00:11:08 <kmc> i wonder what are the consequences of that, besides DoS
00:11:19 <kmc> perhaps it makes traffic analysis or session downgrade attacks easier
00:12:10 <shachaf> Well, as you pointed out, you still need to know the sequence number to send the RST packet.
00:12:43 <kmc> i wonder how long it is before Chromium installs a shim that runs as root and reimplements TCP to their exact preferences using raw sockets
00:14:58 -!- atriq has quit (Quit: Leaving).
00:15:15 -!- Phantom__Hoover has joined.
00:17:14 <kmc> i wonder if people worry about "end-to-end" net neutrality
00:17:31 <kmc> that is, instead of your ISP making certain sites slower, your browser or OS could do so
00:17:51 <kmc> it's already the case that Chrome performs better and is more secure when accessing Google sites, not for particularly nefarious reasons
00:18:38 -!- Phantom_Hoover has quit (Ping timeout: 256 seconds).
00:27:43 <Vorpal> <kmc> it's already the case that Chrome performs better and is more secure when accessing Google sites, not for particularly nefarious reasons <-- or anything else that offers SPDY
00:27:50 <Vorpal> which is starting to appear
00:28:14 <Vorpal> I guess the list of hard coded certificates won't be available to most other sites though
00:28:44 <Vorpal> I think some big non-google sites have that protection as well, like paypal
00:29:30 <kmc> also some vanity domains of the people who worked on the feature at google :3
00:29:45 <kmc> i think there is a process for third parties to get included
00:29:52 <kmc> in the HSTS preload if not the hardcoded certs
00:30:21 <kmc> i mean Chromium is open source right ;)
00:30:45 <Vorpal> the vast majority will be using Chrome not Chromium, just saying
00:31:34 <kmc> doesn't chrome pull from the open source project though
00:32:03 <Vorpal> but they can make their own modifications should they wish to
00:32:27 <kmc> don't think so
00:32:44 <Vorpal> shachaf, what is libpdf.so?
00:32:48 <Vorpal> something from poppler?
00:32:55 <shachaf> A plugin from Chrome to view PDFs in-browser.
00:33:08 <shachaf> I used to use it but it crashes Chromium now.
00:33:32 <kmc> "Pop a Poppler in your mouth when you come to Fishy Joe's / What they're made of is a mystery, where they come from no-one knows"
00:33:47 <Vorpal> that explains the difference in behaviour between Windows and Linux chrom(e|ium) PDF behaviour
00:34:03 <kmc> "You can pick 'em you can lick 'em you can chew 'em you can stick 'em / If you promise not to sue us you can shove one up your nose."
00:34:16 <shachaf> Vorpal: Chrome has the PDF viewer on every platform now, I think.
00:34:36 <Vorpal> shachaf, I thought it was just some compile time option that was off, or the version that was different
00:34:43 <Vorpal> but I guess that was the cause then
00:35:03 <shachaf> Having a PDF viewer in the browser is pretty great. :-(
00:35:50 <Vorpal> shachaf, eh, making evince or whatever load takes like 2 seconds
00:36:18 <shachaf> Vorpal: What if you're "one of those people" who has hundreds of tabs open?
00:36:20 <Vorpal> same for the PDF app I use on windows, which I forgot the name of. It isn't adobe reader
00:36:22 <shachaf> And half of them are PDFs.
00:36:42 <Vorpal> <shachaf> Vorpal: What if you're "one of those people" who has hundreds of tabs open? <-- I used to be one of those, nowdays I'm down to maybe 50
00:36:48 <Vorpal> <shachaf> And half of them are PDFs. <-- lol what
00:36:53 <zzo38> Why do you need so many tabs?
00:37:31 <Vorpal> I'll let shachaf explain that
00:37:36 <Vorpal> I need to sleep, good night
00:37:43 <shachaf> I'll let Vorpal explain that.
00:38:08 <zzo38> No! He needs to sleeps, good night.
00:40:22 -!- copumpkin has quit (Quit: Computer has gone to sleep.).
00:40:45 <shachaf> Needs For Sleeps III: Sweets Dreams
00:42:20 -!- Vorpal has quit (Ping timeout: 246 seconds).
00:42:31 -!- JavaBotN has joined.
00:42:36 -!- JavaBotN has quit (Remote host closed the connection).
00:44:29 -!- Arc_Koen has quit (Quit: It's a ship. It goes through the gate.).
00:58:37 -!- nooga has quit (Ping timeout: 246 seconds).
01:02:26 -!- copumpkin has joined.
01:07:08 -!- Nisstyre-laptop has joined.
01:08:23 <zzo38> I also intend to make up a new computer, as much hardware open source as possible and all built-in software open source, as much simplify as possible, includes a DVD with many games, a complete manual (including pinouts, schematics, detail of built-in software, licenses, jumpers, BIOS calls, instruction set, etc)
01:08:45 <zzo38> Often I find computers today their manual is incomplete.
01:11:31 <zzo38> And they contain mistakes too
01:22:56 -!- nooodl has quit (Ping timeout: 246 seconds).
01:53:35 -!- Frooxius has quit (Ping timeout: 260 seconds).
01:59:36 <Bike> you could just ping, you know.
02:00:40 <zzo38> It depend what you are trying to check; sometimes I use PING command to the server to check connection to server. To check client connection, is different.
02:01:19 <Sgeo_> I wasn't sure about @ping
02:03:14 <zzo38> Do you know if there is any MML compiler for Csound?
02:05:34 <Sgeo_> http://blog.ezyang.com/2012/11/plan-9-mounts-and-dependency-injection/
02:05:34 <Sgeo_> This makes me want to use Plan 9
02:08:33 <Sgeo_> What time did I say that I want to use Plan 9?
02:08:42 <Sgeo_> And I do have some use cases for the sort of thing that that article talks about
02:12:34 <Sgeo_> XChat says my lag is 12.4 seconds, but I saw that almost instantly
02:15:02 <Sgeo_> Taking an IRC bot and making it talk on a different protocol without writing a fake IRC server.
02:15:49 <Bike> why on earth would writing a fake server be the solution there
02:16:01 <elliott> have you actually written an irc bot
02:16:20 <Sgeo_> But I'm thinking more of bots whose code I am too lazy to change
02:16:59 <elliott> do those bots run on plan 9
02:17:36 <Sgeo_> Is there a Tcl implementation that runs on Plan 9? Is there a Python implementation that runs on Plan 9?
02:17:55 <Sgeo_> Actually, hmm, I think the Tcl bot uses some library that uses native code, for XML stuff :(
02:18:39 <Bike> more to the point, does the socket stuff in the irc library work on plan 9 and bla bla bla bla
02:18:41 <elliott> i really dont see how youd be able to use an existing program and do those things
02:18:47 <elliott> youd need an actual file to override
02:18:52 <elliott> so itd have to do networking plan 9-style
02:18:56 <elliott> which a compatibility layer... wouldnt
02:19:40 <Sgeo_> Wouldn't the networking API of a Plan 9 implementation of those languages do Plan 9 style networking when the application does what it thinks are normal networking calls?
02:19:43 <Bike> Maybe it would be easier to write an x86/OS emulator that would automatically translate into appopriate protocol thingamajiggers.
02:19:56 <elliott> well if it does networking plan 9 style
02:20:04 <elliott> then i'm pretty sure you'd have to actually supply the whole dial() device thing
02:20:13 <elliott> i forget how it actually works though
02:20:20 <elliott> but i dont really see how this is viable with any kind of compatibility layer
02:20:31 <elliott> that said i approve of use of plan 9
02:21:31 <Sgeo_> Is Inferno a sort of successor?
02:22:01 <Sgeo_> Less than a minute later. I'm impressed.
02:22:40 <kmc> plan 9 is still based on the idea that C is the end-all of programming languages and the byte is the most perfect data structure, right?
02:22:43 <kmc> i guess inferno is not
02:24:23 <Sgeo_> Am I going to see Plan 9-style beauty by looking at Inferno, is my question I guess
02:26:14 <elliott> inferno is like plan 9 but in a vm but weird and there's limbo and stuff
02:26:20 <elliott> also its deader than plan 9 afaik
02:26:28 <elliott> since some people actually use plan 9 and i dont think anyone really uses inferno
02:26:32 <elliott> so i dont know why youd look at it
02:26:57 <elliott> kmc: it takes the directory tree as the most perfect data structure moreso, also its C language is better than standard C
02:27:02 <elliott> kmc: but yes that is its main problem
02:27:07 <elliott> kmc: oh it also has excellent Unicode support obviously
02:30:18 <Sgeo_> @ping at 12:29PM my time.
02:30:28 <Sgeo_> That was shockingly fast
02:32:14 <Sgeo_> If it takes a while for there to be a connection
02:32:22 <Sgeo_> And it sends all my line squickly
02:32:30 <Sgeo_> I thought I would look like a really fast typer
02:35:09 <kmc> how is its C better than standard C
02:35:55 <elliott> i forget all the changes but all the ones it does have are good changes
02:36:14 <elliott> like its IO library isn't nearly as dumb
02:36:26 <elliott> its linking system is much better (no -l at the command-line)
02:36:32 <elliott> it has the nice struct inclusion thing go has
02:36:38 <elliott> (and GNU C, as an extension nobody uses)
02:37:01 <elliott> also it does cross-compilation much much better
02:37:05 <elliott> plan 9 has excellent cross-architecture support
02:37:12 <elliott> since it was built to handle distributed systems that used multiple architectures
02:37:27 <elliott> and of course it has actual unicode support
02:37:35 <elliott> (BMP only because it predates >16-bit Unicode, unfortunately)
02:39:13 -!- Sgeo_ has quit (Ping timeout: 246 seconds).
02:41:55 <kmc> which struct inclusion?
02:44:14 <elliott> struct foo { struct bar; ... }
02:44:22 <elliott> if bar has element x you can do f.x where struct foo f;
02:44:33 <elliott> it's like "inheritance" but it's actually composition
02:44:47 <Bike> what if you want a struct that actually has a struct as an element?
02:44:48 <kmc> and this just happens automatically in gnu c?
02:44:49 <elliott> (the -l removal thing is done by having pragmas in the header files that specify what library to link with fwiw)
02:44:54 <elliott> Bike: then you do struct bar y;
02:45:08 <elliott> kmc: no it just supports the same syntax (iirc, maybe it is actually different) if you enable the extension with a flag
02:45:34 <kmc> that's pretty gross yet enjoyable, like <insert sexual metaphor here>
02:45:50 <kmc> most gnu c extensions don't require a flag beyond -std=gnu99 or whatever, which is default
02:46:47 <elliott> http://doc.cat-v.org/plan_9/programming/c_programming_in_plan_9 is a decent introduction to Plan 9 C
02:47:07 <elliott> though a practical one rather than one that explains why Plan 9's programming environment feels so well-designed
02:47:31 <elliott> oh right plan 9 also has a networking library that doesn't make me want to die
02:47:34 <elliott> which is a nice improvement over BSD sockets
02:47:36 <elliott> which does make me want to die
02:47:55 -!- Phantom__Hoover has quit (Read error: Connection reset by peer).
02:52:03 <kmc> an important distinction
02:55:29 -!- madbr has joined.
02:57:38 <zzo38> If you want to die, then you should first check if you are sure, and if you are really sure then die.
02:57:51 <kmc> words to live by
02:57:59 <kmc> but can they be expressed in flowchart form
02:58:12 <shachaf> zzo38: What if you checked and found out that you're sure, but then didn't check that you're sure you're sure?
02:59:14 <zzo38> shachaf: Then too bad you have to learn to try again, that is why you should wait a while after you sleep then you can think of that and try again
03:01:12 <madbr> argh, I'm obsessed by instruction set design
03:02:17 <zzo38> What instruction sets do you design?
03:04:52 <madbr> usually they're in the family of early RISCs
03:05:06 <madbr> avoiding out of order execution basically
03:05:41 <madbr> but trying to avoid ending up with something impossible to pipeline like the 6502
03:06:09 <zzo38> Is the ARM2 instruction set OK?
03:06:32 <madbr> often the results look kinda like the ARM instruction set yes
03:08:33 <zzo38> Have you send anything to Famicompo Mini vol.9? I have send some, but I won't tell you which one.
03:09:19 <zzo38> One instruction set I want to have is some instruction set which the Checkout esolang compile into easily and have a hardware implementation (in some hardware description language)
03:09:25 <madbr> nah, this time around I didn't care enough and didn't make anything
03:10:19 <zzo38> Did you download it?
03:10:31 <madbr> didn't really listen to it
03:11:36 <Bike> What's Famicompo?
03:11:51 <madbr> bike: NSF song competition
03:11:52 <zzo38> Bike: Music contest involving .NSF format
03:12:10 <madbr> NSF is music ripped from NES games
03:12:22 <zzo38> You can use any number of expansions (including multiple expansions) or no expansions.
03:12:27 <madbr> playing it actually involves running a small NES emulator
03:13:11 <zzo38> Not always is ripped you can make up your own music using various programs such as ppMCK and FamiTracker.
03:13:44 <madbr> yeah for Famicompo you make your own song
03:13:45 -!- NihilistDandy has joined.
03:14:27 <madbr> you can actually play the songs from famicompo if you have a powerpak cart... or at least, the ones that don't use expansions
03:14:59 <zzo38> I always use ppMCK (although I have made various improvements to ppMCK, so now it has track questioning commands, subroutines, tail recursion, text replacement macros, custom tuning, and various other things)
03:15:01 <madbr> and fit in the powerpak (max 256k data I think)
03:15:05 <Bike> I was confused by all the talk of instruction sets.
03:15:43 <madbr> instruction set design is hard :o
03:16:01 <zzo38> I have tried in before to design instruction set.
03:16:10 <Bike> well yes, but is that related to chiptuning, or what
03:16:12 <zzo38> I have designed some instruction sets for virtual machines, though.
03:16:41 <madbr> not instruction sets :D
03:18:15 <zzo38> Have you used some effect with multi channels with the same notes but with different delay, duty, octave, etc?
03:25:13 <zzo38> Such as in "ctfinal" the M and N channels play the same notes but with a different delay, duty, volume, and octave; same with a and channels, and the C and O channels play the same notes with no delay but they have a different waveform, and uses detune, etc
03:26:52 <zzo38> So I have set up effect at the start like M @@0@v1 and N @@1@v2r16K12 and then when the notes are played like MN l4o4 ce-gc' ^1 b-2gb- ^c'^2 <a->ce-a- ^1 g2f^8a-8 gfe-d it will use effects that are set.
03:26:56 -!- Sgeo has joined.
03:28:11 <zzo38> Although you said you used Impulse Tracker? So you would have to program the effects differently
03:29:50 <zzo38> You would have to write them manually I guess, or use copy/paste
03:48:05 <kmc> every time flash plugin crashes, take a shot
03:49:52 <madbr> I often use fake delay and detuning and layering yes
03:50:25 <madbr> I use copy/paste a lot
03:59:06 <elliott> http://www.salon.com/2012/11/09/white_house_obama_will_veto_any_bill_extending_tax_cuts_for_those_making_more_than_250k/singleton/ journalism
03:59:18 <elliott> i like how it is by <empty string>
04:00:40 <zzo38> How can you program NSF effects in Impulse Tracker anyways?
04:00:51 <Bike> elliott: guess the wire doesn't have much capacity
04:01:50 <madbr> zzo: depends on the converter
04:01:59 <madbr> there are like 3 different converters
04:03:00 <madbr> s3m2nsf is the simples
04:03:25 <madbr> only supports VRC6, no special macros
04:03:38 <madbr> it simply plays the s3m with the NES waveforms as the samples
04:04:11 <madbr> for some freak reason, s3m note frequencies and NES ones are the same
04:06:02 <zzo38> But wouldn't you still have to program in the duty of square waves?
04:06:45 <madbr> it simply uses instrument 1,2,3,4 as duties 12%,25%,50%,75%
04:07:18 <zzo38> OK, although VRC6 supports different duty than 2A03 (and VRC6 also lacks 75% duty)
04:07:19 <madbr> and 5 as triangle wave, 6 noise, 7 short noise, 8+ dpcm
04:07:43 <madbr> zzo: oh yeah forgot about those
04:08:10 <madbr> I can't remember how it maps the VRC6 duties but it's just instrument numbers too
04:09:17 <madbr> the second converter is it2nsf and it supports a lot more stuff
04:09:25 <madbr> essentially all expansions
04:09:45 <madbr> (you have to enter them in the song comments and it parses that)
04:10:17 <madbr> some IT stuff like note off commands
04:11:04 <madbr> third converter is supernsf which is the one that does software mixing for multichannel PCM
04:15:00 <zzo38> Can it use the MMC5 8-bit PCM?
04:16:10 <zzo38> The MMC5 8-bit PCM is the only feature which ppMCK does not support.
04:16:23 <madbr> supernsf uses the same hw as the DPCM
04:16:28 <madbr> it just writes the value
04:16:37 <zzo38> O, so it uses 7-bit PCM, then.
04:22:52 <zzo38> Have you used ppMCK or any other programs to write a .NSF music other than the three you mentioned?
04:26:13 <zzo38> What I would like to have is program to compile MML into MOD/S3M/IT modules; do you know of any such software?
04:26:48 <elliott> kmc: btw plan 9's rc shell is also much better than any unix shell I know of
04:27:00 <elliott> (in terms of the language, not the interactive features, of which it has none; that's all in the terminal in plan 9)
04:27:14 <madbr> no I just use IT + converter
04:27:34 <madbr> I've never tried MML actually
04:29:36 <zzo38> It might be difficult to do because you may have various features in MML which the module formats does not support, including desynchronized loops and having multiple effects running simultaneously and subroutines and various other things
04:29:51 <zzo38> I have used Impulse Tracker and ModPlug Tracker, but I find MML is much better.
04:30:05 <zzo38> Some people don't; and that is OK you can use what you prefer.
04:31:31 <zzo38> Which expansion audio have you used in .NSF?
04:31:32 <madbr> yeah but most of these things are very rare in music
04:32:19 <madbr> you'd think the limit of having only one effect blocks many things but in real life usage it's really not that bad at all
04:32:44 <madbr> most effects are either used on note on or after note on and are more or less mutually exclusive
04:33:07 <madbr> desynchronized loops in not useful in music
04:33:22 <madbr> subroutines I just do with copy paste
04:34:33 <zzo38> Well, I do sometimes have some channels playing a much shorter loop than others
04:35:00 <zzo38> If it is an exact multiple, it could easily be compiled into a format that does not support that, by automatically making as many copies as needed
04:40:50 <madbr> so of course it's all going to be powers of 2
04:41:31 <zzo38> Although the time signature could be 3/4 or something else like that
04:44:05 <madbr> mostly it's much easier to listen to your song as you make it in the tracker
04:45:31 <madbr> that a much larger advantage that anything you'd win from subroutine stuff
04:45:43 <zzo38> Maybe to you it is.
04:46:04 <madbr> it's the different between music and programming :D
04:46:49 <madbr> it's better to copy paste a section of music
04:46:49 <zzo38> But even deaf people can write music.
04:47:32 <madbr> than to have multiple links to one piece of data
04:47:47 <madbr> because if you copy paste it's much easier to modify it
04:47:54 <madbr> without changing the other copies
04:49:25 <madbr> and that's something you want to do a LOT
04:49:42 <madbr> variation is the key
04:49:52 <zzo38> To you it is. To me, well, sometimes I want to make a copy with variation, but other times I don't want.
04:50:09 <madbr> if you don't want variation you just copy paste
04:50:16 <zzo38> But especially if it is a multi-channel effect I don't want to have to change everything in all channels
04:50:42 <madbr> mhm, in that case you're right
04:51:34 <zzo38> Of course in MML you can still copy/paste too (and use any other functions your text editor has, such as regular expressions), and I use that too.
04:52:09 <zzo38> I just find it easier to write music using MML, although this is not the case for everyone.
04:52:59 <madbr> well, that's because you're a programmer
04:54:49 <zzo38> Even things specific to music such as to indicate if I want D sharp or E flat (which are the same note, but written differently), or even double sharp possibly, or to change ostinato, including short drum loops, to just write it once and have it loop
04:55:48 <zzo38> And comments; sometimes I want to keep track of what chords I want and so on
04:58:13 <madbr> why would you want to use a double sharp
04:58:41 <zzo38> It is rare, but there are some cases in which such thing would be used.
04:59:36 <zzo38> Another thing which is rare but sometimes used is to change the time signature in different parts of the music, and what is even more rare is for the right hand and left hand to have different time signatures.
05:00:07 <zzo38> madbr: One case is if your key signature has sharps, and you want to raise one of those notes in one case, you would use a double sharp.
05:03:24 <madbr> that's really only important in staff notation
05:03:39 <madbr> and just really when it's the most readable notation, which is very rare
05:06:01 <zzo38> Well, at least I find it easier to write music when keeping track of things such as this.
05:06:25 <madbr> that's because you're a programmer
05:06:59 <madbr> when you're punching in notes for the sound it doesn't matter if your Gm chord is G A# D
05:07:27 <zzo38> No, it is because I know how to write music.
05:07:45 <zzo38> I don't just make up notes at random to see if it sound right.
05:08:22 <zzo38> I would think of the notes I want and then put them in.
05:08:37 <zzo38> Like anyone who write music would do.
05:08:42 <madbr> what I do is that I think of notes
05:08:54 <madbr> if I like it, keep
05:09:10 <madbr> if I don't like it, try to see if I can make it better
05:09:34 <madbr> also it's easy to get "happy accidents"
05:09:43 <madbr> which you'll want to keep and elaborate on ofc
05:10:59 <madbr> also it helps when you try 2 or 3 voicings for a chord
05:11:03 <madbr> and chose the best one
05:13:00 <zzo38> I would just think of if it is OK or not; there is no need to try it until after it is written.
05:13:36 -!- NihilistDandy has quit (Quit: ["Textual IRC Client: www.textualapp.com"]).
05:14:07 <madbr> do you play your stuff on a keyboard as you compose it?
05:14:11 <zzo38> And some composers are deaf they cannot listen to it at all. However, they are much better composers than I am anyways.
05:14:15 <madbr> or sing to yourself?
05:14:22 <zzo38> madbr: Sometimes, but usually I just think of it.
05:14:34 <madbr> there's beethoven but that was a special case
05:15:13 <zzo38> I do have a piano but usually (not always) I find it sufficient to think of it, and using the classical musical theory.
05:16:31 <madbr> theory won't tell you if a melody works or not
05:16:33 <zzo38> It can often be easier when written down, though, especially the chords.
05:17:05 <zzo38> madbr: Yes, I know, but if I can think of the melody then I could know the melody.
05:17:09 <madbr> or if a given progression of chord voicings don't work even tho they're correct in theory
05:18:59 <zzo38> Yes, with chord voicings I do sometimes find it necessary to try it, although not always; the chords I may play on piano, although I may write them down on a normal musical notation; melody I can usually just think of and write using MML.
05:19:52 <zzo38> But like I said not always, so even when writing the chords directly using MML I will write comments so that I know what chord I want instead of having to read the notes (even when writing on paper, I write the chord symbols so I know what chords I want; this is same idea)
05:21:11 <zzo38> But usually I will know that I want this voice to go up and down, and that voice to go down and up, and then work the chords around that.
05:22:55 <zzo38> So the voices may be written like separate melody or accompaniment
05:24:32 <zzo38> I only know the classical music theory, and it works for me, although I usually don't need any theory to know if the melody is good.
05:25:08 <madbr> do you know jazz theory?
05:25:21 <zzo38> No, I don't know jazz
05:28:34 <zzo38> My sister studied both classical and jazz
05:36:35 <madbr> also if you don't listen to your stuff as you compose it, how do you know when to follow the rules and when not to? :D
05:37:45 <zzo38> I don't know how I know, somehow I usually know, though; some other composers do too and I don't know how they know either.
05:38:38 <madbr> and how do you mix your nsf music?
05:39:23 <madbr> you do multiple conversions adjusting stuff each time?
05:41:09 <zzo38> If I do something I realize I didn't like, then yes I will adjust something each time, but usually I don't need unless I made a mistake. Usually the first thing I would do is to write only a few bars to figure out the effects I want, and then I know what effects and the rest of the music I can write all at once since I can think of that.
05:45:54 <zzo38> Famicompo Mini vol.7 has a few files in original section using multiple expansions, but in vol.9 there is only one file in original section which uses multiple expansions.
05:46:22 <zzo38> Which expansions have you used?
05:47:49 <zzo38> I did write a program to test VRC7 patches, although I wrote it a while ago before I know really how it works, so it is very unlikely to work on a real hardware and runs only on an emulator, but it is good enough for my use.
05:48:21 <zzo38> Although sometimes there is no need and I just know if I want a pure sine wave it is easy and it is not necessary to test it at first.
05:48:26 <madbr> I would probably have done the patches on adlib tracker 2 (OPL3)
05:48:34 <madbr> and translated them
05:48:57 <madbr> but I used only default patches
05:50:24 <zzo38> I do not think it has all features of OPL3, so I don't expect that to work.
05:50:37 <madbr> well, using only the subset that works ofc
05:52:37 <zzo38> I made a .NES file to test the VRC7 patches instead (it test both the built-in and custom patches), although what you suggest might work too.
05:53:36 <zzo38> Have you used VRC6 and VRC7 together?
05:55:03 <madbr> the mix of regular nes channels and vrc7 is already crazy enough :D
05:55:49 <madbr> I should write more opl3 music
05:56:03 <madbr> there's not enough opl3 specific music (using custom patches etc)
05:56:32 <zzo38> Is there Csound program to emulate OPL3?
05:57:16 <madbr> there's an OPL3 emulation in the MAME source code
05:58:02 <madbr> I think some of the features are wrong but very few people used those so in actual use (ie games on dosbox) it works
05:58:41 <zzo38> But Csound has its own programming language for writing music.
05:59:17 <madbr> yes that's why I don't use Csound
06:01:08 <zzo38> The orchestra format is OK, but the score format seem wouldn't be very good to write music and MML would be better; but, Csound only support the numeric score. (A program could be written to convert MML to numeric score, and I might write such a program.)
06:02:05 <madbr> tbh you should learn to use something like ableton
06:02:11 <zzo38> But I do like the Csound orchestra format.
06:02:20 <madbr> or reaper or cubase or fruity loops
06:02:41 <madbr> programs that are designed for, you know, composing music
06:02:45 <zzo38> I have used some of those and I find they don't help.
06:03:30 <zzo38> Especially if you want to write your own instrument program; for this, I find Csound is very good and I have experimented with some instrument sounds using Csound.
06:05:29 <madbr> I'd rather use something like synthedit for designing instruments
06:06:07 <madbr> or use C++ and make a nice tight VST that other ppl can use too
06:09:28 <madbr> for me csound is in the same category as max/msp and puredata and supercollider and whatever else
06:09:49 <madbr> toys until someone can demonstrate a serious musical use
06:10:27 <elliott> er max/msp is used to make actual music that people buy
06:11:30 <madbr> link to a song that is actual music that people buy and that uses max/msp
06:11:46 <monqy> uh oh, elliott ! hope you know your music
06:11:54 <elliott> is this a trick question where you define "actual music" as "not whatever i linked to"
06:12:19 <elliott> right off the top of my head i know autechre use max extensively for example
06:12:38 <madbr> aha yeah ok that definitely counts
06:12:52 <monqy> max costs money so probably people use it for actual things???
06:12:59 <monqy> bad "roi" otherwise
06:13:28 <Bike> akira yamaoka uses max, he scores video games and crap.
06:13:28 <madbr> monqy: it's not about the money
06:13:54 <monqy> you can't buy roi with happiness
06:13:58 <monqy> return on investment
06:14:06 <madbr> you mean money roi or usage roi?
06:14:24 <zzo38> I don't like VST, although Csound can be compiled with support for VST (in both directions) if you need it. Csound can also load MIDI sequences if that would help you too.
06:14:42 <madbr> what's wrong with vst?
06:15:52 <zzo38> It has restrictive licensing, is Windows only, requires a GUI to use, and some others. Nevertheless Csound can use VST if it is helpful to do so.
06:16:53 <madbr> also it's like the least stupid of the plugin APIs
06:17:15 <madbr> AU is like 10 times more restrictive, RTAS like 100 times
06:17:33 <madbr> also it does not require a GUI
06:18:06 <madbr> making GUI-less VSTs is actually really easy
06:18:16 <madbr> easier than with a gui ofc
06:18:37 <zzo38> Well yes but some VSTs requires GUI.
06:18:48 <madbr> what's wrong with that
06:19:45 <zzo38> Nothing much, but still, I think Csound is much better (actually Csound can use GUI too)
06:20:18 <zzo38> Still sometimes some things can be difficult to do when a GUI is required.
06:21:13 <zzo38> Does VST require a different program for Windows and Macintosh computers?
06:22:03 <madbr> haven't tried compiling VSTs on macos
06:22:10 <madbr> so I dunno how different
06:22:42 <madbr> presumably you don't need too much source changes and it might all be handled in the VST header stuff
06:23:15 <madbr> (with preprocessors)
06:23:51 <zzo38> Well, still you need to compile it separately on the other computer, which is not always required with Csound and Pure Data and whatever (although you still do need to if you are using native code, although probably no source changes will be needed, at least with Csound and no GUI, I think)
06:25:46 <zzo38> I still find Csound is a good way to write instrument sounds. Yet if you need VST, you can use that with Csound too.
06:27:00 <madbr> Csound makes sense for generating samples that you'll load up in a more practical composition tool imho
06:27:41 <zzo38> Well, that won't work so well if you need parameters, though. But if you don't need parameters, that works.
06:28:27 <madbr> can you load csound as a vst?
06:28:32 <zzo38> madbr: Yes you can.
06:29:14 <madbr> yeah ok in that case it's potentially useful
06:29:14 <zzo38> And if you do that then you can use parameters too.
06:29:50 <zzo38> Csound can also load MIDI files in case you use a MIDI program (or external MIDI instrument) to write the music.
06:30:09 <madbr> no that's not what you want to do
06:30:18 <madbr> you want to pipe your midi into the VST
06:30:38 <madbr> from your host sequencer, real time
06:31:02 <zzo38> Csound can do that too, whether or not you use VST.
06:33:20 <zzo38> There are also some front-ends available for Csound.
06:34:00 <madbr> I'm not sold on the idea of front-ends
06:34:37 <madbr> dunno for sound programs but for emulators, the ones that use front-ends are the worst
06:35:05 <madbr> much better to have the gui and emu code as one program
06:36:49 <madbr> trying to locate a csound DLL I could load
06:37:51 <zzo38> I don't use any of the front-ends myself, but some day I may write a program to compile MML to Csound numeric score, to write music with it, as the numeric score format seems a terrible way to write any music (although some people have written music this way).
06:38:09 <zzo38> madbr: Yes I think there is DLL the documentation should mention it I think.
06:38:37 <zzo38> But I think for VST, you need to have VST support compiled in; I don't think VST support is compiled in by default, for some reason I don't know.
06:38:51 <madbr> I don't want to compile shit
06:38:57 <madbr> I want a DLL I can play with
06:39:27 <zzo38> Maybe you can download the DLL for VST separately and it will work; I haven't checked. But the documentation does say you can use VST both ways.
06:42:15 <zzo38> It says: CsoundVST is a multi-function front end for Csound, based on the Csound API. CsoundVST runs as a stand-alone graphical user interface to Csound, and it also runs as a VST instrument or effect plugin in VST hosts such as Cubase with the same user interface. CsoundVST is part of the main csound source tree, but is not included in standard distributions, due to licensing limitations of Steinberg's VST SDK.
06:44:58 -!- ogrom has joined.
06:45:38 -!- ogrom has quit (Client Quit).
06:52:12 <zzo38> It seems that CsoundVST binaries are not available just because SourceForge does not allow it; you may be able to download it from elsewhere or to compile it yourself and host it somewhere, or whatever.
07:00:46 -!- ogrom has joined.
07:08:40 <zzo38> But even if you want to load live MIDI data from one program to another, you can use the MIDI Yoke driver which allows this.
07:12:15 <madbr> no I want to avoid that kind of hacks
07:14:12 -!- ogrom has quit (Ping timeout: 255 seconds).
07:46:26 -!- epicmonkey has joined.
07:46:27 -!- elliott has quit (Remote host closed the connection).
08:00:57 -!- madbr has quit (Quit: Radiateur).
08:09:24 -!- variable has quit (*.net *.split).
08:11:31 -!- variable has joined.
08:16:50 <impomatic> zzo38: is your Forth adventure system still online somewhere?
08:19:27 <zzo38> impomatic: I don't think so.
08:20:08 <zzo38> A few things are broken and don't work properly, although I have it on my computer and if you want you can get it and correct these problems.
08:30:59 <impomatic> Mostly I wanted to see how it's done? :-)
08:33:17 <zzo38> The program is also badly written, although I can post it here: http://sprunge.us/RUiH
08:34:12 <shachaf> I’ve written some very ugly Haskell code that creates a vector using destructive updates. It is in fact an imperative algorithm, not a functional one. When the initialization is over the vector is frozen using unsafeFreeze. I wrote my code using read and write functions, tested it using QuickCheck and when the tests passed I switched to unsafeRead and unsafeWrite to make my program faster. Some time later I started getting random segfaults ...
08:34:18 <shachaf> ... when running my tests. This never happened before in any of my Haskell programs so I almost panicked. At first I didn’t had a slightest idea how to even approach this problem. I suspected that this might even be a bug in GHC.
08:46:03 -!- sirdancealot7 has quit (Ping timeout: 252 seconds).
08:48:09 -!- ogrom has joined.
08:49:03 <Jafet> This is why haskell programmers don't write tests, you know
08:57:20 -!- Bike has quit (Quit: dead).
09:08:04 <ion> I take it you tested the changed version with QuickCheck, too?
09:16:22 <Deewiant> (source: http://ics.p.lodz.pl/~stolarek/blog/2012/11/how-to-shoot-yourself-in-the-foot-with-haskell/ )
09:16:50 -!- epicmonkey has quit (Ping timeout: 246 seconds).
09:28:14 -!- Frooxius has joined.
09:38:03 -!- epicmonkey has joined.
09:43:22 -!- zzo38 has quit (Remote host closed the connection).
09:53:45 -!- nooodl has joined.
09:54:11 <fizzie> "The Norwegian convicted of the massacre of 77 people last year has said he is being held in "inhumane" conditions.
09:54:14 <fizzie> Anders Behring Breivik complained in a letter to the prison service that his coffee is served cold, he does not have enough butter for his bread, and he is not allowed moisturiser."
09:54:20 <fizzie> Sounds quite terrible indeed.
09:54:45 <fizzie> Also his cell is "poorly decorated and has no view".
09:55:19 <fizzie> These are possibly things to keep in mind next time you feel like killing 77 people; you might end up in a place where the coffee is often cold.
10:16:14 -!- nooga has joined.
10:17:14 -!- impomatic has quit (Quit: http://about.me/john_metcalf).
10:38:33 -!- monqy has quit (Quit: hello).
11:01:16 -!- epicmonkey has quit (Ping timeout: 245 seconds).
11:08:16 -!- copumpkin has quit (Quit: Computer has gone to sleep.).
11:30:53 -!- DHeadshot has quit (Read error: Connection reset by peer).
12:00:15 -!- Phantom_Hoover has joined.
12:43:10 -!- carado has joined.
12:48:50 -!- epicmonkey has joined.
13:02:48 -!- ais523 has joined.
13:03:20 <ais523> "You have received 87 new messages"
13:22:10 -!- epicmonkey has quit (Ping timeout: 246 seconds).
13:37:05 -!- nooga has quit (Ping timeout: 246 seconds).
13:40:19 -!- nooga has joined.
13:49:07 -!- atriq has joined.
14:20:58 -!- neostream has joined.
14:23:31 -!- neostream has quit (Quit: Ex-Chat).
14:32:36 -!- DHeadshot has joined.
14:58:13 -!- variable has changed nick to constant.
14:58:41 -!- constant has changed nick to trout.
15:09:34 -!- ogrom has quit (Ping timeout: 255 seconds).
15:29:54 -!- Phantom_Hoover has quit (Ping timeout: 265 seconds).
15:36:03 -!- ogrom has joined.
15:37:04 -!- Phantom_Hoover has joined.
15:39:11 -!- ogrom has quit (Client Quit).
15:41:00 <Phantom_Hoover> hey guys should i get the prison architect alpha or wait
16:32:09 -!- zzo38 has joined.
17:04:58 -!- Bike has joined.
17:23:25 -!- elliott has joined.
17:41:11 -!- Sgeo_ has joined.
17:42:35 -!- Sgeo has quit (Ping timeout: 246 seconds).
17:59:28 <kmc> "The Best Groupon Deal Ever: 86.2% Off On GRPN at NASDAQ"
18:07:56 <atriq> Advice: writing a library in Haskell for interfacing with Tumblr is NOT a productive use of time.
18:08:02 <atriq> Although it can prove educational.
18:08:27 <Bike> what's the lesson, apis for web 3.0 services suck?
18:09:40 <atriq> More "these are some good http libraries for Haskell"
18:10:00 <atriq> "here's how to stop them redirecting on a 301 error"
18:10:40 <Bike> 301 isn't an error.
18:11:41 <atriq> It caused errors :(
18:12:54 <Bike> It means that a page has moved. an http client is supposed to request the page moved to to finish the request.
18:13:22 <atriq> Not when it's used in the Tumblr API, it isn't
18:13:59 <Bike> which brings us back to my proposed lesson.
18:14:07 <atriq> It means "Here's some info about the image, but look! The image is actually here!"
18:14:10 <Bike> What's it mean in tumblr?
18:36:21 -!- Sgeo has joined.
18:38:34 -!- Sgeo_ has quit (Ping timeout: 246 seconds).
18:42:57 -!- ogrom has joined.
18:56:43 <atriq> There is over 24 hours of Homestuck music
18:57:46 <atriq> I make it about 26 hours
19:00:05 <atriq> For a silly webcomic only 3 years old, that's a lot
19:00:06 -!- ogrom has quit (Quit: Left).
19:00:13 <atriq> In fact, for a webcomic that's a lot
19:00:18 <atriq> In fact in fact, that's a lot
19:02:34 <atriq> The superlative doesn't need narrowing
19:03:28 <olsner> oh, according to QI, horses are the deadliest animals in australia
19:08:12 -!- Nisstyre-laptop has quit (Quit: Leaving).
19:12:21 -!- atriq has quit (Quit: dinner).
19:26:23 -!- carado has quit (Ping timeout: 246 seconds).
19:30:50 -!- Sgeo has quit (Read error: Connection reset by peer).
19:31:15 -!- Sgeo has joined.
19:32:43 -!- atriq has joined.
19:41:21 -!- carado has joined.
19:43:27 -!- carado has quit (Remote host closed the connection).
19:44:33 -!- carado has joined.
19:44:57 -!- carado has quit (Client Quit).
19:46:27 -!- carado has joined.
19:48:12 -!- AnotherTest has joined.
20:09:47 -!- Phantom_Hoover has quit (Ping timeout: 246 seconds).
20:10:47 -!- monqy has joined.
20:19:12 -!- sirdancealot7 has joined.
20:23:02 * pikhq_ watches the new My Little Pony
20:23:13 <atriq> I'm blaming Gregor for this.
20:23:32 <pikhq_> Actually, it's more the fault of meatspace people.
20:23:35 <atriq> Now, if you excuse me, I've got some Adventure Time to be watching.
20:25:06 <Gregor> pikhq_: SPOILARZ: They all die.
20:25:18 <pikhq_> And suddenly, an 11 episode funeral!
20:33:37 <pikhq_> It's weird, but I am a *lot* more comfortable watching MLP now that I came ultra-mega-hyper-out.
20:34:01 <pikhq_> "Lawl, that's gay" "Yup, I like cock so much I'm dating a woman with one. Problem? *trollface*"
20:34:53 <Jafet> Are you implying FIM isn't for heterosexual men
20:35:17 <pikhq_> No, just that I had slight latent fears of being accused of being gay.
20:35:30 <pikhq_> In spite of actually not being heterosexual.
20:36:13 <pikhq_> Sorry, I've just had rough experiences with people IRL being dicks, and not in the sense that I like. :P
20:36:21 <atriq> Is coming ultra-mega-hyper-out much different from coming regular-plain-ol'-out?
20:36:49 <pikhq_> atriq: Yes, it's about 3 times as much outness.
20:36:54 <Bike> it means pkhq is dating a transsexual pony, apparently
20:37:17 <atriq> I imagine it involves screaming unto the heaven while volcanoes erupt and cities crumble
20:37:39 <pikhq_> But I managed to inform people that I'm bisexual atheist dating a trans woman, not heterosexual Christian single.
20:38:30 <pikhq_> (the last, I would've just neglected to mention the trans bit, and it would've been merely twice-out or such, except it's a bit hard to be *discrete* when she's only started working on her voice.)
20:43:03 <soundnfury> pikhq_: you mean "discreet", not "discrete"
20:43:18 <soundnfury> unless you're trying to say that her voice forces you to be continuous ;)
20:43:38 <pikhq_> Sorry, I usually use the math word, so I sometimes slip up when using the "normal" word. :)
20:44:16 <soundnfury> "Your voice is so bad, it makes me thrice differentiable!"
20:47:26 <pikhq_> Also, not so much "bad" as it is "it's very much male".
20:47:42 <pikhq_> Whiich is a dead giveaway, isn't it?
20:49:53 <pikhq_> I don't know many cis women who speak in a baritone.
20:51:27 <Gregor> As a man with long hair, I can assure you that people are more stupid than you can possibly imagine.
20:51:53 <Gregor> "Uhhh, ma'am?" "I'm at a urinal, you FUCKING IMBECILE"
20:52:06 <pikhq_> Ah, right, I *have* experienced that.
20:52:15 * pikhq_ is not a man with long hair, but was for several years
20:52:34 <pikhq_> Was awesome turning around.
20:52:46 <pikhq_> "Uh, ma'am?" *turns around* "Ah, sorry."
20:52:54 <pikhq_> I also had an epic beard for the latter half of that.
20:53:13 <Gregor> I have... a scruffy goatee. And it's kinda generous to call it a goatee X-D
20:55:19 <pikhq_> Currently, I've got a maybe a couple weeks' growth full beard.
20:55:47 <Gregor> My scruffy goatee is several month's growth full beard.
20:56:12 <pikhq_> Several month's growth on me is a UNIX beard.
20:56:38 <coppro> several months growth on me is a light fuzz
20:58:13 -!- atriq has quit (Quit: Leaving).
21:01:01 -!- AnotherTest has left.
21:02:09 <pikhq_> coppro: Have you considered being more manly?
21:03:37 <Gregor> Several months growth for me is a light fuzz on the upper lip.
21:03:44 <Gregor> But a scruffy goatee on the chin.
21:14:52 -!- Nisstyre-laptop has joined.
21:22:17 <fizzie> I've only been mistaken as a lady (from behind) a few rare times. Once at a beach by a drunken man, that was probably the most memorable of them.
21:22:40 <pikhq_> It's pretty rare if you're tall.
21:22:47 <pikhq_> There's not that many 6' women walking around.
21:23:08 <pikhq_> I think I more commonly was mistaken for a grad student.
21:24:14 <pikhq_> Which was funny. "No, no, I'm a freshman."
21:24:29 <fizzie> I'm not terribly tall, I think 180.5 cm was the official measurement at some routine health checkup last. I suppose that's like 5'11".
21:24:49 <pikhq_> For a man, that's slightly above average. For a woman, that's rather tall.
21:26:45 -!- atriq has joined.
21:26:54 <fizzie> My wife's brother (one of them) and father are something like 6'3" or more; she's probably less than an inch shorter than me. They've certainly got the length genes.
21:27:14 <zzo38> I made a new kind of audio waveform synthesis "IDFX" synthesis, which uses four parameters "initial AND gate", "duty", "final AND gate", "XOR gate".
21:28:39 <atriq> I've been mistaken for female a few times
21:28:44 <atriq> When I had a stupid moustache
21:30:03 <zzo38> Which will work like w[t]=(((t&I)^-(t<D))&F)^X;
21:33:23 <atriq> News update: I am a better version of Simon
21:34:23 <zzo38> If I=X=0 then it makes up a square wave with duty D and volume F. It can also make saw wave and triangle wave with certain parameters.
21:49:05 <pikhq_> It's only been 25 years since the APA stopped viewing homosexuality as a disease.
21:49:42 <atriq> That's only 7 years longer than I've been alive!
21:49:50 <zzo38> The only thing you call a disease is because you want to call them a disease.......!!
21:49:53 <pikhq_> That's only 3 years longer than I've been alive.
21:52:31 -!- Nisstyre-laptop has quit (Quit: Leaving).
21:56:17 -!- atriq has quit (Quit: Leaving).
21:59:45 <olsner> pikhq_: sweden had that too until -79
22:01:22 -!- Phantom_Hoover has joined.
22:01:25 <olsner> iirc they removed it becasue a lot of people called in sick as a protest
22:01:51 <kmc> that is pretty great
22:02:11 <Phantom_Hoover> 21:49:05: <pikhq_> It's only been 25 years since the APA stopped viewing homosexuality as a disease.
22:02:58 <Phantom_Hoover> Because at least having homosexuality being classed as something innate was some kind of improvement.
22:03:23 <olsner> better than it being illegal, I suppose
22:04:19 <pikhq_> Phantom_Hoover: It was used as justification for curing the gay.
22:06:09 <pikhq_> It only became legal nation-wide in the 2003.
22:06:30 <pikhq_> When the SCOTUS ruled that sodomy laws were unconstitutional.
22:06:45 <pikhq_> Thereby legalizing it in 14 states.
22:08:27 <Phantom_Hoover> Was that something they were deliberately trying to do or just a side-effect?
22:08:39 <pikhq_> Deliberately trying to do.
22:08:51 <pikhq_> Heck, 3 states *still* have such laws on the books.
22:09:03 <kmc> it's dumb that we even have to care as a society whether homosexuality is "innate" or not
22:09:04 <pikhq_> (the others repealed on the notion that it was just embarassing now)
22:09:56 <kmc> it speaks to the fact that our legal system is still based on fear of divine judgement, rather than assessing harm to others or lack thereof
22:10:48 <kmc> in other news obama has finally won florida as well
22:12:34 <kmc> yay florida did not matter :)
22:12:46 <pikhq_> Oklahoma, Kansas, and Texas.
22:13:09 <kmc> in fact he would have won without OH, FL, *or* VA
22:16:38 <Phantom_Hoover> so would you say this election was more entertaining than the last
22:16:47 <kmc> definitely not
22:17:07 <kmc> last time we had an epic democratic nomination fight
22:17:23 <kmc> which i made a considerable amount of money betting on
22:17:50 <kmc> also the republican nominee last time was someone who had been written off as dead early on
22:17:55 <kmc> and neither of them was an incumbent
22:18:33 <kmc> 2008 was unusual and history-making in a bunch of ways
22:18:35 -!- quintopia has quit (Ping timeout: 260 seconds).
22:18:40 <kmc> new states coming into play, etc
22:18:52 <kmc> also i got really drunk on election night 2008 so yeah
22:18:56 <kmc> much more entertaining
22:19:54 <kmc> n.b. by "epic" i mean "actually epic" and not "internet word for kind of interesting"
22:23:49 <Bike> you mean "epic" as in "my five-volume poetical envisioning is already at the press", right
22:24:09 <kmc> i would read that
22:24:14 <kmc> actually an opera
22:24:20 <Bike> yes, yes indeed
22:24:32 <kmc> with barack obama and hillary clinton singing in italian
22:24:42 <Bike> terza rima the whole way through
22:24:57 <olsner> epic (from the Ancient Greek adjective ἐπικός (epikos), from ἔπος (epos) "word, story, poem"[1])
22:25:13 <zzo38> Can they speak Italian?
22:25:27 -!- atriq has joined.
22:25:32 <kmc> they'll just have to get copies of Teach Yourself Italian in 24 Hours For Dummies
22:28:02 <zzo38> Do you think /([-+])([0-9]*)(d[0-9]+|)(g-?[0-9]+|)([dk][hl][0-9]*|)/ is OK for dice rolling specification or do you think some features have been missed?
22:28:05 -!- quintopia has joined.
22:28:52 <FireFly> Shouldn't [-+] be optional?
22:29:14 <ais523> annoying things about mailman: I tried to send a message through a mailing list I administered
22:29:24 <zzo38> Actually it adds a + to the beginning implicitly so it is already optional for the first term only.
22:29:31 <ais523> it told me there were too many names on the To: line and that it had been held for moderation, and a separate email asking me to moderate it
22:29:35 <zzo38> For other terms it is required.
22:29:49 <ais523> FireFly: [-+] looks rather like an infinite loop, if it does anything at all
22:30:05 <FireFly> ais523: regex, not brainfuck :P
22:30:27 <ais523> I'd write the regex as [+-], for consistency in escaping character classes
22:30:33 <ais523> if you want ] in a character class it has to come at the start
22:30:37 <ais523> and - has to come at one of the ends
22:30:49 <ais523> so I put the - at the right end out of habit in order to avoid clashes
22:30:51 -!- Vorpal has joined.
22:31:14 <Bike> ais523: hey, I'm supposed to ask you about Feather
22:31:14 <ais523> so, say, something that matches BF comments is [^][<>,.+-]
22:31:31 <ais523> the summary is, it was originally an attempt at a serious esolang
22:31:35 <ais523> but I got mad thinking about it
22:31:45 <ais523> so it became an inside joke (the joke being "ais523 refuses to talk about Feather")
22:31:55 <zzo38> OK but that still doesn't completely answer my question (it only partially)
22:31:56 <Vorpal> <Bike> ais523: hey, I'm supposed to ask you about Feather <-- hah
22:32:04 <ais523> also nobody else really seems to understand the language
22:32:14 <ais523> I mean, I don't either, but I understood it more than other people before it drove me crazy and I stopped
22:32:22 <Bike> next you're going to tell me that i don't have to write a eodermdrome interpreter
22:32:28 <Vorpal> ais523, you gave it up for good?
22:32:44 <Vorpal> Bike, Feather involved retroactive changes to the interpreter of the language iirc
22:32:53 <Bike> that sounds pretty exciting
22:32:56 <ais523> Bike: an eodermdrome interp would be great, go for it
22:33:13 <Bike> well, yeah, I've got the subgraph isomorphism check thing down
22:33:15 <Vorpal> Bike, you don't *have* to do anything, that is up to you
22:33:36 <FireFly> I think the Feather article could use some links to relevant portions of the channel logs
22:33:38 <Bike> really, I'm pretty sure I was told that it was a rite of passage, and otherwise I'd be left to the brainfuck wolves
22:33:49 <Bike> and yeah, I was asking about the logs. since there's nine years in the topic.
22:34:16 <zzo38> I wanted to design a hardware and instruction set which makes certain things possible to do efficiently, including run Checkout programs, decode open source audio/video codecs, and emulate the Famicom APU and PPU.
22:34:16 <ais523> most recently I was poking at the edges of the problem that a Feather interpreter must be entirely written in Feather, also you must be able to interpret a Feather program through an arbitrarily large stack of Feather interpreters with only a constant performance penalty (that is, constant time added, not constant factor multiplied)
22:34:29 <ais523> this is… not really a trivial problem
22:35:03 <Vorpal> does anyone know of a software like teamspeak but that is open source?
22:35:17 <ais523> it's slightly more tractable if you force Feather programs to have a property that's a bit like totality but stronger
22:35:31 <Bike> that sounds like like that 900-page 3-lisp thesis i've never been able to read
22:35:35 <Bike> infinite stack of interpreters that is
22:35:43 <ais523> Feather does this sort of thing all the time
22:35:50 <FireFly> Vorpal: I've heard of Mumble, but haven't tried it
22:35:54 <Vorpal> <Bike> and yeah, I was asking about the logs. since there's nine years in the topic. <-- better start reading!
22:35:55 <ais523> it's no wonder that I got confused
22:35:59 <Vorpal> FireFly, thanks, will look into that
22:36:31 <ais523> (the larger problem, which I don't really know how to solve, is trying to determine, after a retroactive change to the fundamentals of the entire universe, which object before the change corresponds to which object afterwards)
22:36:52 <Vorpal> ais523, why does it require a constant performance penality?
22:37:00 <Vorpal> nobody said it had to be fast
22:37:00 <zzo38> Does someone here know something about design of such hardware and instruction set?
22:37:10 <zzo38> Does OpenCores have any such things?
22:37:26 <ais523> Vorpal: for everything except the 2,3 machine, I consider terminating before the heat death of the universe to be a desirable property
22:37:28 <ais523> especially for testing
22:37:48 <ais523> and even the 2,3 machine proof I managed to test each of the parts individually, and in pairs and threes and fours
22:37:56 <ais523> just fives was beyond the ability of my computer
22:38:10 <Vorpal> ais523, how much beyond?
22:38:20 <ais523> many orders of magnitude
22:38:34 <ais523> we're talking about O(2^(2^n)) performance here
22:38:42 <ais523> (where ^ is exponentiation, not bitwise xor)
22:38:58 <Vorpal> ais523, as long as it isn't up arrow I'm happy :P
22:39:04 <ais523> this is one of the least practical computational orders I've ever seen :)
22:40:06 <ais523> nonetheless, Wolfram was insisting that the machine in question might be useful for implementing DNA computers and such
22:40:14 <ais523> not with this construction, at least, it won't be :)
22:41:12 <ais523> one thing I realised for the first time when I met him in person was that he hadn't actually read or understood the proof
22:41:19 <kmc> heh, one of the rapid transit lines in new york has reopened with two stations deliberately closed to avoid dangerous overcrowding
22:41:20 <ais523> and just cared about the page hits from it
22:42:11 <ais523> well, or whatever the less web-specific version of page hits is
22:42:13 <Bike> proof of what?
22:42:17 <ais523> idea exposure, or whatever
22:42:38 <ais523> Bike: http://wolframprize.com
22:42:46 <Vorpal> Bike, ais523's proof that the 2,3 machine is TC
22:42:46 <ais523> (warning: sets a cookie on every image load, I /hate/ sites that do that)
22:42:54 <kmc> FireFly: PATH from Journal Square to 33rd St; Christopher St and 9th St stations are closed
22:43:01 <Bike> dang, that was you?
22:43:11 <Vorpal> ais523, oh is THAT what causes firefox to go crazy?
22:43:23 <ais523> Vorpal: it's not the only cause, but it's the most common
22:43:29 <ais523> Bike: yes, people didn't believe it for a while
22:43:31 <Vorpal> why does firefox handle it in such a stupid way
22:43:35 <Vorpal> pretty sure chrome does not
22:43:51 <ais523> Vorpal: because technically speaking they're separate requests
22:43:52 <Bike> i'll just assume that I'm talking to an alex smith lookalike then
22:44:03 <Vorpal> ais523, yeah but it is not a sensible way to handle it
22:44:10 <ais523> IMO choosing deny should deny cookies for all indirect loads from the page, until the next refresh
22:44:18 <Sgeo> I think I should read English As She Is Spoke sometime
22:44:23 <ais523> or make it explicit with a "deny for session" link
22:44:32 <Vorpal> ais523, it should collapse the stack of when I check the checkbox to always deny
22:44:40 <ais523> worse is things that set cookies on a timer, though
22:44:53 <ais523> (mostly that happens as the result of adverts or mouseover effects)
22:45:10 <Vorpal> I have noscript so yeah I wouldn't notice
22:45:12 <ais523> you could do it with meta refresh too but nobody does
22:45:32 <ais523> btw, holding down alt-d works and is a huge improvement over trying to close all the windows manually
22:45:50 <ais523> especially as they have to be closed in the correct order, and all overlap each other on the same point of the screen
22:47:00 * kmc is plotting to connect an Arduino to his gas stove; what could go wrong?
22:48:02 <ais523> <Wolfram Science> It is not clear, for instance, whether one should allow a Thue-Morse sequence in the initial conditions. In most cases, it should nevertheless be fairly obvious whether something should be considered a valid encoding for a universal system. But in general there is no firmly established criterion.
22:48:07 <ais523> that was written before the proof
22:48:11 <Vorpal> speaking of which, I should get a raspberry pi at some point
22:48:17 <olsner> kmc: obviously some kind of gas explosion
22:48:33 <olsner> also, didn't people stop using gas stoves along with the introduction of electricity?
22:48:53 <Vorpal> olsner, in Sweden, pretty much yes
22:49:03 <kmc> they are still very popular in the USA
22:49:07 <Vorpal> in UK I believe they remained popular for some time
22:49:35 <kmc> they provide much faster control of temperature
22:49:42 <ais523> Vorpal: they're still popular here
22:49:47 <kmc> and more even heating (depending)
22:49:52 <ais523> in both the house I live in and the house I'm currently online from, there's a gas stove
22:49:54 <Vorpal> olsner, but yeah in Sweden I only seen gas stoves in places without electricity, like my grandparent's kollonistuga (whatever that is in English...)
22:50:11 <ais523> they're pretty popular in houses that are gas heated anywhere
22:50:14 <Vorpal> kmc, even more than induction furnaces?
22:50:29 <Vorpal> ais523, the issue is I don't think there are any gas heated houses either here
22:50:49 <kmc> i don't know about induction furnaces
22:51:00 <Vorpal> err not furnace, stove
22:51:11 <Vorpal> I was thinking about minecraft and messed it up XD
22:51:11 <kmc> also i think in some areas of the US, natural gas is much cheaper than electricity for this purpose
22:51:14 <ais523> Vorpal: gas is almost as important as electricity in the UK
22:51:30 <Vorpal> ais523, I heard you even have gas lines in the streets over there
22:51:32 <ais523> I think this is partly because natural gas is one of the few fossil fuels that the UK (mostly Scotland) has that are usable for heating
22:51:42 <ais523> yes, every now and then they dig up the streets to repair them
22:51:58 <ais523> just like with electricity and water and (sometimes) telephone
22:52:05 <Vorpal> ais523, it sounds terribly unsafe to have those in the street, what happens if they leak?
22:52:12 <olsner> Vorpal: https://en.wikipedia.org/wiki/Allotment_(gardening)
22:52:15 <Vorpal> ais523, also where would the telephone be if not in the street?
22:52:32 <ais523> Vorpal: then the leak is typically quickly detected
22:52:41 <kmc> Vorpal: small leaks outside are not that much of a problem, it will dissipate and not build up concentration
22:52:45 <kmc> but yes, they detect and fix leaks
22:52:50 <kmc> electricity is also quite dangerous if it "leaks"
22:53:08 <kmc> every year a few people are electrocuted by touching streetlight poles or manhole covers that have mistakenly become electrically live
22:53:09 <ais523> Vorpal: there's a chemical added to the gas so that people can smell it
22:53:16 <ais523> and pretty much everyone (well over half the population) recognises it
22:53:23 <kmc> natural gas fires/explosions are a particular concern in earthquakes
22:53:27 <ais523> you smell a small amount of it whenever anyone lights an oven
22:53:34 <ais523> and there's an emergency number for gas leaks
22:53:43 <ais523> kmc: the UK, at least, is particularly tectonically stable
22:53:55 <Vorpal> olsner, are you sure we two didn't travel back in time somehow relative to the rest of the world?
22:54:10 <ais523> Vorpal: a few years ago, there was a large gas odoriser leak
22:54:15 <ais523> it caused chaos with all the false positives
22:54:25 <ais523> (some sort of accident at a factory that was manufacturing gas odoriser)
22:55:05 <olsner> Vorpal: different parts of the world live in different so-called "time zones"
22:55:44 <Vorpal> olsner, ah, that must be it
22:56:38 <fizzie> We had a gas oven where we lived back when I was (more of) a child.
22:57:56 <fizzie> There was a leak -- I forget the details -- and I was home alone, and like somewhere between six-to-eight years old, and I had been told not to go outside alone, but it was so smelly inside, due to that leak.
22:58:34 <fizzie> When my parents came home, they found me standing inside the apartment but with my head outside in the stairway; they asked me what on earth I was doing, and I apparently replied that it's too smelly inside.
22:59:04 <fizzie> From what I have been told, they were kind of relieved-shocked-scared after realizing what had been going n.
22:59:25 <kmc> the punchline is that there was no gas leak, you had just been farting a tremendous amount
22:59:29 <olsner> why didn't you open a window?
22:59:53 <fizzie> olsner: I don't know. I don't really remember any of this, I've just heard about it afterwards.
22:59:59 <olsner> I wonder if that's a good or bad thing to do when you have a gas leak
23:00:00 <kmc> some friends of mine (mathematicians all) lived in a house with a gas fireplace
23:00:11 <kmc> they tried to turn it on but lacked necessary tools so they went at it with pliars and a blowtorch
23:00:26 <kmc> this created a persistent gas leak in the living room, which was noticed by all but not really deemed to be a problem
23:00:29 <atriq> They were never heard from again
23:00:42 -!- sirdancealot7 has quit (Ping timeout: 264 seconds).
23:00:44 <ais523> people tend to take gas leaks seriously around here
23:00:45 <kmc> "so, how bout that gas leak?" "yup." "someone should do something about it"
23:00:51 <ais523> oh, also they install carbon monoxide detectors
23:00:57 <ais523> to allow for another common malfunction
23:01:02 <nortti> did you know that konqueror supports ^u and hjkl?
23:01:03 <kmc> this went on for a few weeks until one of their girlfriends visited the house and she was like "uh, you guys are idiots"
23:01:16 <Vorpal> olsner, how very peculiar the rest of the world seems to be
23:01:27 <ais523> nortti: it also will give a keybinding to as many links on the page as it can if you press and release control
23:01:30 <fizzie> Huh. The backlight of this (win7) laptop just turned to maximum brightness, and the (probably GPU) fan started to go on like anything. Also everything went kind of bright.
23:01:41 <ais523> fizzie: sounds like some sort of total failure of ACPI
23:02:07 <fizzie> ais523: Kind of weird. Perhaps I should rebootsen.
23:02:24 <Vorpal> nortti, ^u? What does that do?
23:02:39 <ais523> fizzie: hmm, if ACPI is that broken, weird stuff might happen on the reboot attempt
23:02:40 <Vorpal> and since hjkl is vi(m) I assume so is ^u
23:02:48 <ais523> if you do, let us know what happened when you come back
23:02:51 <nortti> Vorpal: it is emacs/shell keybindig. empties current line
23:02:58 <fizzie> ais523: Bringing up the "NVIDIA Control Panel" made the brightness go back to normal and the fan quiet down.
23:03:10 <ais523> Vorpal: hjkl is NetHack, and control-u moves north-east as much as possible
23:03:17 <ais523> fizzie: ah, so some sort of GPU reset, perhaps?
23:03:23 <fizzie> (Also the display flicker briefly, but I don't recall whether the control panel always does that.)
23:03:24 <Vorpal> ais523, oh, I use numpad in nethack
23:03:38 <Sgeo> ais523, I think NetHack took those keys from vi(m)
23:03:45 <nortti> Vorpal: also I think readline (or what was it called. bash and irssi use it) has it.
23:04:18 <nortti> ais523: hjkl is from ex
23:04:21 <ais523> nortti: hmm, I use C-a C-k to empty the current line in Emacs (and that also works in bash, at least)
23:04:46 <ais523> elliott: is that a capital U specifically?
23:04:58 <nortti> ais523: ^u and ^U do the same
23:05:01 <ais523> you never can tell with keybindings
23:05:15 <nortti> ais523: 2 most significant bits of 7 bit ascii char are cut out
23:05:48 <ais523> nortti: not on /every/ possible control combo
23:05:53 <ais523> although it's the case with most of them
23:06:32 <elliott> ais523: it's capital U because that's how ^ works :P
23:06:44 <ais523> this is why I use C- notation :P
23:06:55 <elliott> > let control x = chr (ord x - ord '@') in control 'U'
23:06:59 <elliott> > let control x = chr (ord x - ord '@') in control '@'
23:07:02 <elliott> > let control x = chr (ord x - ord '@') in control 'u'
23:07:12 <elliott> i can assure you: it is not 5
23:07:23 <shachaf> > let control x = chr (ord x - ord '@') in control 'U'
23:08:47 <nortti> ^s is xoff but I have never heard or ^r
23:10:52 <FireFly> In what context? in shells it's usually reverse-incremental-search
23:11:53 <Vorpal> FireFly, what is the usual mapping of history-search-backward and history-search-forward
23:11:59 <Vorpal> I use my own custom mappings for them
23:12:04 <Vorpal> so I forget what the default ones are
23:12:58 <FireFly> I don't know what those do
23:14:05 <fizzie> Vorpal: bash (according to the readline parts of the man page) doesn't have a default binding.
23:14:20 <Vorpal> FireFly, allow you to step to the previous/next line in history that matches the currently written line. So if I write cd f<PgUp> (which is how I have it bound) it will go to, say "cd foo" then on PgUp again to "cd faa" or whatever was done before that
23:14:55 <fizzie> I just do that with reverse-incremental-search, generally.
23:15:06 <fizzie> Possibly due to laziness.
23:15:15 <Vorpal> fizzie, how does that work, it isn't something I ever used
23:15:50 <fizzie> You type ^r first, and then 'cd f'; it will keep incrementally-updating the search result.
23:15:58 <fizzie> And then you can keep tapping ^r to go further back.
23:16:08 <Vorpal> and if you went to far, what do you hit then
23:16:13 <Vorpal> to go forward one step
23:16:14 -!- sirdancealot7 has joined.
23:16:17 <fizzie> I forget exactly what. :p
23:16:48 <Vorpal> fizzie, I have to say that history-search-backward/history-search-forward seems like slightly less typing to me
23:17:55 <FireFly> Vorpal: if you do "foo<PgUp>", would it match e.g. "cd foo"?
23:18:04 <fizzie> "<C-r>cd f" has the same number of typed characters as "cd f<PgUp>".
23:18:15 <Phantom_Hoover> Vorpal, ISTR zsh bound going forward to whatever Emacs uses for going down a line.
23:18:17 <Vorpal> FireFly, no it is bound to the start of the line
23:18:24 <Vorpal> Phantom_Hoover, I'm a bash user
23:18:40 <FireFly> Vorpal: I'm pretty sure ^R would match that
23:18:53 <fizzie> The incremental search matches anywhere in the line, yes.
23:18:54 <Vorpal> FireFly, right, and if I don't want that?
23:19:03 <fizzie> It's sometimes useful; sometimes not.
23:19:19 <Vorpal> does C-r support regexp?
23:19:26 <Vorpal> so you just put ^ in front or such
23:19:39 <elliott> <fizzie> You type ^r first, and then 'cd f'; it will keep incrementally-updating the search result.
23:20:00 <FireFly> Vorpal: the zsh version does, apparently
23:20:06 <Vorpal> elliott, so it is C-Shift-r?
23:20:28 <elliott> ^R means press control and the R key
23:20:40 <Vorpal> elliott, and what does ^r mean?
23:20:41 <elliott> because that's what a terminal sends
23:20:45 <elliott> ^r means something nonsense
23:21:13 <Vorpal> elliott, how do you type ^r?
23:21:15 <fizzie> The "UI" of the search is also sometimes somehow iffy; you can keep typing things, but if there are no matches, the input doesn't show up; yet it "exists" in that backspace removes it before it affects the visible part.
23:22:11 <Vorpal> fizzie, well I guess it is nice having both that and the history search style I'm in love with, for different occasions
23:23:26 <fizzie> Bash/readline incremental search doesn't seem to support anything else than just plain (unanchored) string match, sadly.
23:23:29 <Vorpal> my recommendation for your inputrc if pgup/pgdown is easy to reach on your keyboard (as it is on my laptop):
23:23:30 <Vorpal> "\e[5~": history-search-backward
23:23:30 <Vorpal> "\e[6~": history-search-forward
23:25:14 <fizzie> I keep using "<C-r>sprunge" to pull up a "curl -F 'sprunge=<-' http://sprunge.us" command line, because I somehow can't seem to manage to be bothered to make an alias/a script for that. I do it so often that it always seems to be in the history; the fact that I can reach it with that is probably the reason I still haven't managed to make that command.
23:26:04 <fizzie> (It's a vicious circle.)
23:26:13 <elliott> fizzie: i am the exact same
23:26:19 <elliott> i never thought i'd meet someone with that problem
23:26:48 <pikhq_> Weird. I've just had ~/local/bin/sprunge for ages.
23:27:08 <nortti> fizzie: try shell that doesn't save history in a file :P
23:27:10 <pikhq_> Though, I also have ~/local/bin so....
23:27:13 <fizzie> I've had that at home for quite a while, it's just the workstation at wrk that's the problematical one.
23:27:20 <atriq> elliott, when was the last time you actually met someone
23:27:48 <fizzie> And I do have a ~/local/bin/ at work, it's not like I couldn't just type "scp ~/local/bin/sprunge james:local/bin/" at home once and get it done.
23:27:51 <elliott> atriq: 1992 when i died irl
23:28:15 <Phantom_Hoover> i've been hoping to meet a fellow member of the dead community for so long!
23:28:20 <fizzie> (I even typed it as a part of that comment, just not in the right place.)
23:36:34 <ais523> Vorpal: hmm, perhaps the Web would be better off if images couldn't set cookies
23:36:45 <ais523> I only know of one legitimate use for that, and it's workaroundable
23:42:33 <ais523> elliott: Wikimedia's unified login thing
23:42:44 <ais523> for setting cookies on all the domain names they use when you log in
23:42:55 <ais523> they could use iframes for that instead, though
23:43:16 <ais523> or even a huge redirect chain, I guess
23:57:00 -!- DHeadshot has quit (Read error: Connection reset by peer).
23:57:05 -!- DH____ has joined.