00:00:26 * ais523 vaguely wonders what foo would do in void context, but it's probably optimized out
00:02:52 <ehird> 00:02 sproingie: mmmKAY
00:03:08 -!- oerjan has joined.
00:03:35 <ehird> Sgeo[College]: one perl person finds it stupid, I think is more accurate
00:03:35 <ais523> well, it seems that Slashdot achievements are here to stay
00:03:47 <ehird> ais523: achivements?
00:03:55 <ehird> also, what was /.'s ap foos?
00:04:03 <ais523> ehird: there were several
00:04:07 <ais523> but achievements were their on-site things
00:04:14 <ais523> it's basically a concept taken from MMOs
00:04:25 <ais523> http://slashdot.org/~ais523/achievements probably explains it better than I could, anyway
00:04:36 <ehird> that's gotta be not real
00:04:56 <ehird> 00:03 sproingie: my attention span isn't letting me read the texty bits
00:04:56 <ehird> 00:03 sproingie: "on april 1 2009, skynet achieved consciousness. and decided it liked pandas"
00:05:02 <ehird> Update: sproingie has severe ADHD.
00:05:04 <ais523> with Slashdot, you never know
00:05:21 <ehird> Hmm, so nothing I didn't know. #perl is not that much of a good channel.
00:05:25 <ais523> Sgeo[College]: for putting a negative tag on something
00:05:31 <ais523> many of them are really easy, it seems
00:05:43 <ais523> I wonder how many people will try to get as many as possible?
00:06:07 <ehird> http://news.php.net/php.internals/41374/uq ← Wow, php guys have a good april fools for once
00:06:32 <ais523> ehird: that one's genuine
00:06:40 <ais523> I've known about it for a while, anyway
00:06:53 <ehird> That was a gale-force wind trying to go over your head but knocking you over instead.
00:09:35 <ehird> anyone have that perl script that moves the window around in waves?
00:10:17 <ehird> 00:09 Limbic_Region: sparc - no, you should be prematurely worrying about optimization
00:11:14 <ais523> also, wow at the other choices they considered for namespace separators
00:11:47 <ehird> [16:09:18] <@CelloG> i.e. foo::bar might be a short name - class foo, method bar, but it could also be a long name, function foo::bar
00:11:48 <ehird> [16:09:25] <@CelloG> and the engine just assumes it is a short name
00:11:49 <ehird> [16:09:36] <@CelloG> and that is why :: is fatally flawed as a separator
00:11:53 <ehird> YOUR IMPLEMENTATION SUCKS
00:12:27 <ais523> oh, I see the problem, namespaces and classes are separate in PHP
00:12:56 <ais523> still, Perl used to use ' as a namespace separator
00:13:09 <ais523> IIRC it's still accepted for backwards compatibility
00:13:13 <ais523> although :: is more usual
00:14:27 <ais523> use Language'INTERCAL'Parser
00:14:34 <ehird> ais523: and then youc an do
00:14:40 <ehird> Language'INTERCAL'Parser->new()?
00:14:45 <ehird> > _______________________________________________ <
00:15:09 <ais523> use Language'INTERCAL'Parser;
00:15:11 <ais523> my $x = Language'INTERCAL'Parser->new();
00:15:12 <ais523> Usage: new Language::INTERCAL::Parser(SYMBOLTABLE) at - line 2
00:15:14 <ais523> the last line is perl's output
00:15:29 <ais523> VHDL still uses ' to extract metadata from things
00:15:32 <ais523> and also as a cast operator
00:16:01 -!- Sgeo[College] has quit ("http://www.mibbit.com ajax IRC Client").
00:16:12 <ais523> write (l, String'("Hello world!"));
00:16:18 <ais523> <--- a line from the standard VHDL hello world
00:18:08 <ais523> oh, and /also/ to delimit characters
00:18:30 <ais523> the other use is, for instance a'lower to get the lower bound of array a
00:18:34 <ehird> After a lot of research, I discovered the third one was in
00:18:34 <ehird> fact Baudot (the INTERCAL variety, I believe)
00:18:42 <ehird> how come coppro knows everything?
00:18:55 <ehird> maybe he's secretly CADIE
00:19:08 <ais523> well, the only Baudot encoder/decoder I know of is the one that comes with C-INTERCAL
00:19:16 <ais523> oh, and CLC-INTERCAL itself, ofc
00:19:27 <ais523> it's a 19th-century encoding, after all
00:21:07 <ais523> anyway, that Baudot was a polygolt
00:21:17 <ais523> between the traditional and INTERCAL versions
00:21:27 <ais523> why have one when you can have both?
00:21:56 <ehird> ais523: hmm, do you know coppro, perchanc?
00:23:46 -!- olsner has quit ("Leaving").
00:24:12 <ehird> I think cadie's committed suicide or something.
00:24:31 <ais523> maybe they switched her off
00:24:41 * kerlo goes to google.co.uk to check on the ehird version of CADIE
00:24:54 <ais523> hmm... google's pretty slow atm
00:24:57 <ehird> fun fact: I am the only brit
00:25:40 <kerlo> Actually, there are seven Brits. ehird is one, and the six others make up the entirety of the BBC.
00:25:56 <ehird> kerlo: they just swap costumes a lot?
00:26:03 <ehird> also, where will i work when I grow up? The BBC?
00:26:31 <ehird> kerlo: seven brits is more than the 3 finns
00:26:38 <ehird> (of which 5 or 6 are in here; I forget
00:27:16 <kerlo> Actually, I'm mistaken. There are seven people at the BBC, but one isn't officially a member; she just joins them whenever they need an attractive female, because all the actual BBC members are male.
00:28:00 <ais523> Con: INTERCAL extensions such as threading and operator overloading are non-standard and poorly tested, and may therefore have broad unforeseen side effects.
00:28:02 <ais523> Pro: INTERCAL extensions such as threading and operator overloading are non-standard and poorly tested, and may therefore have broad unforeseen side effects.
00:28:04 <ais523> I love that style guide
00:28:14 <kerlo> Also, one of the BBC members is dead.
00:28:34 <ais523> kerlo: am I ehird, or am I at the BBC?
00:28:48 <ais523> I hope I'm not ehird, that would be worrying
00:28:53 <kerlo> ais523: chances are you're at the BBC.
00:28:58 <ehird> ais523: sorry you had to find out this way
00:29:00 <ehird> you're the unifier
00:29:03 <ehird> you're both me, and at the BBC
00:29:10 <ehird> it helps to keep the two realities in bind
00:29:19 <ehird> sort of, subatomic reality glue
00:29:40 <kerlo> ais523 is the female.
00:30:05 <ehird> if you measure the female over time she turns out to be indistinguishable to me at micro levels
00:30:10 <ehird> even though we are separate at the macro levels
00:30:11 <oerjan> i take it you take turns playing the Doctor?
00:30:29 <ehird> http://googleappengine.blogspot.com/2009/04/brand-new-language-on-google-app-engine.html
00:30:34 <ehird> FORTRAN 77 on app engine
00:31:10 <ehird> thanks to CADIE, ofc
00:32:07 <ais523> you have to mail them punch cards to submit the code, though
00:32:34 <ehird> "Checking the HTTP header for google.com today, you will find it’s run by “ELIZA”, “WOPR”, “IIS/3.0”, “Google Operating System (BETA)” or similar... perhaps depending on the time you’re checking"
00:32:53 <ais523> that IIS is very worrying
00:33:36 <ehird> "Google’s image album software Picasa now comes with a special “Auto Red Eye” functionality for version 4.1 (it turns normal eyes into red eyes). I was only able to see this on the Picasa homepage when using a US proxy. "
00:34:06 <ais523> they added an option to google code search to search for programs in lolcode
00:34:32 <ais523> oerjan: Microsoft's webserver
00:34:54 <ais523> losing out quite badly to Apache in terms of market share
00:36:05 -!- sebbu has joined.
00:36:46 <ehird> How did you get started as a designer of programming languages?
00:36:46 <ehird> Ken Iverson showed me APL in 1969 when I was 11.
00:36:52 <ehird> — Kx systems (K developers) ceo
00:37:25 <kerlo> Did APL invent the term "programming language"?
00:37:58 <ais523> kerlo: that's a seriously great question
00:38:04 <ais523> I'd never even thought in those terms before
00:38:10 <ais523> it's such a pity the answer is no, actually
00:39:00 <ehird> http://news.netcraft.com/archives/2009/04/01/deluge_of_browser_security_issues_drives_mass_migration.html
00:39:05 <ehird> Non-lynx browsers are dead, Netcraft confirms it.
00:39:18 <ais523> netcraft did indeed confirm it
00:39:22 <ais523> but does that make it true?
00:39:29 <ehird> Just ask any Slashdot troll!
00:39:42 <ais523> also, that stuff about encoding shellcode into smileys, is it definitely known to be false?
00:39:55 <ais523> after all, it could be true, although would look suspicious
00:40:08 <ais523> http://blog.cr0.org/2009/04/massive-exploitation-of-instant.html
00:40:18 <ais523> of course, actually getting the smileys to run would be harder
00:40:27 <ehird> http://www.cr0.org/misc/smile.rb
00:40:41 <ehird> ais523: well, it just makes shellcode look like smilies
00:41:01 -!- Slereah has quit (Read error: 110 (Connection timed out)).
00:41:23 <ais523> oh, and the resulting smilies aren't executable
00:41:32 <ais523> smiley steganography < smiley/shellcode polyglots
00:42:02 <ehird> Three smilies and ... something.
00:42:04 <ais523> yep, that's referred to as the killer smiley sometimes
00:42:22 <ais523> the last one's someone with glasses, winking
00:42:49 <kerlo> Very small glasses that are below the eyes?
00:42:49 <ehird> Erste said that while the bank is dedicated to providing an accessible online banking experience, some customers still report difficulties when trying to make HTTPS requests through Telnet without the aid of an extended keyboard layout.
00:43:03 <ais523> kerlo: "four-eyes" is a common nickname for glasses-wearers
00:43:06 <ehird> To bolster Lynx's growing footprint in the browser market, Netcraft has released the Netcraft Toolbar for Lynx. This free add-on blends in at the top of every web page, and not only protects Lynx users against phishing attacks, but the beautiful text-based rendition of the Netcraft logo is sure to brighten anyone's day.
00:43:16 <ehird> http://news.netcraft.com/archives/2009/04/01/netcraft-toolbar-for-lynx-resized.png
00:43:26 <ais523> also, I tried using Lynx to access a bank website earlier today
00:43:37 <ais523> although Lynx is so garish and hard to control, I preger w3m I think
00:43:53 <ais523> maybe the first text browser you use is the one that you stick with for life
00:44:27 <ehird> w3m is sort of "gnu lynx"
00:44:42 <ais523> well, it certainly has too many features
00:44:45 <ais523> it even has tabbed browsin
00:44:55 <ais523> which I think is hilariously funny, despite being useful
00:45:33 <ehird> the real pain is links, really — you want to click links with a mouse because they're suited for that sort of addressing
00:45:56 <ehird> given a sufficiently bloated terminal.
00:45:58 <ais523> it uses some sort of terminal mouse-click layer
00:46:04 <ehird> does it work in urxvt? that terminal is excellent
00:46:05 <ais523> also, you have to click twice I think
00:46:16 <ehird> vim lets you move the cursor with one click
00:46:21 <ais523> and I don't know, I've only tried in gnome-terminal
00:46:33 <ais523> ehird: same with w3m, one click moves the cursor, the second follows the link
00:47:04 <ehird> 'moves the cursor'?
00:47:12 <ehird> oh, right, w3m lets you move around text for no reason whatsoever
00:47:22 <ais523> I find it easier than trying to use Lynx
00:47:34 <ais523> but then, I'm used to using the keyboard as a mouse
00:48:01 <ehird> i'm a pretentious mouse user
00:48:45 <ehird> pretention levels: hunt and peck keyboarder and mouser < keyboard-only hardcore 1337 h4ck3r < person who uses keyboard for typing and mouse for precise, vague and absolute addressing
00:49:19 <ais523> I mostly only use the mouse for websurfing
00:49:35 <ais523> and even then, I use w3m if I want to quickly look something up on a local html file rather than do serious surfing
00:49:40 <ais523> because it means I don't have to go to the mouse
00:49:44 <ais523> oh, I use the mouse for games too
00:50:10 <ehird> spend a week with plan 9; you can't even use up/down keys for editing (they are pgup/pgdown)
00:50:23 <ehird> just left/right, typing, and backspace (no shift-selecting, or even DEL key)
00:50:38 <ehird> it's simultaneously agonizing and enlightening
00:51:52 <ehird> Plan 9 isn't an elitist system, just an opinionated one.
00:51:59 <ehird> A very, very, very opinionated one.
00:52:10 <ehird> "Because dwm is customized through editing its source code, it's pointless
00:52:10 <ehird> to make binary packages of it. This keeps its userbase small and
00:52:11 <ehird> elitist. No novices asking stupid questions."
00:52:18 <ehird> ais523: `apt-cache show dwm` would you?
00:52:22 <Robdgreat> "I'm not opinionated, I'm just always right"
00:52:23 <ehird> I bet £50 it has a package
00:52:44 <ehird> [[maybe you should go to school and learn how to spell cuz who the fuck
00:52:44 <ehird> spells marK with a c? musta been a baby of a couple retards]]
00:52:58 <ais523> Please notice that dwm is currently customized through editing its source code, so you probably want to build your own dwm packages. This package is compiled with the default configuration and should just give you an idea about what dwm brings to your desktop.
00:53:05 <ais523> <--- partial output from apt-cache show dwm
00:53:16 -!- sebbu2 has quit (Read error: 110 (Connection timed out)).
00:53:16 <ais523> ehird: I didn't accept the bet
00:53:57 <Robdgreat> let the record show that ais523 tacitly accepted the bet
00:54:10 <ais523> what language is dwm written in?
00:54:14 <ais523> I'm guessing C from the dependencies
00:54:27 <ais523> I was wondering about somehow smalltalkifying it so that you could edit the source while it was running
00:54:29 <ehird> ais523: it's written by the suckless.org people
00:54:41 <ehird> they're a pretentious, people-hating bunch of plan 9 addicts
00:54:46 <ehird> http://dwm.suckless.org/
00:55:05 <ehird> ais523: I think if you did such a modification they'd treat it akin to a death threat and an insult on their mother
00:55:15 <ais523> is that firefox in the foreground in that screenshot?
00:55:36 <ehird> ais523: asking consistency of them is henious.
00:55:40 <ehird> Go to your room and think about what you've done.
00:56:09 <ais523> "dwm has no Lua integration, no 9P support, no shell-based configuration, no remote control, and comes without any additional tools, such as for printing the selection or warping the mouse."
00:56:20 <ais523> I like the way that's in the feature list
00:56:31 <ehird> Yeah, they go on about that a lot: http://suckless.org/common/
00:56:59 <ais523> hmm... also, it uses mercurial as the vcs
00:57:08 <ais523> I'm disappointed, I'd expected them to use something I'd never heard of
00:57:16 <ehird> ais523: regardless of how silly they are, reading the code is nice: http://code.suckless.org/hg/dwm/file/deaa276abac1/dwm.c
00:57:23 <ehird> 1706 lines; and a lot of sane people swear by it
00:57:30 <ehird> modern-software people, that is
00:58:40 <ais523> how does that fit in with their concept of using less code rather than bad code
00:58:47 <ais523> I mean, obviously that section contains macros
00:58:58 <ais523> that's like writing x = y + 2; /* addition */
00:59:00 <ehird> ais523: send a patch just to remove those two lines
00:59:11 <ehird> they'll probably accept it and hail you as a genius in the commit messag
00:59:19 <ehird> 207 /* variables */
00:59:23 <ehird> 130 /* function declarations */
00:59:43 <ais523> 247 /* function implementations */
00:59:57 <ais523> it must have been written by COBOL programmers
01:00:11 <ais523> COBOL requires you to do that, C doesn't but they've kept the habit
01:00:39 <ehird> http://incise.org/tinywm.html i still love this wm
01:00:50 <ehird> I bet I could write a humane WM in ~500 lines
01:00:59 <ais523> ehird: [01:00] <AirRaven> IRP > Should I go to York or Manchester University next year? (U.K.)- if all else fails, mudkips or coin-flips shall suffice.
01:01:08 <ais523> how long's xmonad, by the way?
01:01:11 <kerlo> ehird's hobby is pressing the last letter of his sentence and the enter key simultaneously
01:01:25 <ais523> kerlo: I thought it was me that did that, not ehird
01:01:31 <kerlo> It works roughly half the time/
01:01:43 <kerlo> ...I'm bad at this.
01:03:20 <ais523> ehird: that dwm source is clearly golfed
01:03:28 <ais523> it's doing things like fitting entire loops into the head of a for loop
01:03:44 <ais523> which is always possible, but only used for golfing and showing off AFAIK
01:07:33 <kerlo> Anyway, kerlo's Canonical Programming Rule: The only good way to write a program is the best way to write the program.
01:08:18 <ais523> what's the canonical programming hello world, in your opinion?
01:08:44 <kerlo> Canonical Programming Rule, corollary 1: If you can think of multiple equally good ways of writing that beat all other ways of writing it, you need to come up with an even better way of writing it.
01:09:09 <kerlo> Canonical Programming Rule, corollary 2: If there are multiple best ways of writing a program, the programming language sucks.
01:09:22 <kerlo> ais523: what language?
01:10:08 <kerlo> PutStrLn "Hello, world!"
01:10:55 <ais523> what language is that?
01:11:00 <kerlo> A hypothetical language I've thought about since... a while ago.
01:12:36 <mmorrow> speaking of lines-of-code, i happened to graph loc for each lua version earlier http://moonpatio.com/lua/lua_loc.png
01:12:50 <ais523> [17:20:15] <@lsmith> ok .. this is quite unanimous .. even if we count Stas to be of the same opinion as dmitry
01:12:52 <kerlo> After I finish it, ask the Singularity to upload a description into your mind.
01:12:56 <kerlo> It'll take a while, you see.
01:13:18 <ais523> mmorrow: that's quite a lot for a lightweight scripting language
01:13:21 * kerlo decides that the best thing he could possibly do right now is read the Wikipedia page "Concept"
01:14:26 <mmorrow> ais523: the lua code is some of my favorite C code (wrt clarity/tidyness)
01:14:56 <ais523> [17:34:18] <@andrei_> hey, what about :) separator
01:15:08 <ais523> hmm... I'm not surprised they went with \ in the end
01:15:24 <ais523> kerlo: from the actual discussion between PHP devs about what char to use for namespacing
01:16:33 <ais523> apparently they had to make a decision in a hurry because it was degenerating into an emacs vs. vim holy war
01:19:49 <kerlo> Hey, I think this also solves another of my problems.
01:20:05 <ais523> what, smileys as the PHP namespace separator?
01:21:07 <kerlo> Well, either I've forgotten which problem it solves, or I've realized that it actually doesn't solve that problem.
01:23:13 <ais523> well, what was the solution, anyway?
01:23:18 <ais523> even if it's looking for a problem
01:24:36 <kerlo> Concept splitting.
01:24:57 <kerlo> If you don't know whether an item is an instance of a category or not, split the category into two categories, one containing the item and the other not.
01:25:44 <kerlo> If you forget about one category, great: you've determined that the other category is the one worth remembering.
01:26:08 <kerlo> And now, having told you that, I have to kill you.
01:26:45 <kerlo> Or at least ask you if you'll ever write or otherwise work on an AI that isn't Friendly.
01:27:02 -!- Sgeo[College] has joined.
01:27:14 <Sgeo[College]> ehird: transcript of http://www.youtube.com/watch?v=BeJ9Q40kIR0 ?
01:30:15 <kerlo> I was mainly asking ais523, you know.
01:30:29 <oerjan> that might be more important.
01:30:57 <ais523> that's how TAEB identifies items in NetHack, it's a rather limited application though
01:32:39 <ais523> unrelated to #esoteric, although it's something I work on from time to time
01:32:44 <ais523> not my project, other coders are more active on it
01:32:52 <ais523> I'm just a low-level person who helps out occasionally
01:34:03 <ais523> it's not april fool's any more, and I have no reason to lie to you about that
01:35:01 <kerlo> I want proof, not because I don't believe you, but because proofs lead to understanding that mere statements do not.
01:35:25 <kerlo> Much how the string "IO ()" is much less useful than an actual Haskell program.
01:35:29 <oerjan> kerlo: don't trust him, he could be lying about his timezone
01:35:33 <ais523> kerlo: http://sartak.org/code/TAEB/
01:36:13 <ais523> my AI branch is at http://ais523.sartak.org/TAEB-AI-Planar/
01:40:04 -!- psygnisfive has quit ("Leaving...").
01:41:21 <ais523> heh, http://www.sixxs.net.ipv4.sixxs.org.sixxs.org.ipv4.sixxs.org/main/ actually works
01:42:58 <ais523> wtf, I just got first post on a Slashdot article
01:53:12 -!- psygnisfive has joined.
01:54:30 <kerlo> That URL makes me think of http://normish.org/ihope/root/var/www/ihope/root/home/ihope/root/var/www/
01:54:54 <ais523> Normish is something special, alright
01:55:32 <kerlo> Heck, a better one: http://normish.org/home/ihope/root/var/www/ihope/root/var/www/root/var/www/
02:03:03 -!- zbrown has joined.
02:03:10 -!- zbrown has left (?).
02:09:30 <Asztal_> ais523: but where's your "frost pist!" achievement? :(
02:09:49 <ais523> I hope they don't add that, it'll make things even worse
02:10:04 <Asztal_> also, http://code.google.com/creative/cadie/ :)
02:11:39 <ais523> Asztal_: that code's nonportable
02:11:41 <kerlo> I want a frost pist.
02:11:55 <ais523> I wrote a fixed version, that works in both C-INTERCAL and CLC-INTERCAL
02:30:51 -!- Sgeo[College] has quit ("http://www.mibbit.com ajax IRC Client").
03:11:25 <ais523> by the way, I was reading the report on voting machine security flaws that the state of Ohio commissioned
03:11:30 <ais523> some of it's really ridiculous
03:11:50 <ais523> the flaws, I mean, not the report itself
03:13:05 <ais523> they managed an all-new sort of buffer overflow I've never seen before:
03:13:08 <ais523> 288 _stprintf(&name, _T("\\Storage Card\\%s"), findData.cFileName);
03:13:24 <ais523> (TCHAR's like char, it's a single-character type)
03:15:15 <ais523> that code is even worse than gets
03:15:23 <ais523> at least there are circumstances under which gets doesn't overflow...
03:15:53 <ais523> (if AnMaster were alive, he'd probably complain that _T infringes on implementation namespace or something like that...)
03:16:38 <ais523> also, it seems that that 0x0102030405060708 is accepted as the password on one type of machine they make, even if the password's been changed
03:28:19 -!- Sgeo has joined.
03:31:53 <oerjan> argh, the science subreddit is _still_ in AF mode
03:32:52 <Sgeo> Is CADIE dead, or is she just elsewhere? The video suggests dead, but the blog says elsewhere
03:33:25 <oerjan> http://code.google.com/creative/cadie/ still works
03:35:28 * Sgeo tried to make language accept INTERCAL
03:35:33 <Sgeo> The page just refreshed
03:36:01 <ais523> yep, it's stopped working
03:36:25 <Sgeo> You mean when you put INTERCAL in?
03:37:05 <ais523> you can enter queries but don't get replies
03:40:38 <ais523> with yet another typically INTERCAL-related response
03:41:25 <oerjan> bsmntbombdood: how succinct
03:43:09 <MizardX> Heh. Tried to add an Intercal-choice in the combo-box, but when I pressed "Ask CADIE", the page just refreshed and the form was reset.
03:44:47 <Sgeo> bsmntbombdood, it reminds me of some Runescape quest
03:57:37 -!- Slereah has joined.
04:11:14 <Sgeo> http://ytmnd.com/april/hehe_mix.swf
04:13:48 <kerlo> Proof that materials don't mean anything: there are stone glasses, and there are glass stones.
04:13:55 <kerlo> Though both are pretty weird.
04:15:26 <kerlo> I mean, you *could* make glasses out of stone.
04:31:54 -!- mmorrow has quit (Read error: 110 (Connection timed out)).
04:45:37 <lament> there're also marble stones and stone marbles, which is a little less obvious than it sounds at first
04:46:32 -!- neldoreth has quit (Read error: 113 (No route to host)).
04:49:32 -!- bsmntbombdood has set topic: topic ain't done changed since 2009-04-01 | http://tunes.org/~nef/logs/esoteric/?C=N;O=D.
04:56:15 -!- MizardX has quit (calvino.freenode.net irc.freenode.net).
04:56:16 -!- fungot has quit (calvino.freenode.net irc.freenode.net).
04:56:16 -!- fizzie has quit (calvino.freenode.net irc.freenode.net).
04:56:18 -!- lifthrasiir has quit (calvino.freenode.net irc.freenode.net).
04:56:18 -!- pikhq has quit (calvino.freenode.net irc.freenode.net).
04:56:20 -!- GregorR has quit (calvino.freenode.net irc.freenode.net).
04:56:20 -!- mtve has quit (calvino.freenode.net irc.freenode.net).
04:56:21 -!- Sgeo has quit (calvino.freenode.net irc.freenode.net).
04:56:21 -!- psygnisfive has quit (calvino.freenode.net irc.freenode.net).
04:56:23 -!- ais523 has quit (calvino.freenode.net irc.freenode.net).
04:56:23 -!- Asztal has quit (calvino.freenode.net irc.freenode.net).
04:56:28 -!- ineiros has quit (calvino.freenode.net irc.freenode.net).
04:56:28 -!- ehird has quit (calvino.freenode.net irc.freenode.net).
04:56:28 -!- Robdgreat has quit (calvino.freenode.net irc.freenode.net).
04:56:31 -!- AnMaster has quit (calvino.freenode.net irc.freenode.net).
04:56:33 -!- rodgort has quit (calvino.freenode.net irc.freenode.net).
04:56:33 -!- comex has quit (calvino.freenode.net irc.freenode.net).
04:56:34 -!- Dewi has quit (calvino.freenode.net irc.freenode.net).
04:56:34 -!- sebbu has quit (calvino.freenode.net irc.freenode.net).
04:56:35 -!- SimonRC has quit (calvino.freenode.net irc.freenode.net).
04:56:36 -!- Deewiant has quit (calvino.freenode.net irc.freenode.net).
04:56:38 -!- lament has quit (calvino.freenode.net irc.freenode.net).
04:56:38 -!- Ilari has quit (calvino.freenode.net irc.freenode.net).
04:56:39 -!- GreaseMonkey has quit (calvino.freenode.net irc.freenode.net).
04:56:39 -!- dbc has quit (calvino.freenode.net irc.freenode.net).
04:56:39 -!- Gracenotes has quit (calvino.freenode.net irc.freenode.net).
04:56:41 -!- Asztal_ has quit (calvino.freenode.net irc.freenode.net).
04:56:44 -!- Leonidas has quit (calvino.freenode.net irc.freenode.net).
04:56:44 -!- ski__ has quit (calvino.freenode.net irc.freenode.net).
04:58:23 -!- Sgeo has joined.
04:58:23 -!- psygnisfive has joined.
04:58:23 -!- sebbu has joined.
04:58:23 -!- GreaseMonkey has joined.
04:58:23 -!- MizardX has joined.
04:58:23 -!- Gracenotes has joined.
04:58:23 -!- ais523 has joined.
04:58:23 -!- dbc has joined.
04:58:23 -!- lifthrasiir has joined.
04:58:23 -!- pikhq has joined.
04:58:23 -!- fungot has joined.
04:58:23 -!- GregorR has joined.
04:58:23 -!- ehird has joined.
04:58:23 -!- AnMaster has joined.
04:58:23 -!- comex has joined.
04:58:23 -!- Asztal_ has joined.
04:58:23 -!- Leonidas has joined.
04:58:23 -!- ski__ has joined.
04:58:23 -!- ineiros has joined.
04:58:23 -!- SimonRC has joined.
04:58:23 -!- Robdgreat has joined.
04:58:23 -!- Dewi has joined.
04:58:23 -!- Deewiant has joined.
04:58:23 -!- fizzie has joined.
04:58:23 -!- rodgort has joined.
04:58:23 -!- Asztal has joined.
04:58:23 -!- mtve has joined.
04:58:49 -!- lament has joined.
04:58:49 -!- Ilari has joined.
05:54:51 -!- calamari has joined.
05:57:19 -!- neldoreth has joined.
06:16:18 -!- bsmntbombdood_ has joined.
06:16:41 -!- bsmntbombdood has quit (Read error: 110 (Connection timed out)).
06:27:19 -!- oerjan has quit ("leaving").
06:42:00 -!- Sgeo has quit ("Leaving").
06:47:54 -!- Gracenotes has quit ("Leaving").
06:49:42 -!- Gracenotes has joined.
07:18:51 -!- FireFly has joined.
07:59:59 -!- clog has quit (ended).
08:00:00 -!- clog has joined.
08:00:58 -!- cherez has joined.
08:01:04 -!- cherez has quit (Client Quit).
08:01:12 -!- cherez has joined.
08:01:26 -!- cherez has left (?).
08:03:53 -!- olsner has joined.
08:14:57 -!- tombom has joined.
08:21:43 -!- calamari has quit ("Leaving").
08:25:38 -!- FireFly has quit ("Later").
08:35:16 -!- GreaseMonkey has quit ("X-Chat -> http://xchat.org <- At least when I quit I don't look like a lamer").
08:42:18 <ais523> gah, Google took down their INTERCAL style guide
08:42:36 <ais523> also, the Google cache of it isn't showing, nor is the Yahoo cache, but I saved a copy from the live.com cache, of all places
08:42:45 <ais523> first time ever that live.com has been remotely useful
08:43:20 <ais523> I had the CADIE source already, luckily
08:52:35 <AnMaster> ais523, your ick repo should be up again
08:53:32 <ais523> how are your download logs for C-INTERCAL?
08:53:41 <ais523> if it's inconvenient to get at them, don't worry
08:53:53 <ais523> I'm just mildly curious as to how popular it is
08:56:06 <AnMaster> first: almost all from http://ipv4gate.sixxs.net
08:57:01 <AnMaster> then there is me, and one other persomn
08:57:14 <ais523> how many accesses altogether?
08:57:29 <AnMaster> /usr/www/data/log $ wc -l thttpd_log
08:57:51 <ais523> not bad for just a few hours after release
08:58:12 <ais523> the old MSIE 404 favicon.ico bug strikes again
08:58:17 <ais523> I think it's been fixed nowadays
08:58:43 <ais523> AnMaster: in older versions of IE, if favicon.ico's a 404, it ends up pinging it every time you access a page on the website
08:59:30 <AnMaster> both ick downloads are from that other person with native ipv6 btw
09:00:32 * AnMaster looks in whois (which is usually quite correct for ipv6 IME)
09:01:06 <ais523> hmm... not all that useful
09:01:14 <ais523> although knowing RIPE narrows it down to one continent
09:01:39 <AnMaster> the contact addresses are both in Amsterdam
09:02:03 <fizzie> Your logs should now have another native-IPv6 entry.
09:02:15 <ais523> only Dutch INTERCALer I know, and he's pretty active in terms of submitting patches
09:02:54 <fizzie> 2607:f0d0:3001:36::dead:baba, a rather curious address.
09:03:08 <ais523> fizzie: is that yours?
09:03:22 <ais523> or are you somehow reading AnMaster's logs?
09:03:22 <fizzie> ais523: No, it's that lepton.kuonet-ng.org.
09:03:26 <AnMaster> ais523, it is that of the vps. The person who runs it was a joker.
09:04:05 <fizzie> My hit in the logs should be from 2001:1bc8:102:587b:21d:7dff:fee4:a593 I presume.
09:04:33 * ais523 wonders if anyone has memorized their own IPv6 address
09:04:45 <ais523> the proxy I used to use at the University was 147.188.147.123
09:04:55 <AnMaster> fizzie, no I don't have another native one?
09:04:57 <ais523> but nowadays I normally use wireless, so I show up elsewhere
09:05:04 <AnMaster> fizzie, I had a new one from that one in Amsterdam though
09:07:14 <AnMaster> ais523, I remember mine. It is ::1
09:07:27 <AnMaster> (no I don't remember my full one)
09:09:05 <fizzie> A /64 prefix is relatively easy to remember, so if you put something at prefix::1, that's not too bad either. I used to remember my 6bone prefix. I don't remember this current one, though.
09:09:46 <ais523> it's somewhat interesting that IPv6 was initially only used to create long hostmasks that spelt out words on IRC
09:11:06 <AnMaster> it's odd that SixXS handed out a /48 when you requested a subnet. Presumably it is related to making routing simpler?
09:11:39 <ais523> using lots of subnets of the same size makes routing tables smaller and simpler
09:11:58 <fizzie> Yes, and they give you a /48 so that you can do some subnetting of your own.
09:12:02 -!- olsner has quit ("Leaving").
09:12:04 <ais523> also, a /48 is amazingly large
09:12:11 <ais523> a /64 would be the usual size for IPv6
09:12:11 <AnMaster> fizzie, can't you subnet at /96 or such?
09:12:19 <ais523> anything smaller than that annoys the routers
09:12:24 <fizzie> You can't really make smaller nets than /64.
09:12:35 <ais523> so if you want to subnet something, it should be bigger than /64
09:12:43 <fizzie> The address format specifies that the last 64 bits is the node ID.
09:13:03 <fizzie> And all those stateless-autoconfig things generate EUI-64 format addresses, too, which need a 64-bit network.
09:13:19 <fizzie> I've seen a tunnel provider handing out /60s, though, which means you have 16 subnets to play with.
09:14:08 <fizzie> And if you configure things manually, at least Linux is completely happy to use smaller-than-/64-networks, it's just not pedantically-speaking correct.
09:14:11 <AnMaster> for a /48 you would have 65536 subnets right?
09:14:28 <AnMaster> assuming 2 16^p is the correct way to calculate it
09:14:45 <ais523> well, probably 65534 in practice
09:14:52 <ais523> because IIRC max and min addresses have special meanings
09:15:25 <fizzie> It's max and min addresses in a network, you can still use things like 192.168.0.5, even though there's that 0 byte in there.
09:16:20 <ais523> but would 192.168.0.0 be the min address of 129.168.0.0/16 or 129.1268.0.0/24?
09:16:23 <fizzie> So you can just as well use the 2001:1234:1234:0::/64 subnet.
09:17:00 <fizzie> Obviously it's the min address of both, and it depends on configured netmasks and such what it means.
09:17:28 <AnMaster> what does the min address mean then?
09:17:37 <AnMaster> max is multicast or something isn't it?
09:17:46 <fizzie> It's traditionally "the address of the network, not a host in it".
09:17:54 <fizzie> While all-bits-one is the broadcast address, yes.
09:18:15 <ais523> in theory, sending a packet to 255.255.255.255 sends it to the entire Internet
09:18:19 <AnMaster> what about multicast then? Isn't that "broadcast, to some hosts"
09:18:21 <ais523> in practice, any sane router will drop it
09:18:29 <fizzie> Multicast uses a separate range of addresses.
09:18:35 <fizzie> To identify multicast groups.
09:18:45 <AnMaster> fizzie, mhm and how do you mark such groups?
09:19:11 <fizzie> What do you mean, exactly?
09:19:25 <AnMaster> fizzie, no idea. I hate networking
09:19:55 <AnMaster> "<fizzie> To identify multicast groups." <-- well, how do you identify if a computer is in a group or not
09:20:10 <fizzie> There's the IGMP protocol for that.
09:20:16 <fizzie> Does group-registration and such.
09:21:54 <fizzie> Although in IPv6 the multicast stuff has been taken in ICMPv6. As was ARP.
09:22:30 <fizzie> All IPv6 multicast addresses are in the ff00::/8 block, while the IPv4 range goes from 224.0.0.0-239.255.255.255.
09:23:03 <fizzie> I guess that's 224.0.0.0/4 if I count the bits right.
09:23:44 <AnMaster> fe80:: is "link local" or something like that right? But what is the use of it?
09:25:04 <fizzie> Well, for link-local purposes. Like router-discovery and such, I think those tend to use the link-local addresses at some point.
09:25:26 <AnMaster> fizzie, this ipv6 over facebook thing hm... the page *is* still there http://www.facebook.com/apps/application.php?id=59502659200 ...
09:25:49 <fizzie> Sure, and if the statistics are correct, the app has been there for a while.
09:25:54 <ais523> it's one of those April Fool's RFCs which probably actually works, just is insane
09:26:01 <ais523> fizzie: they were testing it before releasing the RFCs
09:26:26 <fizzie> There's also the fec0::/10 prefix for "site-local" v6 addresses, presumably for rather similar use than the current private-use non-routable IPv4 addresses.
09:27:24 <AnMaster> http://www.facebook.com/topic.php?uid=59502659200&topic=6840 <--- ?
10:04:33 -!- ais523 has quit (Read error: 110 (Connection timed out)).
10:14:08 -!- neldoret1 has joined.
10:15:40 -!- neldoret1 has quit (Client Quit).
10:16:51 -!- neldoret1 has joined.
10:17:43 -!- sebbu2 has joined.
10:21:30 -!- neldoret1 has quit (Client Quit).
10:21:42 -!- neldoret1 has joined.
10:25:15 -!- neldoreth has quit (Read error: 110 (Connection timed out)).
10:34:47 -!- sebbu has quit (Read error: 110 (Connection timed out)).
13:27:10 <AnMaster> btw wikipedia lists http://tools.ietf.org/html/rfc5513 too as a joke rfc this year (yes it lists the facebook one too)
14:33:15 -!- Gracenotes has quit (Remote closed the connection).
14:34:23 -!- FireFly has joined.
14:39:58 -!- Judofyr has joined.
15:17:15 -!- sebbu2 has quit (Read error: 104 (Connection reset by peer)).
15:17:43 -!- sebbu has joined.
15:45:08 -!- BeholdMyGlory has joined.
15:56:27 -!- jix has joined.
16:01:05 -!- Judofyr has quit (Remote closed the connection).
16:10:00 -!- oerjan has joined.
16:27:05 -!- BeholdMyGlory has changed nick to FireFly-.
16:27:15 -!- FireFly- has changed nick to BeholdMyGlory.
16:36:14 -!- Judofyr has joined.
16:41:17 <ehird> anyone know an ipv6 isp in uk?
16:44:23 <fizzie> http://www.sixxs.net/faq/connectivity/?faq=native&country=gb
16:44:56 <ehird> Bogons or Goscomb!
16:45:50 * oerjan expects the spanish vogons
16:46:05 <fizzie> I have no idea how comprehensive or accurate that list is; I know one (1) ISP in Finland who does native IPv6 to customers, and that's Nebula, and coincidentally it's also the only thing listed under the Finnish flag there.
16:46:36 <ehird> fizzie: and you're on nebula?
16:47:03 <ehird> anyway, these uk isps seem very, very obscure
16:47:05 <fizzie> Nebula were polite enough to point the reverse-DNS zone for the /64 they've assigned to me to any DNS server I wanted; that's some good service. The network in the student apartments of my university also does native IPv6, but there are no reverse-DNS entries there.
16:47:10 <ehird> How many IP addresses?
16:47:10 <ehird> Normally a company would received 1208925819614629174706176 addresses to cover up to 65536 sites.
16:47:14 <ehird> that's rather gung-ho
16:47:35 <ehird> fizzie: nice; Orange are in the business of assraping their customers with restrictions
16:48:02 <ehird> thing that irritates me:
16:48:08 <ehird> calling hosting services isps
16:48:11 <ehird> yes, it's technically correct
16:48:14 <ehird> no it is not helpful!
16:48:36 <ehird> Broadband services from Andrews & Arnold Ltd are aimed at the technical and professional customer. We cater for home users right up to large corporate users. Most of our customers are reasonably technical or purchase via an IT consultant/dealer. The services we offer are not the fast food of broadband - we have a range of specialist options including blocks of IP addresses at no extra cost, reverse DNS delegation, IPv6 services, bonded lines with fast fa
16:48:38 <ehird> llback, and much more. We also pride ourselves in having technically competent support staff based in our UK offices.
16:48:41 <ehird> yarr, this sounds hot
16:48:44 <oerjan> which is strange since norway was listed in a wp article as having among the top relative penetrators
16:48:57 <ehird> Our Professional service starts from as little as £20.50 pcm† (£17.83+VAT) including 2GB peak download allowance, so can easily be used by more demanding home users. Professional customers can pick a tariff that matches their needs. The professional service includes a free ADSL2+ ethernet router, fixed IP address blocks, IPv6, a UK domain, web space, and IMAP email (with spam and virus checking).
16:49:06 <ehird> well 2gb download allowance is shite
16:49:11 <ehird> and I don't need the domain/webspace/email
16:49:23 <ehird> Our broadband services are provided using ADSL2+ were possible which allows up to 24Mb/s downlink and up to around 1Mb/s uplink. When not possible we provide using ADSL1 which allows up to 8.128Mb/s downlink and up to 448Kb/s uplink (832Kb/s with the premium option). These speeds are the maximum line rate and depend on the line legth and quality. The availability checker can give an indication of likely downlink speed. We upgrade ADSL1 lines to ADSL2+ wh
16:49:25 <ehird> en possible as part of the 21st Century Network upgrade. Details
16:49:40 <ehird> okay, so if only there wasn't a 2gb download allowance that'd be awesome
16:49:56 <ehird> Peak period download per month.
16:49:56 <ehird> Peak period is 9am-6pm Mon-Fri.
16:50:01 <ehird> Off Peak allowance100GB
16:50:04 <ehird> well that's stupid
16:50:14 <ehird> so 9am-6pm i'm only allowed to d/l 2gb.
16:50:24 <ehird> All other times except night (2am-6am) which is unmetered.
16:50:27 <ehird> this is kind of fucked up
16:50:55 <ehird> i mean, it has everything, i just wish it didn't have the stupid bw limit
16:51:37 <ehird> "A byte is 8 bits of data which is normally one character. A gigabyte or GB is 1,000,000,000 bytes (not to be confused with a Gibibyte or GiB which is 1,073,741,824 bytes)"
16:51:43 <ehird> you're full of shit, andrews & arnold.
16:51:58 <Slereah> Well, technically, it's correct
16:52:04 <Slereah> It's just that no one uses gibi
16:52:35 <fizzie> Gibibibibii! Sounds like a pokey-mon or something.
16:52:47 <ehird> but seriously they give you a decent router and ipv6 and whatnot and the speed looks great
16:52:50 <oerjan> with a terrifying gibibite
16:52:56 <ehird> but 2gb peak allowance?
16:53:03 <ehird> that's like... giving you a rocket and telling you to only fly at 3mph
16:53:26 <ehird> http://www.bogons.net/
16:53:35 <ehird> I SHALL LOOK UP THE "BOGONS"
16:53:41 <oerjan> i suspect pokemons don't bite though, would not be family friendly
16:54:26 <ehird> oerjan: err, they knock each other out and seriously injure them with massive electric attacks and shit
16:54:29 <ehird> I think biting is pretty tame
16:54:33 <Slereah> But it's been... ten years since I played
16:54:40 <ehird> oh you mean the game
16:54:45 <ehird> some have a bite attack
16:54:49 <ehird> I was talking about the shitty show
16:54:51 <ehird> "* Static IP address if required "
16:55:07 <ehird> "* Multicast and/or IPv6 routing over ADSL "
16:55:14 <ehird> "* No enforced transfer cap (subject to reasonable usage) "
16:55:20 <Slereah> I seem to recall biting in the movie
16:55:35 <ehird> ok, bogons sounds fuckin' awesome
16:56:11 <oerjan> ehird: just beware of the clause where they reserve the right to blow up the earth for their information superhighway
16:56:19 <ehird> http://www.enta.net/Customer_Products/Data_&_Connectivity_Products/Broadband/ ← yawn, no thx
16:57:09 <ehird> Do you restrict access to any Bit torrents/news sites?
16:57:10 <ehird> IDNet do not restrict access to any sites or ports etc.
16:57:23 <ehird> the fact that they explicitly state that is nice
16:57:40 <ehird> I might email the v^Hbogons to ask if they filterz
16:58:40 <ehird> idnet home supermax http://www.idnet.net/solutions/home/broadband/
16:58:49 <ehird> vs bogons adsl http://www.bogons.net/
16:58:56 -!- MigoMipo has joined.
16:59:50 <ehird> random opinions welcome, I'd like to escape this shithole of an esp
17:17:37 -!- oerjan has quit ("My esp is rather lousy, too").
17:26:21 -!- jix_ has joined.
17:32:18 -!- tombom has quit ("Peace and Protection 4.22.2").
17:38:09 -!- jix has quit (Read error: 113 (No route to host)).
17:43:55 <ehird> http://redsnake.me/
17:52:38 <ehird> Bogons vs idnet is looking good right now.
18:04:15 <ehird> <!--> <font size="-2">Did you know that <b>your Browser is <em>buggy</em></b>: it can't parse comments correctly.</font> -->
18:37:04 <Asztal_> <!-- -- --->ERROR<!- ------ >
18:39:09 -!- kar8nga has joined.
18:39:16 <Asztal_> the ERROR part may actually be commented, depending on whether or not you live in 1999 or 2005 or elsewhen
18:48:46 <ehird> PRETEND TO SWIM WHY DON'T YOU JIM
18:59:52 -!- tombom has joined.
19:23:59 -!- MigoMipo has left (?).
19:31:49 -!- kar8nga has quit (Remote closed the connection).
19:33:15 -!- Sgeo has joined.
20:31:11 <ehird> I think I just solved syntax.
20:31:32 -!- olsner has joined.
20:35:35 -!- psygnisfive has quit ("Leaving...").
20:42:16 <kerlo> The syntax of what?
21:03:14 -!- Gracenotes has joined.
21:09:41 -!- nooga has joined.
21:12:08 -!- neldoret1 has quit (Read error: 113 (No route to host)).
22:00:46 -!- BeholdMyGlory has quit (Remote closed the connection).
22:18:59 -!- nooga has quit (Read error: 104 (Connection reset by peer)).
22:20:16 <ehird> fizzie: Mr ip wizard, "/29 or /30" will mean ipv4 rite?
22:21:04 <fizzie> Most likely, yes. Those are pretty small blocks.
22:21:31 <ehird> Yar. I'm deciding between http://bogons.net/ and http://idnet.net/, both of which look more or less perfect.
22:22:09 -!- tombom has quit ("Peace and Protection 4.22.2").
22:22:10 <fizzie> Especially since generally the all-bits-0 and all-bits-1 nodes are kept as the network address and the broadcast address, respectively, so a /30 would have just two usable addresses.
22:22:56 <ehird> (Enta) IPStream Max (up to 8M)100kbps448kbps£60.00£35.00
22:23:03 <ehird> That 100kbps second is the download spede column.
22:23:07 <ehird> Yet it says up to 8M?
22:23:51 <fizzie> I think our student housing network had a costs-a-bit add-on service to get your own non-DHCP-allocated /29 (so six hosts) block.
22:24:19 <Deewiant> Well, they do have a /16 of their own.
22:24:35 <ehird> Bogons give you a static ipv6 address with yer own reverse dns so that's just fiiine by m
22:24:37 <Deewiant> (Assuming you're referring to TKY.)
22:24:50 <fizzie> Deewiant: You mean TKY? The old Trinet block at least was just a /20.
22:25:13 <fizzie> I don't really know about the new one. TKK has a /16 block of its own.
22:25:38 <fizzie> ehird: The student union. Acronym of the finnish name.
22:25:40 <ehird> I got a personal reply from Bogons with techy details, which was nice. Delivered via RT too.
22:26:16 <fizzie> I had a map somewhere...
22:26:22 <ehird> say, about that dead:baba ipv6 address - is there liek a service to get a custom one you specify or sth?
22:26:24 <ehird> like license plates
22:26:42 <fizzie> If you have your own network, you can number the nodes any way you like.
22:27:18 <fizzie> Although the stateless autoconfig stuff uses automatically generated addresses that are <first 3 bytes of Ethernet MAC>-ff:fe-<last 3 bytes of MAC>.
22:27:23 <ehird> The good ol' I Have Gratuitous Amounts Of Money option.
22:28:09 <ehird> Why don't OSes come with a dns resolver?
22:28:14 <ehird> I don't see why it's done remotely.
22:29:09 <fizzie> Might be to cut down traffic a bit, so that just the ISP's ns has to wonder where google.com is.
22:30:05 <ehird> fizzie: Well, sure, but it seems like it'd be much, much faster.
22:30:21 <ehird> I doubt it'd be hard to handle the traffic considering how powerful networks are nowadays
22:30:41 <fizzie> The root servers would have a lot more work to do; I think they have quite a lot as-it-is.
22:31:05 <ehird> fizzie: i wonder what the root servers run on
22:31:10 <ehird> probably a mega-optimized bsd
22:31:21 <fizzie> Not sure about speeds, either, since now you can just ask a very nearby server, and it probably knows the answers already, since everyone else's asking about the same things; you don't have to start from the root.
22:31:35 <ehird> fizzie: after using it a while it should be a lot faster, no?
22:31:41 <ehird> since cache retrieval would be instant
22:32:11 <fizzie> Well, yes, caching would be a good idea. I have my local BIND configured so that it forwards all queries to the ISP's server if it doesn't know the answer offhand.
22:32:19 <ehird> http://www.root-servers.org/ Poor greenland.
22:32:26 <fizzie> Deewiant: Oh, I foundeded the (IPv4) map: http://zem.fi/trinet/trinetmap.html
22:32:49 <Deewiant> dig reports around 7 milliseconds on average
22:33:32 <Deewiant> Lot of addresses for those houses
22:33:36 <fizzie> Deewiant: That might be obsoleted already a bit, since I did that back when they updated the network. They might have changed stuff.
22:33:48 <ehird> "2. As root, create UNIX accounts named Gdnscache and Gdnslog. "
22:33:54 <ehird> oh fuck, time for djb to lecture me about how to run a fuckin' unix system
22:34:34 * ehird tries to use local djbdns
22:34:59 <ehird> Hmm, root lookups are quite slow aren't they?
22:35:03 <ehird> As in 30 seconds and still going.
22:35:26 <ehird> Guess I fucked up.
22:35:35 <fizzie> Dnsmasq is probably the cheapest choice if you want a caching-only local DNS thing. (Although it comes integrated with a DHCP server, which might be pretty useless.)
22:36:10 <fizzie> I think I've seen dnsmasq used in some of those Linux-based routery devices too.
22:36:12 <ehird> "8. Set up a public web page saying that your DNS cache is powered by djbdns, so that a Google search for powered djbdns will find your page in a few months. These public statements will encourage other people to deploy djbdns, provide djbdns support services, and develop djbdns-related tools. Please also consider making a donation to the Bernstein Writing Fund."
22:36:18 <ehird> Go fuck yourself, bernstein
22:36:34 <Deewiant> What's wrong with asking people to help?
22:36:49 <ehird> Deewiant: I just like insulting Bernstein because he's pretentious
22:37:10 <ehird> "To install djbdns, ..." GO TO HELL BERNSTEIN
22:38:07 <Deewiant> Give me an obscure domain to lookup, I want something that takes longer than 5ms
22:38:24 <ehird> Deewiant: eso-std.org.
22:38:37 <ehird> That's domain squatters for you
22:38:43 <ehird> Deewiant: agoranomic.org
22:39:00 <Deewiant> I wonder if anything breaks 200
22:39:42 <ehird> Are you gay? Are you a nigger? etc.
22:40:38 <ehird> Deewiant: what about 2ch.net
22:40:44 <ehird> in japan everything is slow.
22:41:05 <Deewiant> I'm trying to think of something .cn and failing
22:41:55 <Deewiant> Is India slower or faster, I wonder
22:42:07 -!- bsmntbombdood_ has quit (Read error: 113 (No route to host)).
22:42:47 -!- bsmntbombdood_ has joined.
22:42:58 <Deewiant> foo.in was below 100, bar.in got no answer and took 850
22:43:04 <fizzie> If you're curious, you can often "dig @server -c ch -t txt version.bind" to find out what a DNS server's running; often even if it's not BIND. Although a lot of people do disable that feature, too.
22:43:29 <Deewiant> version.bind.0CHTXT"This space intentionally left blank"
22:43:50 <Deewiant> version.bind.0CHTXT"NSD 2.3.7"
22:44:39 <ehird> % sudo svstat /opt/local/var/svscan/service/dnscache
22:44:39 <ehird> /opt/local/var/svscan/service/dnscache: up (pid 97416) 15 seconds
22:44:49 <ehird> Deewiant: Any root servers running djbdns?
22:45:08 <Deewiant> Maybe one of these sneaky ones :-P
22:45:17 <ehird> they're so ashamed
22:45:30 <ehird> % dnsq a www.aol.com 127.0.0.1
22:45:33 <ehird> Let's see if it works this time
22:45:45 <ehird> Dayum this is slow.
22:45:51 -!- FireFly has quit ("Later").
22:46:04 <ehird> ok, it doesn't seem to be _doing_ anything
22:46:41 <ehird> ' and if your DHCP client cannot be configured to discard external DNS cache information'
22:46:45 <Deewiant> Most appear to run bind (9.2 through 9.4), two NSD (2.3.7 and 3.0.5), one VGRS4
22:47:20 <Deewiant> One blank, one times out for that request, one with that email address, one "intentionally left blank"
22:48:53 <fizzie> I wouldn't be too surprised if djbdns came with the version info disabled by default.
22:49:25 <ehird> How irritating, I wonder how to ix this
22:50:31 <fizzie> It's a funny query; the "-c ch" means dns record class Chaos, for Chaosnet; normally just about any request is for class IN (Internet).
22:51:22 <ehird> fizzie: what's a funny query?
22:51:31 <fizzie> The version.bind. one.
22:52:18 <fizzie> After all, they can't really use an Internet DNS name of "version.bind", I'm sure we'll end up with a .bind TLD sooner or later. Using a Chaosnet name is rather cleverness.
22:53:33 <ehird> How did you set up your local bind, fizzie?
22:53:40 <ehird> Wanna figure out what I did wrong.
22:55:20 <fizzie> Well, mostly I just have: acl "local" { local; networks; here; }; ... options { ... listen-on { all; private; IPv4; addrs; }; allow-query { "local"; } recursion yes; forwarders { isp's; dns; server; ips; }; ... }; -- and that's about it.
22:55:32 <ehird> fizzie: I didn't mean how you set up the bind in particular
22:55:37 <ehird> I meant how did you then test it
22:55:44 <ehird> Because I'm just getting timeouts
22:56:33 <fizzie> Can you do something like "dig @ns.google.com a google.com" and get results on the box you're running the name server on?
22:56:53 <fizzie> If not, it might be that your friendly ISP is filtering outgoing DNS that's not directed to their nameserver. :p
22:57:28 <ehird> So my ISP is sane in that respect.
22:57:42 <ehird> Then % dig @127.0.0.1 a google.com... and... silence.
22:57:52 <ehird> ;; connection timed out; no servers could be reached
22:58:12 <AnMaster> ehird, why are you setting up a local bind?
22:58:19 <ehird> AnMaster: a local djbdns.
22:58:21 <ehird> and because I want to.
22:58:32 <AnMaster> You said "too much work" and such before
22:58:41 <ehird> I do things that require too much work when I am bored.
22:58:57 <fizzie> That's rather strange. Maybe you could "netstat -nlp" or something to check whether your djbdns is actually listening?
22:59:04 <ehird> AnMaster: djbdns's dnscache server
22:59:16 <AnMaster> ehird, I would use dnsmasq for that. Or bind.
22:59:20 <ehird> fizzie: what would the protocol name be?
22:59:24 <ehird> AnMaster: what's wrong with using dnscache?
22:59:59 <ehird> fizzie: er, can you tell I've never used netstat?
23:00:02 <AnMaster> just dnsmasq almost works out of box (read through a config and edit an option or two, start daemon
23:00:06 <fizzie> The /etc/services name here is "domain".
23:00:22 <fizzie> ehird: Of course if this is OS X netstat, it'll have a different syntax.
23:00:23 <AnMaster> <ehird> fizzie: er, can you tell I've never used netstat? <-- :D
23:00:34 <ehird> Usage:netstat [-AaLlnW] [-f address_family | -p protocol]
23:00:34 <ehird> netstat [-gilns] [-f address_family]
23:00:43 <ehird> Well, I don't see why that wouldn't work.
23:00:48 <ehird> I could just nmap myself. :-D
23:01:00 <fizzie> Everyone's netstat seems to be rather different. Just "netstat -nl" is probably enough.
23:01:02 <ehird> no command, AnMaster
23:01:19 <ehird> Discovered open port 53/tcp on 127.0.0.1
23:01:24 <ehird> PORT STATE SERVICE
23:01:24 <ehird> 53/tcp open domain
23:01:31 <AnMaster> ehird, oh? It is bshish though, so I thought it would make sense on OS X
23:01:32 <fizzie> The "-nlp" was for Linux's netstat, for which "-p" means "show PID and program name for whatever's there".
23:02:21 <fizzie> Anyway, that sure sounds like it's listening; strange that it doesn't want to talk to you. Maybe start it with enough debugging and verbosity flags, then.
23:02:56 <ehird> Initiating Service scan at 23:02
23:02:56 <ehird> Scanning 2 services on localhost (127.0.0.1)
23:03:15 <ehird> fizzie: Maybe it's talking to me but it can't find anything.
23:04:03 <fizzie> Well, that's possible, of course. Although since you do DNS queries yourself, I see no reason why dnscache couldn't.
23:04:22 <ehird> Stats: 0:01:02 elapsed; 0 hosts completed (1 up), 1 undergoing Service Scan
23:04:22 <ehird> Service scan Timing: About 50.00% done; ETC: 23:03 (0:00:11 remaining)
23:04:25 <ehird> Something is horribly wrong here.
23:04:36 <ehird> PORT STATE SERVICE VERSION
23:04:37 <ehird> 53/tcp open domain?
23:05:05 <ehird> % telnet 127.0.0.1 53
23:05:05 <ehird> Trying 127.0.0.1...
23:05:06 <ehird> Connected to localhost.
23:05:08 <ehird> Escape character is '^]'.
23:05:17 <ehird> I might need to learn dns.
23:05:33 <ehird> I'm sure 'look up google.com' can't be hard.
23:05:51 <fizzie> Well, you can talk DNS over TCP too.
23:05:58 <ehird> Well, no I can't :-)
23:06:01 <fizzie> Normally it's done only for zone transfers, though.
23:06:17 <fizzie> I wonder if they're still doing that DNS-based DDOSing that's been very popular lately. At least my DROP rule for that has seen 298K packets since I last reloaded the iptables rules.
23:07:26 <AnMaster> fizzie, zone transfers in djbdns are different iirc
23:07:48 <fizzie> Yes, I seem to recall too that it did something different.
23:08:33 <ehird> Anyone know how to speak DNS?
23:08:51 <fizzie> It's a rather binary protocol, not too comfortable to speak manually.
23:12:03 <ehird> Well, I can connect.
23:12:05 <ehird> Which is imporatnt.
23:12:22 <fizzie> With a normal program you could just run it with some sort of "don't daemonize, absurdly verbose output" command line options to see what's happening, but I have no idea how to do something like that to dnscache.
23:13:03 <ehird> It uses djb's wacky daemon runner thing.
23:13:21 <fizzie> Yes. Well, there should be some sort of log file. Wanna bet it's empty?
23:14:05 <ehird> It just says it started up.
23:14:36 <ehird> @4000000049d531bd380b501c starting
23:14:37 <ehird> @4000000049d535b439b2b2d4 tcpopen 00000000000000000000ffff7f000001:cd55
23:14:39 <ehird> @4000000049d535b439c3b64c tcpclose 00000000000000000000ffff7f000001:cd55 connection reset
23:14:43 <ehird> djbdns, you doofus.
23:14:46 <ehird> I don't have ipv6.
23:15:08 * ehird checks dnscache/root/servers/@
23:15:14 <fizzie> That looks like a ipv6-mapped ipv4-address, though; ::ffff:127.0.0.1, to be exact.
23:15:14 <ehird> Hmm, it's just ipv4 addresses
23:15:19 <AnMaster> 00000000000000000000ffff7f000001?
23:15:40 <ehird> Right, it's just me.
23:15:54 <AnMaster> ehird, @4000000049d535b439b2b2d4 is a timestamp
23:15:55 <ehird> Okay, so we're having TCP acknowledgment.
23:16:10 <ehird> Can you make dig, uh,
23:16:15 <fizzie> Some dual-stacked systems do so that if you listen on the wildcard ipv6 socket, you get also ipv4 connections using that funky mapped format.
23:16:44 <fizzie> It's a +tcp somewhere in the command line.
23:16:56 <ehird> +[no]vc (TCP mode)
23:16:56 <ehird> +[no]tcp (TCP mode, alternate syntax)
23:17:04 <ehird> % dig @127.0.0.1 -4 +vc google.com
23:17:07 <ehird> We have successory.
23:17:13 <ehird> ;; Query time: 1103 msec
23:17:21 <ehird> ;; Query time: 1532 msec
23:17:25 <ehird> It is the slow at first.
23:17:39 <ehird> OK, it's tcp vs udp.
23:17:47 <ehird> So something's up with udp port 56.
23:17:53 <ehird> The question is, what.
23:18:22 <ehird> This is really odd.
23:18:28 <ehird> I wonder what's up with it.
23:18:32 <fizzie> Yes. Your system is the odd.
23:19:25 <ehird> Yes, Linux is wooing me atm.
23:19:59 <fizzie> Hm, that DNS-DDOS doesn't seem *so* popular any more; I reset those counters more than 15 minutes ago, and haven't seen a single incoming packet.
23:20:30 <ehird> I'm not too sure I like this 100msec initial lookup thing.
23:20:51 <ehird> Heh, my router doesn't seem to do tcpy dns.
23:20:54 <ehird> % dig +vc en.wikipedia.org [hang]
23:21:26 <ehird> fizzie: Got an obscure domoniker?
23:26:16 <ehird> Anyhooways, it is annoying having a choice between two equals,.
23:45:09 <ehird> "Brandon Butterworth is a Principal Technologist in the BBC's research and development team and the man who first registered the bbc.co.uk domain"
23:45:21 <ehird> That's the guy that responded to the question I asked bogons.net
23:46:02 <ehird> AnMaster: http://www.bogons.net/ resolves for me.
23:46:15 <ehird> bogons.net doesn't resolve
23:46:24 <ehird> AnMaster: no, it'll be how they've set up their servers
23:46:27 <ehird> since they're an isp
23:46:29 <ehird> a separate webserver
23:46:43 <ehird> it's annoying but understandable
23:46:56 <AnMaster> ehird, why have it not resolve
23:47:03 <ehird> because it didn't cross their minds?
23:47:04 <AnMaster> they could do a CNAME or A record
23:47:06 <ehird> anyway they're one of the isps i'm considering
23:47:18 <ehird> they do ipv6, don't have bandwidth limits, don't shape traffic
23:47:22 <ehird> and give you a static ip for free
23:47:42 <ehird> I asked whether they shaped traffic and got a technical response from the guy who first registered the BBC's domain.
23:47:46 <ehird> So that is pretty awesome.
23:48:22 <AnMaster> "We have over 15 years experience of the Internet - dating from the start of the Internet in the UK in 1989. If you feel that your clue level isn't quite high enough to do-it-yourself yet,"
23:49:05 <ehird> the other isp i'm considering is http://www.idnet.net/, but I'm leaning more to bogons now for the techness and smallness.
23:49:19 <ehird> AnMaster: are you sure xs4all does ipv6?
23:49:26 <ehird> I suppose they might, I'm just not aware of it
23:49:53 <AnMaster> ehird, well they do newszilla6.xs4all.nl (free for sixxs customers too!)
23:50:03 <AnMaster> somehow I then doubt they don't do ipv6 elsewhere...
23:50:24 <ehird> I wonder if bogons provides usenet
23:51:53 <ehird> AnMaster: heh, bogons is hard to google
23:51:56 <ehird> it's an internet-related term
23:52:01 <ehird> http://en.wikipedia.org/wiki/Bogon_filtering
23:52:56 <ehird> they charge £10 more a month for 800Kbps upload
23:53:03 <ehird> Surely it can't be that different?
23:54:05 <ehird> I like how bogons don't bundle a shitrouter though
23:54:10 <ehird> I can get a linksys and put openwrt on it
23:54:44 <AnMaster> ehird, I'd prefer if they bundled a Cisco for those £10
23:54:49 -!- oerjan has joined.
23:55:00 <ehird> I've always heard recommendations for linksys
23:55:21 <AnMaster> ehird, high end? More expensive?
23:55:32 <ehird> cisco owns linksys
23:55:47 <ehird> linksys.com redirects to http://www.linksysbycisco.com/
23:56:12 <ehird> That'd be a pain to get working... vs something like openwrt
23:56:13 <AnMaster> ehird, my router. A pentium 3, running openbsd
23:56:20 <ehird> AnMaster: That's not a "router".
23:56:41 -!- olsner has quit ("Leaving").
23:56:58 <ehird> how do you plug it into the phoneline?
23:57:05 <ehird> I don't recall what cable it is, I'll go check
23:57:16 <AnMaster> switch - openbsd router - adsl modem - phone
23:57:16 <ehird> Wait, it's just ethernet isn't it?
23:57:30 <ehird> AnMaster: most modern routers have an adsl modem built in
23:57:35 <AnMaster> yes ethernet all the way except modem-phone which is special cable
23:58:09 <ehird> http://www.linksysbycisco.com/UK/en/products/Routers ← I like how they're all oh-so-stylish up to the WAP54G, when they turn into megauglys.
23:59:36 <ehird> http://www.linksysbycisco.com/UK/en/products/WRT54GL
23:59:41 <ehird> Compatibility: two windows logos
23:59:53 <ehird> Minimum Requirements
23:59:53 <ehird> Internet Explorer 6.0 or Firefox 1.0 or Higher for Web-based coniguration
23:59:57 <ehird> Windows 2000, XP, or Vista
23:59:59 <ehird> Network Adapteror Wireless Network Adapter