←2009-10-12 2009-10-13 2009-10-14→ ↑2009 ↑all
00:03:16 <SimonRC> it seems to be bed-time for me
00:15:26 <madbr> well, you could make a video-game system for small games and possibly allow some more classic operations
00:32:22 <Ilari> Wow... T_c 254K superconductor. I have seen ordinary freezers that can go below that temperature...
00:36:24 <madbr> mostly you could "hardware accelerate" some graphics or sound operations, and there's ample space for craziness in there :D
00:50:05 <madbr> 254k semiconductor? what? where?
00:52:54 <coppro> 254K
00:53:09 <coppro> or 254 K, to be pedantic
00:59:42 <madbr> wikipedia lists superconductors with temperatures up to 138k
01:00:12 <coppro> neat fact: Wikipedia does not reflect reality
01:00:59 <madbr> and I can't find a news item on it
01:02:04 -!- puzzlet has quit (Read error: 104 (Connection reset by peer)).
01:02:05 <coppro> neat fact #2: Wikinews doesn't reflect reality either
01:02:09 -!- puzzlet has joined.
01:02:10 <coppro> Try /.
01:02:29 -!- puzzlet has quit (Read error: 104 (Connection reset by peer)).
01:05:27 <madbr> yeah ok i got one
01:05:43 <madbr> still looks a bit rough, the kind of results that need confirmation
01:07:09 -!- puzzlet has joined.
01:10:08 -!- fungot has quit (Read error: 148 (No route to host)).
01:10:11 -!- fizzie has quit (Read error: 148 (No route to host)).
01:10:55 <Ilari> Well, pushing T_c higher and higher is fun. Then there are the actually useful superconductors. Like Magnesium diboride. One of the very few non-ceramic superconductors that don't need LHe cooling (LH2 cooling suffices).
01:13:46 <Ilari> Non-ceramic => No lattice to worry about.
01:14:55 <coppro> lattice?
01:15:07 <coppro> also, lolboron
01:17:14 <Ilari> Some wild proposals have pipe that transports both electricity and hydrogen by having MgB2 to transport electricity and use liquid hydrogen for cooling while it is being transported.
01:18:17 <coppro> :D
01:19:42 <Ilari> Well, there are all kinds of wild proposals that "it only takes some engineering". Including 5000km train tunnel that crosses the atlantic...
01:21:00 <Ilari> At least that wouldn't require any new materials, unlike stuff like space elevator...
01:21:23 <coppro> that's actually a brilliant idea
01:21:35 <coppro> (the superconducting hydrogen pipe)
01:22:35 <Ilari> Actually, it would be more like cable than pipe...
01:22:42 <coppro> yeah, I guess
01:22:54 <coppro> still, it's a great idea
01:23:19 <Ilari> Wonder what B_c that superconductor would have at LH2 boiling point...
01:23:57 <coppro> B_c?
01:24:35 <Ilari> Critical magnetic flux strength.
01:24:47 <Ilari> *flux density
01:26:40 <Ilari> That determines how much current superconductor can carry.
01:31:42 -!- fizzie has joined.
01:40:23 -!- oerjan has joined.
01:58:19 <oerjan> !interps
01:58:35 <oerjan> !languages
01:58:40 <oerjan> !help
01:58:41 <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>.
01:58:46 <oerjan> !help languages
01:58:46 <EgoBot> languages: Esoteric: 1l 2l adjust asm axo bch befunge befunge98 bf bf8 bf16 bf32 boolfuck cintercal clcintercal dimensifuck glass glypho haskell kipple lambda lazyk linguine malbolge pbrain perl qbf rail rhotor sadol sceql trigger udage01 underload unlambda whirl. Competitive: bfjoust fyb. Other: asm c cxx forth sh.
01:59:02 <oerjan> no lua i see
02:00:04 <madbr> !bf_txtgen a
02:00:21 <madbr> hm
02:00:22 <oerjan> madbr: i thought of a bitstring representation that should make quines easy
02:00:23 <EgoBot> 39 ++++++++++++[>++++++++>+>><<<<-]>+.>--. [34]
02:00:38 <madbr> oerjan: oh?
02:01:07 <oerjan> (<,$,(<,) <,$,(>,) etc...)
02:01:22 <oerjan> then you just have to pass it the function of what to do with each bit
02:01:26 <oerjan> iiuc
02:04:35 <Ilari> Closely associated with quines are self-f's. Including self-hashes (program which prints hash of its source code).
02:05:17 <oerjan> then print (, the representation above for each bit, ), then the raw bitstring
02:06:44 <madbr> ah, make bit array, print encoded form and then raw form of the bitstring?
02:06:46 <oerjan> and the bitstring is the string for that printing program
02:06:51 <oerjan> yep
02:07:48 <coppro> !help bf_txtgen
02:07:48 <EgoBot> Sorry, I have no help for bf_txtgen!
02:08:20 <oerjan> coppro: it just makes a bf program to print the string you give
02:08:37 <coppro> Okay.
02:09:23 <Ilari> !bf_txtgen foobar
02:09:25 <EgoBot> 66 ++++++++++[>++++++++++>+++++++++++>+><<<<-]>++.>+..<----.-.>+++.>. [310]
02:10:25 <oerjan> madbr: otoh a shorter bit representation will make the quine shorter, probably
02:10:38 * Sgeo wrote something like txtgen for use with PSOX
02:10:47 <Sgeo> However, I did it crappily, with no loops
02:11:23 <Sgeo> Well, actually, it does use [-]
02:11:24 <Sgeo> http://trac2.assembla.com/psox/browser/trunk/utils/TEXT2BF.py
02:11:32 <oerjan> afk
02:18:07 <madbr> ok I see
02:18:18 <madbr> I've never written a quine before
02:19:51 <Ilari> madbr: Writing quine in Lua is quite simple. First version that I got to syntax-check worked correctly.
02:20:17 <oerjan> Ilari: he's not writing a quine in lua, he's writing a quine in his esolang written in lua
02:20:20 <oerjan> afk again
02:21:42 <Ilari> madbr: The basic idea: Write code that takes some kind of string representation of itself, then prints that string representation as it was and then the string representation expanded into code.
02:23:48 <Ilari> madbr: Its easier to first write the code, then work backwards to encode it as data.
02:26:54 <madbr> yeah, a lua quine is not very hard: s=[[io.write("s=[".."["..s.."]".."]"..s)]]io.write("s=[".."["..s.."]".."]"..s)
02:29:07 <madbr> let's see, the symbol set is &<>,$()!?
02:29:22 <madbr> but I don't need ? so that makes it 8
02:29:59 <madbr> yeah I can represent those as (<<<,) etc
02:43:08 <madbr> yeah, I can see how you can do it, but it will generate a fairly huge program
02:46:42 * Sgeo has never written a quine before :(
02:49:33 -!- Pthing has joined.
02:50:31 <oerjan> madbr: oops, that representation above should be (<&,$,(<,) <&,$,(>,) etc...)
02:51:41 <oerjan> i wasn't thinking about compressing to just commands though, just characters, since that would be easier (but more verbose)
02:52:02 <oerjan> 3 vs. 8 bits per command
02:52:03 <madbr> that's equivalent to (, , etc) no?
02:52:27 <madbr> hmm
02:52:29 <oerjan> yes, but you need it only for printing
02:52:32 <madbr> aha
02:52:36 <madbr> neat
02:53:49 <oerjan> otherwise i guess ($,(<,) $,(>,) ...) would be more flexible
02:54:14 <oerjan> um wait
02:54:47 <oerjan> actually with your tuple closures that could work...
02:55:05 <madbr> yeah but your first idea sounded nicer
02:56:46 <oerjan> if you want to handle them three at a time, (<&,$,&&(<,)(<,)(>,) ...)
02:57:49 <oerjan> (for a whole compressed command)
02:59:06 <oerjan> !haskell map fromEnum "<>,&$!()"
02:59:11 <EgoBot> [60,62,44,38,36,33,40,41]
02:59:52 <oerjan> that is only 5 bits varying actually
02:59:55 <oerjan> fwiw
03:01:17 <oerjan> oh wait
03:01:41 <oerjan> (<<<,) (<<>,) etc. for your command representations
03:02:00 <oerjan> much easier to branch on
03:09:44 <madbr> it's fairly easy to do a version that prints bit per bit though
03:09:55 <oerjan> yeah it should be
03:10:31 <oerjan> you don't have to do any explicit looping, i think
03:12:05 <madbr> print (, then enter data function with parameter=print "<&,$,(" then switch between "<" and ">" - there's only 1 bit difference! - then ",)"
03:12:56 <madbr> then print ), then enter data function again but with (!,) as a parameter :D
03:13:50 <oerjan> :)
03:14:06 <madbr> this is too easy
03:16:01 <oerjan> it's 9*8 data chars to encode one command though, so i think you can easily trade ease for compression here
03:17:11 <madbr> yeah but then I have to decode commands into particular ascii chars, which sounds like it would take more code
03:17:21 <oerjan> true
03:19:29 <madbr> the resulting added complexity might reduce the data compression advantage by a lot
03:20:35 <oerjan> otoh printing "<&,$,(>,)" itself is going to take some room
03:22:23 <oerjan> !(>) or !(<) for each bit in those, 4*8*9 commands
03:22:28 <oerjan> er
03:22:41 <oerjan> !(>,) or !(<,) for each bit in those, 5*8*9 commands
03:22:53 <oerjan> that's going to be the main part of the code
03:23:48 <oerjan> !haskell 5*8*9 * 9*8
03:23:50 <EgoBot> 25920
03:23:55 <oerjan> wow
03:24:06 <Sgeo> !haskell (unsafeCoerce 5)::String
03:24:17 <oerjan> surely not imported
03:24:23 <Sgeo> Why is EgoBot trying to chat with me over DCC?
03:24:35 <oerjan> to give you a long error message, i presume
03:24:44 <Sgeo> An error message consisting of "b"
03:24:45 <oerjan> everything longer than 1 line is sent over DCC
03:24:47 <Sgeo> and a blank line
03:24:52 <oerjan> huh
03:24:52 <Sgeo> Try it
03:24:58 <oerjan> !haskell (unsafeCoerce 5)::String
03:25:18 <oerjan> 04:25 =EgoBot> /tmp/input.20929.hs:1:0: Invalid type signature
03:25:28 <Sgeo> I must not have gotten it soon enoguh
03:25:35 <Sgeo> Did EgoBot pm you?
03:25:40 <Sgeo> !haskell (unsafeCoerce 5)::String
03:25:48 <oerjan> DCC. i have it set up to autoaccept EgoBot
03:26:03 <Sgeo> Now I get the typesafe thingy
03:26:15 <Sgeo> Why did I get b before?
03:26:23 <oerjan> heck if i know
03:26:35 <Sgeo> !haskell repeat 1
03:27:02 <Sgeo> !haskell take 5 $ repeat 1
03:27:04 <oerjan> iirc EgoBot has trouble with infinite output without newlines
03:27:06 <EgoBot> [1,1,1,1,1]
03:27:09 <Asztal> ghci> unsafeCoerce 5 :: String
03:27:11 <Asztal> ""
03:27:17 <Asztal> that was sort of unexpected.
03:27:22 <oerjan> heh
03:28:44 <oerjan> madbr: that 25920 above is my minimum estimate on the character length of the quine by this method btw
03:29:51 <oerjan> hm ! returns its argument doesn't it. so you can compress equal runs of bits fwiw
03:30:22 <oerjan> so maybe not that large
03:31:31 <oerjan> oh and maybe do something with the 3 common most significant bits
03:31:41 <madbr> hmm
03:40:39 <madbr> I think I have it
03:41:06 <madbr> the quine and the result have the same filesize - 39 276 bytes :D
03:41:34 <oerjan> good, good
03:41:38 <madbr> wait
03:41:39 <madbr> haha
03:41:54 <madbr> I messed up
03:42:11 <madbr> It turned < into > and vice versa :D
03:48:40 <oerjan> how much harder is $,(>,) i wonder, should be shorter
03:48:42 <madbr> can't see any difference in the start or end, so it probably worked :D
03:48:50 <madbr> 40006 byte quine :D
03:48:53 <oerjan> you don't have diff?
03:49:06 <madbr> in windoze?
03:49:08 <Gregor> ZEE
03:49:23 <oerjan> alas
03:49:52 <Gregor> Observation: Images carefully taken losslessly will interpolate decently up to about 10x in terms of pixel count (roughly 3x each dimension)
03:49:54 <oerjan> <- windows too, but has a linux shell account when he needs it
03:50:15 <Gregor> Further observation: Images not taken losslessly will interpolate decently up to about 0.95x :P
03:50:34 <madbr> yeah it worked, it's just that the quine is too large to post on the wiki :D
03:50:49 <oerjan> indeed
03:51:32 <madbr> but yeah it uses 72 bytes of data for each original byte :D
03:52:35 <coppro> madbr: get diffutils
03:53:02 <madbr> http://s.engramstudio.com/src/quineprogram.txt
03:55:24 <madbr> http://s.engramstudio.com/src/quineresult.txt
03:55:34 <madbr> yeah, I checked, there is no difference
03:56:09 <madbr> (switching between 2 notepads with the same window settings... yes I know that is incredibly ghetto :D)
04:11:21 <madbr> The quote is "Please type 'The quote is ', the quote inside quotation marks, a period and a space.". Please type 'The quote is ', the quote inside quotation marks, a period and a space.
04:13:12 <madbr> Wait, no, that doesn't work :(
04:13:21 <oerjan> you're missing the final part
04:13:56 <madbr> yeah
04:14:19 <madbr> haha, I have a much better one :D
04:14:26 <madbr> Please type this.
04:14:38 <Asztal> this
04:14:42 <Asztal> (Had to be done.)
04:14:55 <augur> squeee
04:14:57 <augur> :3
04:15:04 <augur> i made my first type calculator
04:15:09 <madbr> <mad> Please type this.
04:15:09 <madbr> <qualo> reduz, nobody's playing
04:15:09 <madbr> <reduz> oh :(
04:15:09 <madbr> <coda> Please type this.
04:15:09 <madbr> <reduz> so no go?
04:15:10 <madbr> <qualo> mad, Please type this.
04:15:12 <madbr> <reduz> Please type this.
04:15:56 <oerjan> that's a cheating quine
04:16:01 <Asztal> the dreaded IRP quine :(
04:16:04 <madbr> Well, yeah :D
04:17:11 <augur> :|
04:21:31 <madbr> X is ". Type 'X is', X inside quotation marks and X.". Type 'X is', X inside quotation marks and X.
04:22:28 <oerjan> X is". Type 'X is', X inside quotation marks and X.". Type 'X is', X inside quotation marks and X.
04:22:39 <madbr> hahaha :D
04:22:47 * oerjan whistles innocently
04:22:49 <madbr> :D
04:24:07 <oerjan> ^ul (:aSS):aSS
04:24:16 <oerjan> dammit fungot
04:24:25 <oerjan> !underload (:aSS):aSS
04:24:26 <EgoBot> (:aSS):aSS
04:24:33 <madbr> haha what language was that again :D
04:24:46 <oerjan> underload, naturally
04:25:56 <oerjan> also very simple, functional concatenative. with text representations so quines are super easy.
04:27:02 <madbr> woot
04:27:04 <oerjan> no input though
04:27:19 <madbr> yeah it's a lot like the kind of language I'm trying to do
04:27:48 <madbr> especially my first one which was a stack functional language too... underload is better though :D
04:42:00 -!- coppro has quit ("I am leaving. You are about to explode.").
04:55:38 -!- coppro has joined.
05:33:24 -!- oerjan has quit ("leaving").
05:37:13 -!- coppro has quit (Remote closed the connection).
05:46:02 -!- Slereah has quit (Read error: 110 (Connection timed out)).
06:00:13 <fizzie> There's a diff in Windows too; it's called "fc", and it's crummy, but it does compare two files.
06:02:02 -!- calamari has joined.
06:21:09 -!- Asztal has quit (Read error: 110 (Connection timed out)).
06:35:02 -!- Slereah has joined.
06:39:19 -!- Sgeo has quit (Read error: 110 (Connection timed out)).
06:40:05 -!- Gracenotes has joined.
06:44:44 -!- calamari has quit ("Leaving").
06:55:38 -!- madbr has quit ("Radiateur").
06:57:54 -!- FireFly has joined.
07:03:50 -!- MigoMipo has joined.
07:06:52 -!- sebbu has joined.
07:12:38 -!- sebbu2 has quit (Read error: 60 (Operation timed out)).
07:49:55 -!- MigoMipo has quit.
07:59:59 -!- clog has quit (ended).
08:00:00 -!- clog has joined.
08:22:02 -!- FireFly has quit ("Later").
08:23:19 -!- ehird has joined.
08:23:33 <ehird> 15:25:12 <madbr> you know what would be cool? an esoteric video game system
08:23:34 <ehird> 15:26:32 <SimonRC> not really
08:23:34 <ehird> story of asiekierka's life
08:26:52 <ehird> anyway, good morning!
09:03:29 -!- adam_d has joined.
09:10:15 <ehird> "WHY Buy NEW?...When USED Will DO ! ™"
09:11:19 <ehird> I'm going to try going Dvorak cold-turkey again.
09:11:48 <ehird> First thing first, set it as part of my desktop background in the bottom-left so I can glance at it.
09:12:16 <ehird> I wonder if I should try Programmer Dvorak.
09:40:36 -!- adam_d_ has joined.
10:00:08 -!- adam_d has quit (Read error: 110 (Connection timed out)).
10:46:31 -!- ehird has quit.
11:00:55 -!- ehird has joined.
11:07:24 -!- MigoMipo has joined.
11:07:33 <ehird> Why isn't there a USB→Bluetooth adapter? Just plug in a little stub instead of the USB cord to a device, and it's a Bluetooth device.
11:10:21 -!- BeholdMyGlory has joined.
11:17:09 <fizzie> There might be some problems getting power to the stub, since I think the USB specs sort of say the "host" side of the connection has to provide it.
11:18:54 <ehird> Wireless power!
11:19:07 <ehird> fizzie: Well, Bluetooth devices have batteries.
11:19:11 <ehird> So the little stub has batteries, obviously.
11:19:19 <ehird> So it can provide power.
11:19:31 <ehird> (Presumably different sizes for different battery capacities based on device power usage.)
11:21:23 <ehird> fizzie: Am I crazy, by the way? I have the urge to buy a $325 product and brutally hack it up to my needs, and thus of course voiding that wonderful warranty and any possibility of returning.
11:21:30 <ehird> Don't answer that, I know the answer.
11:21:54 <fizzie> Presumably you can't really fit very much power in there before the stub starts to become cumbersome, especially since wired-USB devices probably haven't really been designed with power savings in mind. I guess you could get some sort of "working" solution, but it doesn't sound so obviously marketable.
11:22:22 <ehird> Also, um, I'm pretty sure you could power most USB stuff with two AA batteries.
11:22:37 <ehird> For weeks. I'm talking stuff like keyboards and mice.
11:23:02 <ehird> (Presumably with a specially-shaped one for different types of mice :P)
11:24:55 <fizzie> I'm not so completely sure about that if the mouse in question doesn't bother with power-saving so much; wireless mice in general are a lot more aggressive in turning things off when the mouse doesn't seem to be moving. And two AA batteries sounds more in the "blob with maybe a wire" category than "small stub".
11:25:06 <ehird> Well, alright then.
11:25:20 <ehird> Anyway! Someone ask me what I want to hack up, and what the hack in question is.
11:25:46 <fizzie> I've seen at least one "make a device 'wireless'" solution in the form that there's a USB stick receiver which talks the wireless-USB standard to a four-port USB hub you can plug things in. It doesn't say in this description whether the hub is battery-provided, or whether it uses a power brick, which makes the wirelessness a bit silly.
11:26:16 <fizzie> What do you want to hack up, and what the hack in question is? (See, I can follow instructions.)
11:26:18 <ehird> Could be useful for controlling, say, a media PC in another room.
11:27:09 <ehird> fizzie: The http://www.kinesis-ergo.com/advantage.htm keyboard, and to add a TrackPoint (including buttons and maybe even a scroll wheel). Also, possibly replace the keyswitches with something nicer; I know they're mechanical and I don't think they're clicky, but who knows.
11:27:42 <ehird> Technically I guess it's $299, since I don't want or need hardware QWERTY/Dvorak switching.
11:31:37 <fizzie> There certainly would be ample space to stick those every-finger trackpoints oklopol wanted on that huge slate.
11:31:55 <fizzie> But yes, it sounds crazy. Wait, I wasn't supposed to answer that. Never mind.
11:32:40 <ehird> fizzie: I don't even know how I'd even think about going about doing it; I'm pretty sure I'd have to carve out the eight keys (four around each nub (two nubs, one per side))
11:32:45 <ehird> as in, the edges
11:33:16 <fizzie> You can melt the keys appropriately; as a bonus, it then doubles as a modern-art peice.
11:33:19 <ehird> And replacing the switches would be *major* surgery...
11:33:22 <ehird> fizzie: :D
11:33:44 <ehird> fizzie: With rubber dome keyboards you can melt it BEYOND REPAIR.
11:33:51 <ehird> Well, that applies to all things I guess.
11:33:53 <ehird> Entropy and all.
11:35:02 <fizzie> They could have some sort of "will it melt" youtube show, but maybe that's a bit silly, since I guess just about anything melts if you just try hard enough.
11:36:23 <ehird> Exactly :P
11:37:04 <ehird> Will It Eventually Disperse Into Nothingness As The Universe Further And Further Approaches Infinite Disorder
11:42:26 <ehird> "Will it Exist'
11:42:29 <ehird> *Exist"
11:46:12 -!- bsmntbombdood has quit (Read error: 145 (Connection timed out)).
11:56:33 -!- bsmntbombdood has joined.
11:58:11 -!- BeholdMyGlory has quit (Remote closed the connection).
12:02:20 -!- MigoMipo has quit ("Page closed").
13:48:19 -!- Pthing has quit (Remote closed the connection).
13:50:29 -!- MigoMipo has joined.
14:10:51 -!- ais523 has joined.
14:12:42 -!- oerjan has joined.
14:13:30 <oerjan> <ehird> story of asiekierka's life
14:13:34 <oerjan> erm wait?
14:13:57 * oerjan looks at madbr suspiciously. or would, if he were here.
14:20:28 -!- ais523 has quit (Read error: 104 (Connection reset by peer)).
14:21:02 -!- ais523 has joined.
14:28:56 -!- ais523_ has joined.
14:29:02 -!- ais523 has quit (Read error: 104 (Connection reset by peer)).
14:31:09 -!- ais523_ has changed nick to ais523.
14:33:36 <ais523> wow, they made a superconductor that works at -19 degrees C
14:37:10 <oerjan> or so they claim, i recall reddit comments casting doubt on it
14:39:05 <Ilari> And even if it was true, it might not be very useful...
14:42:41 <ais523> it's probably affected by stray magnetic fields, or whatever
14:43:03 -!- Sgeo has joined.
14:47:03 <oerjan> yeah that was mentioned, although it seemed even more dubious than that
14:47:44 <oerjan> someone said the graph looked like random noise rather than a successful experiment
14:49:03 <ais523> heh
14:53:29 -!- Pthing has joined.
15:07:19 -!- MigoMipo has quit ("Page closed").
15:11:40 <Gregor> Wowsa.
15:11:43 <Gregor> -NickServ- 6 failed logins since last login.
15:11:43 <Gregor> -NickServ- Last failed attempt from: GrEgOr!n=GrEg@95.235.58.21 on Oct 11 19:04:36 2009.
15:11:53 <Gregor> Apparently now that I have the nick "Gregor", people try to steal it :P
15:17:05 -!- Sgeo has quit (Read error: 110 (Connection timed out)).
15:35:24 -!- FireFly has joined.
15:48:57 -!- fax has joined.
15:49:05 <ehird> 12q33w4e5dt67y12345e6r76t87y89u9i234w5ed6fr7t78gyh8u90iJ_0Okp322222222222wsss4ed5rf6tg77hy8u99ji00ko--23w4se5drrrrrft67ghu89i0-oooop123w4e5dgt67uy89j0ik-ol=;23w4e5dt67ghu89j0iko-p=[
15:49:12 <ehird> Yo.
15:49:57 <oerjan> your ehird impersonation needs work, bot
15:50:10 <ehird> FUCK YOU AND ALL YOUR FAMILY
15:50:22 <oerjan> ok, getting closer
15:54:43 -!- coppro has joined.
16:03:10 <ehird> "The delete key is located directly above the enter key; the key normally found there is the second rightmost key on the row above it. Furthermore, this is an actual delete key, not a historically named backspace. Backspace is accessible through Fn+Delete."
16:03:26 <ehird> They seriously think Delete is more common than Backspace?
16:03:36 <Deewiant> Context?
16:03:54 <ehird> Happy Hacking Keyboard.
16:04:09 <ehird> I have to research it; it has topres!
16:04:18 <Deewiant> :-P
16:05:01 <ehird> http://imgur.com/cxMKL.png
16:05:37 <oerjan> wth is/are topres
16:05:51 <ehird> oerjan: Capacitive keyswitches.
16:06:02 <ehird> Combination of spring + rubber dome.
16:06:15 <ehird> Deewiant: How's those Browns, btw?
16:06:37 <Deewiant> They're alright
16:06:53 <Deewiant> WTF is that pic :-D
16:07:22 <oerjan> keyboard cat a few years later
16:07:42 -!- MigoMipo has joined.
16:07:43 <fax> nothing can beat humble hacker
16:07:53 <fax> best keybeard I ever saw
16:08:01 <fax> <http://www.humblehacker.com/Keyboard/>
16:08:02 <Deewiant> Evidently from _why's guide
16:08:21 <oerjan> keybeards for your wearable computer
16:08:23 -!- Pthing has quit (Remote closed the connection).
16:10:00 <ehird> Deewiant: I made that pic
16:10:07 <ehird> Deewiant: In the poignant guide, s/keyboard/cigarette/
16:10:15 <ehird> fax: keybeard? :D
16:10:15 <Deewiant> Well yes, the edit there is obvious
16:11:40 <ehird> The Humble Hacker Keyboard - geekhack forums
16:11:40 <ehird> 19 Jun 2009 ... A project that I've been working on off and on for more than a year now, the Humble Hacker Keyboard (i.e. a keyboard for humble hackers like ...
16:11:40 <ehird> geekhack.org/showwiki.php?title=Island:6292 - Cached - Similar -
16:11:47 <ehird> I'm pretty sure you can't call yourself humble.
16:12:12 <ehird> Nice apple.com rip, though...
16:13:06 * oerjan is far too humble to call himself humble
16:14:34 <ehird> How to beat the DataHand II Professional in cost for a keyboard: $359 Kinesis Advantage Pro + like $200 Topre keyswitches + $50 replacement ThinkPad keyboard to extract TrackPoint from + $50 same = $659. Close, close.
16:15:43 <ehird> I guess the Topres would be more like $400-$600 without a bulk deal...
16:17:18 -!- BeholdMyGlory has joined.
16:17:35 <ehird> Such fun, anyway.
16:18:39 <ehird> I should see what Apple Extended Keyboard[ II]s are going for today. Argh, I'm too addicted! Help help help help help help help help help help help help help
16:19:08 <ehird> Wow, a new AEKII with adapter for £80.
16:19:14 <ehird> Now that's something you don't see every day. Unopened...
16:19:20 <oerjan> i'm afraid the only cure is to cut off your hands. no hands, no keyboard.
16:19:54 <fizzie> Some major kbd manufacturer should randomly make a dozen keyboards that'd type "Help, I'm trapped in a keyboard factory." every time you press F7.
16:20:32 <ehird> Why F7?
16:20:41 <ehird> oerjan: I'd just obsess over voice and foot interfaces.
16:20:50 <ehird> "Ooh, nice speech actuation amplifier..."
16:20:54 <fizzie> I just picked a random not-so-common key.
16:20:59 <ehird> "Hmm... more use out of the tendons? Gimme!"
16:21:32 <oerjan> ehird: hm at that point i think only buddhism helps
16:21:33 <coppro> fizzie: Try Ab
16:21:44 <ehird> coppro: Ab?
16:21:50 <ehird> oerjan: :(
16:21:53 <coppro> a random, not-so-common key!
16:22:02 <ehird> aaaaaaaaargh
16:22:05 <ehird> oerjan: Swat him quick!
16:22:10 * coppro ducks
16:22:15 <ehird> DUCKS CANNOT HELP YOU
16:22:17 <ehird> THEY WILL JUST QUACK
16:22:53 <oerjan> i cannot possibly swat for a crime i do not understand
16:23:03 <ehird> oerjan: wut
16:23:07 <oerjan> oh wait
16:23:27 * oerjan swatteth coppro -----###
16:23:50 <ehird> <ducks> quack
16:23:53 <ehird> <ducks> quack quack
16:23:58 <ehird> <ducks> quaaaaaaaaaaaack
16:24:28 * oerjan swats one of the ducks at ehird -----###
16:24:37 <ehird> Hey!
16:24:42 <ehird> I helped your swatting!
16:24:48 * ehird gets a duck to bite off oerjan's nose
16:24:52 <ehird> <duck> quack
16:25:14 <oerjan> ow by dose
16:25:17 -!- coppro has changed nick to walter.
16:25:24 <walter> neat, nick is available
16:25:31 <walter> now if only I had my bot to use it :(
16:25:33 -!- walter has changed nick to coppro.
16:25:41 <ais523> hmm... people here, try to guess which language the code I've been asked to maintain is written in
16:25:46 <ais523> (as part of my job)
16:25:47 <ehird> ais523: Java
16:25:49 <ehird> COBOL
16:25:50 <ehird> INTERCAL
16:25:51 <ehird> C++
16:25:52 <coppro> MUMPS
16:25:55 <ais523> elisp
16:25:57 <ehird> coppro: IT'S CALLED M :P
16:25:58 <ais523> yes, really
16:26:01 <coppro> ...
16:26:03 <ais523> and no, the code has nothing to do with Emacs
16:26:05 <ehird> ais523: hmm, right so, that job, I'd quit it?
16:26:10 <ehird> that is what i would do
16:26:17 <ais523> ehird: it's not that bad, at least I know elisp
16:26:24 <coppro> My job is Python and C++
16:26:25 <ehird> I know Brainfuck too
16:26:25 <fax> who the hell writes elisp that isn't to do with emacs??
16:26:32 <ais523> although using it as a CGI script is one of the weirder uses of it that I've seen
16:26:34 <ais523> fax: no idea
16:26:36 <ehird> ais523: WHAHT
16:26:40 <ehird> ais523: Seriously, man
16:26:42 <ais523> somehow I doubt the code is going to be high-quality
16:26:47 <ehird> Duuuuuuude
16:26:54 <fax> lol
16:26:54 <ais523> I haven't seen it yet
16:26:58 <fax> I also doubt
16:27:07 <ehird> I'm pretty sure McDonalds is preferable to maintaining elisp CGIs
16:27:09 <ehird> :-P
16:27:20 <coppro> No. No it's not.
16:27:42 <ais523> actually, I was given three choices by the people in charge:
16:27:43 <ehird> well, almost
16:27:51 <ais523> maintain the code so it works (they haven't tried to run it yet)
16:27:55 <ais523> rewrite it from scratch
16:27:57 <ais523> or do the work it does by hand
16:28:01 <ehird> 2
16:28:02 <ais523> the third would take me a couple of days
16:28:08 <ais523> 2 seems potentially an interesting one
16:28:14 <ehird> unproductive, but less self-hating
16:28:18 <coppro> Can you legally tell us what it does?
16:28:21 <ehird> and whoever has elisp cgis doesn't deserve productivity
16:28:46 <ais523> coppro: several different things, mostly related to trying to fairly allocate things with first/second/third choices and keep track of them after they've been allocated
16:28:57 <ehird> wat
16:29:00 <coppro> oh, ok
16:29:01 <ehird> you mean
16:29:04 <ehird> maintain lists?
16:29:07 <coppro> shouldn't be too hard to rewrite then
16:29:17 <ais523> as far as I can tell, most of what it's meant to do is trivial
16:29:35 <ais523> which is why I'm encouraged enough to at least have a look at the elisp
16:29:38 <ais523> once they figure out where it is
16:30:11 <ehird> Why isn't reddit's css loading for me
16:30:54 <ehird> http://www.reddit.com/static/reddit.css?v=79f52686b6e52d3db3bddb54e3852e84
16:30:54 <ehird> ends at
16:30:55 <ehird> .leftpad { padding-left:1em }
16:30:57 <ehird> .nomargin { margin:0px }
16:30:57 <ehird> .nopadding { padding:0px }
16:30:57 <ehird> .hover a:
16:31:02 <ehird> same for anyone else?
16:31:03 <ehird> what follows it
16:31:25 <fizzie> .hover a:hover { text-decoration:underline }
16:31:25 <fizzie> .selected { font-weight:bold; }
16:31:26 <fizzie> ...
16:31:32 <fizzie> Quite a lot follows afterwards.
16:31:43 <ehird> stupid 3g
16:31:47 <ehird> being broken 'n shit
16:31:58 <fizzie> In other news, that's one big CSS file.
16:32:04 <ehird> how big
16:32:13 <ehird> argh
16:32:14 <fizzie> 14761 bytes.
16:32:15 <ehird> it works with curl
16:32:16 <ehird> maddening
16:32:23 * ehird clears cookies, restarts safari
16:33:15 <ehird> yay
16:35:07 -!- oerjan has quit ("leaving").
16:54:17 * ehird attempts to figure out what he can remove from his system to make room for snow leopard
16:54:26 <coppro> nothing
16:54:30 <coppro> Snow Leopard has bugs
16:54:39 <ais523> doesn't all software have bugs?
16:54:47 <ais523> even the traditional hello world doesn't check the return value from printf
16:55:23 <coppro> and do what with an error?
16:55:35 <coppro> http://www.theregister.co.uk/2009/10/12/snow_leopard_data_eating_bug/
16:56:16 <ehird> ais523: ignore coppro; he's an anti-Apple zealot
16:56:17 <ehird> that bug, fyi
16:56:21 <ehird> is nowhere near as simple as portrayed
16:56:29 <coppro> no, of course not
16:56:44 <coppro> a bug very rarely is
16:56:45 <ehird> You must upgrade from Leopard, to the Snow Leopard developer seeds, to Snow Leopard; and furthermore, have enabled the Guest account while still in Leopard, and never disabled it since
16:56:46 <ais523> from what I've heard on Slashdot, it causes massive data loss but only in a rather obscure corner case
16:56:52 <ais523> which nevertheless can happen by accident
16:57:03 <ehird> Yes, it's a bad bug... so fucking what? That's no reason not to install Snow Leopard.
16:57:04 <coppro> oh, didn't know you had to use a dev version
16:57:20 <ehird> That's what I've read; others contradict it, but the extra detail makes me think someone thought about this.
16:57:27 <coppro> I personally would wait until a fix was deployed to upgrade
16:57:28 <ehird> (You don't just add that upgrade step in a misrememberance)
16:57:30 <coppro> Apple or not
16:57:34 <ehird> coppro: I don't use the Guest account
16:57:42 <ehird> It isn't enabled by default, and I have no use for it
16:57:43 <ehird> furthermore, I will not be upgrading
16:57:47 <ais523> I heard that the upgrade step wasn't relevant, but you had to hard-reboot while in the Guest account, then log in as Admin with the next login
16:57:50 <ais523> which is also relatively unlikely
16:57:56 <ehird> ais523: I'm pretty sure that's false
16:58:02 <ais523> well, upgrade from Leopard is relevant
16:58:06 <ais523> but from the dev version isn't
16:58:30 <ais523> I think what all this row means is, people aren't entirely sure how to reproduce it reliably
16:58:32 <ehird> coppro: and congratulations for citing El Reg as a reliable source, you either have never read it before or are seriously deluded
16:58:51 <ehird> *especially* since it has a known strong anti-Apple slant
16:59:00 <coppro> ehird: I was just grabbing the quickest link I could find
16:59:29 <ehird> Anyway, I think a bug, serious or not, that only appears in an obscure, non-applicable to me in every way corner case is no reason not to install it.
16:59:46 <ehird> Even if I used the Guest account, I'd just disable it and re-enable it, which fixes the bug.
17:00:55 <ais523> El Reg has a strong anti-everyone slant, AFAICT
17:01:03 <ais523> sort of like the lecturer I had today
17:01:09 <ais523> who appears to hate everyone, but Americans in particular
17:02:00 <ehird> Oh, of course, but they're particularly ridiculous about Apple, who are the cause of everything bad, regardless of whether it happened or not, about everything even slightly related to something with their name on it
17:02:02 <ehird> Naturally.
17:02:08 <ais523> ah
17:02:17 <ehird> s/(?!\.)$/./, grr.
17:02:29 <ehird> (Most complex regexp to ever be used to fix a line on IRC? You decide!)
17:04:15 <ehird> someone ought to start a non-profit online tech newspaper without stupid FUD...
17:04:34 <fizzie> It would be better if it worked; that one adds the . always. You must have meant s/(?<!\.)$/./ which will only add it if necessary.
17:04:37 <ehird> without reader submission (although tips are fine)
17:04:46 <ehird> fizzie: ?! is what it is in Ruby
17:04:47 <ehird> iirc
17:04:51 <ehird> it means "not"
17:04:54 <ais523> ehird: your lookahead goes the wrong way
17:05:00 <ehird> [^.] would also work, technically.
17:05:09 <ais523> you're saying "if there's no full stop after the end of the string"
17:05:11 <ais523> which is always true
17:05:27 <fizzie> Yes. You'll always get a matching negative zero-width lookahead for . there if you anchor it at the $. You need the look-behind behaviour of <!.
17:05:38 <ehird> ah, oops
17:06:01 <ais523> to put it another way, $ and (?!) are both zero-width assertions
17:06:06 <ais523> so it doesn't matter what order they're in
17:06:15 <ais523> and s/$(?!\.)/./ is obviously nonsensical
17:06:17 <ehird> irb(main):001:0> "a".gsub /(?!\.)$/, "."
17:06:18 <ehird> => "a."
17:06:18 <ehird> irb(main):002:0> "a.".gsub /(?!\.)$/, "."
17:06:18 <ehird> => "a.."
17:06:19 <ehird> irb(main):003:0> "a.".gsub /(?<!\.)$/, "."
17:06:19 <ehird> => "a."
17:06:20 <ehird> irb(main):004:0> "a".gsub /(?<!\.)$/, "."
17:06:22 <ehird> => "a."
17:06:24 <ehird> yar
17:07:25 <fizzie> I did think about trying to fix it with another s///, but came into the conclusion that someone else would point out the problem while I would still be constructing the fix. (Or maybe not, since in retrospect it's just s/\?!/?<!/.)
17:07:58 <ehird> regexps are so ridiculous :)
17:09:06 <fizzie> But delicious!
17:11:25 <ais523> does anyone here know a reliable way to use up about 90% of your CPU in a busyloop?
17:11:37 <ais523> I'm trying to reproduce a CPU-speed-dependent bug
17:11:51 <ehird> ais523: hmm
17:11:56 <ehird> ais523: 100% user + nice(1)
17:12:02 <ehird> user = thing that uses
17:12:07 <ehird> toy with values
17:12:52 <ehird> Also, news: Dyson introduced a new product which happens to consist entirely of things people already know about and have combined; claims it as his own invention.
17:13:01 <ehird> Hmmmmmmmmmmmmmmmmm.
17:13:05 <ehird> Right, yeah, that's nothing new.
17:14:40 <fax> http://en.wikipedia.org/wiki/Dyson_sphere -- reinventing the (generalized) wheel
17:15:05 <ehird> fax: your pattern-matching ability is stellar
17:15:09 * ais523 nice -n 1 sh -c "while true; do :; done;"
17:15:31 <ais523> for some reason, sh didn't like "do done;"
17:15:45 <fax> that pun was sweet
17:16:47 <ais523> `wolfram dyson
17:16:48 <Deewiant> What's :?
17:16:52 <ais523> Deewiant: a null command
17:16:54 <ehird> Do nothing
17:16:59 <HackEgo> dyson \ \ Input interpretation: \ \ Dyson Group DYS \ Last close: \ \ £ 0.16 $ 0.26 \ Recent returns: \ \ DYS London Stock Exchange Monday 12:00 am EDT \ \ 36 hrs ago \ \ day 0.00 \ \ month 0.00 \ \ YTD 71.05 \ \ year 75.84 \ \ 5 year 95.27 \ \ Price history: \ £ 0.70 £ 0.60 £ 0.50 £ 0.40 £ 0.30 £ 0.20 £ 0.10
17:17:02 <ais523> IIRC it always returns true
17:17:04 <Deewiant> Is that actually POSIX?
17:17:05 <ehird> do; done works in zsh, btw
17:17:09 <ehird> Deewiant: Yes
17:17:14 <Deewiant> Alright, good to know
17:17:18 <Deewiant> ehird: do done as well.
17:17:23 <ehird> `wolfram dyson's age
17:17:24 <ais523> most shells allow redefinition of :
17:17:28 <HackEgo> $Failed \ \
17:17:29 <ais523> which is how the killer smiley works
17:17:47 <ais523> also, that's a worrying price histor
17:17:49 <ais523> *history
17:17:59 <ais523> presumably it's a PDF-scraping error, rather than an actual sequence
17:19:00 <ehird> http://www.wolframalpha.com/input/?i=dyson
17:19:03 <ehird> A lot more info her.
17:19:04 <ehird> *here
17:19:15 <ehird> (you're already violating the TOS by using a bot)
17:19:26 <ais523> no, HackEgo is
17:19:38 <fizzie> Yes, the "price history" is just the Y axis labels for a plot.
17:19:39 <ais523> I'm just typing stuff in an IRC channel...
17:19:41 <ehird> ais523: "No, your web browser is"
17:19:48 <ehird> You're just typing stuff in a form field
17:20:07 <ehird> you have clear intent to use W|A with those lines, just like you do with a browser
17:20:10 <ais523> ehird: I didn't make a connection to the Wolfram servers at all
17:20:24 <ehird> Indeed, you wouldn't with a browser, your ISP would
17:20:40 <ais523> I'd at least be sending the IP address to connect to a website the usual way
17:20:53 * ais523 is actually surprised Alpha still exists
17:20:56 <coppro> here's a better question: are such TOS enforceable?
17:21:07 <ais523> but then, Cuil still exists, too
17:21:08 <fax> ais523: why are you surprised?
17:21:32 <ais523> fax: just because everyone's mostly forgotten about it
17:21:32 <ehird> coppro: no.
17:21:37 <ehird> contract under duress
17:21:40 <ehird> like EULAs
17:21:42 <ais523> there's a court case in the US saying it is
17:21:43 <ehird> also, most people never see it
17:21:44 <ais523> but it's under appeal
17:21:50 <fizzie> ais523: "You may not in effect use Wolfram|Alpha through an alternate user interface presented by another website." Doesn't say anything about requiring you to make direct connections. (Also, curiously, doesn't say right *there* that you may not actually make such a website, but I guess it's covered by some other phrase.)
17:21:50 <coppro> ais523: which case?
17:22:00 <ais523> the one about driving a girl to suicide over MySpace
17:22:14 <coppro> can you find me a link please?
17:22:15 <ais523> for some reason, they decided to prosecute it as a violation of MySpace's TOS rather than any other way
17:22:30 <coppro> also, EULAs/TOS are different if you actually have to agree
17:22:53 <ehird>
17:22:55 <ehird> oops
17:23:15 <ais523> http://www.law.com/jsp/legaltechnology/pubArticleLT.jsp?id=1202434043364&Will_SCOTUS_Rule_on_the_Computer_Fraud_and_Abuse_Act
17:23:28 <coppro> there's still the factor of duress, as ehird pointed out, as well as whether the thing would be enforceable (many disclaimers wouldn't be enforceable in any circumstances), but I don't think W|A's TOS could be enforced at all
17:24:00 <ehird> yes, but ais523 never violates any law even possibly. well, okay, he's violating the TOS here due to "You may not in effect use Wolfram|Alpha through an alternate user interface presented by another website."
17:24:05 <ehird> but we're all fallible, right?
17:24:43 <coppro> hmm... that's a specific law, though, don't know how it's applicable in the general terms
17:24:50 <ais523> ehird: violating the TOS of a website that you aren't using is like violating the rules of a nomic you aren't playing
17:24:59 <ehird> you are patently using it
17:25:01 <coppro> (i.e. where cybercrime laws don't exist like in the US)
17:25:01 -!- AnMaster_ has joined.
17:25:16 <ais523> ehird: suppose you asked me what 2+2 was, and I looked it up on Alpha and said 4
17:25:17 <ehird> unless you want to throw intent out of the window
17:25:20 <ais523> would that be you using Alpha?
17:25:32 <ehird> in which case; congrats, you've neutered the law system
17:25:34 <coppro> on another note, I don't think software licenses are, strictly speaking, enforceable in Canada
17:25:58 <ehird> 222222222222222222222222222222222222222222
17:26:03 -!- AnMaster has quit (Nick collision from services.).
17:26:09 -!- AnMaster_ has changed nick to AnMaster.
17:26:17 <coppro> there's a Statue of Frauds-like requirement that copyright licenses be signed by both parties in Canada
17:27:06 * ais523 reads Cuil's privacy policy
17:27:10 <ais523> it claims to have an opt-out for adverts
17:27:12 <ais523> but I can't see it
17:27:19 <ehird> i'll find it
17:27:26 <ehird> http://www.cuil.com/prefs
17:27:31 <ehird> "Advertising Preferences"
17:27:39 <ais523> they don't appear on my screen
17:27:43 <ehird> took me 5 seconds and one click to find it after hitting cuil.com<enter>
17:27:47 <ais523> heh, I wonder if AdBlock blocked the preferences?
17:27:47 <ehird> ais523: that's your fault for disabling js
17:27:49 <coppro> ais523: want to know a true legal joy?
17:27:52 <ais523> (I tried enabling JS)
17:27:58 <ehird> WFM
17:28:03 <ehird> it's a link
17:28:35 <coppro> I have to explain to Lenovo/Microsoft at some point in the near future why, despite clicking the 'Accept' button and starting up Windows, they are nonetheless required to refund it under the terms of the license agreement as which I have not agreed to
17:28:45 <ais523> ehird: it worked when I turned JS on /and/ AdBlock off
17:28:47 <ais523> it seems to need both
17:28:57 <ehird> that blind-search-of-google-yahoo-and-bing thing should replace yahoo with cuil, now that yahoo=bing
17:29:00 <ehird> :D
17:29:01 <ais523> coppro: why did you click the Accept button?
17:29:06 <ehird> i bet it turns out to bee not actually terrible
17:29:21 <ehird> ais523: presumably he wants to use windows because he's crazy
17:29:27 <coppro> ais523: I didn't realize that Lenovo's idea of a Windows install disc is a "reimage your computer" disc
17:29:46 <ais523> actually, the thing that confuses me is, that the bit of the agreement that says you can send the software back for a refund if you don't agree is /in the agreement/
17:29:55 <ais523> so, if there's no agreement, they don't have to honour it...
17:30:02 <coppro> it's a unilateral contract
17:30:13 <coppro> independent from the main agreement
17:30:55 <ehird> hmm... http://www.cuil.com/search?q=cuil+sucks gives crap results
17:30:59 <coppro> in any case, I have two grounds to void the contract - one being misunderstanding (the contract clearly states there is a reinstallation disc), the other being that I'm a minor
17:31:10 <ais523> coppro: heh, I like the second reason there
17:31:29 <coppro> yeah, being under 18 has fun implications
17:32:13 <ehird> LOL; Microsoft have plastered a Bing ad over their acquired-and-then-left-to-die http://powerset.com/, basically saying "This is shit, use Bing"
17:33:19 <ais523> wow, they have as well
17:33:29 <ehird> "as well"?
17:34:02 <ais523> figure of speech
17:34:04 <ais523> I'm not sure what it means
17:47:23 -!- AnMaster_ has joined.
17:49:28 -!- Asztal has joined.
17:55:18 <ehird> so, corporate affairs just spat on the h2g2 trilogy's grave with the new book not written by douglas adams' publication.
17:56:00 <ehird> (I love long nested expressions in English.)
17:57:43 -!- AnMaster has quit (Connection timed out).
18:05:43 -!- kar8nga has joined.
18:07:05 <fizzie> Oh, they got that Artemis Fowl guy book published?
18:08:02 <ais523> beh, more spam phone calls
18:08:05 <ais523> they happen a lot in this office
18:08:15 <ais523> I just leave the phone off the hook in an attempt to waste their money
18:11:11 <ehird> fizzie: Ye.
18:11:12 <ehird> *Yes
18:11:21 <ehird> A travesty.
18:14:52 <ais523> gah, I've just spent half an hour reading a website about shoelaces
18:17:26 <ehird> haha, is it that really comprehensive one?
18:17:29 <ehird> that was linked from geekhack.org
18:17:36 <ehird> with the huge hieerarchical sidebar?
18:17:39 <ehird> *hierarchical
18:18:30 <ais523> http://www.fieggen.com/shoelace
18:19:47 <ehird> yep
18:20:02 -!- AnMaster_ has changed nick to AnMaster.
18:20:05 <ehird> ais523: http://www.fieggen.com/shoelace/IanKnot16.gif
18:20:09 <ehird> warning: hypnotising
18:20:11 <Deewiant> I spent a few hours reading that site once and didn't change my shoelacing habits at all
18:20:35 <ehird> Dyson should make a shoe that doesn't need laces because it uses magnets
18:20:37 <ehird> :-P
18:20:43 <ehird> also, it's made out of metal
18:20:57 <ais523> incidentally, I invented the security shoelace knot independently
18:21:15 <ais523> I've never been able to tie shoelaces, so I reinvented the shoelace knot from first principles
18:21:48 <ehird> ooh, the shoelace guy is double crazy
18:21:48 <ehird> This new section was created mainly to promote my breakthrough JPG image optimization program, JPGExtra. All of my programming is done in Assembly Language, which results in minimalist software programs that are tiny yet very powerful.
18:22:00 <Deewiant> Haha :-D
18:22:18 <Deewiant> http://www.fieggen.com/software/assembly.htm
18:22:35 <ehird> Every computer program is a sequence of instructions that the processor chip of the computer understands. Instructions such as: ADD THESE TWO NUMBERS, or PRINT THIS MESSAGE, or STORE THIS VALUE, or QUIT IF THE RESULT IS ZERO.
18:22:48 <ehird> print this message?
18:22:54 <ehird> is he sure he isn't programming in python instead?
18:23:11 <ehird> [[To me, this is a modern example of the old saying: The end justifies the means.]]
18:23:13 <ais523> in DOS, at least, it's two bytes of ASM to "print this message"
18:23:15 <ehird> warning— this guy may be a serial killer
18:23:16 <ais523> you just do an interrupt
18:23:19 <Asztal> He's cheating with INT 21h :(
18:23:26 <ehird> that's not an assembly instrucution, ais523
18:23:38 <ehird> and the processor doesn't understand it as that
18:23:41 <ais523> no, it doesn't
18:24:27 <ehird> "Despite the fact that Assembly Language can produce the most powerful and efficient programs, the majority of today's software (even "Windows" itself) in written in high-level programming languages.
18:24:27 <ehird> Software developers have many reasons for this, some quite legitimate, but mostly to do with maximising profits."
18:24:30 <ehird> whaaaaaaaaaaaaaaaaaaaaaaaaat
18:25:17 <ais523> if you write everything in asm, C is high-level
18:25:25 <ehird> ais523: see the last line
18:25:28 <ehird> is what i was talking about
18:25:34 <ais523> oh
18:25:45 <ais523> well, asm programmers need to be paid more than C programmer on avergae
18:25:45 <AnMaster> ehird, what an idiot. Sadly I seen people like that before.
18:25:46 <fax> lol
18:25:47 <ais523> *average
18:25:52 <ais523> *C programmers
18:25:56 <ais523> and also, the programs take longer to write
18:25:59 <ais523> which also reduces profits
18:26:18 <ehird> well, yes, but...
18:26:20 <ais523> (put another way: if writing everything in asm did produce the maximum profits, people would actually do it)
18:26:35 <ais523> maximising profits isn't necessarily a bad thing
18:26:36 <Deewiant> ehird: That's completely correct?
18:26:38 <ais523> sometimes it just means being sane
18:26:46 <ehird> It's not the reason it's done, though
18:26:55 <ehird> It's done because, you know, writing complicated software in asm is near-impossible
18:27:04 <ehird> Which, sure, you can abstract as "maximising profits"
18:27:04 <Deewiant> Exactly
18:27:11 <ehird> But then you can say that everything is done because of that, which is true and useless
18:27:15 <Deewiant> ---> it costs more to produce complicated software in asm
18:27:21 <ehird> And even more useless when given for a justification for any given action by a company
18:27:25 <ehird> Because it's always true
18:27:30 <ais523> I agree, the statement in question is true but useless
18:27:32 <Deewiant> Pretty much, yep
18:27:36 <ais523> so why are we attacking it? this is #esoteric!
18:27:42 <ehird> So, I wasn't denying it; it's just a ridiculous statement
18:27:51 <ehird> ais523: because he's an idiot :P
18:29:03 <ehird> I need a good keyboard fast :(
18:30:05 <AnMaster> ehird, about that "QUIT IF THE RESULT IS ZERO"... I don't think that the concept of quitting even exists at the hardware level
18:30:13 <AnMaster> so yeah an idiot
18:30:19 <ehird> Quit = ret, obviously
18:30:23 <ehird> *ret, obviously
18:30:39 <ehird> But the simplification doesn't mark him as an idiot
18:30:45 <ehird> The blaisé assertions do
18:30:48 <AnMaster> yes
18:30:57 <AnMaster> indeed
18:32:30 <ais523> ooh, new esolang on the wiki
18:32:57 -!- augur has quit (Read error: 113 (No route to host)).
18:33:04 <ais523> doesn't actually look that esoteric, tbh
18:33:14 <ais523> it's basically a stock imperative language that hasn't been thought through
18:33:49 <ehird> http://esolangs.org/wiki/Image:Zetaplex.png
18:33:52 <ehird> another OS X esolanger!
18:34:03 <ais523> OS X is common amongst computery people
18:34:12 <ehird> not overly so, windows far more
18:34:17 <ehird> and in some circles, linux far more
18:34:21 <ais523> this one, for instance
18:34:23 <ehird> OS X is relatively uncommon outside of mac circles
18:34:31 <ais523> actually, I wonder if Windows or Linux is more common in #esoteric?
18:34:34 <fizzie> Misread "OS X is common amongst the conspiracy people".
18:34:35 <ehird> ais523: me, lament and (doesn't really count) Corun
18:34:40 <ehird> are the OS Xers here
18:34:42 <ais523> I'd expect there to be several mac users here, but nowhere near a majorit
18:34:44 <ais523> *majority
18:35:06 <ehird> oh, incidentally, Corun's mac app got bought out by a bigger mac vendor
18:35:09 <ehird> read that earlier today
18:35:13 <ehird> (well, I think another person worked on it too)
18:35:20 * ais523 CTCP VERSIONs the whole of #esoteric
18:35:31 <ehird> I do that occasionally
18:35:42 <ais523> [18:35] [CTCP] Received CTCP-VERSION reply from Gregor: xchat 2.8.6 .
18:35:43 <ehird> oops
18:35:43 <ais523> [18:35] [CTCP] Received CTCP-VERSION reply from Gregor: Microsoft IRC# 2010 64-bit (Windows 7 Beta, x64, 2GB RAM).
18:35:44 <ehird> I pinged instead
18:35:47 <ais523> there's something odd there...
18:35:52 <ehird> ais523: obviously a joke
18:35:54 <ehird> a bad one, too
18:35:55 <ais523> yes
18:36:06 <ehird> why is windows 7 x64 with 2GB of RAM odd?
18:36:12 <ehird> it's not silly like the rest
18:36:18 <ais523> ehird: I meant, using both xchat and IRC#
18:36:21 <ais523> over the same connection, somehow
18:36:26 <ehird> IRC# doesn't exist
18:36:31 <fizzie> ais523: I think I should be replying with multiple version replies too.
18:36:32 <ais523> again, I suspected that
18:36:32 <ehird> And definitely not 2010
18:36:33 <ais523> but didn't know
18:36:41 <fizzie> ais523: One for the bouncer, one connected irssi, one connected XChat.
18:36:47 <ehird> Oh, I forgot that fax uses OS X too
18:36:57 <ais523> fizzie: ah yes, you said both irssi and xchat
18:37:03 <fizzie> ais523: And "bip"?
18:37:05 <ais523> and bip
18:37:06 <ais523> earlier
18:37:09 <fizzie> Right.
18:37:28 <ehird> bip is an in-irssi bouncer, I gather
18:37:32 <ehird> from fizzie's talkings
18:37:38 <fax> I use ubunut
18:37:47 <ehird> fax CTCP REPLY <CTCP>VERSION X-Chat Aqua 0.17.0-rc1 (xchat 2.8.6) Darwin 9.8.0 [i386/2.66GHz/SMP]<CTCP>
18:37:48 <fizzie> No, it's a separate program. I did use the irssi bouncer too for a while there.
18:37:49 <ehird> demonstratably false
18:37:57 <ehird> *demonstrably
18:38:19 <ais523> yes, how could you get a 386 to run at 2.66GHz?
18:38:34 <ehird> ais523: no, I meant his ubuntu clame
18:38:36 <ehird> and don't be silly
18:38:45 <ehird> i386 is a blanket classification
18:38:57 <ehird> i686 being i686-and-better, etc
18:39:18 <fax> I use Ubuntu
18:39:23 <fax> GNU/Linux
18:39:25 <ehird> fax: your client says you use OS X
18:39:30 <ehird> :P
18:39:33 <fax> I know that
18:39:48 <ehird> also, "GNU/Linux"? did the silly drones infect your brain or something?
18:39:56 <fizzie> Bip's a bit silly; it multiplexes the different networks over a single port, uses the server password (given as "user:pass:network") to determine which one to connect. There's a slight trick needed to use it with Irssi, which normally refuses to connect multiple times to the same host:port pair.
18:40:14 <ais523> I say GNU/Linux when I want to mention both parts, but that's rare
18:40:15 <fax> GNU/Linux is the correct tern
18:40:19 <ais523> the GNU stuff is normally irrelevant
18:40:48 <ehird> i'm pretty sure fax is just trolling, since anyone whose client identifies as OS X and defends the use of GNU/Linux is clearly too insane to use IRC
18:40:52 <ehird> as in the temr
18:40:53 <ehird> *term
18:40:57 <ehird> not GNU/Linux itself
18:41:23 <ais523> Aqua/Darwin!
18:41:32 <ehird> aaaargh
18:41:52 <ais523> I say GNU/Linux if I need to draw a comparison with BusyBox/Linux for whatever reason
18:41:54 <ais523> but that's rare
18:42:17 <ehird> Gun/Linux, for the esr in you.
18:42:24 <ehird> (Pun shamelessly lifted from ELER.)
18:42:40 <ais523> incidentally, are there famous people who support one name over the other?
18:42:44 <ais523> RMS prefers GNU/Linux, obviously
18:42:52 <ais523> but IIRC, even he's given up trying to get people to use it
18:43:03 <ehird> I don't think anyone but rms cares
18:43:04 <fizzie> How is it with Ubuntu, does it identify itself as "GNU/Linux" anywhere?
18:43:04 * ais523 comes to a decision
18:43:04 <AnMaster> heh
18:43:08 <ehird> at least, anyone relevant
18:43:09 <AnMaster> hm what about windows?
18:43:11 <ehird> fizzie: No.
18:43:13 <AnMaster> is that. Areo/NT?
18:43:15 <ais523> a system is GNU/Linux if and only if Emacs is used as the main shell on it
18:43:17 <ehird> It's "Ubuntu Linux".
18:43:17 <ehird> Areo!
18:43:20 <ais523> (and it runs Linux)
18:43:31 <ehird> It's a layer of GUI in between a chocolate biscuit.
18:43:38 <ais523> Emacs is clearly a central part of the GNU operating system
18:43:44 <fizzie> Debian still says "GNU/Linux" in many places, like the default motd and issue files.
18:43:45 <ais523> based on all the discussions about it
18:43:48 <fax> GNU/Emacs
18:43:55 <ais523> therefore, an OS must be based on Emacs to be GNU-based
18:43:55 <AnMaster> hm
18:43:59 <ehird> fax: your trolling powers are weak
18:44:00 <AnMaster> someone should make BSD/Hurd
18:44:03 <AnMaster> just for the hell of it
18:44:03 <ehird> ais523: no no, GNU have two operating systems, GNU/Linux and GNU/Emacs
18:44:05 <AnMaster> ehird, ^
18:44:09 <ehird> /Linux is better
18:44:12 <ehird> it has multitasking, for instance
18:44:15 <ais523> ehird: Emacs/Linux would make more sense than GNU/Emacs
18:44:20 <ehird> and more flexible windowing systems, and program compatibility
18:44:22 <AnMaster> ehird, you forgot Hurd
18:44:25 <ais523> there's "GNU Emacs", which is a version of Emacs
18:44:26 <ehird> ais523: it runs GNU software on top of Emacs
18:44:26 <fax> elinux
18:44:28 <ais523> but that's different
18:44:32 <AnMaster> ais523, you too
18:44:39 <ais523> ehird: the OS still needs a kernel
18:44:43 <ehird> AnMaster: BSD/Hurd is impossible; BSDs' kernels are strongly linked with their software
18:44:45 <ais523> AnMaster: I didn't forget it, I just disregarded it
18:44:50 <ehird> ais523: /Emacs is portable to other kernels
18:44:51 <AnMaster> ehird, hm true :/
18:45:02 <ehird> but it's mostly used on top of /Linux
18:45:14 <ehird> making it basically a useless toy, only usable on top of another, better system
18:45:29 <fizzie> Well, gnu.org front page (third paragraph from top-left, quite a prominent spot) is still all "Sometimes this combination is incorrectly called Linux."
18:46:06 <ais523> how many actual GNU/Linux distributions are there, though/
18:46:18 <ais523> pretty much no modern distro cares about userland command-line tools
18:46:22 <ais523> as the major focus of the distro
18:46:26 <ais523> so it would have to be something with no GUI
18:46:26 <ehird> Slackware!
18:46:32 <ehird> Well, Slackware has KDE.
18:46:46 <ais523> also, http://www.gnu.org/distros/free-distros.html still amuses me
18:46:58 <ehird> heh, they list BLAG first
18:47:09 <ehird> I read BLAG's forums a while back after stumbling upon them
18:47:21 <ehird> the distro is redhat-derived (iirc via fedora)
18:47:26 <ehird> but they're all anarchists, basically
18:47:40 <ehird> makes for an interesting user group...
18:48:09 <fax> why?
18:48:37 <ehird> why is the sky brown
18:49:10 <ais523> oh, another random programming question: does anyone here know how to get curses (in general, or ncurses in particular) to output a nonprintable character raw?
18:49:24 <ehird> um
18:49:28 <ehird> addch or whatever it is
18:49:31 <ehird> just pass the code
18:49:35 <ehird> addch(33)
18:49:35 <ais523> that only does printable
18:49:36 <ais523> *printables
18:49:37 <ehird> or whatever it is
18:49:39 <ais523> I've tested
18:49:43 <ehird> ais523: print it to stdout
18:49:46 <ehird> then flush stdout
18:49:57 <ehird> that might work
18:49:59 <ais523> then curses will overwrite it at random because it doesn't know about it
18:50:23 <Deewiant> Why do you need to output a nonprintable?
18:50:51 <ais523> Deewiant: I'm trying to make a client-server architecture for NetHack
18:51:00 <ais523> with the interface separate from the game logic
18:51:08 <ais523> (don't ask me why I want to do /that/, this is #esoteric)
18:51:14 <ais523> anyway, some people use specialist fonts for NetHack
18:51:22 <ehird> ais523: just output it all as lisp
18:51:26 <ais523> in which all the character codes from 128-255 are printables with special meanings
18:51:27 <ehird> oh wait, that already exists.
18:51:30 <ais523> ehird: that's what I am doing
18:51:33 <ais523> I used that code and tweaked it
18:51:39 <ais523> what I'm talking about is, how to write the client code
18:51:45 <ehird> ah
18:51:49 <ehird> well, don't use curses
18:51:50 <ehird> just use termios
18:51:59 <ehird> and Ctrl-V ←
18:52:03 <ehird> and all those sorts of characters
18:52:09 <ais523> that's a possibility
18:52:12 <ehird> (termios to make input raw)
18:52:14 <ais523> it would be nicer to use curses if I could, though
18:52:21 <ais523> so it's portable to lots of different terminals
18:52:45 <ehird> termios is
18:52:56 <fizzie> Steal the wintty code out of Nethack too?
18:52:58 <ehird> it's just the control chars might not be; ais523: or do you want to support non-VT?
18:53:00 <ehird> if so, you're crazy
18:53:04 <ehird> but i knew that
18:53:16 <ais523> I want to support everything, reall
18:53:18 <ais523> *really
18:53:29 <fizzie> You're supposed to take the control chars from termcap/terminfo; it's really bad form to hard-code *those*.
18:53:31 <ais523> atm I'm limiting myself to things that claim to support the ANSI terminal codes
18:53:41 <ais523> mostly for sanity reasons
18:53:47 <ais523> but everything does nowadays, even the non-VT stuff
18:53:54 <ehird> ais523: then termios will be fine
18:54:01 <ehird> it just lets you set terminal input as raw
18:54:05 <ehird> so that you get every keypress directly
18:54:08 <ehird> and immediately
18:54:13 <ais523> I want to do cursy stuff too, though
18:54:29 <ehird> yes, but the arrow keys are ANSI stuff, no?
18:54:37 <ais523> it's not the arrow keys I want, strangel
18:54:38 <ehird> to reposition
18:54:39 <ais523> *strangely
18:54:43 <ehird> what then
18:54:46 <ais523> besides, arrows on a VT + nethack cause all sorts of amazing problems
18:54:53 <ehird> erm
18:54:54 <ehird> i mean
18:54:57 <ais523> things like delayed-update rendering
18:54:58 <ehird> ctrl-v (arrow key)
18:55:01 <ehird> that gives you a keycode
18:55:07 <ehird> that repositions the cursor
18:55:08 <fizzie> So what's wrong with using the existing TTY "window system" routines to present the interface? Then it'll be the one people are used to, after all.
18:55:09 <ehird> in that direction
18:55:10 <ehird> when printed
18:55:22 <ais523> fizzie: they're too tightly bound to NetHack
18:55:25 <ais523> and also, a mess
18:55:34 <ais523> I decided rewriting would be easier than reimplementing half of NetHack on the client side
18:55:38 <ehird> ais523: redundant
18:55:41 <ehird> you could have just said the first line
18:55:47 <ais523> heh
18:56:46 -!- augur has joined.
18:56:49 <fizzie> It doesn't look *that* tightly bound, to be honest; though admittedly I've just looked at it for two minutes or so. A horrible mess it of course is.
18:57:29 <ehird> we should form an esoteric english assocation
18:57:36 <ehird> to celebrate and cultivate sentences like "A horrible mess it of course is."
18:57:47 <ais523> something is wrong; ncurses appears not to have official online documentation
18:57:53 <ais523> it has manpages, but no info, despite being a GNU thing
18:58:00 <ais523> and this is one case where info would really help
18:58:04 <ehird> ncurses isn't really much of a gnu thing
18:58:06 <ehird> for instance, it's mit-licensed, IIRC
18:58:12 <AnMaster> <ais523> so it would have to be something with no GUI <-- most server distros *should* fall into that category. Probably "should" != "do" in this case :(
18:58:17 <ehird> it's a project that's de facto gnu, but de jure ncurses
18:58:29 <ais523> ah
18:58:33 <ais523> is that backwards?
18:58:42 <ehird> Erm, yes
18:59:54 <ais523> ooh, more activity in Bilski
19:00:57 <ais523> Microsoft are arguing for the same verdict that the EFF are, but on completely different reasoning
19:00:59 <ehird> "In re Bilski"; wonderful name
19:01:15 <ehird> <3 eff
19:01:21 <ehird> I should donate to the EFF.
19:01:40 <ais523> all this effort will probably be pointless
19:01:46 <ais523> because the patent in question isn't actually a software patent
19:01:55 <ais523> so the Supreme Court don't have to rule on them
19:02:02 <ehird> heh
19:02:43 <fizzie> ais523: If you are ncurses-specific, you can use legacy_coding(2) -- "the library ignores isprintf for codes in the range 128-255."
19:02:56 <fizzie> (In this case "2" is a value, not a man page section number.)
19:03:11 -!- Gracenotes has quit (Remote closed the connection).
19:03:39 <ais523> fizzie: that's pretty much what I was looking for
19:03:53 <fizzie> "AUTHOR: Thomas Dickey (to support lynx's font-switching feature)."
19:04:38 <fizzie> Er, and I mean "use use_legacy_coding(2)"; one "use" got left off.
19:05:08 -!- kar8nga has quit (Remote closed the connection).
19:05:10 <Deewiant> "isprintf"?
19:05:14 <ais523> yes, "isprintf"
19:05:17 <ais523> it seems to be a curses thing
19:05:38 <Deewiant> It doesn't seem to be explained much
19:05:43 * ehird toys around with the idea of modding this keyboard with his Model M to produce an unimaginable horror: The Topre of scissor-switches!
19:05:45 <ais523> also, yay "keyname", that'll help this code quite a bit
19:06:00 <ais523> I'm planning to have customizable keybindings
19:07:24 <fizzie> In this particular context, they may have also meant just plain old "isprint"; at least my man page is pretty broken anyway. The NAME section says "use_legacy_coding - use terminal's default colors".
19:07:38 -!- AnMaster_ has joined.
19:07:38 <ehird> Or I could mod it with my old Apple keyboard; then I could produce the plastic-vessel-bashing-scissor-switches horror.
19:07:54 <fizzie> (Probably copied from use_default_colors.)
19:08:55 <Deewiant> Presumably it is just a prevailing typo
19:09:07 <Deewiant> grepping for isprintf in all of ncurses gives only that manpage
19:09:13 <ais523> yes
19:09:39 <ais523> int isprintf(int (*f)(char*,...)) { return f == printf; }
19:09:51 <Deewiant> :-D
19:10:00 <Deewiant> Seems like a useful function
19:10:58 <ais523> it probably has some ues
19:11:00 <ais523> *use
19:11:28 <ehird> ais523: nonono, you know __attribute__((printf))?
19:11:30 <ehird> it should detect that somehow
19:11:43 <ehird> (gcc; makes it parse the format argument to warn about extraneous or omitted arguments)
19:11:51 <ehird> (I think it has an argument to tell which one's the format, but eh)
19:12:11 <ais523> there are gcc attributes for scanf and strftime too, IIR
19:12:12 <ais523> *IIRC
19:13:54 <pikhq> ehird: It also does type-checking on arguments, IIRC.
19:14:03 <ehird> yes
19:14:22 <ehird> ais523: could one detect it with crazy gcc functions, I wonder?
19:14:26 <ehird> in a macro
19:14:36 <ais523> I don't know
19:14:41 <pikhq> I don't think GCC provides sufficient attribute introspection.
19:15:14 <ehird> also: http://www.jwz.org/gruntle/transubstantiation.html
19:15:22 <ehird> i lol'd
19:15:33 -!- AnMaster- has joined.
19:17:51 -!- AnMaster_ has quit (Read error: 145 (Connection timed out)).
19:19:17 <AnMaster-> ...
19:20:23 -!- AnMaster has quit (Nick collision from services.).
19:20:25 -!- AnMaster- has changed nick to AnMaster.
19:21:53 <pikhq> ehird: :)
19:24:10 -!- adam_d_ has changed nick to adam_d.
19:25:28 <ais523> wow, .se was down for half an hour yesterday
19:25:30 <ais523> the entire TLD
19:25:36 <ais523> it's rare for a TLD to crash...
19:25:45 <ais523> (only the DNS, though, so it isn't all that big a deal)
19:26:45 <ais523> the reason is fun, too
19:26:50 <fizzie> "-- due to a bug in the script that generates the SE zone file. The SE tld has close to one million domains that all went down due to missing the trailing dot in the SE zone file --"; news like this are comforting, in the sense that it seems more likely that they're still humans running the Internet. Or maybe that's just what they want us to believe.
19:27:33 <Deewiant> It's just cosmic rays messing up the AIs
19:27:54 <ais523> yes, they pushed a typo directly to production
19:28:03 <ehird> isn't all that big a deal? Sure it is
19:28:16 <ais523> ehird: the DNS was cached in most places
19:28:22 <ais523> top-level DNSes often are
19:28:24 <ehird> ah, at ISPs
19:30:35 <AnMaster> <AnMaster_> <AnMaster> <ais523> ooh, more activity in Bilski <-- what is bilski?
19:30:35 <AnMaster> <AnMaster_> before I lost connection
19:30:35 <AnMaster> <AnMaster_> very bad connection today
19:30:42 <AnMaster> I have no clue if that got through
19:31:00 <ais523> AnMaster: a US supreme court case
19:31:06 <ais523> (and no it didn't get through)
19:31:06 <AnMaster> ais523, about?
19:31:12 <ais523> it's about a business method patent
19:31:17 <AnMaster> ais523, heh?
19:31:28 <ais523> and a huge horde of people are trying to persuade the supreme court to settle the software patent issue at the same time
19:32:39 <AnMaster> ais523, about the dns issue for .se. Didn't notice anything. Though yesterday I was a bit busy IRL so could easily have missed it.
19:32:57 <ais523> it was only for half an hour
19:39:20 <ehird> also, your ISP had it cached.
19:57:37 <AnMaster> ehird, well not sure what isp my uni uses
19:57:48 <ehird> It's almost certain.
19:57:55 <ehird> Like p=99%.
19:58:01 <AnMaster> actually they are hooked up to sunet I think
19:58:11 <AnMaster> which isn't exactly an ISP
19:58:20 <AnMaster> but more like university backbone
20:00:57 <ehird> Tee hee, I love notebooks with accelerometers.
20:02:49 <AnMaster> ehird, oh?
20:03:01 <ehird> Yes!
20:03:06 <AnMaster> ehird, you bought a netbook recently?
20:03:16 <ehird> I said notebook
20:03:29 <AnMaster> oh, misread
20:03:31 <AnMaster> but then:
20:03:40 <AnMaster> you finally bought a notebook?
20:03:48 <AnMaster> and my thinkpad has an accelerometer
20:03:54 <ehird> No, I just love notebooks with accelerometers
20:04:11 <fizzie> Welll... my calculator has an accumulator!
20:04:16 <ais523> desktops with accelerometers would be more fun
20:04:41 <ehird> GPU giant NVIDIA has confirmed that the company is putting the brakes on the Nforce chipset line because of legal wranglings with Intel.
20:04:41 * ehird boggles
20:04:42 <fizzie> Accelerators and accumumometers.
20:04:53 <ehird> so now the only real option for non-shitty notebook graphics is ati
20:05:12 <AnMaster> <ehird> GPU giant NVIDIA has confirmed that the company is putting the brakes on the Nforce chipset line because of legal wranglings with Intel.
20:05:13 <AnMaster> what
20:05:13 <AnMaster> the
20:05:14 <AnMaster> hell
20:05:22 <ehird> yep...
20:05:35 <AnMaster> ehird, they will lose LOTS of money by that
20:05:37 <AnMaster> also
20:05:46 <AnMaster> since that is their main product
20:05:48 <ehird> well, they said legal wranglings; I doubt they have much choice
20:05:50 <ehird> no it's not
20:05:53 <ehird> GeForce is their main product
20:06:06 <ehird> nForce might be more successful, but GeForce is where their millions of bux go
20:06:55 <AnMaster> oh right hm
20:07:01 <AnMaster> misread it as geforce
20:07:11 <AnMaster> and yeah nforce chipset. Has that
20:07:14 <AnMaster> well had.
20:07:15 <ehird> nForce is used in low-end desktops but, most imrp
20:07:18 <AnMaster> on an older computer
20:07:20 <AnMaster> this one is via
20:07:22 <ehird> *importantly, notebooks
20:07:26 <AnMaster> ehird, yep
20:07:38 <fizzie> I guess they're still going to do mobile GPUs, though, just not chipsets?
20:07:38 <ehird> for instance, I have no idea what Apple will do now
20:07:45 <ehird> they used to use Intel chipsets, but, err, they suck
20:07:53 <ehird> fizzie: nForce includes the GPUs
20:07:54 <ehird> So no
20:08:01 <ehird> (The rest of the article clarifies)
20:08:10 <ehird> No more GPUs for now
20:08:24 <ehird> (Still produced, just no new architectures)
20:09:09 <AnMaster> hm
20:10:51 <ehird> Ehh, every now and then the latency on this thing shoots up and I'm reminded I'm on a 3G modem.
20:10:55 <ehird> (Apart from that it's insanely good.)
20:13:04 <ehird> "While the Unihan file still isn’t part of UnicodeChecker due to its size, the Unicode 5.2 data files are included. There are plenty of additions in this revision. They even introduced additional Snowman codepoints! Black Snowman (U+26C7) and Snowman without Snow (U+26C4) OMG ☃☃☃11!!!!!1☃☃☃11! Now I’ll just need a font with glyphs for these new codepoints."
20:13:13 <ehird> Someone do it, quick
20:13:16 <ehird> s/$/!/
20:13:17 <fizzie> There are those gazillion mobile GeForce variants (8xxxM, 9xxxM, 1xxM, 2xxM), though; are you saying those are all nForce-chipset-integrated-only things? That doesn't seem like it'd be the case. (Though those seem to be used more on the high-end not-so-laptop-any-mores.)
20:14:07 <ehird> fizzie: Those are going, that's the point. And Apple uses them on all notebooks.
20:15:02 <ehird> http://earthlingsoft.net/ssp/blog/graphics/UnicodeChecker%20114%20QuickLook.png ;; Spotlight plugin for an app that contains every Unicode character + Finder's Cover Flow = Ridiculously glitzy fun that may or may not be useful in the slightest
20:20:04 <fizzie> I guess I'm not just getting it; I don't see any references in the pcmag.com or zdnet.com articles to anything else than "placed its Nforce chipset line on hiatus"; they still have those mobile Tecra and ION things, and from what I can tell they do discrete-but-still-intended-for-mobile-use GPUs in the GeForce line; just not CPU chipsets.
20:20:46 <ehird> Hmm
20:20:55 <ehird> Oh well
20:20:57 <ehird> We'll see
20:21:17 <ehird> fizzie: But, FYI, a lot of Apple's notebooks use the non-discrete GeForce GPUs
20:21:26 <ehird> (The others use them + an extra discrete GeForce)
20:22:08 <fizzie> Yes, I guess there's a lot of hardware that uses Intel CPUs and would prefer to have an integrated solution.
20:22:13 <fizzie> Those are a bit of out of luck.
20:25:34 -!- cal has joined.
20:26:02 -!- cal has changed nick to Guest76637.
20:26:17 <ais523> hi Guest76637
20:26:17 -!- Guest76637 has changed nick to iamcal.
20:26:34 <iamcal> hi
20:29:17 <ehird> I am cal. Wait, what?
20:31:06 <ais523> !clc-intercal DO WRITE OUT #12345 DO GIVE UP
20:31:13 <ais523> !help
20:31:13 <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>.
20:31:17 <ais523> !help languages
20:31:17 <EgoBot> languages: Esoteric: 1l 2l adjust asm axo bch befunge befunge98 bf bf8 bf16 bf32 boolfuck cintercal clcintercal dimensifuck glass glypho haskell kipple lambda lazyk linguine malbolge pbrain perl qbf rail rhotor sadol sceql trigger udage01 underload unlambda whirl. Competitive: bfjoust fyb. Other: asm c cxx forth sh.
20:31:23 <ais523> !clcintercal DO WRITE OUT #12345 DO GIVE UP
20:31:24 <ehird> clcintercal
20:31:31 <EgoBot> *000 DO WRITE OUT #12345
20:31:35 <ais523> aagh
20:31:38 <ais523> !clcintercal DO READ OUT #12345 DO GIVE UP
20:31:44 <EgoBot> xiiCCCXLV
20:31:51 <ais523> that's better
20:32:22 <AnMaster> ais523, you fell for the write out thing XD
20:32:27 <AnMaster> you of everyone
20:32:31 <ehird> What?
20:32:37 <ais523> AnMaster: INTERCAL is full of traps like that
20:32:40 <ais523> and I haven't written it for a while
20:32:42 <ehird> ah
20:34:22 <ais523> !clcintercal DO CREATE _1 ?VERB ,WRITE, ,OUT, ?RVALUES AS GER + #13 + ROU + !RVALUES #1 + ?RVALUES #1 DO WRITE OUT #12345 PLEASE DO GIVE UP
20:34:28 <EgoBot> *000 DO CREATE _1 ?VERB ,WRITE, ,OUT, ?RVALUES AS GER + #13 + ROU + !RVALUES #1 + ?RVALUES #1 DO WRITE OUT #12345
20:34:49 <ehird> my brain hurts
20:34:50 <ais523> hmm... CLC-INTERCAL doesn't like me mixing INTERCAL and IACC
20:35:01 <ais523> although, the location where that statement ended is interesting
20:35:14 <ais523> !clcintercal DOREADOUTDOREADOUTDOREADOUT
20:35:20 <EgoBot> *000 DOREADOUTDOREADOUTDOREADOUT
20:35:32 <ais523> !clcintercal DOREADOUTDOREADOUTDOREADOUT#5
20:35:38 <EgoBot> *000 DOREADOUTDOREADOUT
20:35:45 <ais523> likewise, that's also interesting
20:35:54 <ais523> (J-INTERCAL had a rather spectacular misparsing of that particular string...)
20:37:11 <ehird> my brain-hertz
20:55:08 -!- AnMaster has quit (Read error: 60 (Operation timed out)).
20:56:50 -!- AnMaster has joined.
21:08:38 -!- nooga has joined.
21:08:55 <nooga> ehird: vim or emacs?
21:09:17 <ehird> Mu; the question is incorrect.
21:09:22 <ehird> False dichotomy.
21:10:39 -!- sebbu2 has joined.
21:10:42 <ais523> nooga: for what?
21:10:46 <ais523> emacs is better for CGI scripting
21:12:03 <nooga> i was just curious what that little RMS would say
21:12:32 <nooga> basically there is lisp in emacs - it counts
21:13:19 <ehird> congratulations, two more incoherent lines
21:13:47 <ehird> that little RMS = me? more like polar opposite...
21:13:51 <ehird> and I have no idea what the second line maens
21:14:42 -!- sebbu has quit (Read error: 60 (Operation timed out)).
21:15:12 <nooga> you're just like rms
21:15:19 <ehird> \\\\\\\\\\\\\\\\\\\\\\\\\\\\\
21:15:20 <ehird> oops
21:15:22 <ehird> nooga: no, I'm not
21:15:29 <ehird> I disagree with rms on just about everything
21:15:33 <nooga> "it's not X, i like only X, i won't come any closer"
21:15:37 <ehird> also, I don't have a beard; and I'm not homeless
21:16:01 <nooga> but i have beard and i'm almost homeless ;D:D:D
21:16:08 <ehird> nooga: you continue to pioneer new heights of awkward english. but that sentiment you invent is more commonly expressed by vim and emacs uesrs.
21:16:09 <ehird> *users
21:16:21 <ehird> I'd say I have no time for this tediousness, but I do.
21:17:24 <nooga> i was trying to annoy you because if you'd say vim i''d say that you're inconsistent in your opinions because the lisp is in emacs, not vim
21:17:44 <nooga> and i still remember how you love genera and other weird lisp contraptions
21:17:55 <ais523> adding Lisp to something does not automatically make it good
21:18:06 <ehird> You're full of shit. First of all, liking something with the property P does not mean my quality metric solely consists of P.
21:18:25 <ehird> Secondly, Genera is indeed good because of Lisp, but again this in no way implies the generic goodness of Lisp, let alone it being the only metric of quality.
21:20:29 -!- MigoMipo has quit ("Page closed").
21:20:38 <nooga> my bad ;\
21:28:30 <ehird> About "half" of a "letter size" paper —Happy Hacking Keyboard site
21:28:31 <ehird> "half"
21:29:34 <nooga> ppl tend to put " everywhere
21:33:15 <pikhq> People like scare quotes.
21:33:22 <pikhq> They're all 'scary'.
21:33:28 <Deewiant> http://www.unnecessaryquotes.com/
21:33:30 <pikhq> And "useless".
21:33:43 -!- Asztal has quit (Read error: 110 (Connection timed out)).
21:36:54 <SimonRC> ais523: what dareIask is IACC?
21:37:04 <nooga> "maybe "we" should think about some "esolang" based on "unnecessary" quotes"
21:37:09 <ais523> SimonRC: the language in which CLC-INTERCAL is implemented
21:37:24 <ais523> and IACC is implemented in IACC, but runs on a VM written in Perl
21:37:29 <ais523> which is where most of the coding actually is
21:37:30 <ehird> To be fair the site is all Japanese
21:37:38 <ehird> Not language, that is
21:37:52 <SimonRC> I was hoping it was YACC that outputted INTERCAL
21:37:58 <ehird> Anyway, looks like it'll be useless to me, as they only sell the Lite 2, whereas I want (if I do want a HHKB at all) the Professional 2
21:38:15 <ais523> SimonRC: the name's clearly a variation on YACC
21:42:54 <SimonRC> ehird: I reply to your transubstantiation link: http://www.smbc-comics.com/index.php?db=comics&id=1664#comic
21:43:17 <ehird> Deewiant: you could say that unneccessaryquotes.com is full of unneccessary quotes
21:43:19 <ehird> hur hur
21:43:25 <ehird> a*unnecessary
21:43:30 <ehird> **unnecessary
21:43:40 <ehird> SimonRC: :D
21:44:25 <SimonRC> don't forget the alt-image
21:44:45 <SimonRC> (mouse-over the red circle below)
21:45:06 <SimonRC> took me 1500 comics to find that button's purpose...
21:45:07 <SimonRC> d'oh
21:45:13 <ehird> SimonRC: don't be condescending to me with your whole assuming I don't read smbc stuff :|
21:45:14 <ehird> :P
21:45:39 <ehird> well technically you had to vote for smbc in some random webcomics ranking before you could access it in the old days iirrc
21:45:40 <ehird> *iirc
21:46:14 <SimonRC> uhhhh, that sounds a bit circular
21:46:25 <ehird> eh?
21:46:44 <ehird> you clicked the votey (thus the name), it'd ask "DO YOU CONFIRM THAT YOU WANT TO VOTE FOR: SATURDAY BLARGHFEST", confirm, bam, image
21:47:36 <ehird> why am i so tired
21:47:42 <SimonRC> ah, that "it"
21:49:10 <ehird> ah heh
21:49:29 <ehird> i briefly considered that interpretation for a few seconds before dismissing it as too silly
21:49:56 <ehird> i like the idea of a link to a thing you have to do to see the link, though
21:50:13 <SimonRC> nah, just the comic image, I thought
21:51:56 <ehird> heh
21:53:42 -!- AnMaster_ has joined.
21:54:43 -!- AnMaster has quit (Read error: 60 (Operation timed out)).
21:57:36 <Gregor> I didn't know how to make the default version string go away :P
21:57:39 <Gregor> So it sends both.
22:00:58 <ehird> http://www.addictinggames.com/fingerfrenzy.html ;; argh, addicting (rather trivial game: type a→z as fast as possible)
22:01:05 <ehird> 2.767 with crappy keyboard and qwerty is my record...
22:01:20 <ehird> oh, no
22:01:24 <ehird> got some 2.6s
22:01:44 <ehird> 2.4!
22:02:00 <ehird> 2.532!
22:02:58 <fizzie> Gregor: /set irc_hide_version on
22:03:07 -!- nooga has quit (Read error: 104 (Connection reset by peer)).
22:03:27 <fizzie> (Note: I'm not completely sure "on" is what you type in the /set command to set a boolean value; I just looked at the source.)
22:04:29 <ehird> eh, and i've lost the skill
22:07:29 <ehird> 2.394!
22:08:32 -!- Asztal has joined.
22:08:39 <ehird> 2.107!
22:14:56 <ehird> 2.011, fuck yes
22:19:55 -!- AnMaster_ has changed nick to AnMaster.
22:23:12 <Deewiant> Best non-cheating score of this month seems to be 1.635
22:24:07 <ehird> There are YouTube videos of people doing it in <1sec
22:24:25 <Deewiant> Oh, okay
22:25:08 <Deewiant> I guess I'm happy with 2.283
22:25:19 <ehird> Hah, I am superior to you with QWERTY on a scissor-switch
22:25:30 <Deewiant> This was also QWERTY
22:25:46 <Deewiant> I used to type a-ö just for fun occasionally
22:26:38 <ais523> abcdefghijklmnopqrstuvwxyz
22:26:39 <ais523> takes me ages
22:26:45 <ais523> because I'm used to typing words, not letters
22:26:48 <Deewiant> On Colemak I keep typing qp instead of pq
22:27:53 <ehird> My bord
22:28:03 <ehird> *board's all sweated up, yuck
22:28:13 <ehird> (OK, I'm officially a keyboard nut; saying "board"...)
22:28:18 <ehird> ais523: consider it as one big word
22:28:27 <ehird> like supercalifragilisticexpialidocious
22:28:40 <ehird> omg, which gets past spellcheck in os x
22:28:41 <ehird> <3
22:28:45 <ais523> yes, it's not a word I've finger-learnt yet
22:30:41 <ehird> you have to learn every individual word to touch-type it?
22:30:42 <Deewiant> And I think QWERTY has a bit of an advantage here with easily-rollable sequences: de, fghjkl, mn, op. Colemak has only mn and rst.
22:30:43 <ehird> that's worrying
22:32:33 <Deewiant> http://www.youtube.com/watch?v=4jHze_NYbGs 0.265 seconds?
22:32:43 <AnMaster> night ⇄
22:33:10 <ehird> Deewiant: wow that typing style is awful
22:33:33 <ehird> also that video was so faked
22:33:36 <ehird> at least from the looks of it
22:38:35 <AnMaster> hm make a keyboard layout that allows you to roll completely
22:38:46 <AnMaster> non-faked record solved.
22:40:01 <AnMaster> night really now... ──────►
22:44:41 <ehird> cheating.
22:44:49 <ehird> only qwerty,dvorak,colemak allowed.
22:47:51 -!- ehird has quit.
23:10:38 -!- BeholdMyGlory has quit (Remote closed the connection).
23:14:34 -!- adam_d has quit (Read error: 110 (Connection timed out)).
23:18:13 -!- AnMaster_ has joined.
23:22:45 -!- Sgeo has joined.
23:26:21 * SimonRC goes to bed
23:30:07 -!- AnMaster has quit (Connection timed out).
23:34:57 -!- ais523 has quit (Remote closed the connection).
23:37:25 -!- FireFly has quit ("Later").
←2009-10-12 2009-10-13 2009-10-14→ ↑2009 ↑all