in a pretty-printed manner
07:06:13 <\oren\> █ ▅█▂█ ▅ ▅ ▂█
07:06:14 <\oren\> ▂ ▅█ █▅ ▅████ █▂█ █▂ █ ██▅
07:06:14 <\oren\> █▂█▂██▅█████████▅████▅██████▅███
07:06:18 <\oren\> more digits
07:06:31 <oerjan> @pretty pi*r^2
07:06:31 <lambdabot> pi * r ^ 2
07:06:40 <oerjan> bah
07:08:39 <\oren\> http://www.orenwatson.be/grph.htm
07:10:23 <\oren\> that's the program I used
07:14:55 <\oren\> ▇▇▇▇▆▆▅▄▄▃▁ ▁▃▄▄▅▆▆▇▇▇▇▇▇▇▆▆▅▄▄▃▁
07:14:58 <\oren\> ████████████▇▆▅▃▂▁ ▁▂▃▅▆▇███████████████████████▇▆▅▃▂▁
07:15:02 <\oren\> ███████████████████▇▆▄▃▃▂▁▁ ▁▁▂▃▃▄▆▇█████████████████████████████████████▇
07:15:06 <\oren\> cosine
07:17:19 <\oren\> hmm unison font supports these pretty well too
07:17:41 <\oren\> oerjan: what backward font are you using?
07:18:43 <oerjan> courier new
07:19:18 <\oren\> oh, I see. poor oerjan
07:19:20 <oerjan> it's the default.
07:19:36 <oerjan> i'm not a font-maniac like the rest of this channel hth
07:20:35 <oerjan> although for the last one the bigger problem is that irssi wraps.
07:33:57 <hppavilion[1]> \oren\: Do it in binary and it might actually mean something.
07:34:00 <hppavilion[1]> Probably not, though.
07:34:25 <hppavilion[1]> I don't know, maybe it'll help towards discovering new pimes.
07:35:25 <hppavilion[1]> Which would, of course, contribute towards discovering new primes
07:45:13 -!- Moose has quit (Quit: Page closed).
07:51:53 <\oren\> █████▇▇▇▇▆▆▅▅▄▄▃▂▂▁ ▁▂▂▃▄▄▅▅▆▆▇▇▇▇████ 1.000000
07:51:57 <\oren\> █████████████████████▇▆▆▅▄▃▂▂▁ ▁▂▂▃▄▅▆▆▇████████████████████
07:52:01 <\oren\> ████████████████████████████████▇▆▆▅▄▄▃▃▂▂▁▁▁▁ ▁▁▁▁▂▂▃▃▄▄▅▆▆▇███████████████████████████████ -1.000000
07:52:07 <\oren\> I modified it to add a scale to the side
07:53:18 <\oren\> █ ▅█▃█ ▅ ▅ ▃█ 9.000000
07:53:18 <\oren\> ▃ ▅█ █▅ ▅████ █▃█ █▃ █ ██▅
07:53:19 <\oren\> █▃█▃██▅█████████▅████▅██████▅███ 0.000000
08:01:25 <HackEgo> [wiki] [[THRAT]] https://esolangs.org/w/index.php?diff=46255&oldid=46252 * Oerjan * (+3) Undo revision 46252 by [[Special:Contributions/|]] ([[User talk:|talk]]) I find no evidence of this claim.
08:02:37 <oerjan> hppavilion[1]: discovering new primes is easy. the hard part is finding _all_ of them.
08:03:00 <hppavilion[1]> oerjan: Yes, that was the point
08:03:33 <hppavilion[1]> Someone should make a language called Sea of Brackets
08:11:30 <myname> why
08:14:24 <HackEgo> [wiki] [[Talk:Minebit]] M https://esolangs.org/w/index.php?diff=46256&oldid=46253 * Oerjan * (+47) unsigned begone
08:14:31 <hppavilion[1]> myname: https://esolangs.org/wiki/Special:RecentChanges
08:16:56 <oerjan> <hppavilion[1]> How about a non-isolating programming language? <-- see Perligata
08:18:37 <hppavilion[1]> oerjan: If it can note that you should sign your comments with ~~~~, why can't it just autosign your comments with ~~~~?
08:19:28 <oerjan> hppavilion[1]: because your edit might not be a complete comment
08:19:44 <hppavilion[1]> oerjan: Oh right??
08:19:54 <hppavilion[1]> s/\?\?/\?/
08:20:10 <oerjan> also, ask whoever made mediawiki. i'm sure they could have handled it better.
08:20:14 <hppavilion[1]> But it can detect if you didn't sign your edit with ~~~~ when you should have
08:20:50 <oerjan> there's probably a plugin for it.
08:21:57 <oerjan> ask fizzie to enable it hth
08:54:37 -!- andrew has joined.
08:58:15 <hppavilion[1]> CombinatorOS?
08:58:20 <hppavilion[1]> SKOS
09:00:40 -!- andrew has quit (Remote host closed the connection).
09:03:54 -!- andrew has joined.
09:07:07 <hppavilion[1]> You can clearly tell I don't understand OSes
09:07:32 <hppavilion[1]> I mean, /maybe/ it makes sense as a format for executables?
09:11:11 <hppavilion[1]> Should I use vi or emacs?
09:11:14 <hppavilion[1]> (>:))
09:11:37 <zzo38> Use what you prefer
09:11:41 <zzo38> I use vi
09:12:36 <hppavilion[1]> ...
09:12:37 <hppavilion[1]> Dammit
09:22:58 -!- Celeritas has joined.
09:26:10 <HackEgo> [wiki] [[Special:Log/newusers]] create * Celerition * New user account
09:27:16 <HackEgo> [wiki] [[BadScript]] N https://esolangs.org/w/index.php?oldid=46257 * Celerition * (+15) Created page with "This is a stub."
09:28:20 -!- Celeritas has changed nick to Celerition.
09:43:13 <HackEgo> [wiki] [[BadScript]] https://esolangs.org/w/index.php?diff=46258&oldid=46257 * Celerition * (+104)
09:43:36 <Celerition> I'm doing an esoteric language
09:43:46 <Celerition> what should I add?
09:45:51 <Celerition> I'll make an IRC bot with it.
09:51:26 <HackEgo> [wiki] [[BadScript]] https://esolangs.org/w/index.php?diff=46259&oldid=46258 * Celerition * (+152)
09:52:47 -!- Celerition_ has joined.
09:54:30 <HackEgo> [wiki] [[BadScript]] M https://esolangs.org/w/index.php?diff=46260&oldid=46259 * Celerition * (+67)
09:55:00 -!- Celerition has quit (Ping timeout: 252 seconds).
10:00:32 -!- andrew has quit (Remote host closed the connection).
10:06:53 -!- oerjan has quit (Quit: LAter).
10:11:34 -!- hppavilion[1] has quit (Ping timeout: 276 seconds).
10:12:26 <HackEgo> [wiki] [[BadScript]] https://esolangs.org/w/index.php?diff=46261&oldid=46260 * Celerition * (+711)
10:13:42 -!- Celerition_ has changed nick to Celerition.
10:13:53 <Celerition> !help
10:13:53 <zemhill> Celerition: I do !zjoust; see http://zem.fi/bfjoust/ for more information.
10:13:53 <EgoBot> ​help: General commands: !help, !info, !bf_txtgen. See also !help languages, !help userinterps. You can get help on some commands by typing !help <command>.
10:14:15 <Celerition> !bf_txtgen hello
10:14:19 <EgoBot> ​56 ++++++++[>+++++++++++++>+>><<<<-]>.---.+++++++..+++.>++. [228]
10:14:35 <Celerition> !bf ++++++++[>+++++++++++++>+>><<<<-]>.---.+++++++..+++.>++.
10:14:35 <EgoBot> hello
10:17:59 <Celerition> !bf -[--->+<]>-----.++.---------.>+[--->++<]>.---------.++++++.------------.+[->++++<]>.+++.-[->+++<]>-.[--->+<]>----.----.+++++.+++[->+++<]>.+++++++++++++.---------.------.-[------>+<]>-.-[----->++<]>-.+[-->+<]>++.---[->+++<]>+.[->+++++<]>-.++[->+++<]>+.++++++++.------.+++++.-------.-[--->+<]>--.+++++[->+++<]>.---------..-[->+++<]>-.>++++++++++.++[++++>---<]>+.+++++.------.+++++.++[----->++<]>.+++.-[->+++<]>.++++++++++++.------
10:18:00 <EgoBot> PRIVMSG #esoteric:I'm going off. \ JOIN #fr
10:36:01 <myname> wat
10:44:30 -!- Celerition has quit (Ping timeout: 252 seconds).
10:46:31 <HackEgo> [wiki] [[BadScript]] https://esolangs.org/w/index.php?diff=46262&oldid=46261 * Celerition * (-35)
10:54:43 -!- boily has joined.
11:07:23 -!- sebbu has quit (Ping timeout: 264 seconds).
11:10:03 -!- Phantom_Hoover has joined.
11:11:29 <boily> good Phantom_Hoovorning.
11:20:17 <Phantom_Hoover> boily
11:22:03 -!- sebbu has joined.
12:05:56 -!- boily has quit (Quit: COVARIANT CHICKEN).
12:12:58 -!- J_Arcane has quit (Ping timeout: 250 seconds).
12:16:52 -!- J_Arcane has joined.
12:58:09 -!- impomatic_ has joined.
13:04:43 -!- LexiciScriptor has joined.
13:06:55 -!- Vodkode has quit (Ping timeout: 240 seconds).
13:07:36 -!- mroman has joined.
13:07:53 <mroman> aw firefox you gotta be fucking kidding me :(
13:08:16 <mroman> my parent element has font-size 18px
13:08:25 <mroman> code { font-size: inherit; }
13:08:39 <mroman> but firefox manages to compute that code should be displayed in 12px font
13:08:45 <mroman> this is insane
13:08:50 <mroman> this is like basic stuff :(
13:11:11 <mroman> font-size: 1em; doesn't work either
13:13:38 <myname> does font-size: 1em make sense?
13:15:15 <mroman> it sometimes works as a hack
13:15:23 <mroman> but font-size: 100%; works neither
13:15:43 <mroman> maybe it's some weird settings on this machine
13:16:24 <LexiciScriptor> does it work in a different context?
13:16:34 <mroman> it works in IE :D
13:16:41 <mroman> I guess i'll just switch to using IE
13:20:11 <mroman> http://i.imgur.com/bCAMGni.png
13:20:22 <mroman> ^- main (parent of pre) has computed font-size 18px
13:20:38 <mroman> pre (child of main) has font-size: 100%; but computed font-size is 12px
13:27:10 <mroman> let me see if I can reproduce this in a simple html file :)
13:29:44 <mroman> http://codepad.org/vrSs25pC
13:29:51 <mroman> can any of you guys reproduce that?
13:30:05 <mroman> just look at the computed sizes for body and pre
13:30:18 <mroman> the computed sizes of pre should be the same as those for body since pre should inherit the size from body
13:30:24 <mroman> but it doesn't inherit it properly on my firefox
13:32:37 <LexiciScriptor> 16px and 13px :/
13:32:50 <LexiciScriptor> on firefox
13:45:13 <int-e> the height is mismatched, but <pre style="font-size: 1em"> produces the same font size
13:46:01 <int-e> what does font-size: 1em mean anyway...
13:46:56 <LexiciScriptor> it's a problem of courier
13:47:07 <LexiciScriptor> if you change the font, the font-size is the same
13:53:24 -!- lleu has joined.
13:53:24 -!- lleu has quit (Changing host).
13:53:24 -!- lleu has joined.
13:58:54 <int-e> mroman: check edit -> preferences -> fonts and colors -> advanced and note that "monospace" has a smaller default size than the other fonts (16 vs 12); that corresponds to your observed difference
14:03:09 <int-e> (but I don't know *why* that default size is smaller... if you find a reference for that I'd be interested, but not interested enough to search for it myself)
14:20:38 -!- Lord_of_Life has quit (K-Lined).
14:32:06 <fizzie> I looked up a couple of source code references, but none of them have seen fit to put in comments as to why it's like that.
14:33:19 <fizzie> "The basic argument goes that your prefrerred serif font can be relatively larger to your preferred sans-serif font for example." https://bugzilla.mozilla.org/show_bug.cgi?id=84398
14:39:36 -!- Lord_of_Life has joined.
14:41:24 -!- MDream has changed nick to MDude.
14:51:43 -!- LexiciScriptor has quit (Quit: LexiciScriptor).
15:20:20 -!- rifter has quit (Quit: Leaving).
15:20:54 -!- rifter has joined.
15:22:46 -!- LexiciScriptor has joined.
15:22:47 -!- LexiciScriptor has quit (Client Quit).
15:42:15 <mroman> int-e: but I can't set font-sizes for specific fonts through CSS?
15:42:39 <mroman> also... shouldn't the font-size apply to whatever font currently is used?
15:48:22 <fizzie> It does, there's just a "thing".
15:49:00 <fizzie> If you do "font-family: monospace,monospace;" on at least Chrome (I think also on Firefox) you'll get the logical sort of behavior.
15:49:47 <mroman> true
15:49:50 <fizzie> Yeah, also on Firefox (well, Iceweasel).
15:49:52 <mroman> wtf is this o_O
15:50:00 <fizzie> http://code.stephenmorley.org/html-and-css/fixing-browsers-broken-monospace-font-handling/
15:50:05 <fizzie> It's just general stupidity.
15:50:27 <fizzie> But you can see how it looks "too big" (like, objectively) when you don't do the mangling.
15:51:15 -!- jaboja has joined.
15:53:44 <mroman> well it's wider than non-monospace font
15:53:54 <mroman> that's true.
15:54:55 <fizzie> On my fonts, it's also taller for the lowercase letters, which is the main thing.
15:55:09 <fizzie> https://developer.mozilla.org/en-US/docs/Web/CSS/font-size-adjust
15:55:35 <fizzie> Defined in CSS 2.0, dropped in CSS 2.1, defined again in CSS 3 and implemented by almost nobody.
15:56:18 <fizzie> https://www.w3.org/TR/REC-CSS2/fonts.html#propdef-font-size-adjust has fancy sample images and all.
15:57:47 <mroman> hm
15:57:58 <mroman> but if the parent element has it's font-size defined in px it behaves differently again
15:58:04 <fizzie> Yes.
15:59:44 <mroman> I can see that it somehow makes sense that browsers do this
16:00:07 <mroman> different fonts look different for different font-sizes
16:00:26 <mroman> so having a reference default font-size for different fonts kinda makes sense
16:00:41 <mroman> that way a user can choose what default font-size pre should have
16:01:28 <impomatic_> Fonts are bugging me at the moment :-(
16:01:28 <mroman> it just looks a bit off if you use it inline :)
16:01:45 <mroman> because then suddenly that thing in <code> is only half as high as the rest of the text
16:02:23 <mroman> but I'll use tho double monospace hack for this then
16:02:33 <impomatic_> My main content and sidebar use the same font size. The text renders the same size on Chrome for Windows, but the sidebar text is bigger than the main text on Chrome for Android :-( http://corewar.co.uk
16:03:05 <fizzie> https://googledrive.com/host/0B4J9OAzXNfZAQy1NWEVHUURaVmM on my browser -- the 13px monospace looks much closer to the 16px proportional than the "same size" 16px monospace does. (The difference isn't quite as dramatic on whatever fonts and defaults Iceweasel happens to go with, where the monospace indeed looks a bit "too small".)
16:04:20 <fizzie> https://googledrive.com/host/0B4J9OAzXNfZAeGQzbHduMG1CajQ -- yeah, on that thing it looks more than a bit too small.
16:06:12 <fizzie> It does 16/12 by default (instead of Chrome's 16/13) and manages to have a much "larger-looking" proportional font.
16:10:29 <mroman> your monospace font isn't as wide as mine :)
16:10:41 <mroman> mine's a bit wider.
16:12:40 <fizzie> impomatic_: That's weird.
16:13:21 <fizzie> Also weird: Chrome's remote inspector window for a page in an Android device lacks the "computed" tab for styles.
16:20:14 <fizzie> FWIW, the sizes are identical (but the page overall doesn't look too good) if the sidebar is also position: absolute;.
16:21:50 -!- myname has quit (Ping timeout: 260 seconds).
16:22:31 <fizzie> (Because it will drop out of the body max-width and #content background in that case.)
16:24:13 <impomatic_> stack overflow said I can fix it by putting "max-height: 999999px;" on everything :-)
16:24:49 <fizzie> Sounds very elegant.
16:26:52 <fizzie> I don't really do webdev; I'm sure there's some reasonable-looking CSS'es that would render nicely, but I don't know what they are. (Though having a @media-query rules that would just not use a sidebar at all on "narrow" devices might be worth it even if you can get the sidebar to work. It's a bit unwiedly as-is on a phone.)
16:26:55 -!- myname has joined.
16:32:32 <HackEgo> [wiki] [[Talk:Quiney]] N https://esolangs.org/w/index.php?oldid=46263 * * (+100) /* Working on an interpreter. */ new section
16:40:50 <mroman> I don't do webdev professionally
16:40:56 <mroman> because I'd suck at that.
16:41:04 <mroman> I prefer very simple designs
16:41:06 <Taneb> I changed the CSS on my esolangs page recently
16:41:21 <mroman> http://mroman.ch/exps/1.html <- designs like that
16:42:15 <mroman> but there are some ingenious CSS hacks around there
16:42:24 <mroman> you can completely reorder/rearrange tables with CSS
16:44:26 <mroman> there's js on your esolangs page though
16:44:59 <impomatic_> mroman: that design looks nice and clean
16:47:41 <mroman> It looks pretty crappy on a standard installation of IE though :D
16:47:56 <mroman> at least in my opinion
16:48:07 <mroman> but that's only because it has a sucky default font set
16:48:10 -!- J_Arcane has quit (Ping timeout: 256 seconds).
16:48:20 <mroman> and a very small font-size
16:48:28 <mroman> I'm using 18px as my default font-size in Firefox
16:49:09 -!- J_Arcane has joined.
16:49:38 <mroman> I don't know but I'm having trouble reading font-sizes such as 12,13px on LCD screens
16:50:12 <mroman> and IE's font rendering generally looks a bit smudged/blurry around the edges
16:53:03 <mroman> IE has no font size setting o_O
16:53:09 <mroman> probably uses some windows setting then
17:09:06 <HackEgo> [wiki] [[Talk:Quiney]] https://esolangs.org/w/index.php?diff=46264&oldid=46263 * * (+2179) An interpreter!
17:10:09 <HackEgo> [wiki] [[Talk:Quiney]] https://esolangs.org/w/index.php?diff=46265&oldid=46264 * * (-15)
17:15:03 <myname> http://llvm.moe/ because.
17:17:51 -!- ais523 has joined.
17:21:20 -!- dcentral has joined.
17:22:09 -!- Treio has joined.
17:23:38 -!- mroman has quit (Quit: Page closed).
17:26:24 -!- dcentral has quit (Max SendQ exceeded).
17:26:57 -!- dcentral has joined.
17:31:55 <\oren\> myname: lol, it's a moe compiler development book
17:34:42 <\oren\> what other kinds of graphs should my grph command support?
17:36:12 <\oren\> Maybe labelled bar graphs?
17:39:12 <\oren\> horizontal graphs is probably a good idea
17:46:31 <\oren\> OOH! 2-dimensional scatterplots using the braille characters!
17:56:39 <\oren\> well, horizontal is kinda working, but I should add zebra striping
18:00:49 <J_Arcane> speaking of compiler books, anyone have some good suggestions for non-C books about building compilers.
18:01:17 <J_Arcane> I am currently oppressed by a mad plan, and it involves me being on the other side of knowing how to write one.
18:04:09 <myname> depending on the level, how about the dragon book?
18:05:54 <J_Arcane> I thought the dragon book was in C?
18:06:25 <myname> define "in C"?
18:06:40 <myname> as far as i know, it is not really bound to any language
18:09:50 <Taneb> J_Arcane, there's a book I'm borrowing, which I've forgotten the name of, that I believe does it in C and Haskell
18:10:07 <J_Arcane> myname: I may've had a wrong impression then.
18:11:49 <J_Arcane> Taneb: Cool. I've been impressed thus far with the stuff I have read about language implementation in ML-family langs. I did at one point try to work through the first few chapters of Write Yourself a Scheme but I didn't have enough Haskell skills to get much out of it.
18:13:03 <myname> it's really not that hard
18:13:20 <myname> we wrote a compiler at university
18:13:23 <myname> it was fun
18:15:25 <fizzie> \oren\: 5-level 2D heatmaps with " ░▒▓█".
18:15:42 <Taneb> Sometimes I have to take a step back and realise that a) I'm implementing insertion sort, b) I'm doing it in Agda, and c) I've spent over an hour trying to prove that ¬ (A <= B) => B <= A
18:16:14 <\oren\> ██████████ Trump
18:16:14 <\oren\> ████▏ Kasich
18:16:14 <\oren\> ███▌ Cruz
18:16:14 <\oren\> ███▏ Rubio
18:16:14 <\oren\> ██▋ Christie
18:16:16 <\oren\> ██▌ Bush
18:16:19 <\oren\> █▎ Paul
18:16:21 <\oren\> █▎ Fiorina
18:16:24 <\oren\> ▊ Carson
18:16:26 <\oren\> ▎ Huckabee
18:16:29 <\oren\> ▎ Santorum
18:17:07 <\oren\> New Hampshire polling data
18:17:53 <\oren\> fizzie: good idea, heatmaps
18:19:04 <fizzie> You can also do double-dense horizontal with the quadrants, though you couldn't have labels for them like that.
18:19:32 <fizzie> And the horizontal resolution would be worse.
18:19:37 <Taneb> fizzie, might be useful for comparing two sets of data
18:20:07 <myname> Taneb: truth tables aren't THAT hard
18:20:35 -!- jaboja has quit (Ping timeout: 264 seconds).
18:20:39 <Taneb> myname, <= is less than or equals, not reverse implication
18:20:43 <int-e> > 2^31
18:20:45 <lambdabot> 2147483648
18:20:52 <fizzie> Our compiler course used that book where the compiled language is called "MiniJava", and is really silly.
18:20:54 <myname> ah
18:20:59 <int-e> \oren\: urgh
18:21:02 <fizzie> "System.out.println" is a single lexical token in it.
18:21:25 <\oren\> hehehe
18:21:40 <myname> fizzie: well, in most languages it is called something like print
18:22:08 * Taneb away
18:22:28 <fizzie> Yeah, but the MiniJava language wasn't really all that Java-like (no objects or anything), it just copied "surface forms" like that.
18:22:32 <fizzie> http://www.amazon.com/Modern-Compiler-Implementation-Andrew-Appel/dp/052182060X <- I think it's this one.
18:24:33 <fizzie> http://www.cambridge.org/us/features/052182060X/grammar.html -- hmm, that looks otherwise familiar, except it does have some sort of classes and inheritance.
18:24:49 <fizzie> Maybe I'm remembering that part wrong.
18:25:26 <fizzie> It does have the System.out.println thing, and the other thing where it's still "public static void main" even though public, static and void have no independent meaning.
18:27:51 -!- mauris has joined.
18:29:57 <myname> so it is basically C that looks like java?
18:37:09 * Taneb back
18:38:23 <Taneb> J_Arcane, the book is called "Implementing Programming Languages"
18:42:54 <J_Arcane> fizzie: Hmm, I think the ML version of that book is one of the ones I was looking at.
19:19:06 -!- p34k has joined.
19:19:43 -!- LexiciScriptor has joined.
19:51:23 <\oren\> ██████████ Clinton
19:51:23 <\oren\> ███████▍ Sanders
19:51:23 <\oren\> 0.000000 ▕ 51.200001
19:52:50 <\oren\> this program doesn't do labeling, I did that with paste.
19:53:38 <HackEgo> [wiki] [[FISHQ9+]] N https://esolangs.org/w/index.php?oldid=46266 * Erinius * (+4402) Created page with "'''FISHQ9+''' is a simple interpreted language created by combining the functionality of [[Deadfish]] and [[HQ9+]]. FISHQ9+ has complete backwards compatibility with deadfish..."
19:54:39 <HackEgo> [wiki] [[FISHQ9+]] https://esolangs.org/w/index.php?diff=46267&oldid=46266 * Erinius * (+0) Fixed categories
19:55:33 <myname> what's up with these "i mix up useless languages and call it a day" people?
19:55:34 <HackEgo> [wiki] [[Language list]] https://esolangs.org/w/index.php?diff=46268&oldid=46241 * Erinius * (+14) /* F */
19:57:25 <FireFly> myname: yes, basically C tha looks like Java.
19:58:00 <FireFly> We used the same language (with some minor changes) for the compiler course I took
19:58:32 <HackEgo> [wiki] [[Joke language list]] https://esolangs.org/w/index.php?diff=46269&oldid=46246 * Erinius * (+14)
20:01:04 <HackEgo> [wiki] [[Joke language list]] https://esolangs.org/w/index.php?diff=46270&oldid=46269 * Erinius * (+58) /* General languages */
20:01:13 -!- Veltas has joined.
20:02:19 <ais523> myname: it doesn't even increment the accumulator :-(
20:02:25 <ais523> or maybe it does
20:02:33 <ais523> what does + do in that language?
20:02:34 * ais523 looks
20:03:01 <ais523> right, + is just an alias for i
20:03:09 <myname> ais523: the most useless one is a mixup of brainfuck and hq9+ where the + operator of hq9+ is not present
20:03:10 <ais523> this means it's not useless, which misses half the point of HQ9+
20:03:22 <ais523> that's not useless, it embeds bf
20:03:42 <ais523> it's pointless, rather than useless
20:03:50 <myname> well, it still is more hq9 and bf rather than hq9+ and bf
20:04:18 <ais523> let me think about how I'd mix HQ9+ and BF if I wanted to do it properly
20:04:46 <ais523> I guess I'd try to move in the golfing language direction
20:04:57 <ais523> use BF's 8 characters, encoded as octal digits
20:05:15 <ais523> then have hello-worldy, quiny, etc. behaviour on invalid programs
20:05:21 <ais523> such as < or [[+]
20:05:42 -!- jaboja has joined.
20:06:50 <myname> if i wouldn't always mess up the wiki syntax for talk pages, i'd complain there
20:07:22 <myname> but then again, he probably wouldnkt even read that
20:12:40 -!- jaboja has quit (Ping timeout: 250 seconds).
20:21:51 <\oren\> `1234567890-=][oiytrewqasdfghjkl;'/.,mnbvcxz
20:22:05 <HackEgo> ​/home/hackbot/hackbot.hg/multibot_cmds/lib/limits: line 5: /hackenv/1234567890-=][oiytrewqasdfghjkl;'/.,mnbvcxz: No such file or directory
20:22:12 <myname> i agree
20:26:08 -!- Treio has quit (Quit: Leaving).
20:53:15 -!- jaboja has joined.
20:58:50 <\oren\> `?
20:58:52 <HackEgo> ​? ¯\(°​_o)/¯
20:58:54 <\oren\> ``
20:58:55 <HackEgo> No output.
20:58:57 <\oren\> ``help
20:58:58 <HackEgo> ​/home/hackbot/hackbot.hg/multibot_cmds/lib/limits: line 5: exec: `help: not found
20:59:00 <\oren\> `help
20:59:00 <HackEgo> Runs arbitrary code in GNU/Linux. Type "`<command>", or "`run <command>" for full shell commands. "`fetch <URL>" downloads files. Files saved to $PWD are persistent, and $PWD/bin is in $PATH. $PWD is a mercurial repository, "`revert <rev>" can be used to revert to a revision. See http://codu.org/projects/hackbot/fshg/
20:59:36 <\oren\> `fetch http://www.orenwatson.be/grph.c
20:59:38 <HackEgo> 2016-01-23 20:59:37 URL:http://www.orenwatson.be/grph.c [1835/1835] -> "grph.c" [1]
20:59:56 <\oren\> ``gcc grph.c -o grph
20:59:57 <HackEgo> ​/home/hackbot/hackbot.hg/multibot_cmds/lib/limits: line 5: exec: `gcc: not found
21:00:01 <\oren\> ``cc grph.c -o grph
21:00:02 <HackEgo> ​/home/hackbot/hackbot.hg/multibot_cmds/lib/limits: line 5: exec: `cc: not found
21:00:09 <\oren\> `` cc grph.c -o grph
21:00:09 <fizzie> You want "`` gcc"
21:00:11 <HackEgo> ​<stdin>:1:5: error: expected ‘=’, ‘,’, ‘;’, ‘asm’ or ‘__attribute__’ before ‘.’ token \ compilation terminated due to -Wfatal-errors.
21:00:14 <fizzie> Or cc, for that matter.
21:00:29 <fizzie> Also remember you won't get multiple lines of output.
21:00:35 <\oren\> `` gcc grph.c -o grph
21:00:58 <HackEgo> No output.
21:01:17 <\oren\> `` ./grph <<<"0 1 2 3 4"
21:01:19 <HackEgo> ​ ▂█ \ ▄██ \ ▆███
21:01:29 <\oren\> `` ./grph -h 1 <<<"0 1 2 3 4"
21:01:30 <HackEgo> ​ \ ▎ \ ▌ \ ▊ \ █
21:01:40 <\oren\> `` ./grph 1 <<<"0 1 2 3 4"
21:01:41 <HackEgo> ​ ▂▄▆█
21:02:06 <fizzie> `` sed -i -e 's/int h = 3/int h = 1/' grph.c
21:02:08 <HackEgo> No output.
21:02:22 -!- hppavilion[1] has joined.
21:02:23 <fizzie> `` gcc grph.c -o bin/grph
21:02:27 <\oren\> good idea
21:02:27 <HackEgo> No output.
21:02:59 <fizzie> Although for even more HackEgo compatibility, you might make it (or a wrapper script) take inputs from the command line.
21:03:20 <\oren\> `` ./grph -z 1 <<<"0 1 2 3 4"
21:03:20 <HackEgo> ​[32m [31m▂[32m▄[31m▆[32m█[0m
21:03:27 <\oren\> `` ./grph -zc <<<"0 1 2 3 4"
21:03:28 <HackEgo> ​[44m[32m [31m [32m [31m▂[32m█[0m \ [44m[32m [31m [32m▄[31m█[32m█[0m \ [44m[32m [31m▆[32m█[31m█[32m█[0m
21:03:37 <\oren\> `` grph -zc <<<"0 1 2 3 4"
21:03:37 <HackEgo> ​[44m[32m [31m▂[32m▄[31m▆[32m█[0m
21:04:12 <\oren\> `` grph -c <<<"0 1 2 3 4"
21:04:13 <HackEgo> ​[44m ▂▄▆█[0m
21:04:20 <\oren\> `` grph -cl <<<"0 1 2 3 4"
21:04:20 <HackEgo> ​[44m ▂▄▆█[0m 4.000000 0.000000
21:04:34 <\oren\> options seem to work ok
21:04:54 <fizzie> The colors might be more portable with mIRC-style ^c escapes.
21:06:36 <\oren\> i don't know those
21:07:12 <\oren\> the blue color is mainly to show where the top and bottom of the graph is
21:07:32 <\oren\> the red and green show the bounda between values
21:08:35 -!- jaboja has quit (Ping timeout: 264 seconds).
21:08:39 <fizzie> Yes, it's very festive.
21:08:45 <fizzie> Although the holiday season is already over.
21:12:35 <fizzie> `` printf 'they \x033,2go \x035like \x033this\x03 -- that is to say, byte 03 followed by decimal "F" or "F,B" for foreground or foreground/background colors, with a somewhat different color palette, and a plain \\x03 without digits to reset.'
21:12:35 <HackEgo> they go like this -- that is to say, byte 03 followed by decimal "F" or "F,B" for foreground or foreground/background colors, with a somewhat different color palette, and a plain \x03 without digits to reset.
21:13:46 -!- dcentral has quit (Ping timeout: 250 seconds).
21:13:57 <FireFly> It's also worth mentioning that the control bytes are mnemonic: ^Color, ^Original, etc
21:14:22 <fizzie> Yes. (So ^o also resets the color, among other formatting.)
21:16:03 <\oren\> but it appears that hackego alread translates ANSI to mIRC colors
21:16:50 <fizzie> No, it doesn't.
21:17:05 <fizzie> Something else along the way from HackEgo might, but HackEgo doesn't.
21:17:32 <fizzie> They're ANSI escapes in my client logs.
21:17:37 <\oren\> wait, irssi jsut lets ANSI sequences through?!?!
21:17:55 <fizzie> I seem to recall it has a whitelist.
21:18:16 <fizzie> For "safe" escapes.
21:19:05 <\oren\> `` printf '\e[Afoo\e[Bbar'
21:19:06 <HackEgo> ​[Afoo[Bbar
21:19:40 <fizzie> Or possibly even, instead of passing them through, it might've parsed them into some intermediate form, and then back again to ANSI escapes by the text frontend.
21:19:46 <FireFly> \oren\: yes, I believe irssi accepts both mIRC and ANSI escape sequences
21:19:54 <FireFly> It looks very messy for me in weechat FWIW
21:20:11 <\oren\> `` printf 'foo\e[Dbar'
21:20:12 <HackEgo> foo[Dbar
21:20:13 <FireFly> http://xen.firefly.nu/up/2016-01-23_222003.png
21:20:44 <\oren\> `` printf 'foo\e[38,126mbar'
21:20:44 <HackEgo> foo[38,126mbar
21:20:54 <\oren\> `` printf 'foo\x33[38,126mbar'
21:20:55 <HackEgo> foo3[38,126mbar
21:21:02 <FireFly> Don't you want a semicolon rather than comma?
21:21:08 <\oren\> `` printf 'foo\e[38;126mbar'
21:21:09 <HackEgo> foo[38;126mbar
21:21:10 <fizzie> Yes, the strip_codes function just parses them into a thing.
21:21:19 <fizzie> It also has a 24-bit truecolor escape code, which I didn't know about.
21:21:31 <\oren\> `` printf 'foo\e[3mbar'
21:21:32 <HackEgo> foo[3mbar
21:21:51 <\oren\> `` printf 'foo\e[7mbar'
21:21:51 <HackEgo> foo[7mbar
21:22:03 -!- idris-bot has quit (Quit: Terminated).
21:22:10 <\oren\> `` printf 'foo\e[7;1;42mbar'
21:22:10 <HackEgo> foo[7;1;42mbar
21:22:20 <\oren\> `` printf 'foo\e[7;1;42;30mbar'
21:22:20 <HackEgo> foo[7;1;42;30mbar
21:22:29 <\oren\> `` printf 'foo\e[7;1;32;40mbar'
21:22:29 <HackEgo> foo[7;1;32;40mbar
21:22:42 <\oren\> `` printf 'foo\e[7;32;40mbar'
21:22:43 <HackEgo> foo[7;32;40mbar
21:22:53 <FireFly> fizzie: yeah, the 24-bit one is from Konsole I believe, but has become rather popular nowadays
21:23:13 -!- oerjan has joined.
21:23:51 -!- idris-bot has joined.
21:25:26 <hppavilion[1]> I'm building a GCC cross-compiler with cygwin
21:25:28 <\oren\> `` printf 'foo\e[38;2;0;191;255mbar'
21:25:28 <HackEgo> foo[38;2;0;191;255mbar
21:25:29 <hppavilion[1]> I have touched grace
21:25:48 <fizzie> FireFly: I mean, yes, terminals have that, but irssi also has an irssi-specific format for that, \x04#<something>.
21:26:09 <\oren\> either irssi or my teminal doesn't have the ansi 24 bit sequence
21:26:23 <FireFly> fizzie: oh. huh.
21:26:28 <fizzie> I don't think it supports the escape sequence form.
21:26:39 <fizzie> Though I didn't read get_ansi_color that closely.
21:27:04 <fizzie> The custom one is only compiled in if TERM_TRUECOLOR is set, which seems to need a not-on-by-default --enable-true-color to configure.
21:27:06 <\oren\> `` printf 'foo\e[38;5;88mbar'
21:27:07 <HackEgo> foo[38;5;88mbar
21:27:26 <\oren\> `` printf 'foo\e[48;5;88mbar'
21:27:27 <HackEgo> foo[48;5;88mbar
21:27:50 <\oren\> `` printf 'foo\e[5mbar'
21:27:50 <HackEgo> foo[5mbar
21:27:54 <\oren\> bah
21:28:02 <\oren\> `` printf 'foo\e[6mbar'
21:28:02 <HackEgo> foo[6mbar
21:28:07 <J_Arcane> goddammitalltohell, I hate when I do that. Always bookmark a link you want to see again.
21:28:32 <\oren\> I always bookmark or download
21:28:45 <fizzie> Oh, it does support the escape form too, but that also needs the TERM_TRUECOLOR flag on.
21:28:55 <fizzie> The blinking works fine for me, incidentally.
21:29:01 <fizzie> Well, "fine" is maybe not the right word.
21:29:04 <J_Arcane> I found a book on writing compilers in F# the other night, and now I can't find it again.
21:29:16 <J_Arcane> I miss full-text history searching. :(
21:29:28 <fizzie> They're all decomposed to irssi's own flags (GUI_PRINT_FLAG_BLINK and so on), so it's not just passing escape codes through.
21:30:05 <hppavilion[1]> J_Arcane: If you do find it, could you send me the linky?
21:30:11 <J_Arcane> Happily.
21:30:25 <hppavilion[1]> Thanks
21:30:37 <hppavilion[1]> All I need to do now is learn F# xD
21:32:29 <J_Arcane> Hah! Found it. Retracing my steps found this: http://www.itu.dk/~sestoft/plc/
21:35:14 <J_Arcane> I also discovered the full text of Bornat's "Understanding and Writing Compilers: A do-it-yourself guide" from 1977. It's in BCPL!
21:35:52 <\oren\> yay bcpl!
21:37:02 <fizzie> Hm. It's supposed to read 24-bit colors in all cases and interpolate (with quite a lot of code, in fact) back to the 256-color format. But I can't make that work.
21:38:04 <fizzie> Oh well. I'm using this rxvt-unicode anyway, and it does the 88-color thing, which always confuses everything.
21:38:48 <J_Arcane> BCPL is officially 50 years old this year.
21:38:59 -!- impomatic_ has quit (Quit: http://corewar.co.uk).
21:39:13 <oerjan> before c's parent language
21:41:16 <J_Arcane> https://projects.exeter.ac.uk/BCPL/WindowsBCPL.htm
21:42:15 <oerjan> i think the neighbors are partying about twice a week.
21:44:05 <fizzie> ls
21:44:08 <fizzie> Whoops.
21:44:41 <oerjan> i think there are no files in the channel. unless someone smuggled one in with a cake.
21:45:29 * mauris ----###
21:46:03 <oerjan> that file looks suspiciously like a swatter
21:46:16 -!- Sgeo has quit (Ping timeout: 250 seconds).
21:48:30 <hppavilion[1]> [Ugh
21:48:30 <hppavilion[1]> http://pastebin.com/mvJa36Sa
21:48:43 <hppavilion[1]> Can't get binutils to compile with make
21:48:48 <hppavilion[1]> Or whatever the correct thing to say would be
21:48:59 <hppavilion[1]> I type make and it spits out the above error
21:49:08 <ais523> fizzie: I'm not aware of any terminal emulator but Konsole that does 24-bit color
21:49:14 <ais523> and actually renders it as such
21:51:02 -!- jaboja has joined.
21:51:57 <\oren\> yay I won a tournamant in angry birds
21:53:00 <mauris> > (\d -> text [" ▁▂▃▄▅▆▇█" !! (round ((x - m) / (maximum d - m) * 8)) | let m = minimum d, x <- d]) $ map (cos.(/2)) [0..32]
21:53:02 <lambdabot> ██▆▄▂▁ ▁▃▅▇██▇▅▃▂ ▁▂▄▆▇██▆▅▃▁
21:53:32 <mauris> > (\d -> text [" ▁▂▃▄▅▆▇█" !! (round ((x - m) / (maximum d - m) * 8)) | let m = minimum d, x <- d]) $ map (^2) [0..32]
21:53:34 <lambdabot> ▁▁▁▁▁▂▂▂▂▃▃▃▃▄▄▄▅▅▆▆▇▇██
21:54:01 <oerjan> darn int-e is probably not here
21:55:18 <\oren\> Ok so my terminal only supports the 38;5;Nm sequence
21:55:23 <fizzie> ais523: https://gist.github.com/XVilka/8346728#now-supporting-truecolour lists 8 + everything on libvte since 0.36.
21:55:39 <ais523> thanks
21:56:07 <oerjan> @ask int-e does lambdabot have a permanent function for single line function graphs or was that just something someone made on the fly
21:56:08 <lambdabot> Consider it noted.
21:56:32 <fizzie> I can verify that at least 'st' does; don't know of the rest.
21:56:45 <\oren\> liars! mintty doens't support 38;2;R;G;B!
21:57:10 <fizzie> I kind of considered switching to st, but it's got that "recompile it to configure" thing, and I'd rather install it from the Debian package.
21:57:11 <\oren\> at least, not on windows
21:57:36 <\oren\> anyway I like pietty
21:58:01 <fizzie> Also it doesn't support at all the "make 'bold' stuff just bright, don't switch to a bold font" option, which I've already gotten quite used to.
21:58:34 <\oren\> wow that sucks
21:58:42 <\oren\> even putty supports that iirc
21:59:02 <fizzie> rxvt-unicode "supports" it by letting you explicitly define the font used for bold.
21:59:20 <mauris> comic sans
21:59:51 <fizzie> Wonder if fontconfig can do per-program rules. It's complicated enough to make that a possibility.
22:01:03 <fizzie> Well, there's the FONTCONFIG_FILE envvar.
22:01:08 <fizzie> Gets a bit complicated that way.
22:02:09 <fizzie> Also the "light gray" color is too bright, and they don't believe in resources, just in a config.h.
22:03:57 <fizzie> Truecolor in a terminal looks weird anyway. Like if you had a black-and-white comic and then some parts were in color.
22:04:53 <\oren\> that's really common though
22:05:12 <\oren\> most comics have a color part in the first few pages
22:05:43 <\oren\> or sometimes just the first page
22:06:00 -!- graue has joined.
22:06:25 <fizzie> I meant more like within an otherwise black-and-white panel.
22:06:36 <fizzie> Gah, these colors look so garish.
22:06:55 <ais523> fizzie: xkcd does that sometimes, it doesn't look too bad
22:07:39 <fizzie> And the block-drawing characters don't form solid surfaces, there are thin lines showing the character cells.
22:07:51 <\oren\> agh
22:08:08 <\oren\> well that's partially dependent on your font
22:08:19 <oerjan> hi graue!
22:08:28 <fizzie> Yes, but the same font works just fine in rxvt-unicode.
22:10:02 -!- Sgeo has joined.
22:10:07 <\oren\> urgh. sounds like they haven't done much UAT
22:11:35 <FireFly> <fizzie> ais523: https://gist.github.com/XVilka/8346728#now-supporting-truecolour lists 8 + everything on libvte since 0.36. ← also anything based on libvterm (which is basically pangoterm and neovim's built-in terminal, AFAIK), and ToaruOS's terminal emulator
22:12:05 <\oren\> ABCDEF
22:12:18 <ais523> what about PuTTY? thats a big one
22:12:32 <ais523> ah right, it doesn't
22:12:37 <ais523> there's a list of terminals that don't too
22:12:52 <fizzie> https://googledrive.com/host/0B4J9OAzXNfZAWnRpMTBfV0JEUE0
22:12:54 <\oren\> nope. although pietty supports the 38;5;Nm
22:13:30 <int-e> oerjan: that looked like self-contained code to me, fwiw
22:13:59 <ais523> also I assume there's no way to write a truecolor code that will be ignored by all commonly used terminals that don't support it?
22:14:08 <ais523> the colons version of the code might work for that I guess
22:14:24 <fizzie> Maybe, but the colons version seems less supported as well.
22:14:37 <FireFly> If anything, then it'd be the colon version I think
22:14:39 <oerjan> int-e: not what mauris did, something i remember from earlier
22:15:08 <oerjan> it had lines not blocks
22:15:11 <\oren\> fizzie: duuuude you have a ginormous monitor
22:15:28 <fizzie> It's just 1920 by 1200, that's not big these days.
22:16:01 <FireFly> oh, those look nice
22:16:08 <fizzie> They're selling 3840x2160 ones now.
22:16:11 <fizzie> For $500.
22:16:22 <FireFly> I have a line-plot thingy in J using braille
22:16:27 <FireFly> Hmm
22:16:38 <\oren\> well it looks big on my 1366x768
22:17:04 <int-e> oerjan: doesn't ring a bell
22:17:09 <FireFly> [ plot sin i:4j80 NB. does this work?
22:17:10 <j-bot> FireFly: |value error: sin
22:17:10 <j-bot> FireFly: | plot sin i:4j80
22:17:16 <FireFly> [ plot 1&o. i:4j80 NB. does this work?
22:17:17 <j-bot> FireFly: ⠢⢄⡀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⣀⠤⠔⠒⠉⠉⠉⠉⠑⠒⠤⣀⠀⠀⠀⠀⠀⠀
22:17:17 <j-bot> FireFly: ⠀⠀⠈⠑⠢⢄⡀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⢀⡠⠒⠉⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠉⠒⠤⣀⠀⠀
22:17:17 <j-bot> FireFly: ⠀⠀⠀⠀⠀⠀⠈⠑⠢⠤⣀⣀⣀⣀⡠⠤⠒⠊⠁⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠉⠒
22:17:21 <FireFly> Neat
22:17:29 -!- jaboja has quit (Remote host closed the connection).
22:17:41 <fizzie> I think it's a bit sad that they went all 16:9 instead of 16:10, because 16:10 lets you have a full-width 16:9 video + some space left over for controls.
22:18:26 <\oren\> works
22:18:41 -!- bb010g has quit (Quit: Connection closed for inactivity).
22:18:45 <FireFly> fizzie: hmm, good point
22:19:29 <fizzie> FireFly: I guess the argument is that you won't keep the controls visible all the time anyway, because they'd be distracting, so if they need to be auto-hide anyway, they can just be on top.
22:20:44 <tswett> [ plot 1&o. i:4j80
22:20:45 <j-bot> tswett: ⠢⢄⡀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⣀⠤⠔⠒⠉⠉⠉⠉⠑⠒⠤⣀⠀⠀⠀⠀⠀⠀
22:20:45 <j-bot> tswett: ⠀⠀⠈⠑⠢⢄⡀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⢀⡠⠒⠉⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠉⠒⠤⣀⠀⠀
22:20:45 <j-bot> tswett: ⠀⠀⠀⠀⠀⠀⠈⠑⠢⠤⣀⣀⣀⣀⡠⠤⠒⠊⠁⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠉⠒
22:20:57 <tswett> [ plot 1&o. i:2j80
22:20:58 <j-bot> tswett: ⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⣀⣀⠤⠤⠔⠒⠒⠊⠉⠉⠉⠉⠉⠉⠉⠉
22:20:58 <j-bot> tswett: ⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⢀⣀⡠⠤⠒⠒⠉⠉⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀
22:20:58 <j-bot> tswett: ⣀⣀⣀⣀⣀⣀⣀⣀⣀⠤⠤⠤⠒⠒⠊⠉⠁⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀
22:21:11 <tswett> [ plot 1&o. i:40j80
22:21:12 <j-bot> tswett: ⠀⠠⢁⠀⢀⠡⠀⠀⠒⠀⠀⠊⠀⠀⡈⡀⠀⠈⠂⠀⠐⠁⠀⠠⢁⠀⢀⠡⠀⠀⠑⠀⠀⠊⠀⠀⠌⡀⠀⠈
22:21:12 <j-bot> tswett: ⠀⡀⠀⠀⠀⠀⡀⠐⠀⠂⠠⠀⠁⠀⠀⠀⠀⠁⠠⠀⠂⠐⠀⡀⠀⠀⠀⠀⡀⠐⠀⠄⠠⠀⠁⠀⠀⠀⠀⠁
22:21:12 <j-bot> tswett: ⢄⠀⠀⢁⠂⠀⢀⠄⠀⠠⡀⠀⠐⡈⠀⠈⡐⠀⠀⡠⠀⠀⢄⠀⠀⢁⠁⠀⢀⠄⠀⠠⠄⠀⠐⡈⠀⠈⡐⠀
22:21:21 <FireFly> very high-frequency
22:21:30 <zzo38> Still I think it would help to have a separate space for the OSD so that it does not overlap the movie picture
22:21:31 <FireFly> [ i:4j80
22:21:32 <j-bot> FireFly: _4 _3.9 _3.8 _3.7 _3.6 _3.5 _3.4 _3.3 _3.2 _3.1 _3 _2.9 _2.8 _2.7 _2.6 _2.5 _2.4 _2.3 _2.2 _2.1 _2 _1.9 _1.8 _1.7 _1.6 _1.5 _1.4 _1.3 _1.2 _1.1 _1 _0.9 _0.8 _0.7 _0.6 _0.5 _0.4 _0.3 _0.2 _0.1 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 2 2.1 2.2 2.3 2.4 2.5 2.6 2.7 2.8 2.9 3 3.1 3.2 3.3 3.4 3.5 3.6 3.7 3.8 3.9 4
22:21:35 <tswett> [ plot 1&o. i:1639j80
22:21:36 <j-bot> tswett: ⠂⠁⠁⠁⠁⠂⠄⠀⠀⠀⠠⠐⠐⠈⠈⠈⠐⠠⢀⠀⠀⠀⡀⠄⠂⠁⠁⠁⠂⠄⡀⠀⠀⢀⠠⠐⠈⠈⠈⠐
22:21:36 <j-bot> tswett: ⠀⠀⠀⠀⠀⠀⠀⢁⠒⡈⠀⠀⠀⠀⠀⠀⠀⠀⠀⡐⠢⠁⠀⠀⠀⠀⠀⠀⠀⠀⢀⠡⠌⡀⠀⠀⠀⠀⠀⠀
22:21:36 <j-bot> tswett: ⠠⢀⢀⢀⠠⠐⠈⠀⠀⠀⠁⠂⠄⡀⡀⡀⠄⠂⠁⠀⠀⠈⠐⠠⢀⢀⢀⠠⠠⠐⠀⠀⠀⠀⠂⠄⡀⡀⡀⡀
22:21:39 <FireFly> Might help to see what's happening
22:22:00 <fizzie> Aliasing!
22:22:00 <FireFly> i:4 is _4 _3 _2 _1 0 1 2 3 4
22:22:35 <FireFly> bit hacky, but if the parameter is complex then the imaginary part gives the #elements to use instead of the default amount
22:22:37 <tswett> So what number do you have to choose so that it makes one phith of a full circle each time?
22:23:05 <tswett> [ plot 1&o. i:4j160
22:23:06 <j-bot> tswett: ⠢⠤⠤⣀⣀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⣀⣀⠤⠤⠔⠒⠒⠊⠉⠉⠉⠉⠉⠉⠉⠉⠑⠒⠒⠢⠤⠤⣀⣀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀
22:23:06 <j-bot> tswett: ⠀⠀⠀⠀⠀⠉⠉⠒⠒⠤⢄⣀⡀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⢀⣀⡠⠤⠒⠒⠉⠉⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠉⠉⠒⠢⠤⢄⣀⡀⠀⠀⠀⠀
22:23:06 <j-bot> tswett: ⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠈⠉⠑⠒⠒⠤⠤⠤⣀⣀⣀⣀⣀⣀⣀⣀⣀⠤⠤⠤⠒⠒⠊⠉⠁⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠈⠉⠑⠒⠒
22:23:25 <tswett> Well... that looks kind of interesting.
22:23:33 <tswett> [ plot 1&o. i:4j4
22:23:34 <j-bot> tswett: ⠂⠈
22:23:34 <j-bot> tswett: ⠀⠂
22:23:34 <j-bot> tswett: ⢀⠀
22:23:38 <tswett> [ plot 1&o. i:4j8
22:23:39 <j-bot> tswett: ⠂⠀⠈⠁
22:23:39 <j-bot> tswett: ⠠⠀⠂⠐
22:23:39 <j-bot> tswett: ⠀⣀⠀⠀
22:23:46 <tswett> Parfait
22:23:47 <mauris> i can't see anything :(
22:23:58 <\oren\> needs to have vertical lines
22:24:04 <fizzie> It's braille -- maybe try touching your screen?
22:24:14 <FireFly> \oren\: right.. never bothered with that :\
22:24:30 <tswett> This j-bot functionality should be perfect for people like mauris who can't see anything!
22:24:39 <\oren\> mauris: try using a font that has braille
22:26:12 <\oren\> ms gothic has braille
22:26:25 <\oren\> ... for some reason
22:26:28 <zzo38> Which aspect ratios do you think I should need to be supported in Digi-RGB? At first I had only 4:3 and 16:9 but now I can think also to include 5:4 and 16:10 as well; do you expect that is sufficient?
22:26:43 <FireFly> What is Digi-RGB?
22:27:11 <zzo38> A new kind of video interface connection
22:27:12 <\oren\> er, no wait, it's falling back. uh...
22:28:25 <FireFly> zzo38: what is it meant for?
22:28:38 <FireFly> what are the design goals? it's hard to answer your question without knowing
22:30:06 <zzo38> The design goal is to be simple and one-way communications for digital RGB signal; the signal is four red, four green, four blue, one clock, one sync, and each pixel is two clock cycles. Electrical specification and physical specification are not yet written though, only the data is. It is meant to be alternative of VGA and HDMI and DisplayPort (because HDMI is stupid)
22:32:03 <FireFly> Hmm, well it's probably good to include 5:4 and 16:10
22:32:12 * oerjan suddenly realizes someone might think the channel topic refers to a misspelled candidate
22:33:29 <zzo38> There is also Digi-RGB-Plus which adds stereo analog audio and one reverse signal for commands, but it is a requirement that you can use a cable between Digi-RGB and Digi-RGB-Plus in both ways and that it is still guaranteed to work if the command signal is not used or if some commands are not implemented.
22:33:49 <int-e> oerjan: well, it doesn't say "Tromp for president!" ... yet!
22:35:05 <int-e> The depressing files were not so popular, I guess.
22:35:23 <zzo38> FireFly: Yes but do you expect these four aspect ratios are sufficient?
22:36:31 <FireFly> Probably. At least, I can't think of any other common aspect ratio off the top of my head
22:36:40 <FireFly> At least for desktop/laptop use
22:36:57 <FireFly> With portable game consoles/phones I imagine it could vary a bit more
22:37:06 <zzo38> It could be used with television as well as computer
22:43:58 <fizzie> The DCI 2K/4K movie standards have an 256:135 aspect ratio (4096x2160 for the 4K variant), but I don't know if that's very relevant for monitors. I did some with that resolution for sale, though.
22:44:58 <zzo38> For simplicity I do not want to add too many aspect ratios, and anyways the picture could be letterboxed if necessary.
22:46:09 <tswett> Use "end of row" and "end of frame" markers in order to support any aspect ratio!
22:46:12 <tswett> NB: may be a bad idea.
22:46:21 <zzo38> In addition, some aspect ratios cannot be added, because it is necessary for each aspect ratio to have a different squarefree core.
22:47:21 <zzo38> The "end of frame" marker already exist it is what I called the "sync" signal above
22:51:32 <tswett> I just thought of the perfect response to mauris.
22:51:48 <tswett> "It's Braille; I don't understand the problem."
22:51:56 -!- jaboja has joined.
22:52:29 <mauris> ha
23:07:43 <zzo38> Do you do Magic: the Puzzling? I hope you do, because I would need help with the codex
23:12:45 -!- jaboja has quit (Remote host closed the connection).
23:17:12 * oerjan briefly imagines trump vs. sanders
23:22:09 <ais523> oerjan: the possibility of that is definitely significantly nonzero
23:22:14 <ais523> err, *probability
23:22:29 <ais523> however, that doesn't mean it's all that high
23:24:03 -!- Phantom__Hoover has joined.
23:24:16 -!- singingb1yo has quit (Ping timeout: 256 seconds).
23:27:49 <\oren\> Trump vs. sanders would be the most ideologically polarized election since like, i dunno... did Hitler even run against a communist?
23:27:49 -!- Phantom_Hoover has quit (Ping timeout: 276 seconds).
23:28:10 <\oren\> s/even/ever/
23:28:37 <tswett> Huh. Opal Titan as printed says it activates whenever an opponent "successfully casts a creature spell".
23:28:51 <tswett> Which raises the obvious question of what "successfully casts" means.
23:29:46 <tswett> Like, what does "successfully throw a ball at Tom's head" mean? Does it merely mean that you've succeeded in thrown the ball, or does it mean you've actually hit Tom's head?
23:29:54 <zzo38> I think "successfully cast" is the last step of casting a spell
23:30:01 <tswett> The answer, it turns out, is that "successfully casts" means "casts".
23:30:54 <\oren\> so it activates even if the played creature is immediately removed from the game?
23:31:27 -!- Phantom_Hoover has joined.
23:31:29 <tswett> It activates even if the opponent somehow loses the game immediately after casting the spell.
23:32:00 -!- Phantom__Hoover has quit (Ping timeout: 250 seconds).
23:32:10 -!- singingboyo has joined.
23:32:58 <ais523> can we construct a circumstance in which casting a spell causes a player to lose the game?
23:33:11 <ais523> actually it's easy, you cast a spell where you pay life as an additional or alternative cost
23:33:23 <ais523> you pay /all/ of it and then lose as an SBA just after it finishes casting
23:33:29 <tswett> Yup, there you go.
23:34:06 <tswett> Now, state-based actions are faster than the enstacking of triggered abilities, right?
23:34:27 <tswett> (Or EoTA.)
23:36:03 <ais523> I think they're slower, but am not sure if there's ever a situation where it matters
23:36:19 <ais523> you don't have abilities that say "whenever a player casts a spell, as long as this ability is on the stack, players don't lost the game for having 0 or less life"
23:36:41 <ais523> mostly because most players wouldn't understand them
23:40:59 <\oren\> when I played Magic in middle school, we used a literal stack of instant cards
23:45:02 <\oren\> and occasionally when a creature card's ability was played we pu that whole creature card on the stack
23:53:49 -!- hppavilion[1] has quit (Ping timeout: 276 seconds).
←2016-01-22 2016-01-23 2016-01-24→ ↑2016 ↑all