00:00:02 <elliott> pikhq: How's about building xirssi>?
00:00:45 <Gregor> pikhq: I don't think I ever use the mouse for XChat ...
00:00:54 <elliott> Gregor: How do you change channel?
00:01:00 <elliott> Apart from ctrl+pgup/down.
00:01:02 <Gregor> elliott: Ctrl+Pgup/down
00:01:11 <Gregor> ... not keyboard enough for you?
00:01:22 <Gregor> Want everything to be obscure vim control statements?
00:01:36 <elliott> Gregor: No, I just don't want it based on an arbitrary list :P
00:01:46 <elliott> Having to press more keys to get to a channel just because it's further down the alphabet = lame
00:02:20 <Gregor> Lesse, I have twelve channels on four servers plus one query window open.
00:02:22 <Gregor> Assign each of those to a key.
00:03:23 <pikhq> Gregor: Well, my 19 windows in irssi each have their own key combo.
00:04:30 <elliott> What are the packages in Debian to get the version of autotools that works? x_x
00:06:06 <elliott> pikhq: Feel like seeing if xirssi works? :P
00:07:17 -!- pikhq_ has joined.
00:07:52 <elliott> (sleep 2; scrot -bs ~/xirssi.png)
00:08:00 <pikhq_> Nope, smuxi. Trying to see if I can get it to not-suck.
00:08:09 <elliott> pikhq_: WHAT DO YOU HAVE AGAINST XIRSSI :P
00:08:13 <elliott> It's exactly what you asked for.
00:08:33 <pikhq_> And not in the package manager.
00:08:48 <elliott> pikhq_: Oh noes, you have to compile irssi with --with-glib2 and then compile xirssi
00:08:51 <elliott> That could take minutes :P
00:09:10 <pikhq_> Anyways, it seems that smuxi is very, very unconfigurable.
00:09:22 <elliott> pikhq_: It has pages of config options
00:09:39 <elliott> you're clearly opening the wrong menu or something :p
00:09:57 <pikhq_> I can't even change the font.
00:10:16 <elliott> pikhq_: Screenshot, try different menus.
00:11:23 <pikhq_> It's all... Nice-like.
00:11:40 <pikhq_> ... OH MY FUCKING GOD IT DOESN'T HANDLE MY IME AT ALL.
00:12:03 <pikhq_> THAT MOTHER-FUCKING SUCKS MOTHER-FUCKING ASS.
00:12:17 <elliott> right click on text box etc
00:12:49 <pikhq_> Yes, it's set to the system input method.
00:13:31 <pikhq_> Okay, set it to the Scim bridge method and it works.
00:18:14 -!- pikhq has quit (Quit: leaving).
00:18:17 -!- pikhq_ has changed nick to pikhq.
00:22:07 <pikhq> This is a much more comfortable experience than irssi.
00:24:16 <pikhq> And if I paste in some RTL text, it'll display reasonably.
00:25:38 <Zuu> looks arabic
00:26:51 <Zuu> What does it say?
00:28:35 -!- Sasha has joined.
00:34:27 <pikhq> Aaaand I get leſs shitty rendering of þings þat aren't exactly going to be in your average monoſpace font. Hooray.
00:40:48 <elliott> pikhq: Ooh, I'll help you test your new client! :P
00:42:41 <elliott> Sasha: have you been here before?
00:42:44 <Sasha> how do I drain big basins in Minecraft?
00:43:07 <elliott> we don't have any minecraft players here.
00:43:11 <elliott> only fizzie who played for a total of two days
00:43:27 <Sasha> you might not know that I was a player of it
00:43:51 <elliott> right, well, asking questions about it here is unlikely to work :p
00:45:14 <elliott> Gregor: What's the esotericlogs hg again? x_X
00:48:35 <pikhq> Aaand it seems that GUI Emacs sucks even more than my terminal at font rendering. Who would've guessed?
00:48:43 <pikhq> And it can't even use the system IME.
00:48:53 <pikhq> But because it's Emacs, it *of course* has a builtin IME.
00:49:30 <elliott> pikhq: GUI Emacs is nicer than terminal emacs for like two reasons only :P
00:49:41 <elliott> And I'm unsure what those reasons are, exactly.
00:50:12 <elliott> http://en.wikipedia.org/wiki/Hobo_sign#Hobo_code
00:54:37 <pikhq> I do believe this is the first time I've ever wanted a tab bar on the left side of the window.
00:54:48 <augur> why is hobo code the best thing ever
00:54:51 <elliott> pikhq: tab bar on the left - i.e. buffer list
00:55:10 <elliott> augur: because (1) i had no idea anything like that existed and i'm not sure it even does here
00:55:44 <pikhq> Oh, wait, I wanted that just the other day.
00:55:50 <elliott> pikhq: And every other day.
00:55:59 <elliott> pikhq: NOW HOW DO I MAKE XFWM STOP FOCUSING A WINDOW WHEN I SCROLL IN IT
01:01:54 -!- Mathnerd314 has joined.
01:01:56 <elliott> Hmm. If writing to $browser/location loads a page, how do you reload?
01:02:09 <elliott> touch $browser/location? x=$(cat $browser/location); echo "$x" >$browser/location?
01:02:14 <elliott> (The latter would be super-weird.)
01:04:24 <pikhq> Midori is such a nice browser.
01:04:57 <elliott> pikhq: Except that it freezes all the time.
01:09:51 <pikhq> Problematic, that.
01:14:05 <elliott> pikhq: Kayak = my web browser! It displays web pages already!
01:19:44 <elliott> pikhq: It's going to be filesystem'd up the wazoo. If you've heard of uzbl, like that but with less Arch idiocy.
01:19:57 <pikhq> And with a sane UI?
01:20:09 <pikhq> How're you doing the filesystem thing?
01:20:18 <elliott> <elliott> Hmm. If writing to $browser/location loads a page, how do you reload?
01:20:18 <elliott> <elliott> touch $browser/location? x=$(cat $browser/location); echo "$x" >$browser/location?
01:20:19 <elliott> <elliott> (The latter would be super-weird.)
01:20:21 <elliott> TOTALLY NEED AN OPINION ON THIS
01:20:32 <elliott> pikhq: have you ever looked at what you have to do to create a fuse filesystem?
01:20:47 <pikhq> touch $browser/location, definitely.
01:20:52 <pikhq> It just makes sense.
01:21:28 <pikhq> FUSE may be awful, but what about the alternatives?
01:21:30 <elliott> pikhq: True, it does. But then, having a file identifying a browser's location cause the browser to reload just because its modification time changed? What?
01:21:38 <elliott> Also, I could do it with 9p. linux has support for that. :p
01:22:04 <elliott> pikhq: Or I could just have it as regular files...
01:22:37 -!- augur has quit (Ping timeout: 245 seconds).
01:23:09 <elliott> pikhq: Opinions welcome :P
01:23:56 <pikhq> Gamin is revolting.
01:24:07 <elliott> pikhq: FAM, sure, but gamin?
01:24:09 <pikhq> Compared with, y'know, *just implementing a filesystem*.
01:24:21 <elliott> pikhq: Yes, but it's also a lot less work for me. :p
01:24:30 <pikhq> It seems alright for file notification stuff, but what you want is an actual, honest-to-god filesystem.
01:24:58 <pikhq> It could bite you in the ass later if you don't.
01:25:49 <Mathnerd314> I dunno if filesystems are right... how do you treat it if the server's down?
01:26:01 * elliott tries to understand wtf Mathnerd314 is trying to say
01:26:15 * elliott suspects he misunderstands the plan...
01:26:20 <elliott> pikhq: I'd *like* to expose the entire DOM as a filesystem... but that sounds like hell to do with webkit.
01:26:34 <Mathnerd314> filesystem... something like /http/google.com/ goes to http://google.com
01:26:51 <elliott> Mathnerd314: yeah no not that.
01:27:45 <Gregor> elliott: Maybe you could expose the FUSE API to WebKit's JavaScript and write the FS in JS >: )
01:27:55 <Mathnerd314> elliott: yay, just lost all interest in your project.
01:28:03 <elliott> pikhq: here's all the shit you have to do to get just a plain in-memory filesystem with fusepy: http://code.google.com/p/fusepy/source/browse/trunk/memory.py
01:28:25 <elliott> Mathnerd314: so wait, first you said "Hey, that won't work." and then I said "Well, I'm not doing that." and then you said "Oh, your project sucks then>"
01:28:36 <elliott> Mathnerd314: see the mindboggling inconsistency here?
01:28:55 <Mathnerd314> I just want filesystems to be more like websites, rather than the opposite
01:29:17 <elliott> that sounds like the worst idea i've ever heard.
01:29:22 <elliott> anyway i'm not exposing websites as filesystems
01:29:28 <elliott> or said anything related to that
01:29:34 <pikhq> elliott: So, your complaint is that you need to... Handle filesystem operations?
01:29:52 <pikhq> I'm sorry that filesystems suck.
01:30:19 <elliott> pikhq: Oh yeah, and that filesystem DOESN'T EVEN SUPPORT DIRECTORIES.
01:31:14 <elliott> pikhq: I am sort of tempted just to have $(browser)/ctl where you can send commands, and then have the other files be read-only. :/
01:33:54 <elliott> Wow, yeah, because I have an informational file with some bytes in it that's totally like a web page.
01:34:03 <elliott> Are you sure you understand what I'm trying to do here...?
01:34:18 <pikhq> You seem to not have the foggiest clue.
01:34:27 <pikhq> elliott: That would make FUSE significantly easier.
01:34:40 <elliott> pikhq: It would. It'd also be a copout, but.
01:34:51 <elliott> pikhq: OTOH, it would be obvious what reloading would be. "reload".
01:34:58 <elliott> pikhq: Or even "go (current url)"; at least that's a proper verb.
01:35:14 <elliott> pikhq: OTOOH (other other hand), it'd be like uzbl and uzbl sucks. :)
01:35:56 <elliott> pikhq: I know, I'll implement something else to procrastinate a bit.
01:41:42 <elliott> pikhq: It appears that my kernel has 9p support.
01:43:29 <elliott> pikhq: http://bitbucket.org/f2f/py9p/src/tip/examples/simplesrv.py otoh, it looks more complicated than fuse :p
01:46:48 <elliott> pikhq: Woo, it can now embed into other windows (*cough* tabbed(1) *cough*)
01:49:19 <Mathnerd314> ok, I still stand by my earlier statements as being relevant.
01:50:25 <elliott> it's just a control interface exposed via fs
01:51:00 <Mathnerd314> but there's nothing to control, except the dom and various commands
01:51:34 <Mathnerd314> unless you want a simultaneous GUI along with the fs
01:51:45 <pikhq> You really seem to be missing the point.
01:51:52 <elliott> Mathnerd314: consider everything you click on in the process of using the browser
01:51:55 <elliott> that's an element of control
01:51:58 <elliott> usually it is exposed via gui widgets
01:52:05 <elliott> there is no reason it cannot be exposed via a logically-organised file system.
01:52:49 <Mathnerd314> ok... explain how url completion works in this fs
01:53:46 <elliott> Mathnerd314: ok, i give up, learn the unix philosophy and plan 9 and then ask again :)
01:54:04 <elliott> Mathnerd314: (this is not an admission of my design's inferiority, any more than trying to explain a certain programming language feature to someone who hasn't programmed before is)
01:54:13 <elliott> i might explain later when i'm not busy coding
02:00:19 <elliott> pikhq: WOO MY WINDOW HAS AN ICON
02:01:09 <elliott> <a href="http:/intl/en/options/" class=gb2>even more »</a>
02:01:23 <Mathnerd314> ok, I fail to see how a filesystem would be smaller than curl + a parser
02:04:16 <elliott> Mathnerd314: you *really* need to read up on the unix/plan9 philosophy before trying to understand this, nothing that you're saying has any relevance to the design
02:05:09 <Mathnerd314> ok, wait an hour-2 while I regain my sanity
02:05:57 <elliott> pikhq: Quick, what user agent should I send?
02:06:43 <elliott> pikhq: whatever it is, it probably has to start with Mozilla/5.0 :p
02:07:13 <Gregor> Mozilla/13.5 (Microsoft Brows-o-mat .NET# 2022)
02:07:47 * elliott tries to figure out what makes Google realise I'm not a stupid browser
02:08:01 <elliott> Mozilla/5.0 (X11; U; Linux x86_64; en-gb) AppleWebKit/531.2+ (KHTML, like Gecko) Version/5.0 Safari/531.2+
02:08:39 <elliott> i don't like the idea of exposing all that
02:09:01 <elliott> [Default value: "Mozilla/5.0 (X11; U; Linux x86_64; c) AppleWebKit/531.2+ (KHTML, like Gecko) Safari/531.2+"]]
02:11:14 <elliott> Gregor: Wow. Google actually looks for Mozilla/5.0.
02:11:19 <elliott> Despite EVERY BROWSER EVER claiming to be Mozilla/5.0.
02:11:36 <Gregor> Bots and downloaders don't.
02:11:45 <Gregor> So that's a good way of detecting that the client /is/ a browser.
02:11:46 <elliott> Gregor: But it just gives you a slightly less rich UI if you don't.
02:11:57 <elliott> Gregor: Specifically, "more" is a link, not a JS dropdown menu.
02:11:59 -!- Mathnerd314 has quit (Quit: ChatZilla 0.9.86-rdmsoft [XULRunner 1.9.2.11/20101012113537]).
02:12:06 <elliott> Gregor: Woot, now I get to figure out what qualifies me for the GOOGLE INSTANT LOVE
02:12:45 <Gregor> Mozilla/5.0 (X11; U; Linux i686 (x86_64); en-US) AppleWebKit/534.7 (KHTML, like Gecko) Chrome/7.0.517.41 Safari/534.7
02:13:33 <elliott> Gregor: ...and gives me an older version of the homepage
02:13:40 <elliott> Gregor: Yeah, but I want to not have to embed the version like that.
02:14:03 <elliott> self.webkit_settings.props.user_agent = 'Mozilla/5.0 AppleWebKit KHTML Gecko Chrome Safari'
02:14:04 <Gregor> You should just find all the foo/bar strings of every major browser and paste them together.
02:14:12 <elliott> Gregor: Way ahead of you :P
02:14:30 <elliott> GOOGLE GIVE ME YOUR MOTHERFUCKING NEW LOGO
02:14:36 <Gregor> Firefox does admit to being Firefox. Stupidly, Iceweasel also admits to being Iceweasel.
02:14:58 <elliott> I like how user agents are fucked up.
02:15:04 <Gregor> Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv:1.9.1.12) Gecko/20100911 Iceweasel/3.5.12 (like Firefox/3.5.12) <-- a lot like it, in fact.
02:15:27 <Gregor> What with it being the same software with s/Firefox/Iceweasel/g
02:15:30 <elliott> Gregor: I have a proposal.
02:16:04 <Gregor> I accidentally made WebSplat work on Opera. Now it works everywhere :(
02:16:20 <elliott> Gregor: Any modern, standards-compliant browser -- baseline basically being IE9 in standards compliance mode -- can have the string "sane", with word boundaries around it, in their User-Agent. If a website detects "sane", it shall not detect anything else beyond that.
02:16:48 <elliott> Gregor: ...and when the next generation of web standards comes out, know what that'll be?
02:17:01 <elliott> User-Agent: Safari (sane (compatible; sane/2))
02:17:09 <Gregor> Opera will claim to be sane immediately.
02:17:19 <Gregor> Then if you want to support Opera properly, you'll have to check for sane-but-not-Opera.
02:17:35 <elliott> Gregor: Another rule: Opera may not identify as sane.
02:17:36 <Gregor> Plus, IE9 can only report whether it's sane or not AFTER it's downloaded the file, making the user-agent worthless.
02:17:51 <elliott> Gregor: Congrats, you have made sanity insane.
02:18:39 <elliott> Gregor: when the stack overflow guys found out, via reddit, that their site would willingly copy out a script tag onto their careers side, constituting a massive XSS hole:
02:18:40 <elliott> [[Um, thanks? No one is perfect. Would kindly ask for a white-hat approach in the future and not a firehose from reddit: team@stackoverflow.com]]
02:18:46 <elliott> Gee, guys, it's not easy to prevent against XSS.
02:18:49 <elliott> You have to escape strings and shit.
02:19:05 <elliott> One of our company's cofounders wrote a long article on how to name variables so that it's obvious when it's safe to inline things in a page or not.
02:21:05 <elliott> self.webkit_settings.props.user_agent = 'Kayak Mozilla/6.0 WebKit/9.9.9.9 Safari/600'
02:22:38 <elliott> Kayak (Mozilla/5.0; WebKit)
02:23:20 <elliott> "You are currently viewing Google Mail in basic HTML. Upgrade your browser for faster, better Google Mail"
02:24:30 <pikhq> elliott: You know you want to send the user agent *properly*.
02:24:38 <pikhq> elliott: Kayak (WebKit)
02:24:43 <elliott> pikhq: Dude, I fucking did.
02:24:57 <elliott> pikhq: Google thinks I'm running Browser: The 1995 Edition, gmail doesn't let me in...
02:25:12 <elliott> pikhq: tl;dr no, i'm not sending a reasonable user agent :P
02:25:25 <pikhq> Bug report. Google should know better than user agent scanning.
02:25:32 <elliott> pikhq: Yes, but they don't.
02:25:40 <elliott> pikhq: And half the shitty-fucking web doesn't either.
02:25:47 <elliott> pikhq: Google preach against user agent sniffing but do it anyway.
02:25:52 <pikhq> Motherfucking hell.
02:27:56 <elliott> Kayak (Mozilla/5.0 X11; U; Linux x86_64; en-gb AppleWebKit/531.2+ KHTML, like Gecko Version/5.0 Safari/531.2+)
02:28:23 <pikhq> Could you make it have any more ridiculous claims?
02:29:26 <pikhq> I dunno, add Mozilla/4.0 (compatible; MSIE 6.0; X11; compatible)
02:29:39 <pikhq> Yes, it claims to be Mozilla 4 compatible IE6 compatible on X11.
02:30:31 <pikhq> Hell, stick every one of these strings in there.
02:30:36 <pikhq> http://www.user-agents.org/
02:30:47 <pikhq> And this tea sucks. :(
02:30:58 <elliott> pikhq: I wonder what the fuck Version/5.0 even MEANS
02:31:17 <pikhq> elliott: Netscape 5.0. Honest.
02:31:27 <elliott> pikhq: No, that's Mozilla/5.0.
02:33:35 <elliott> pikhq: It's Version, version 5.0.
02:34:30 <elliott> Kayak (Mozilla/5.0; KHTML, like Gecko; WebKit/531.2+; Safari/531.2+)
02:34:33 <elliott> This better motherfucking work.
02:34:51 <elliott> pikhq: IT APPEARS WE ARE NOT COOL ENOUGH TO GET THE FANCY NEW GOOGLE IMAGES
02:35:28 <elliott> http://www.user-agents.org/ <-- this is amazingly terribly designed
02:35:44 <elliott> pikhq: I can probably get rid of "KHTML, like Gecko", right?
02:36:10 <pikhq> No. That is to convince some stupid things that you're using KHTML.
02:36:23 <elliott> pikhq: No website sniffs for KHTML. I *refuse* to believe that.
02:36:35 <pikhq> I am sure they exist.
02:36:41 <elliott> pikhq: There are also IE-only websites.
02:37:13 <pikhq> That's why you add Mozilla/4.0 (compatible; MSIE 6.0; X11; compatible)
02:37:35 <elliott> pikhq: that makes websites rather reasonably give me the IE-sucks-compatibility page :p
02:38:20 <elliott> pikhq: Version/5.0 makes google images work
02:38:50 <elliott> pikhq: You also need Mozilla/5.0 and Gecko or WebKit or whatever to get google's frontpage at the maximal version.
02:39:16 <pikhq> Couldn't it just... Check for the needed functions?
02:40:21 <elliott> pikhq: Kayak (Mozilla/5.0; Gecko; WebKit/531.2+; Safari/531.2+; Version/5.0)
02:40:24 <elliott> This appears to do the trick.
02:41:19 <elliott> pikhq: Getting rid of the Safari version breaks things though; sweet!
02:42:15 <pikhq> From this I can conclude that we need to commit genocide against morons.
02:42:26 <pikhq> They are the primary reason for everything sucking.
02:43:46 <elliott> pikhq: Is there a fancy Buddhist term for becoming one with everything?
02:44:01 <pikhq> None that I know of.
02:44:16 <pikhq> Though I know very little about Buddhist terminology.
02:44:23 <elliott> pikhq: But I need one. It's going to be the name of the function that sets the user agent to claim to be everything. :)
02:44:24 <Sasha> Achieving Nirvana.
02:44:44 <elliott> Sasha: I mean a fancy term for the act of achieving Nirvana.
02:44:57 <pikhq> hù'kiȳô is Japanese for Buddhism! That's about as far as I go.
02:45:08 <Sasha> Grand Harmonization?
02:45:31 <elliott> can't they just have a word for it like
02:45:39 <elliott> i want my function to be incomprehensibly-named
02:45:42 <elliott> for it does an incomprehensible thing
02:45:54 * elliott notes that Sasha definitely hasn't been in here before with that nick
02:46:26 <pikhq> Bodhi? I know someone IRL with that name...
02:46:47 <Sasha> it's sanskrit for enlightenment or something
02:46:48 * elliott gets the feeling that Sasha is a buddhist :p
02:47:10 <Sasha> I dunno exactly, my sanskrit's a wee bit rusty.
02:47:18 <Sasha> I'm a devout atheist
02:47:28 <elliott> nobody's ever said bodhi here either, so presumably Sasha was referring to my query
02:47:28 <pikhq> My sanskrit consists of a single loanword in Japanese!
02:48:12 <Sasha> I was. Bodhi is the word you want, maybe.
02:48:15 -!- Sasha has changed nick to Gluttony.
02:48:28 <Gluttony> maybe this is the nick you're more familiar with?
02:49:00 <pikhq> Can't even recall any regulars from Phoenix, AZ.
02:49:06 <Gluttony> I was one of the folks that migrated in from an XKCD channel.
02:49:20 <elliott> I can easily name all of those ones though.
02:49:30 <elliott> Because I have them on my HIT LIST because I'm EVIL and CRUEL to EVERYONE.
02:49:48 <Gluttony> Yeah, Branan, Caitlin.. one other, and me.
02:49:52 <elliott> 19:42:15 --- join: Chachi (~WHAT@97-124-45-71.phnx.qwest.net) joined #esoteric
02:50:03 <pikhq> That matches the whois.
02:50:10 <elliott> There was also some CatSandwich guy.
02:50:24 <elliott> I've driven them all away because of my IMMENSE POWERS OF ASSHOLERY apparently
02:50:24 -!- Gluttony has changed nick to Chachi.
02:50:34 <Gregor> elliott: Congratulations!
02:50:36 <Chachi> wasn't sure what nick I was using
02:50:41 -!- Chachi has changed nick to Sasha.
02:50:49 <Sasha> this is the one I like the best, though.
02:51:09 <Sasha> Gluttony's another common one, it's what sin I embody.
02:51:10 <elliott> pikhq: http://sprunge.us/HhYA kayak(1) 0.00000001
02:51:22 <Gregor> A friend of mine made the UNIX interface for XKCD ... I should get him to draw XKCD-hat-guy as a WebSplat character and have him put that in as an April fools joke :P
02:51:28 <elliott> pikhq: It can view Google and go from there. It does nothing else.
02:51:33 -!- yorick has quit (Remote host closed the connection).
02:51:35 <elliott> Gregor: I think you should disown your friend forever. :p
02:52:25 <Gregor> Uhh, yes? As opposed to who?
02:52:36 <elliott> Gregor: Anyone else who likes xkcd in any way :P
02:52:50 <elliott> pikhq: DUDE BEHOLD MY SOFTWARE ENGINEERING SKILLZ
02:52:54 <Gregor> XKCD is occasionally sufferable!
02:52:59 <elliott> pikhq: I CALLED LIKE *SEVEN* API FUNCTIONS
02:53:05 <elliott> Gregor: Maybe a year or two ago.
02:53:12 <Sasha> I like the people that read it more than it, usually
02:53:16 <elliott> I am pretty sure I recall... three mediocre comics in the last year.
02:53:23 <elliott> Sasha: that's funny, the people who read it are generally worse :p
02:53:48 <elliott> Hmm, hey, xkcd recently went over 50% crappy.
02:53:55 <elliott> (400 being around the time it started going really downhill)
02:54:14 <pikhq> elliott: That appears to be a very good argument for higher-level languages right there.
02:54:28 <elliott> pikhq: NOW ANY MORON CAN MAKE A TERRIBLE BROWSER
02:54:39 <elliott> pikhq: If you look carefully that thing even has XEmbed support.
02:54:51 <elliott> pikhq: You have to put the window ID as an argument to Browser though :P
02:54:53 <pikhq> elliott: Sure, it's more complex than it should be, buuuut that's still not bad.
02:55:24 <elliott> pikhq: Just because it doesn't support tabs, windows, URL entry, or keyboard bindings!
02:56:17 <elliott> whoops that code doesn't run :D
02:56:23 <Sasha> elliott: Eh. Some of them are rather good.
02:56:27 <elliott> achieve_nirvana needs a self argument
02:56:30 <elliott> Sasha: the people or the comics?
02:57:02 <elliott> Sasha: How apropos! http://xkcd.com/169/
02:57:39 <elliott> (Is that actually irony? Dare I say so for fear of the hordes of reddit users itching for a chance to quote Alanis Morisette?)
02:57:49 <elliott> *Morissette, google tells me
02:57:53 <Sasha> er, it's a completely fine response
02:58:05 <Sasha> You asked an OR question
02:58:17 <elliott> OR in natural language = either list elements, or XOR
02:58:35 <Sasha> Natlangs are rather screwy
02:58:49 <Sasha> so, to restate my answer, both.
02:58:59 <elliott> Sure they are, but only when you deliberately go out of your way to override your brain's natural processing of them :P
02:58:59 -!- Quadrescence has joined.
02:59:12 <Sasha> Hah, natural processing
02:59:36 <Sasha> if I was to allow it to just do its own thing, there would me much fornication and little rejoicing
02:59:45 <Sasha> instead of little fornication and much rejoicing
03:00:03 <Sasha> "Ooh, opposite gender. Allow them to fornicate with you."
03:00:55 <elliott> Sasha: You'd also have to fornicate with every member of the opposite gender :P
03:01:22 <Sasha> Exactly. And some of them aren't so nice-looking.
03:01:42 <Sasha> which is my hardwiring telling me that they have relatively little disease
03:02:09 <elliott> And if we go past every layer of hardwiring, turns out we don't need to fornicate at all!
03:02:17 <Gregor> "Ooh, same (biological) class? Allow them to fornicate with you."
03:02:17 <Sasha> psychology is a funny thing.
03:02:19 <elliott> pikhq: Apparently my browser supports HTML 5 video except it only plays the audio.
03:02:58 <Gregor> elliott: Uhh, maybe it just doesn't support the particular codec.
03:03:11 <Gregor> Oh, your browser as in the browser you're hacking together :P
03:03:13 <elliott> Gregor: I've tried it on like three examples.
03:03:18 <elliott> Gregor: My browser is currently WebKit.
03:03:21 <elliott> So if WebKit does it it should work :P
03:03:55 <Gregor> Idonno, does it actually ship in source form with H.264 support, or does it require some lower-level system support?
03:03:55 -!- wareya has quit (Read error: Connection reset by peer).
03:04:11 <Sasha> Gregor: Not really picky, so much as societal norms. I'm still not sure of my own sexual orientation.
03:04:29 <elliott> i love how there's this idea that you have to FIGURE OUT YOUR RIGIDLY-DEFINED SEXUAL ORIENTATION
03:04:33 -!- wareya has joined.
03:04:43 <Gregor> elliott: SPEAKING OF SOCIETAL NORMS HYUK
03:04:44 <elliott> "Do I find this person attractive? Like do I *really*? But my sexual orientation says I don't! WHAAT"
03:04:57 <elliott> <Gregor> Idonno, does it actually ship in source form with H.264 support, or does it require some lower-level system support?
03:05:03 <elliott> p. sure i have all the evil things necessary to play h.264 but
03:05:09 <elliott> BECAUSE THEORA IS SO AWESOME RIGHT?
03:05:18 <Sasha> elliott: See: Unsure of own sexual orientation
03:05:18 <Gregor> Well it certainly doesn't ship with theora support.
03:05:23 <Sasha> I'll try anything once.
03:05:28 <elliott> Sasha: Are you unsure of who you're attracted to?
03:05:36 <elliott> If not, who the fuck cares? If so, your brain is broken.
03:05:39 <elliott> Sasha: No, I mean, individual people.
03:06:04 <pikhq> Sasha: Here's a hint: sexual orientation is nowhere near as well-defined as you may think it is.
03:06:10 <elliott> I refuse to believe that you can see someone and not have an impulse reaction as to their attractiveness because that's, like, one of the most fundamental snap-decisions the brain makes.
03:06:14 <pikhq> Like most everything, shit is *complicated*.
03:06:23 <elliott> Gregor: What support does it ship with then? >_<
03:06:38 <Sasha> pikhq: I am aware of this
03:06:43 <Gregor> elliott: Probably only support for relying on the underlying OS to support shit :)
03:07:09 <Sasha> elliott: I'm diagnosed with some autism.
03:07:27 <Sasha> so, in a way, my brain could be considered "broken"
03:07:37 <elliott> Sasha: You and every other fucking person on the internet.
03:07:51 <Gregor> I have not been diagnosed with and do not have autism.
03:07:52 <Sasha> I've got a paper saying so
03:08:00 <elliott> Fun fact: Shit means nothing.
03:08:05 <Sasha> from a registered psychologist with an MD
03:08:27 <elliott> As we all know, psychology is the most reliable, scientific and well-grounded profession on earth.
03:08:33 <Gregor> elliott: You're so awesome sometimes :P
03:08:36 <elliott> And autism the least vaguely-defined mental illness in the universe.
03:08:42 <elliott> Gregor: It's because I have autism
03:09:07 <Sasha> also a couple of really nasty things
03:09:41 <elliott> Sasha: In fact, they thought I had so much autism that I should be put in a mental unit for almost a year!
03:09:43 <Sasha> in addition to this, I tend to supress emotions. It's second nature.
03:09:45 * elliott legally certified insane person
03:10:02 <Sasha> because insane is a lawyer-term
03:10:22 <elliott> Or at least they certainly threatened to make it one.
03:10:28 <elliott> I totally need a certificate
03:11:08 <pikhq> Sasha: The fundamental problem with just going and saying "well, I'm autistic" is this: it's a valid excuse for *hardly anything*.
03:11:28 <pikhq> Shame, too, it'd be handy to have an easy excuse for everything.
03:11:32 <Sasha> I am mentioning it as a cause for an effect.
03:11:47 <Sasha> I never use it as an excuse.
03:11:50 <elliott> when i was younger i would kick up fits whenever anything didn't go my way because i was a terrible person and then just blamed it on assburgers
03:11:55 <pikhq> Not parsing it that way. Actually... I'm parsing it as an almost non-sequitur.
03:12:26 <Sasha> The effect being, not making snap judgements about the attractiveness of people.
03:12:36 <elliott> That is not a characteristic of autism.
03:12:43 <pikhq> Sasha: Being autistic does not in any way change how your boner works.
03:12:49 <Sasha> hrm, I was told it was.
03:12:53 <elliott> It's not even your, ahem, boner :P
03:12:59 <elliott> I'm sure even asexual people do it.
03:13:06 <elliott> They make the snap decision, they just don't feel any compulsion to act on it or anything.
03:13:24 <elliott> Pretty sure it's one of the most basic recognitions the brain makes.
03:13:31 <Sasha> However, I lead with my head instead of my heart. Emotion tends to get in the way of reasoning and logic.
03:13:32 <pikhq> elliott: Yeah yeah yeah.
03:13:40 <elliott> Sasha: omg ur like mr. spak
03:14:02 <pikhq> Sasha: Currently, I'm tempted to say you're either asexual or full of shit.
03:14:26 <Sasha> Well, seeing as how the emotions I usually feel are horrible terrible things as defined by society, I tend not to express them.
03:14:37 <Sasha> pikhq: Probably the first one.
03:14:52 <elliott> "But I can't express my inner paedophilia ;__;"
03:15:10 <elliott> Oh! #esoteric! Don't stop! Ohh!
03:15:10 <Sasha> But, you can make your own judgements. I won't tell you what to think.
03:15:12 <elliott> --Gregor's next, untyped line
03:15:14 <pikhq> elliott: You know full well it's pædophilia.
03:15:38 <Sasha> No, elliott, it's more like severe anger and depression.
03:15:55 <Gregor> Your search - wikipædophelia - did not match any documents. D-8
03:16:11 <elliott> Gregor: http://encyclopediadramatica.com/Wikipedophile
03:16:18 <pikhq> Wikipædophilia ← There.
03:17:06 <Gregor> I think that not being able to spell "pædophilia" counts as a win :P
03:17:12 <pikhq> Þou fail'ſt ſo hard.
03:17:25 <elliott> "I hate paedophaeila so much, I can't even type it!"
03:17:27 <Sasha> the a-e digraph looks rather silly in monospace
03:17:30 <elliott> "Why, does it disgust you that much?"
03:17:31 -!- zzo38 has joined.
03:17:34 <elliott> "No, I'm just terrible at spelling it."
03:17:50 <pikhq> Sasha: Ligature, not digraph, and that's your problem, isn't it?
03:17:54 <Sasha> however, I like the thorn and long-s.
03:18:10 -!- Mathnerd314 has joined.
03:18:11 <Gregor> http://encyclopediadramatica.com/File:So_cool.jpg <-- in a situation with any relevance whatsoever, this would be hilarious.
03:18:29 <Sasha> pikhq: Still looks silly, like an a and an e smooshed together.
03:18:38 <elliott> Gregor: Yeah referencing xkcd irl is hilarious :p
03:18:45 <zzo38> Did someone in here have a question about whether or not Enhanced CWEB can be used to create webpages?
03:18:46 <Gregor> elliott: ... referencing XKCD?
03:18:55 <elliott> Gregor: I *envy* your xkcd ignorance.
03:18:58 <elliott> Gregor: http://xkcd.com/285/
03:19:00 <pikhq> Saſha: Problem þine.
03:19:12 <elliott> Gregor: Also: click any "citation needed" link on Wikipedia. See that comic.
03:19:31 <Gregor> elliott: As with reddit and digg, I use people on the web as my XKCD filter :P
03:19:43 <Sasha> pikhq: It is. I am pointing out an opinion.
03:19:55 <elliott> Gregor: But it is also ridiculously super-famous :P
03:20:04 <elliott> To the point of being internet-iconic.
03:20:04 <pikhq> elliott: That [citation needed] picture is from a #xkcd meetup, I'm pretty sure.
03:20:25 <Gregor> elliott: I've PROBABLY seen it before.
03:20:26 <Sasha> or even a normal xkcd meetup
03:20:33 <Sasha> not necessarily the channel
03:20:46 <Gregor> elliott: I think my terrible memory is a blessing :P
03:21:11 <elliott> pikhq: WANNA WRITE FUSE CODE?
03:21:12 <Sasha> I need to start using fun letters too.
03:21:27 <pikhq> I think I'm adopting long s into my romanisation scheme for no good reason. Other than that it looks nice.
03:21:35 <Gregor> elliott: YOU SHOULD EXPOSE THE FUSE API TO JAVASCRIPT THEN WRITE THE *kills self*
03:22:03 <Mathnerd314> elliott: oh yeah, I feel much saner now. maybe you could try to re-explain, if you're not too busy coding
03:22:09 <pikhq> koriȳa ſô miſeteru ne.
03:22:51 <pikhq> {こりゃそう見せてるね。}[korya sō miseteru ne.]
03:23:26 <elliott> Mathnerd314: tl;dr there's webkit in a window. The browser exposes a directory on the filesystem tree. It has files like "location", containing the current browser location, and the like. (You can imagine the fully-extended form of this, a "dom" directory, containing a representation of the entire document as an FS tree -- but that's a bit ridiculous).
03:23:32 <elliott> Mathnerd314: It can be controlled by writing to these files.
03:23:47 <elliott> Mathnerd314: For example, "echo http://reddit.com/ >location" would make the browser load reddit.com.
03:25:14 <elliott> pikhq: How much will you charge to convert a simple Python program into C? >:)
03:25:17 <Mathnerd314> (if texting while driving is as bad for you as drinking while driving, and you shouldn't drink and derive, then texting (IRC-ing) and simultaneously following the derivation of the Mean Value Theorem is bad)
03:25:50 <elliott> Sasha: I don't use Windows.
03:25:51 <pikhq> elliott: All the money.
03:26:04 <elliott> pikhq: It's only 44 lines!
03:26:09 <zzo38> What is the point of a romanisation scheme that uses unicode?
03:26:10 <Sasha> Oh, right. Someone that doesn't like games.
03:26:24 <elliott> Sasha: I play plenty of games. (well, not right now, but that's for other reasons)
03:26:48 <pikhq> zzo38: Better matching the native orthography.
03:26:50 <elliott> Sasha: Of course most of them are either native, work in Wine (which supports many things perfectly), or can be easily run in a VM at a pinch.
03:27:06 <Sasha> whenever I attempt using Wine, my box fails.
03:27:09 <elliott> Sasha: Also: seemingly unlike you, I value other things above the absolute easiest game experience, such as being able to program without wanting to kill myself.
03:27:19 <pikhq> zzo38: Especially handy for demonstrating oddities of the orthography while speaking to an audience that's not going to learn a completely new script.
03:27:20 <Sasha> It hasn't got enough memory.
03:27:27 <elliott> I've never understood how some people put being able to game in one or two clicks above EVERYTHING ELSE EVER.
03:27:41 <Sasha> elliott: Just because I am not a computer programmer means very little
03:27:45 <Mathnerd314> elliott: so, an near as I can follow, you're writing something of a CLI for the web?
03:27:52 <Sasha> ypu could say I am a polymath
03:27:59 <elliott> Sasha: you *do* realise this channel is about programming languages?
03:28:13 <Sasha> I came here to learn more about them
03:28:21 <elliott> Sasha: You do realise it's about *esoteric* programming languages? :P
03:28:27 <elliott> Your education here will be ... skewed.
03:28:32 <elliott> Mathnerd314: Something like that... except the filesystem metaphor goes deeper than just a CLI.
03:28:44 <elliott> Mathnerd314: A filesystem represents a piece of information, a resource.
03:28:46 <pikhq> zzo38: How else could I discuss tiȳônnhų, tàkutenn, or hanntàkutenn with someone who can't read Japanese?
03:28:52 <Sasha> I mean, hell, I don't learn useful natlangs.
03:28:55 <elliott> Mathnerd314: Consider a filesystem tree like a programming language namespace.
03:29:19 <elliott> pikhq: We're talking the SIMPLEST python script here
03:29:23 <Sasha> There was a period there where I was fluent in /ESPERANTO/
03:29:38 <elliott> a few people here know esperanto
03:29:45 <pikhq> Sasha: "was fluent"? Meþinkſt þou ſuckeſt.
03:29:54 <elliott> ...but yeah, you cannot be ex-fluent
03:29:57 <pikhq> elliott: Jes, sed mia Esperanto malbonas.
03:30:02 <elliott> fluency is pretty much defined as the point where you have it internalised :P
03:30:24 <Sasha> pikhq: Mine isn't so great either, I rarely use it.
03:31:41 <pikhq> でも僕の日本語が良い！{tèmo, hòku no nihonnkò kà î.}[demo, boku no nihongo ga ii.](But, my Japanese is good.)
03:31:44 <Sasha> I'm also currently learning German as part of school and am /okay/ at ASL, but not quite fluent.
03:31:52 <elliott> pikhq: Thought: Arch is for people who want to pretend to understand and follow the minimalist Unix philosophy without actually having to give up any of their stupid trinkets.
03:32:03 <pikhq> elliott: Possibly.
03:32:11 <zzo38> elliott: Is it? Maybe it is.
03:32:25 <elliott> zzo38: You'd be proud of me, I'm making a web browser.
03:32:48 <Sasha> hmm. Would the German word for Japanese be "Japanisch"?
03:32:56 <Mathnerd314> so what has everyone who knew Esperanto switched to instead? Lojban?
03:33:03 <pikhq> Sasha: You tell me.
03:33:10 <zzo38> elliott: O, you are? Are you using a existing system or making a new one?
03:33:19 <elliott> zzo38: Well, uh, I'm using WebKit, but that's it.
03:33:21 <pikhq> Mathnerd314: Esperanto is still the most viable auxlang.
03:33:44 <Sasha> Mein Japanisch kennen bist null.
03:33:46 <pikhq> Mathnerd314: Mostly by merit of being the only one that's viable.
03:33:47 <zzo38> (Currently I use Vonkeror it is a existing Mozilla system, it is not best though)
03:33:48 <elliott> zzo38: I could, theoretically, write my own rendering engine, but considering it's almost a hundred thousand lines of code to write one that works with most web pages properly, that sounds un-fun.
03:34:01 <elliott> But yeah, I'm writing everything but the rendering engine.
03:34:07 <Sasha> pikhq: It's not so widely used any more.
03:34:23 <Sasha> 3rd time's the charm?
03:34:36 <Mathnerd314> so maybe interest in auxlangs just died down generally as people learn English?
03:34:46 <pikhq> Sasha: There exist *no other* auxlangs with any hint of popularity, though.
03:34:47 <Sasha> Mathnerd314: Seems that way.
03:35:07 <pikhq> Mathnerd314: Actually, interest in auxlangs died down when Hitler killed 3/4ths of the speakers.
03:35:34 <Sasha> I am also wondering if ASL would be considered an auxlang.
03:35:42 <pikhq> Sasha: Not really.
03:35:55 <zzo38> Best is a completely different way, using only an existing Javascript engine and writing render engine it doesn't have to work correctly, as long as web pages are workable. The JavaScript DOM also has to be completely different; image loading must be different; etc. And there must be a command-line access of services.
03:35:58 <Sasha> It's derived from a few other SLs, IIRC
03:36:14 <pikhq> Mathnerd314: At the time, a very large number of Esperanto speakers were in Poland. And Jewish.
03:36:21 <elliott> zzo38: THINGS ARE NOT BETTER MERELY BY VIRTUE OF BEING DIFFERENT FROM OTHER EXISTING THINIGS
03:36:29 <elliott> ESPECIALLY WHEN NOTHING ELSE WORKS WITH THE DIFFERENT THING
03:37:24 <pikhq> Mathnerd314: In particular, Lidia Zamenhof (Lidja ZAMENHOF)
03:37:48 <elliott> Is it Esperanto policy to uppercase SURNAMES?
03:37:56 <pikhq> Sasha: Most languages are that way.
03:38:15 <Sasha> seems as if he was uppercasing to emphasize
03:38:16 <elliott> pikhq: are you sure you wanted to address that to Sasha?
03:38:31 <pikhq> Sasha: "Derived from other languages".
03:38:42 <Sasha> all these highlights
03:38:51 <Sasha> all this bright glaring orange on my screen
03:38:59 <elliott> SASHA SASHA SASHA SASHA SASHA SASHA SASHA SASHA SASHA SASHA SASHA SASHA SASHA SASHA SASHA SASHA
03:39:13 <Sasha> HAS A CHARISMA MODIFIER OF -6
03:39:33 <pikhq> elliott: The uppercasing of surnames is somewhat more common in Esperanto than elsewhere. It's actually a convention in Latin-script-using languages where the surname might be ambiguous.
03:39:46 <elliott> Sasha: that's what she said
03:41:03 <Mathnerd314> man, I thought freenode was safe from shouting. add to ignore list...
03:41:33 <Sasha> I am developing an a priori language, with an individual system of syntax and grammar.
03:41:40 <elliott> Sasha: YOU SHOUTED "I" THERE
03:41:43 <pikhq> elliott: It's *especially* useful for Japanese names, where the actual surname order *might be either way* when written in English.
03:41:50 <Sasha> Its method of indicating names is a special punctuation mark.
03:41:52 <elliott> pikhq: YOU SHOUTED THAT SURNAME BACK THER
03:42:01 <pikhq> s/Japanese/East Asian/
03:42:20 <Sasha> elliott: I am merely using some English conventions in my typings
03:42:22 <pikhq> For instance, it's KIM Jong-Il but Nobuo UEMATSU.
03:42:48 <Sasha> which does not help folks like me. I write in allcaps.
03:43:44 <Mathnerd314> elliott: do you consume some sort of drug?
03:43:51 <Sasha> No, my manual handwriting is allcaps. It's a reflex on keyboards to use the shift key.
03:44:07 <Sasha> I consume some powerful antidepressants. Does that count?
03:44:13 <pikhq> Sasha: That's fairly ridiculous.
03:44:28 <pikhq> Though my manual handwriting has long ceased to be readable by others, so who am I to judge?
03:44:46 <Sasha> pikhq: Eh, it works.
03:44:54 <elliott> Mathnerd314: well, yes, caffeine makes its way into my body
03:44:59 <Sasha> I also cross Zs, 7s, and 0s
03:45:05 <elliott> Mathnerd314: but i'm perfectly calm right now if that's WHAT YOU'ER THIMNKING WOOOOOOO
03:45:21 <Sasha> so they don't look like 2s, 1s and Os
03:45:54 <Mathnerd314> elliott: I'm just jealous of your seemingly boundless energy. I myself am tired all the time, thus the lowercase
03:46:10 <elliott> Mathnerd314: YES PRESSING THE CAPSLOCK IS VERY ENERGY-USING
03:46:14 <elliott> Mathnerd314: i'm tired all the time too :p
03:46:18 <pikhq> Sasha: I no longer use an orthography anyone else knows. :P
03:46:30 <Sasha> pikhq: Come again?
03:46:38 <Sasha> I do no use the capslock key
03:46:47 <Sasha> I use some amazing shift key skills
03:47:30 <pikhq> My typical handwriting is nowadays English written using Chinese characters and a Latin-alphabet-derived script with an absurd number of ligatures.
03:47:54 <pikhq> High school was boring, what can I say?
03:48:41 <elliott> pikhq: I want to see that.
03:48:46 <Sasha> no, I like some of what I write to be legible.
03:49:09 <pikhq> elliott: I don't have a scanner handy, sadly.
03:49:22 <pikhq> elliott: If I did, I would be quite happy to oblige.
03:49:38 <elliott> pikhq: Try and represent it in Unicodeish? You won't get most of the ligatures and the fancy Latin script but you can get some of them and the freaky half-Chinese stuff.
03:50:21 <pikhq> elliott: Uh, the basic gylph forms themselves are highly modified. But I can maybe approximate it...
03:50:44 <elliott> To be honest, I don't even handwrite.
03:51:01 <elliott> I have a computer. I can type many times faster than any handwriting. It breaks down a bit with drawings etc. though :)
03:51:49 <Sasha> elliott: I can type at 75WPM on a computer, 65WPM on a typewriter, and about 70WPM handwritten.
03:52:03 <Sasha> However, using shorthand, I can get about 120WPM
03:52:04 <elliott> i can type at 110 wpm or so when i'm just copying text
03:52:12 <elliott> including error correction
03:52:17 <elliott> but i think slower than that
03:52:34 <elliott> typing which is walrus kitten bark
03:53:11 <Sasha> I can type at a decent speed
03:53:11 <elliott> since they appear on your screen in something approximating real-time,
03:53:16 <elliott> it is a vaguely useful estimate.
03:53:22 <Sasha> if I have an idea of what I am going to say
03:53:33 <Sasha> but my hands are rather large and this keyboard is small
03:53:38 <elliott> my brain freezes up for long periods of time to figure out what to say first :p
03:54:28 <pikhq> elliott: u`,t`-基字方t`-m自s有高的m°d!f!-d. b;t私可m@yb- a\pr°x!m@t- it...
03:54:40 <elliott> pikhq: What... does that translate as?
03:55:02 <pikhq> elliott: "Uh, the basic glyph forms themselves are highly modified. But I can maybe approximate it..."
03:55:16 <elliott> pikhq: you are a crazy fucker
03:55:28 <elliott> pikhq: why is i written as !
03:55:40 <elliott> why is a written as @ mid-sentence
03:55:43 <pikhq> Note that each consonant-vowel combination would be a written as a ligature.
03:55:53 <elliott> and how do you know whether to use latin or chinese?
03:56:09 <pikhq> elliott: If I know the Chinese for the morphemes of the word I use that, otherwise I don't.
03:56:20 <elliott> pikhq: So you could have written the entire previous line as Chinese.
03:56:26 <elliott> If you expanded your knowledge.
03:56:50 <pikhq> Well, except "the" doesn't really have an analog in Chinese.
03:56:59 <pikhq> But that doesn't matter too much; it's a single glyph.
03:57:03 <Sasha> use Gregg shorthanf
03:57:30 <pikhq> h gets turned into a grave diacritic, you see.
03:57:52 <Sasha> it's a '-like thing in Gregg shorthand
03:58:09 <pikhq> elliott: Oh, yeah, and each ligatured glyph can be written as a single stroke.
03:59:00 <pikhq> Diacritics, sadly, can't really fit into that well.
04:00:56 <Sasha> I'm going to sleep not
04:02:25 <pikhq> elliott: It's really quite nice writing Latin-origin or Greek-origin words using that.
04:03:01 <pikhq> elliott: "Orthographic"? 正書的.
04:12:53 <pikhq> Several hours and Midori has yet to crash on me.
04:37:02 -!- augur has joined.
04:50:20 <zzo38> Can you write shorthand?
04:53:30 <zzo38> I have a book about Pitman shorthand
04:57:50 <catseye_> < zzo38> Did someone in here have a question about whether or not Enhanced CWEB can be used to create webpages?
04:57:51 <storkbot> catseye_: cpressey told me to tell you: Boo!
04:59:40 <catseye_> Also, this terminal does not do unicode, so there's a lot of ????'s
05:00:37 <zzo38> catseye_: You can make CGI programs in Enhanced CWEB, you could also make TeX macros to make printed formatting and HTML; you could also use make file mode and @{ ... @} even, or make a C program to put things together, or whatever.
05:01:13 <zzo38> catseye_: yesweb can do multiple files, but it doesn't do much and is mostly only the proof of concept. yesweb can be improved, however. It is public domain.
05:01:15 <catseye_> could I make a C/HTML polyglot that displays acceptably in a web pages, but also compiles with a C compiler?
05:01:48 <catseye_> (that is not really a Enhanced CWEB-specific question, I guess)
05:02:43 <zzo38> catseye_: I don't know. Try? You might need CSS and/or JavaScript used, and you probably need to set the MIME type for sure.
05:04:29 * catseye_ considers starting out with "#include <html>"...
05:04:52 <catseye_> (I'm not very good at polyglots)
05:05:18 -!- sftp has quit (Remote host closed the connection).
05:06:11 <pikhq> catseye_: It almost certainly wouldn't be valid HTML.
05:06:29 <pikhq> catseye_: But you could probably abuse a tag soup parser into compliance.
05:06:32 -!- catseye_ has changed nick to catseye.
05:06:43 <catseye> pikhq: that would be the plan, yeah.
05:07:00 <catseye> "it doesn't look awful" in some mainstream web browser.
05:07:09 <zzo38> pikhq: It is why you need to set the MIME type and use CSS
05:18:41 <catseye> eventually i'd like to write my own x86-subset assembler
05:20:38 <catseye> it would mainly do checking and stuff, and could use nasm as its backend
05:28:14 <zzo38> Finally I fixed two problems with internationalization.wi program. (I probably never need text internationalization in most programs I write (especially command-line programs), but perhaps someone will need a program from me that supports it?)
05:28:27 -!- Quadrescence has quit (Ping timeout: 245 seconds).
05:29:07 <zzo38> Do you know what these two problems are?
05:32:47 -!- augur has quit (Quit: Leaving...).
05:33:07 -!- augur has joined.
05:33:31 -!- Quadrescence has joined.
06:14:50 <zzo38> First problem is you can't have categories/comments. Now an optional category/comment can be placed as the second parameter to intl(). Both together are used to uniquely identify, and the category/comment (called the auxiliary) is placed in the default.lang file separated from the main text by ~00 which represents a null character, and is therefore ignored when the file is loaded.
06:15:57 <zzo38> Second problem is if you write a translation and the program changes, the index numbers for the strings might change. This is dealt with by a separate program called intlconv (no change is needed in internationalization.wi to correct this second problem).
06:17:09 <zzo38> Basically, let's say you want to make a Klingon translation and you already have one, but that the program has changed and you need to update it. Three steps as follows:
06:17:37 <zzo38> Step 1: Run "intlconv default.lang klingon.xlang" to update your .xlang file with the changes.
06:17:56 <zzo38> Step 2: Edit the .xlang file to add new translations.
06:18:16 <zzo38> Step 3: Run "intlconv default.lang klingon.xlang klingon.lang" to create the .lang file which is loadable by the program.
06:19:55 <zzo38> Is this reasonable?
06:20:33 <catseye> That seems reasonable to me.
06:21:39 <zzo38> And internationalization.wi is still only 219 lines long (and this includes things such as chapter headings, which GNU programs do not have)
06:22:02 <zzo38> Both files are public domain (which is compatible with the GNU GPL).
06:23:18 <zzo38> Do you now think this program is complete, with the changes I specified?
06:24:12 -!- Rugxulo has joined.
06:24:29 <catseye> Hard to say if it's complete, but I can't think of any other features that I want ;)
06:24:59 <catseye> Woo, I'm building NetBSD from source...
06:25:06 <catseye> Maybe I will let this go while I sleep
06:25:18 <zzo38> catseye: Exactly! I don't think there are any other features that are needed.
06:26:46 <zzo38> If you want to use _() as an abbreviation for intl() C macros won't work. Instead do: @{ void _$() { intl$(); } @}
06:29:11 <zzo38> Rugxulo: Can be checked same account by: NS INFO catseye
06:30:10 <zzo38> Except that both "catseye" and "cpressey" are logged in now
06:31:01 <zzo38> And only "cpressey" is the one having the 330 line ("is logged in as") in the WHOIS message
06:31:06 -!- augur has quit (Ping timeout: 252 seconds).
06:32:52 <zzo38> Rugxulo: Does that help you at all?
06:33:18 <Rugxulo> all I'm saying is that if it's not the same person, somebody chose a confusing nick!!
06:34:03 <zzo38> Yes and if that is the case, they took a nick that is registered under someone else's account
06:35:42 <Rugxulo> well, I don't hear cpressey complaining, so who cares I guess
06:53:38 -!- augur has joined.
07:13:42 <Rugxulo> good thing I didn't miss much :-P
07:14:46 <zzo38> Later on I can write other programs in Enhanced CWEB which is also much smaller than GNU software, and much more efficient, and can be made into a book, even.
07:15:42 -!- zzo38 has quit (Remote host closed the connection).
07:16:04 <Rugxulo> smaller and more efficient? how?
07:24:15 -!- MigoMipo has joined.
07:36:41 -!- Rugxulo has left (?).
07:46:58 -!- Rugxulo has joined.
07:48:03 <Rugxulo> eight tabs in a row is weird, I don't know how people think that's reasonable
07:56:38 -!- FireFly has joined.
07:57:23 <fizzie> Rugxulo: As far as I know, it is the same person, yes.
07:58:09 <Rugxulo> BTW, I found Ben Olmstead on the 'Net recently, he mentioned a JIT interpreter for Befunge by John Colagioia (sp?), ever heard of it?
07:58:29 <fizzie> I've heard of one; it was rather more limited than what jitfunge is aiming at.
07:58:50 <Rugxulo> basically, he said he didn't have MCBC anymore :-( but that John's JIT version was better anyways
07:58:53 <fizzie> On the other hand, I haven't made any progress on that front; too many things again.
07:59:02 <Rugxulo> yeah, real life (tm) ... I hear ya
07:59:59 -!- clog has quit (ended).
08:00:00 -!- clog has joined.
08:07:40 -!- MigoMipo has quit (Remote host closed the connection).
08:21:21 <Rugxulo> "As for 12GB of RAM? “Because I can” is one reason" ... lol !!!
08:22:10 <Rugxulo> ironic that he's blogging about memory errors
08:24:10 -!- evincar has joined.
08:36:24 -!- Rugxulo has left (?).
08:46:23 -!- tombom has joined.
08:50:50 -!- Sasha2 has joined.
08:52:32 -!- Sasha has quit (Read error: Connection reset by peer).
09:08:05 -!- FireFly has quit (Quit: swatted to death).
09:16:06 <fizzie> "In August 1990 an unemployed French nuclear physicist named André Gardes attempted a singlehanded invasion of Sark, armed with a semi-automatic weapon. The night Gardes arrived he put up signs declaring his intention to take over the island the following day at noon. He was arrested while sitting on a bench, changing the gun's magazine and waiting for noon to arrive, by the island's volunteer Constable."
09:16:13 <fizzie> There's always something wonky about physicists.
09:49:47 -!- atrapado has joined.
09:58:50 -!- evincar has quit (Quit: ChatZilla 0.9.86 [Firefox 3.6.11/20101012113537]).
10:09:03 -!- augur has quit (Remote host closed the connection).
10:28:50 -!- tombom has quit (Ping timeout: 240 seconds).
10:31:41 <cheater_> i love this page: http://www.cyberciti.biz/tips/howto-linux-unix-bash-shell-setup-prompt.html
10:31:48 <cheater_> How do I add colors to my prompt?
10:31:48 <cheater_> You can change the color of your shell prompt to impress your friend or to make your own life quite easy while working at command prompt.
10:32:18 <cheater_> yes. it's that obvious. anyone who gives a shit about colorful prompts has got only one friend.
10:40:57 -!- cheater99 has quit (Ping timeout: 245 seconds).
10:43:19 -!- wareya has quit (Ping timeout: 252 seconds).
10:52:50 -!- elliott has quit (Ping timeout: 255 seconds).
12:20:43 -!- tombom has joined.
13:01:17 -!- wareya has joined.
13:02:10 -!- nooga has quit (Ping timeout: 252 seconds).
13:03:07 -!- nooga has joined.
13:18:43 -!- nooga has quit (Ping timeout: 240 seconds).
13:20:36 -!- nooga has joined.
13:30:56 -!- augur has joined.
13:56:57 -!- ais523 has joined.
14:25:49 -!- FireFly has joined.
14:38:59 -!- augur has quit (Remote host closed the connection).
14:39:11 -!- augur has joined.
14:53:25 -!- augur has quit (Read error: Connection reset by peer).
15:17:50 <catseye> @tell Rugxulo it wasn't John Colagioia who wrote the JIT Befunge interpreter, it was Kevin Vigor.
15:18:42 <catseye> cheater_: at work I made my prompt a RAINBOW
15:26:43 <catseye> @tell elliott NetBSD takes about 2 hour to build on this machine. haven't installed it yet.
15:44:47 <Vorpal> catseye, hm compiling netbsd from another OS or?
15:49:55 <cpressey> Vorpal: no, rebuilding it on NetBSD. going to install it after I confirm by reading the manual how to do that
15:50:14 -!- Phantom_Hoover has joined.
15:50:17 <cpressey> It's the same version as I installed, so no change :)
15:50:31 <Vorpal> cpressey, just doing it to make sure it works then?
15:50:52 <cpressey> pretty much. practice for when i update the system sources to actually build an updated version.
15:51:14 * cpressey briefly considered running NetBSD-CURRENT
15:51:50 <cpressey> i'm too old for that sort of thing :)
15:52:07 <Vorpal> cpressey, why is netbsd called netbsd btw?
15:52:15 <Vorpal> my best guess would be same reason as nethack
15:55:21 <cpressey> It's BSD hacked on by people on the Net
15:55:37 -!- augur has joined.
16:01:18 <coppro> my prof just 'accidentally' proved the fundamental theorem of calculus
16:03:48 -!- evincar has joined.
16:21:10 <evincar> Wow, it's more dead in here now than it was in the wee hours of the morning.
16:24:28 -!- augur has quit (Remote host closed the connection).
16:25:05 -!- augur has joined.
16:34:53 <Vorpal> Phantom_Hoover, isn't it the one about the relationship between integration and differentiation?
16:35:22 <Vorpal> well, google agrees with me at least
17:16:24 -!- zzo38 has joined.
17:17:05 <zzo38> I found a map of London Underground in the basement yesterday. Now, I just need to make a larger copy......
17:35:49 <cpressey> coppro: spontaneous human computation.
17:46:21 <cpressey> A SIMPLE ALGEBRAIC FORMALIZM FOR WEB DEVELOPMENT THAT'S ALL I ASK IS THAT SO HARD???
17:51:59 <cpressey> I wonder if there is a role for "lightly mutable" data structures in otherwise pure functional programming -- like, in ways that don't affect referential transparency
17:53:05 <cpressey> I'm thinking of memoization, I think
17:53:45 -!- sftp has joined.
17:54:27 <Zuu> memoization is just an optimization
17:54:57 <cpressey> what i was thinking of what just an optimization, too
17:54:57 <Zuu> essentially a fanzy word for 'caching of function calls'
17:56:07 <Gregor> Errr ... I don't think that something that can actually change the computational complexity of an operation should be considered a mere "optimization"
17:57:14 <cpressey> I don't see why you can't use that word, it's just that compiler writers have a limited notion of it :)
17:57:19 <Zuu> well, it's usually transparent to the user, and its intention is to not semantically affect the result in any way, so yes, i'd call it an optimization
17:57:39 <Zuu> just as swapping a bubble sort out with a merge sort is an optimization :)
17:59:04 <cpressey> oh i had some sort of insight about proofs and compilers and assembly language last night!
17:59:14 <cpressey> i'll try to remember it while i go for lunch
18:03:25 -!- atrapado has quit (Quit: Abandonando).
18:12:12 -!- Phantom_Hoover has quit (Ping timeout: 245 seconds).
18:13:36 <pikhq> Mmm, good cheddar cheese. So very delicious.
18:15:13 <Zuu> you ate pacman!
18:17:27 <cheater_> catseye: that would be totally like you.
18:28:45 <zzo38> What is the square root of your godfather's telephone number?
18:29:02 <zzo38> How many bibles have you stolen? (Use roman numerals)
18:39:02 -!- aloril has joined.
18:44:38 <cpressey> I con't know; is the "Codex of Shared Amplification" a bible?
18:44:55 <zzo38> cpressey: You decide.
18:46:49 <cpressey> ok, then only XXVII BUT I HAD A GOOD REASON
18:56:09 <cpressey> anyway, it wasn't much of an insight. it was something to the effect of: purely functional programs are (comparatively) easy to reason about, while arbitrary assembly language programs are difficult to reason about. it's hard to compile a functional program in a way that is efficient; it's (comparatively) easy to write efficient arbitrary assembly code. And, the more you restrict yourself to non-arbitrary (that is, codifiable, repeatable patterns) a
18:57:41 <cpressey> nothing really new here except that i was considering "reason about" in a formal way: e.g. if you wanted to include a machine-readable proof along with your code.
19:00:08 <pikhq> Gregor: GHC's optimization changes the complexity of operations all the *time*.
19:00:21 <pikhq> Gregor: ... Heck, even *Egobfc* changes the complexity of operations all the time.
19:00:32 <pikhq> Gregor: It swaps out an O(n) addition operation for an O(1) addition. :)
19:01:16 <Gregor> Yeahyeah, I just think maybe we should have different terms (or at least more specific terms) for constant-changing optimizations than for order-of-complexity-reducing optimizations.
19:02:21 <pikhq> Aren't the constant-changing ones pretty much all peephole optimizations?
19:02:29 <cpressey> Gregor: some would use "micro-optimization" for the first kind.
19:03:29 <cpressey> That one maybe has yet ANOTHER connotation to it, though.
19:03:37 <Gregor> pikhq: Any sort of improvement to a register allocation scheme is going to be a constant-changing optimization, but it won't necessarily be a peephole optimization.
19:04:00 <cpressey> (That is, something a programmer has manually written in the source code, not a compiler.)
19:04:05 <Gregor> (Assuming uniform but slow access to memory)
19:04:43 <pikhq> (which is a wrong assumption and has been for several decades now)
19:09:33 <Gregor> All I'm doing is removing one constant-changing optimization from the equation to focus on a different constant. It's still not going to affect the order of complexity.
19:10:14 <cpressey> But what if my caches are RAM and my main memory is a tape? Huh? What then?
19:10:42 <Gregor> cpressey: Then you're an idiot.
19:18:14 -!- cheater99 has joined.
19:28:03 -!- distant_figure has quit (Quit: underflow).
19:28:16 -!- distant_figure has joined.
19:35:19 <cpressey> http://webkit.org/projects/webkit/index.html
19:35:30 <Gregor> XCode + WebKit = if you want a header to be project-wide but not public, --- how the hell were we both just about to mention WebKit?
19:35:31 <cpressey> that's just... beautiful *snif*
19:38:42 <cpressey> also i love how the sample code on developer.apple.com is all light grey on light blue.
19:41:24 <cpressey> i should set my text editor to use those two colours. MAN. the joy my eyes would feel!
19:43:56 <olsner> hmm, light grey on light blue, isn't that pretty poor contrast?
19:44:15 -!- ais523 has quit (Ping timeout: 272 seconds).
19:45:25 <cpressey> olsner: you be the judge: http://developer.apple.com/library/mac/#documentation/Cocoa/Conceptual/DisplayWebContent/Tasks/SimpleBrowsing.html#//apple_ref/doc/uid/20002025-CJBEHAAG
19:46:07 <Gregor> XCode + WebKit = if you want a header to be project-wide but not public, mark it "private". If you want it to be private, mark it "project".
19:46:32 <Gregor> WebKit is proof that XCode is a shitty utility, since half the build system is designed to work AROUND XCodec.
19:48:02 <olsner> cpressey: that's not so bad actually
19:50:23 <cpressey> olsner: to my eyes, it's a hair's breadth away from being invisible
19:51:51 <zzo38> I should make a larger copy of this London Underground map, and then make some cards and tokens for the game
19:53:12 <zzo38> Some stations are on two zones
19:54:33 -!- ais523_ has joined.
19:54:34 -!- ais523_ has quit (Changing host).
19:54:34 -!- ais523_ has joined.
19:54:34 -!- ais523_ has quit (Changing host).
19:54:34 -!- ais523_ has joined.
20:04:16 <Vorpal> I looked in some detail at the standard C99 math functions and found some curious stuff.
20:04:42 <zzo38> What curious stuff did you find?
20:04:53 <Vorpal> for example: double nan(char* s); // returns a nan represented by the string s
20:04:59 <Vorpal> which is a bit strange way to do it IMO
20:05:11 <Vorpal> and double remqou(double x, double y, int *quo);
20:05:15 <Vorpal> well that one is just weird
20:05:35 <Vorpal> "[...] compute the remainder and part of the quotient upon division of x by y. A few bits of the quotient are stored via the quo pointer. The remainder is returned as the function result."
20:05:41 <Vorpal> "The value stored via the quo pointer has the sign of x / y and agrees with the quotient in at least the low order 3 bits."
20:07:44 <ais523_> elliott for the logs, other people who are interested: there's a currenty hilarious proggit submission http://www.reddit.com/r/programming/comments/dwtou/stack_overflow_how_do_i_leave_my_site_vulnerable/ about an XSS vulnerability in Stack Overflow; one of the Stack Overflow admins posted to say they'd corrected it (and were annoyed at the way it was disclosed), and someone replied to him using Jeff Atwood's account saying
20:07:46 <ais523_> "I've tried being white hat with an SO employee before. All he did was change his password on one site but not any others."
20:09:46 <Vorpal> ais523_, who is Jeff Atwood?
20:09:57 <ais523_> one of the two founders of Stack Overflow
20:10:16 <ais523_> they're famous for self-promoting to relatively absurd degrees
20:10:32 <ais523_> also, one of them doesn't know much about programming; the other does, but I can never remember which is which
20:13:28 <cpressey> Oh, Jeff Atwood is the Coding Horror guy
20:13:53 -!- zzo38 has quit (Remote host closed the connection).
20:15:16 <cpressey> Therefore Joel must be the one who knows about programming :)
20:16:15 <cpressey> No, Joel actually used to be pretty sane. He just likes to talk about how he came to the deep realization that the sky is blue
20:17:44 <cpressey> "I am now quite comfortable in VB.NET or C#, despite the evils of case sensitivity." -- Jeff Atwood
20:19:47 <cpressey> "I consider the Wumpus my power animal"
20:19:48 <ais523_> hmm, now I'm curious enough to actually read some recent Coding Horror to see what it's like
20:20:22 <Vorpal> cpressey, "power animal"?
20:20:40 <olsner> Vorpal: they are mentioned in fight club
20:20:46 <cpressey> Vorpal: It links to this: http://www.codinghorror.com/blog/2006/04/your-personal-brand.html
20:20:58 <cpressey> I assume it's similar to "spirit animal" in meaning
20:21:37 <cpressey> But it's also, like, apparently his 1st experience with Hunt the Wumpus was with a *graphical* version of the game?
20:21:54 <Vorpal> cpressey, are there graphical versions of it?
20:22:11 <evincar> Vorpal: There certainly are.
20:22:16 <evincar> I wouldn't bother with them.
20:22:17 <ais523_> my first experience with it was the Linux port of the old BSD version
20:22:24 <Vorpal> cpressey, oh and, ever tried that "be the wumpus" game? Which isn't really graphical
20:22:38 <ais523_> although I knew about it before from the famous Befunge-93 program
20:22:58 <cpressey> http://www.videogamehouse.net/huntwumpus.html GRAPHICAL BASTARDIZATION
20:23:32 <cpressey> My first exposure to it was in a book compilation of programs from Creative Computing
20:24:11 <cpressey> ANYWAY, to the OTHER thing: http://en.wikipedia.org/wiki/Power_animal
20:24:12 <Vorpal> cpressey, it's quite fun. The idea is that you play the wumpus, which is a blind animal and thus you have to try to avoid the arrows and locate the hunter to eat em by sound alone. Playable with headphones. Completely impossible with speakers.
20:24:41 <Vorpal> cpressey, list for what
20:26:01 <evincar> List of master programmers whose hearts to eat to gain their powers.
20:26:49 <cpressey> Also on list: ESR and Roger Penrose.
20:26:51 <ais523_> I can think of several sorts of List you could plausibly put Jeff Atwood on
20:26:58 <olsner> oh, is atwood worthy of membership on such a list? why?
20:27:28 <cpressey> olsner: in case you missed it: "I consider the [graphical] Wumpus my power animal" -- Jeff Atwood
20:27:37 <evincar> Well, I'd wager he's a very good programmer, but he's really a brilliant programming philosopher.
20:27:58 <evincar> He knows what programmers need to make them awesome at what they do.
20:28:01 <olsner> cpressey: hmm, that doesn't impress me... what's a wumpus?
20:28:30 <evincar> olsner: Read back a ways. :P
20:29:49 <ais523_> I'm not particularly impressed by Coding Horror so far
20:29:56 <ais523_> half of what I've read is hardware reviews, for no apparent reason
20:30:07 <ais523_> written as if they're authority passed down from God rather than subjective advice
20:30:19 <cpressey> "The power animal may also lend its ward or charge the wisdom or attributes of its kind. For example, a hawk power animal provides hawk attributes, such as hawk-eye."
20:30:35 <cpressey> Need I remind anyone what the powers of a Wumpus are.
20:30:52 <cpressey> Sticky feet to cling to sides of pits with, and smelling awful.
20:31:38 <cpressey> Oh yes, can't forget that. And eating adventurers.
20:32:02 <cpressey> All in all a really noble set of attributes.
20:33:36 <cpressey> Coding Horror has hardware reviews? Oh, this just won't stop getting "better".
20:34:24 <cpressey> Yup, a whole rant about netbooks.
20:36:25 <cpressey> There was a book that used that "Coding Horror" icon as one of its "margin icons"... i used to have a copy. It's popular and I learned nothing from it.
20:37:53 -!- Phantom_Hoover has joined.
20:38:01 <cpressey> I don't think it's the same book, but I associate it with one that was about code performance analysis, that explicitly eschewed asymptotic analysis as uselessly abstract, and dealt instead with concrete time measurements.
20:38:39 <cpressey> Phantom_Hoover: Efficiency is inversely proportional to ease of formal reasoning of the code.
20:39:21 <ais523_> cpressey: if that were true, Haskell would be a lot slower than Befunge
20:39:33 <Phantom_Hoover> So easy-to-reason-with compiled code is inefficient, and efficient code is hard to reason with?
20:40:30 <ais523_> in asm or C, it might make more sense
20:40:39 <ais523_> microoptimisations often make code harder to read
20:40:43 <cpressey> Try to factor compilation out.
20:41:05 <cpressey> ais523_: A Haskell interpreter of the same complexity as your average Befunge interpreter would be a lot slower than Befunge.
20:41:50 <ais523_> hmm, I thought you meant "interpreter written in Haskell" vs "interpreter written in Befunge"
20:42:36 <cpressey> Maybe I'll just repeat my original summary without trying to compress it
20:42:37 <cpressey> 12:55 < cpressey> anyway, it wasn't much of an insight. it was something to the effect of: purely functional programs are (comparatively) easy to reason about, while arbitrary assembly language programs are difficult to reason about. it's hard to compile a functional program in a way that is efficient; it's (comparatively) easy to write efficient
20:42:42 <cpressey> arbitrary asembly code. And, the more you restsrict yourself to non-arbitrary (that is, codifiable, repeatable patterns) assembly code, it becomes easier to reason about, but harder to make efficient.
20:43:26 <cpressey> < cpressey> nothing really new here except that i was considering "reason about" in a formal way: e.g. if you wanted to include a machine-readable proof along with your code.
20:44:04 <cpressey> Phantom_Hoover: Anyway, forget this! Let's talk about Jeff Atwood's power animal instead!
20:44:29 <cpressey> Phantom_Hoover: It's the Wumpus!
20:46:04 <ais523_> cpressey: well, it's easier to embed higher-order typed lambda calculus into hardware using the method I'm using in my research, than it is to embed typical asms with GOTO
20:46:13 <ais523_> the second would require large, inefficient workarounds
20:46:42 <cpressey> ais523_: I should add "on present models of hardware computation" to my insight, then
20:46:43 <ais523_> so perhaps the insight is that the sort of asm today's processors require isn't very suited for formal reasoning
20:47:06 <ais523_> I think it becomes relatively clear why it is, if you do that
20:47:37 <ais523_> asm in general is hard to reason about; subsets may be easier, but it would be a huge coincidence if any given algorithm's fastest implementation happened to fall into one of those subsets
20:49:53 <cpressey> Now I want to see some proofs that extremely gory assembly programs are correct
20:50:37 <cpressey> OK, well, n/m... I have too many interests
20:51:00 <Gregor> Write a formal semantics for assembly, choose what you want to proof, feed it all into Cock^H^Hq, voila :P
20:53:56 <cpressey> That's kind of it, though. Any practical assembly language will have a horrible, horrible semantics that will make Coq... unhappy.
20:54:14 <cpressey> Anything with a nice semantics, otoh, will be too slow :)
21:06:45 <coppro> http://pleasingfungus.com/
21:22:29 <pikhq> My IPv6 tunnel no longer works. I know not why.
21:28:12 <cheater99> if it is impossible to patent protocols..
21:28:27 <cheater99> would it be possible to create a language which is expressed through protocols..
21:31:10 <evincar> cpressey: Like anything else, you can only patent it if you a formal specification of it.
21:31:37 <evincar> But it brought me back an asterisk.
21:40:56 <Vorpal> <olsner> cpressey: hmm, that doesn't impress me... what's a wumpus? <-- did you seriously not know what a wumpus is?
21:41:41 <Vorpal> olsner, next you will claim to not know what colossal cave adventure is!
21:41:51 -!- evincar has quit (Quit: ChatZilla 0.9.86 [Firefox 3.6.11/20101012113537]).
21:42:01 <olsner> Vorpal: indeed I don't
21:42:17 <olsner> I assume it's an adventure game, since the wumpus game is (right?)
21:42:32 <olsner> but even though I have heard of the wumpus game, I have no idea what a wumpus is
21:43:35 <cpressey> the wumpus is the antagonist of the wumpus game.
21:44:15 <olsner> (line 1 and 2 are in response to the colossal cave adventure, line 3 verifying I don't know what a wumpus is)
21:45:48 <cpressey> OH GOD all the top google hits for online versions are graphical.
21:46:00 <Gregor> cpressey: Seems to me like YOU'RE the antagonist. The wumpus is just trying to live its life, and you're hunting it.
21:46:13 <cpressey> That's better: http://scv.bu.edu/cgi-bin/wcl
21:46:18 <Gregor> <evincar> cpressey: Like anything else, you can only patent it if you a formal specification of it. // remove the word "formal" or I'll remove your face.
21:46:33 <cpressey> Gregor: Indeed so. I see a humanties paper here somewhere.
21:46:39 <olsner> I think YOU'RE by definition the protagonist, and it's always the OTHER thing that's the antagonist
21:48:28 <cpressey> olsner: read: http://www.atariarchives.org/bcc1/showpage.php?page=247
21:49:27 <ais523_> Gregor: "you can only patent it if you a specification of it"?
21:51:17 <Vorpal> <olsner> I assume it's an adventure game, since the wumpus game is (right?) <-- uh...
21:51:39 <ais523_> one with quite a high chance of failure even with perfect play
21:52:01 <Vorpal> olsner, and colossal cave adventure was the first text adventure game.
21:52:38 <Vorpal> <Gregor> cpressey: Seems to me like YOU'RE the antagonist. The wumpus is just trying to live its life, and you're hunting it. <-- go play "be the wumpus" then
21:53:49 <Vorpal> ais523_, don't forget the pits
21:54:03 <ais523_> they are considerably more fatal than the bats
21:54:12 * ais523_ wonders what the optimal HtW strategy is
21:54:46 <Vorpal> ais523_, I seem to remember that the b93 htw was quite easy, compared to the classical one
21:54:56 <Vorpal> probably due to lack of being able to make anything complex in b93
21:55:08 <ais523_> the classical one's very simple too
21:55:28 <Vorpal> ais523_, hm well the bsd one then
21:55:35 <ais523_> it's that one I was talking about
21:55:40 <ais523_> it's relatively hard, but very simple
21:55:59 <Vorpal> the b93 one has unlimited arrows iirc
21:56:18 <ais523_> ah, that would help quite a lot
21:57:58 <cpressey> oh... I don't even know where to begin
21:58:44 <cpressey> People who don't believe in attributes-beginning-with-an-underscore in Python because "nothing is really private anyway".
21:59:40 <cpressey> The term "script kiddie for life" comes to mind, I guess.
22:00:17 <Vorpal> hm, from the "be the wumpus" website: "WARNING -- this game does contain small amounts of sounds of people speaking words which some people might consider profane. If you are apt to be offended by the numbers in a .wav file when run through a D/A converter, and then the analog signal amplified then run through a loudspeaker of some kind, well, you may have a problem with my numbers then, I suppose. Thin
22:00:17 <Vorpal> g is, it's your problem, not mine. You've been warned. And what sorts of things would you expect people trapped in a dark cave with a hungry wumpus to say?"
22:15:34 <cpressey> olsner: so you've read that article now, RIGHT?
22:17:49 <cpressey> And now you know what a Wumpus is!
22:18:25 <olsner> yes, turns out it's COMPLETELY FICTIONAL!
22:19:11 <Vorpal> ais523_, do you have any idea why man 3 pow has such a *huge* list of special casing the behaviour?
22:19:26 <ais523_> Vorpal: because floating-point is evil
22:19:38 <ais523_> IEEE floating point inherently has quite a lot of special cases
22:19:39 <Vorpal> ais523_, yes but pow() is a lot worse than, say, sqrt() or even log()
22:19:56 <ais523_> so you have to consider all the special cases for each arguments
22:20:20 <cpressey> olsner: Not only fictional, but also kind of a side-effect of the design of the game!
22:20:27 <Vorpal> ais523_, and even worse for my intended application
22:20:54 <ais523_> why are you intending an application that cares about the corner cases of exponentiation?
22:21:12 <Vorpal> ais523_, which would be a formally speced math.h so I can use that prove an app. However that involves proving that the different behaviour cases are pair-wise disjoint...
22:21:18 <Vorpal> which will be absurd for pow()
22:21:42 <cpressey> "proof" and "floating point" in the same sentence
22:21:51 <Vorpal> cpressey, it is possible actually.
22:22:11 <Vorpal> cpressey, look at frama-c + why + gappa
22:23:03 <Vorpal> cpressey, I do not intend to prove pow() but I intend to spec it's behaviour according to the C standard so I can use this as an "axiom" to prove the code I'm interested in proving
22:23:35 <ais523_> with floating point, not even addition is commutative
22:23:58 -!- ais523_ has quit (Quit: Page closed).
22:26:03 -!- augur has quit (Remote host closed the connection).
22:26:27 -!- augur has joined.
22:40:58 <Phantom_Hoover> cpressey, wasn't there a library for Coq that worked with FP?
22:51:35 <cpressey> Phantom_Hoover: No se habla Coq.
22:52:10 <cpressey> (Nor Spanish, so that is probably ungrammatical)
22:53:45 -!- FireFly has quit (Quit: swatted to death).
22:57:12 -!- Phantom_Hoover has quit (Ping timeout: 245 seconds).
22:59:13 <cpressey> I want to prove rewriting things -- using a rewrite system to do it :)
23:17:15 <olsner> I was thinking about making a compiler for sed
23:18:24 <olsner> don't have any problems that are interesting enough to allow optimizations to be useful, while at the same time being simple enough that anyone can be bothered solving it in sed
23:20:44 -!- Sgeo has joined.
23:21:14 -!- tombom has quit (Quit: Leaving).
23:26:48 -!- elliott has joined.
23:27:52 <elliott> oh shit my ignore didn't carry over the upgrade
23:27:53 <storkbot> elliott: catseye told me to tell you: NetBSD takes about 2 hour to build on this machine. haven't installed it yet.
23:28:19 <elliott> 00:21:21 <Rugxulo> "As for 12GB of RAM? “Because I can†is one reason" ... lol !!!
23:28:19 <elliott> 00:22:10 <Rugxulo> ironic that he's blogging about memory errors
23:28:27 <elliott> anyway, bsmntbombdood has 12GiB :p
23:29:12 <Vorpal> elliott, that quote looked fucked up, was it from clog messing it up?
23:29:31 <elliott> Vorpal: rather from tunes.org not sending character encoding headers
23:29:43 <elliott> although i suppose irc has no real encoding :)
23:33:13 <cpressey> Oh, "Code Complete" is the name of that book. And he stole the icon *from* it, not vice-versa or anything.
23:34:10 <elliott> i can say this without even having seen the other qusetion because it's so obvious what you're referring to :)
23:34:19 <elliott> cpressey: he stole it with PERMISSION, though!
23:34:29 <elliott> cpressey: i had an insane dream and you were the central character!
23:34:34 <cpressey> elliott: If you read the logs you'll discover... the List.
23:34:49 <elliott> cpressey: well you weren't being a very good cpressey.
23:35:01 <Sgeo> Choose my language: Haskell, Factor, or Smalltalk
23:35:33 <cpressey> Sgeo: http://www.fuxi.org/EN/index.htm
23:36:10 <Sgeo> Is using Flash on the front page supposed to make me like it?
23:36:14 <elliott> cpressey: I mean, you had made catseye.tc show a page about how you and your wife were going to commit suicide on the [day] with the help of some, uhh, guy, and that guy was also one of the leaders of uh something you were trying to protest by doing this. I went "NOOOOOO" and the like on IRC and since this is a dream apparently IRC is in-person and you decided that this was the perfect time to have us meet your wife, who is a mermaid except more
23:36:14 <elliott> like a dolphin-mermaid and also she's a (demi-?)god and this is why we'd never met her before.
23:36:21 <elliott> cpressey: Then the whole thing turned into a parody of Scooby Doo.
23:36:31 <Sgeo> Those in the market for languages are not mainstream audiences
23:36:58 <Sgeo> Dear Fuxi website: Please don't use Flash for navigation
23:37:02 <cpressey> elliott: Yes! That is my real life in a nutshell.
23:37:13 <elliott> cpressey: I thought so. So when are you going to kill yourself?
23:37:31 <cpressey> elliott: I have it written down somewhere
23:38:02 <Sgeo> "it takes a conventional syntactic system, that makes the programmers with some experiences of C++, JAVA, use the language without any training;"
23:38:29 <Sgeo> It better not have C++ and Java as subsets
23:38:40 <elliott> cpressey: Sgeo is going to decide that fuxi has interesting aspects now.
23:38:57 <Sgeo> elliott, um, my first impression so far is negative
23:39:09 <cpressey> Sgeo: You should try reading the spec. It's in Chinese.
23:39:42 <cpressey> Sgeo: But, uh, what am I choosing your language for?
23:39:42 <elliott> 21:04:29 * catseye_ considers starting out with "#include <html>"...
23:39:56 <Sgeo> cpressey, I assume there's an English translation?
23:40:05 <Sgeo> cpressey, to be my returning love affair
23:40:14 <cpressey> Sgeo: Of the spec? Not that I've found
23:40:19 <elliott> Sgeo: one of those languages is going to have a baby one day
23:40:19 * Sgeo is currently thinking quite a bit about Haskell
23:40:22 <elliott> and you'll have to pay child support
23:40:40 <elliott> /*<style>body *{display:none} #foo {display:block}</style><div id=foo>
23:40:52 <elliott> might take some more element coercion
23:41:17 <Sgeo> "it collects active, remote, mobile and persistent objects in a transparent way, through some orthogonal styles;"
23:41:27 <Sgeo> That sounded good right up to "orthogonal"
23:41:40 <Sgeo> Oh, actually, it makes sense
23:43:15 <Sgeo> Variables store both a value and a pointer?
23:43:45 <cpressey> Now I feel bad for throwing that in your path, Sgeo
23:43:49 <Sgeo> Unless it can't do both
23:43:53 <cpressey> I've barely looked at it myself
23:43:58 <Sgeo> Maybe I should just read the damn spec
23:44:21 <Sgeo> If it can't do both, that's ok. Except the explanation is a bit WTFy
23:45:05 <Sgeo> Yeah, they don't do both. Unless test() is some quantum weirdness
23:45:31 <Sgeo> I keep seeing bad thing, then realizing that it's not actually that bad
23:46:56 <Sgeo> Whoever wrote this needs to be hit on the head for making the language look more incomprehensible than it actually is
23:47:14 -!- augur has quit (Ping timeout: 264 seconds).
23:47:19 <Sgeo> "But expressions in Fuxi can terminate naturally, without any delimiters."
23:47:36 <Sgeo> I'd headdesk, but I'm sure an explanation will be forthcoming eventually
23:47:58 -!- pikhq_ has joined.
23:48:21 <Sgeo> Oh, well, if an operator takes two arguments, and functions, etc etc., what do you need statement delimeters for
23:48:28 <Sgeo> And this is where pikhq_ hits me
23:48:43 -!- pikhq has quit (Ping timeout: 252 seconds).
23:48:47 <elliott> pikhq is just punching you because he hates you
23:49:00 <elliott> cpressey: We should try and devise a language so horrible that Sgeo will still like it.
23:49:21 <Sgeo> Fuxi keeps being not as horrible as I imagine
23:49:30 <Sgeo> It just needs decent documentation
23:50:10 <elliott> cpressey: Literally impossible.
23:50:16 <cpressey> Sgeo: No, YOU need to learn Chinese.
23:50:31 <Vorpal> "Your search - Fuxi esolang - did not match any documents. "
23:50:33 -!- augur has joined.
23:51:03 <Vorpal> just two flash animations
23:51:08 -!- wareya has quit (Read error: Connection reset by peer).
23:51:10 <Sgeo> Fuxi {} blocks are fucked up
23:51:26 <Vorpal> cpressey, I did. Just two flash ones again
23:52:02 <Sgeo> {}, which is syntactic sugar for a bunch of &&, should absolutely NOT be described as similar to blocks in C/C++
23:52:05 <Vorpal> elliott, want some possible ideas for horrible languages?
23:52:11 <cpressey> Vorpal: the one on the right is a menu
23:52:18 <elliott> Vorpal: No, C++ will provide plenty.
23:52:21 <Vorpal> cpressey, well I don't have flash
23:52:23 -!- pikhq_ has changed nick to pikhq.
23:52:25 <cpressey> Vorpal: http://www.fuxi.org/EN/whitepaper.htm
23:52:40 <Sgeo> I cannot forgive the documentation now
23:52:41 <Vorpal> elliott, my first idea was "Malbolge2EE"
23:52:57 <elliott> I think I'm going to uninstall my web browser and then act like people should have to provide me content without having one.
23:53:01 <Sgeo> Actually... I kind of see the rationale
23:53:10 <Sgeo> Assignments are boolean expressions
23:53:20 <Vorpal> cpressey, it uses god damn double width font
23:53:25 <Sgeo> If one fails, the block returns false... that still kind of is a sucky way to deal with errors
23:53:26 <cpressey> Vorpal: You may prefer http://www.fuxi.org/ZH/whitepaper.htm
23:53:38 <Vorpal> well I can't read that
23:53:40 <elliott> Vorpal REALLY hates double width fonts.
23:53:55 <Vorpal> elliott, I find it VERY hard to read
23:54:21 <cpressey> Not as hard as Han characters, I imagine
23:54:22 -!- wareya has joined.
23:54:33 <Sgeo> ......I hate Fuxi
23:54:36 <elliott> Not unless your brain is completely broken.
23:54:40 <Sgeo> let{ <variable-list> } in <sequence> (10)
23:54:41 <Vorpal> anyway, I worked around it. firebug is an awesome extension
23:54:48 <Vorpal> just add lang="en" to the body element
23:54:50 <elliott> HardER, yes. VERY HARD, no.
23:54:59 <elliott> Sgeo: ...what's wrong with that?
23:55:00 <Vorpal> that makes firefox switch over to proper rendering of English
23:55:00 <Sgeo> Basically, assignments INSIDE the block aren't scoped to stay in the block. To do that, you use a let statement that contains the block
23:55:12 <elliott> omg a let statement how awful
23:55:20 <elliott> providing an alternative to imperative assignment
23:55:29 <elliott> cpressey: it's not *that* bad, afaict
23:55:34 <elliott> since assignment is an imperative thing
23:56:05 <cpressey> elliott: It's mainly the Hey What It's All In Chinese thing.
23:56:10 <Sgeo> elliott, I'm ok with imperative assignment. I'm not ok with imperative assignments having global scope, while telling people that the language is similar enough to C++ and Java
23:56:13 <elliott> cpressey: Oh no, the language is terrible.
23:56:16 <elliott> cpressey: The assignment isn't.
23:56:23 <elliott> Sgeo is just excellent at hating the opposite things that he should.
23:56:27 <Vorpal> cpressey, the grammar in that link is.... really interesting
23:56:40 <elliott> Vorpal: "has"/"had"/"have" -- you, in all the wrong situations
23:57:01 <Vorpal> elliott, yes but that one I don't notice so much :P
23:57:34 <cpressey> I had the impression "let statement that contains the block" meant the block was contained in one of the *bindings* of the let statement, or something.
23:57:46 <cpressey> But then, it's been a long day.
23:57:58 <elliott> cpressey: Nope, just that a variable assignment persists beyond a block, and if you want to scope it to a block, you have to use let.
23:58:03 <elliott> Which, really, isn't *that* horrible.
23:58:08 <Sgeo> '" Fuxi does not provide constructs for looping, like for-statement or while-statement."
23:58:16 <elliott> Sgeo: Neither does Haskell.
23:58:18 <Sgeo> Can I curse Fuxi for false advertising again?
23:58:25 <cpressey> Not compared to say Python, which does that, except without the benefit of a "let" to reign it in
23:58:28 <Sgeo> elliott, Haskell does not claim to be easy for Java and C++ programmers
23:58:37 <elliott> Sgeo: If they provide an equivalent construct in the library, that does not invalidate their statement.
23:59:16 <Vorpal> um, tail recursion isn't hard to learn for an imperative programmer
23:59:18 <cpressey> Sgeo: If you fall in love with Haskell, I'll be slightly less creeped out than if you fall in love with Factor or Smalltalk.
23:59:41 <elliott> cpressey: As long as he doesn't make out with it in here.
23:59:53 <Sgeo> "1) it takes a conventional syntactic system, that makes the programmers with some experiences of C++, JAVA, use the language without any training;"