00:02:20 -!- sebbu has quit (Ping timeout: 246 seconds).
00:05:29 -!- ZombieAlive has joined.
00:10:31 -!- augur_ has changed nick to augur.
00:20:13 -!- sebbu has joined.
00:20:51 -!- sebbu has quit (Changing host).
00:20:51 -!- sebbu has joined.
00:55:59 <fizzie> Hmm. I wonder what all these are: http://sprunge.us/Ghhc
00:56:45 <fizzie> Or maybe "why" is more appropriate -- "what" is easier to answer: DNS queries, with the destination address referring to a computer that's currently offline.
00:57:46 <fizzie> All the domains (or at least wudi128.com, wudi999.com and wdcp3.com) are registered to a "Wang KangJun" at "QiongShanQuJiuZhouZhen, HaiKouShi, HuNanSheng", China.
00:59:18 -!- sebbu has quit (Ping timeout: 245 seconds).
01:00:54 <oren> haikou city hunan province
01:01:27 <fizzie> Some Googling located vague references to a paper called "DNS for Massive-Scale Command and Control", but nothing very clear.
01:05:14 <oren> qiongshan is a district of haikou
01:29:52 <oren> Anyway what about the source ip's? Maybe they're from a botnet?
01:30:59 <fizzie> The source addresses are all unique.
01:31:54 <fizzie> I logged 1943 of these things over 5 minutes.
01:32:03 <fizzie> From http://sprunge.us/YcfN
01:33:39 -!- Phantom_Hoover has quit (Read error: Connection reset by peer).
01:33:47 <fizzie> I'm just puzzled as to why they are addressed to this destination, given that the computer it's pointing up hasn't been online for quite a while, and last I did my random "wonder what the traffic weather looks like" peek, nothing like this was going on.
01:33:59 <oren> The website itself presents a username field and a captcha, along with a link to download a client program
01:34:05 <fizzie> Well, that sounds botnetty enough.
01:34:57 <oren> All the given links pointed to identical websites, in fact
01:35:11 -!- sebbu has joined.
01:35:48 -!- sebbu has quit (Changing host).
01:35:49 -!- sebbu has joined.
01:36:42 <fizzie> For completeness, here's also counts of the suffixes I saw: http://sprunge.us/JcDf
01:40:14 <oren> The javascript on the page, after login success, redirects the user to
01:40:18 <oren> http://tb.53kf.com/webCompany.php?arg=10095449&style=1
01:42:51 <quintopia> a chat room in chinese? must be a command and control room?
01:47:12 -!- ZombieAlive has quit (Remote host closed the connection).
01:47:21 <oren> Well lets see if I can make an account
01:47:47 -!- ZombieAlive has joined.
01:50:28 <quintopia> are you doing this from a VM? be careful man.
01:52:11 <oren> Hmm seems the chat room is the customer support, not the service itself
01:54:03 <oren> Now lesse what's in this RAR?
01:54:37 <oren> (the "client Program")
01:55:07 <oerjan> R.I.P. oren's computer
01:55:08 <oren> If it's an actual program I'll run in on a AWS windows instance
01:56:39 <oren> WDGameSetup.msi
01:57:03 <fizzie> Hm. The copy I got was over 20 MB.
01:57:30 <oren> Ok firefox is lying to me, ls says 23 MB
01:58:07 <fizzie> I gave it to a random upload-files-and-get-virus-scan-reports-from-N-AV-systems web service, but it wasn't really flagged as known malware. Although one (out of 57) said it matches some test that sounded like a generic keylogger heuristic.
01:58:27 <fizzie> It's quite possible that the software is not related to the DNS traffic at all, of course.
01:58:50 <oren> signing into AWS so I can test thsi
02:08:11 <fizzie> FWIW, the rate they're hitting me has dropped to maybe one query every 5 seconds.
02:14:09 <oren> how long does windows have to take "initializing..." blah
02:23:01 <oren> So now I guess I'll go to the site, download the program and try to install it
02:26:28 <oren> "content from thsi site is being blocked by internet explorer"
02:30:45 <oren> http://postimg.org/image/rbl9i9k79/
02:32:37 <oren> the download link doesn't seem to work well
02:38:35 <oren> cloudflare error
02:43:34 <fizzie> Yes, I got that now too.
02:48:57 <oren> http://postimg.org/image/i9lbg3lst/
03:06:34 <zzo38> Now I invented such URIs as <xurn:pokemon:25> and <xurn:bible:exodus:15:11> and <xurn:icao:cyvr> and <xurn:rdn:com.apple.ostype> and <xurn:mime:subject> and so on.
03:12:12 -!- oerjan has quit (Quit: leaving).
03:13:10 <oren> So I intalled the msi sucessfully
03:13:36 <oren> It is some sort of dedicated browser, which goes directly to...
03:13:55 <oren> the very same website
03:14:10 <zzo38> What is that? Why it does that?
03:15:28 <zzo38> If you don't know, then you must learn.
03:55:24 -!- FreeFull has quit (Ping timeout: 255 seconds).
03:56:45 -!- FreeFull has joined.
04:35:16 -!- f|`-`|f has quit (Ping timeout: 256 seconds).
04:35:22 -!- f|`-`|f has joined.
05:37:36 <zzo38> Which one do you like to disagree?
06:07:49 -!- zadock has joined.
06:12:18 -!- teakey has joined.
06:17:26 -!- doesthiswork has quit (Quit: Leaving.).
06:37:17 -!- trout has quit (Ping timeout: 272 seconds).
06:58:02 -!- teakey has quit (Ping timeout: 272 seconds).
06:59:59 -!- ^v has quit (Read error: Connection reset by peer).
07:00:51 -!- Tritonio has joined.
07:10:56 -!- teakey has joined.
07:43:33 -!- zadock has quit (Quit: Leaving).
07:47:07 -!- teakey has quit (Ping timeout: 246 seconds).
07:59:30 -!- teakey has joined.
08:22:38 * impomatic_ pressed the button https://www.reddit.com/r/thebutton/
08:32:17 -!- Phantom_Hoover has joined.
08:34:18 -!- Patashu has quit (Ping timeout: 272 seconds).
08:44:24 -!- teakey has quit (Remote host closed the connection).
09:17:52 -!- fowl has quit (Ping timeout: 256 seconds).
09:20:23 -!- Vorpal has joined.
09:20:32 -!- Vorpal_ has quit (Ping timeout: 272 seconds).
09:31:01 -!- notfowl has joined.
09:31:04 -!- f|`-`|f has quit (Ping timeout: 256 seconds).
09:46:41 -!- copumpkin has quit (Quit: My MacBook Pro has gone to sleep. ZZZzzz…).
09:47:02 -!- f|`-`|f has joined.
10:06:38 <impomatic_> Is there any way to view deleted wikipedia pages. There's some deleted programming stuff I want to recover and put back online (not on Wikipedia).
10:13:06 -!- Frooxius has quit (Ping timeout: 272 seconds).
10:29:09 <mroman> heartbeat detectors that work over 9m
10:32:06 <mroman> 16 year old kids injecting anabolica
10:32:09 <mroman> that's also pretty neat
10:34:35 -!- S1 has joined.
10:53:38 <elliott> impomatic_: contact an admin
10:54:01 <elliott> there might be a page for requesting it, I forget
11:14:06 -!- SopaXT has joined.
11:17:09 -!- SopaXT has changed nick to SopaXorzTaker.
11:19:34 <Melvar> ( \foo : String => interpolate "Hello, ${foo}!"
11:19:34 <idris-bot> \foo => prim__concat "Hello, " (prim__concat foo "!") : String -> String
11:31:51 -!- SopaXorzTaker has quit (Remote host closed the connection).
11:35:07 -!- notfowl has quit (Excess Flood).
11:36:08 -!- teakey has joined.
11:36:22 -!- notfowl has joined.
11:45:37 -!- teakey has quit (Quit: Leaving).
11:50:04 -!- boily has joined.
11:55:26 -!- teakey has joined.
12:21:39 <fizzie> oren: FWIW, over the night the DNS traffic suffix has changed to "jijizy.com", which seems to be registered by the same guy, but contains something that looks like a list of pirated movies in Chinese.
12:30:30 <fizzie> http://sprunge.us/fGDE <- top 5 suffixes for each block, with the full log divided into 20 blocks.
12:33:42 -!- Patashu has joined.
12:41:55 <impomatic_> eBay must know something I don't. They've just increased my selling allowance to 13000000.00!
12:43:41 <fizzie> That's an odd sum. (At least if you count millions.)
13:01:31 -!- GeekDude has joined.
13:02:59 -!- doesthiswork has joined.
13:07:40 -!- nszceta has joined.
13:09:47 -!- teakey has quit (Ping timeout: 256 seconds).
13:22:07 -!- teakey has joined.
13:41:13 <mroman> I wonder how safe simple crypto things are
13:41:48 <mroman> while data: key = sha1(password); encrypt block with key; key = sha1(key);
13:42:36 <mroman> where encrypt is just XOR
13:42:39 -!- Phantom_Hoover has quit (Ping timeout: 256 seconds).
13:44:46 <mroman> let's just hope there aren't many fixpoints in sha1
13:44:48 -!- teakey has quit (Read error: Connection reset by peer).
13:44:55 -!- S1 has quit (Quit: S1).
13:45:45 <HackEgo> [wiki] [[SNUSP]] http://esolangs.org/w/index.php?diff=42814&oldid=42807 * IanO * (+61) /* Square root */ radical, dude!
13:47:06 <mroman> probably weak against known plaintext
13:47:23 <mroman> which allows you to extract sha1(password) which you probably can brute-force somehow
13:48:27 -!- nszceta has quit (Quit: Textual IRC Client: www.textualapp.com).
13:56:30 <fizzie> mroman: Re known plaintext, it also has the problem that guessing any plaintext block lets you decrypt everything that comes after it.
14:02:03 <fizzie> I think the "normal" way to construct a stream cipher out of a hash is to use H(key || counter) -- with 'counter' starting from a never-reused-with-same-key IV -- to generate the keystream. It has the extra benefit that you can parallelize easily.
14:03:22 <fizzie> If "normal" is the right word, since I don't think that sort of thing is really used. Possibly because all the analysis for hash functions has been done with collision resistance in mind, and perhaps also "normal" block ciphers I guess tend to be faster.
14:05:52 <fizzie> A quick search found at least one published-in-an-IEEE-magazine-so-must-be-trustworthy paper about something like that: http://dx.doi.org/10.1109/ICCIAS.2006.295277
14:07:38 <fizzie> (To be honest, I'm not terribly impressed by that paper.)
14:08:53 -!- Patashu has quit (Ping timeout: 264 seconds).
14:11:57 <int-e> I love this sentence. "Firstly the algorithm was introduced detailedly."
14:22:39 <boily> int-ello. what dows it mean?
14:23:13 <int-e> I don't know. It's from the abstract of the paper fizzie linked to.
14:23:40 <int-e> Presumably it means that they introduce their algorithm in detail in said paper.
14:28:12 <fizzie> "The implementation of our algorithm is described detailedly in Section 2."
14:28:20 <fizzie> (From the introduction of the paper.)
14:28:34 <fizzie> It doesn't have quite the same effect without the "firstly".
14:28:52 * boily detailedly disappear to make some coffee. “My brain hurts.”
14:29:06 -!- Phantom_Hoover has joined.
14:30:02 <fizzie> I'm also not entirely sure this part has the right word choice: "-- the change would be diffused to the digest (keystream block) evenly by the core hash function, which means however small the change is, the keystream blocks will be totally different and irrelevant."
14:31:14 <HackEgo> [wiki] [[SNUSP]] http://esolangs.org/w/index.php?diff=42815&oldid=42814 * IanO * (+0) /* Square root */ typo
14:32:55 -!- doesthiswork has quit (Quit: Leaving.).
14:53:08 -!- oerjan has joined.
14:58:15 <HackEgo> [U+FFFD REPLACEMENT CHARACTER]
15:00:45 <Melvar> ( \foo : String => interpolate "Hello, ${foo}!"
15:00:45 <idris-bot> \foo => prim__concat "Hello, " (prim__concat foo "!") : String -> String
15:02:46 <Phantom_Hoover> https://twitter.com/DavidLivey/status/596629805417811968
15:03:02 <Phantom_Hoover> important lesson about coding for edge-cases your election graphics there
15:05:30 <oerjan> ( let s = "Hello, ${foo}!" in \foo : String => interpolate s
15:05:30 <idris-bot> \foo => prim__concat "Hello, " (prim__concat foo "!") : String -> String
15:06:38 <oerjan> ( let s = "Hello, ${foo}!"; f t (foo : String) => interpolate t in f s
15:06:38 <idris-bot> (input):1:1: error: expected: ":",
15:06:38 <idris-bot> let s = "Hello, ${foo}!"; f t (foo : String) => interpolate t in f s<EOF>
15:06:57 <oerjan> ( let s = "Hello, ${foo}!"; f t (foo : String) = interpolate t in f s
15:06:58 <idris-bot> (input):1:1: error: expected: ":",
15:06:58 <idris-bot> let s = "Hello, ${foo}!"; f t (foo : String) = interpolate t in f s<EOF>
15:07:37 <Melvar> let isn’t letrec and indeed can only take one equation.
15:10:19 <Melvar> ( :let f : String -> String -> String; f t foo = interpolate t
15:10:29 -!- idris-bot has quit (Quit: Terminated).
15:13:31 <Melvar> I think it should have complained about a stuck term in a reflected elaborator script.
15:16:37 -!- idris-bot has joined.
15:16:40 <Melvar> Weird, it just eats memory.
15:17:03 <oerjan> dependently typed language: a language which shuffles all termination problems into the type system.
15:18:26 <Melvar> Well, this isn’t about the types really, it’s about the elaborator.
15:19:15 <oerjan> elaborator: a subsystem of a dependently typed language that tries to remove blame from the type system
15:20:38 <Melvar> It’s the thing that turns Idris into TT, which involves expanding all the sugar, qualifying all the names, and inserting all the implicit arguments.
15:25:03 <Melvar> Here we have an implicit argument to interpolate that has a default value which invokes an elaborator script, a program written in idris that runs in the Elab monad, that recieves a reflected version of the environment and produces a reflected version of the desired term.
15:26:42 <Melvar> Here’s the code: http://lpaste.net/132218 and the first more-than-half of that is noodling about with the parsing library.
15:29:29 <oren> Is a tagged union an example of a dependent type?
15:30:19 <Melvar> Depends on what you mean maybe? Haskell has Either just fine non-dependently.
15:31:17 <Melvar> But if you say, have a type of tags, and a function that assigns the tags types, and put that in a dependent pair, …
15:32:53 -!- PinealGlandOptic has joined.
15:36:06 <oren> Hmm... Based on reading idris's website, this seems to mean that type checking can depend on arbitrary computations
15:37:08 <Melvar> arbitrary known-terminating computations.
15:37:56 <Melvar> Expressions with things that aren’t known to be total just fail to reduce in types.
15:41:20 <Melvar> (The termination checker is rather conservative, it requires that recursive calls be made on a subterm of the input.)
15:43:19 <Melvar> For example, in the paste I posted, the merge function at the top takes a Vect because it doesn’t fulfill that requirement if it takes a List.
15:48:08 <HackEgo> [wiki] [[Talk:Gibberish/JavaScript]] http://esolangs.org/w/index.php?diff=42816&oldid=42706 * Esowiki201529A * (-831) Replaced content with " ''$''{fuck:[>]}·test"
15:48:46 <boily> replaced content with fuck?
15:49:12 <HackEgo> [wiki] [[Talk:Gibberish/JavaScript]] http://esolangs.org/w/index.php?diff=42817&oldid=42816 * Esowiki201529A * (+30)
15:49:52 -!- hilquias has joined.
15:49:55 <HackEgo> [wiki] [[Talk:Gibberish/JavaScript]] http://esolangs.org/w/index.php?diff=42818&oldid=42817 * Esowiki201529A * (-1)
15:50:24 <boily> what the fungot is going on...
15:50:24 <fungot> boily: then by all means! i want my source code, then
15:51:10 <boily> fungot: even the source doesn't help hth
15:51:10 <fungot> boily: call another program from this perspective, you'll find a slightly more optimal format.) emacs uses a stack ( ring, rather)
15:51:58 <HackEgo> [wiki] [[Talk:Gibberish/JavaScript]] http://esolangs.org/w/index.php?diff=42819&oldid=42818 * Esowiki201529A * (-12)
15:55:53 <oerjan> boily: i'm not sure Esowiki201529A quite understands talk pages. although e's the only contributor to that one.
15:59:13 <HackEgo> [wiki] [[User talk:Esowiki201529A]] http://esolangs.org/w/index.php?diff=42820&oldid=42716 * Esowiki201529A * (+44) /* Disruptive edits */
16:00:18 -!- hilquias has quit (Ping timeout: 264 seconds).
16:00:50 <HackEgo> [wiki] [[User talk:Esowiki201529A]] http://esolangs.org/w/index.php?diff=42821&oldid=42820 * Esowiki201529A * (+55) /* Quiney */
16:01:33 <HackEgo> [wiki] [[User talk:Esowiki201529A]] http://esolangs.org/w/index.php?diff=42822&oldid=42821 * Esowiki201529A * (+47) /* Quiney */
16:02:43 <HackEgo> [wiki] [[User talk:Esowiki201529A]] http://esolangs.org/w/index.php?diff=42823&oldid=42822 * Esowiki201529A * (+56) /* Quiney */
16:03:00 <oren> I don't think I understand talk pages anymore
16:04:35 -!- oerjan has quit (Quit: Later).
16:04:36 <oren> "the title is incorrect due to technicallimitations. the correct titleis talk page"
16:05:04 <HackEgo> [wiki] [[User talk:Esowiki201529A]] http://esolangs.org/w/index.php?diff=42824&oldid=42823 * Esowiki201529A * (+174) /* Quiney */
16:07:44 <HackEgo> [wiki] [[User talk:Esowiki201529A]] http://esolangs.org/w/index.php?diff=42825&oldid=42824 * Esowiki201529A * (+120) /* Quiney */
16:08:25 <HackEgo> [wiki] [[User talk:Esowiki201529A]] http://esolangs.org/w/index.php?diff=42826&oldid=42825 * Esowiki201529A * (-1) /* Quiney */
16:10:18 <HackEgo> [wiki] [[User:Orenwatson]] M http://esolangs.org/w/index.php?diff=42827&oldid=41324 * Orenwatson * (+22)
16:10:34 -!- hilquias has joined.
16:15:08 -!- kline has changed nick to homel1nen.
16:19:52 -!- homel1nen has changed nick to kline.
17:02:12 -!- aretecode has joined.
17:11:03 -!- aretecode has quit (Ping timeout: 240 seconds).
17:11:35 <oren> Gah. I hate this trend of using special fonts to put logos and pretty-ass buttons on web pages
17:12:16 <boily> I like it for pretty-ass buttons. small semantic icons are helpful hth
17:12:47 <shachaf> boily: seems like your attitude is more like kth hth
17:13:44 <oren> They should use the unicode emojis
17:13:53 -!- aretecode has joined.
17:14:06 <b_jonas> oren: yeah. I don't load the fonts, and I hate icons (even when they're picture-based) in first place.
17:14:20 <b_jonas> I can never figure out what the icons mean. They should just use text.
17:14:31 <boily> shachaf: shellochaf. what's a kth twh
17:14:36 <b_jonas> On saner websites, there's at least title text explaining what they mean.
17:14:51 <boily> oren: no. unicodemojis are... they disturb my feng shui.
17:15:35 <b_jonas> It will get worse when colored font support will get commonly implemented in browsers by the way.
17:15:45 <boily> b_jonas: the icon-text combo is the best imo. small visual cues to guide your eyes to common operations, with text for stuff you can't remember.
17:16:23 <boily> coloured font support is a thing? (well, it existed a long time ago with bitmap tiles, but with current technology?)
17:16:44 <boily> shachaf: they do, even when they get excised. ice creaaaaaaam!
17:16:53 <b_jonas> for me, text only would be enough. it can be two pieces of text, a short one as a label and a longer title text for explanation. But I know icons help some other people, so if they want them, sure.
17:17:02 <oren> Like they don't even use the standard code point for icons that are standard like floppy disk and gear
17:17:06 <b_jonas> boily: not really current tech yet, but some people are working on it
17:17:42 <b_jonas> boily: apparently people want it for colored smiling face emojis or something
17:17:43 <HackEgo> U+1F4BE FLOPPY DISK \ UTF-8: f0 9f 92 be UTF-16BE: d83ddcbe Decimal: 💾 \ 💾 \ Category: So (Symbol, Other) \ Bidi: ON (Other Neutrals)
17:18:54 <HackEgo> U+23F8 - No such unicode character name in database \ UTF-8: e2 8f b8 UTF-16BE: 23f8 Decimal: ⏸ \ ⏸ (⏸) \ Uppercase: U+23F8 \ Category: Cn (Other, Not Assigned)
17:19:08 <shachaf> They have so many code points and they can't find room within their cold dark hea^H^H^Hplanes for play/pause/fast forward/rewind?
17:19:12 <HackEgo> U+23F8 - No such unicode character name in database \ UTF-8: e2 8f b8 UTF-16BE: 23f8 Decimal: ⏸ \ ⏸ (⏸) \ Uppercase: U+23F8 \ Category: Cn (Other, Not Assigned)
17:19:48 <Melvar> Anyway, that’s DOUBLE VERTICAL BAR with an alias “pause”.
17:20:18 <oren> If they used the standard code points, it would at least whow up semi-properly, and you could even have fonts that diplay them as SAVE, SETTINGS, PAUSE etc
17:20:28 <Melvar> The next two are BLACK SQUARE FOR STOP and BLACK CIRCLE FOR RECORD.
17:20:28 <b_jonas> boily: I'm not specifically against colored font support, firstly because it could be useful for decorative header fonts, secondly because if people get it working, then maybe they'll also improve the rendering and font-making tools for grayscale bitmap (not vector) fonts.
17:21:21 <HackEgo> U+23CF EJECT SYMBOL \ UTF-8: e2 8f 8f UTF-16BE: 23cf Decimal: ⏏ \ ⏏ \ Category: So (Symbol, Other) \ Bidi: ON (Other Neutrals)
17:21:26 <b_jonas> I don't like colored smiley face nonsense, especially when it's used with auto-replace like, you know, Data:☺umper
17:22:44 <oren> Well yeah that tends to cause problems when discussing code on facebook
17:23:58 <b_jonas> Mibbit and Skype are two particular offenders. It's possible to turn it off in both, though the option is well-hidden, and the other party has to do it as well.
17:24:04 <shachaf> don't discuss code on facebook hth
17:24:28 <Melvar> b_jonas: As in, it turns if *either* side has it turned on?
17:24:29 <b_jonas> There are also some web forums doing autoreplacements, though some of them only use longer replace codes, not short ones like :) or :P
17:25:05 <b_jonas> Melvar: no, I think the message goes through as you type, and how it shows up in your window is determined by your setting, no matter who sent the message.
17:25:14 <oren> My project group moved to just typing our discussion into a shared google doc for that reason
17:25:19 <b_jonas> but I'm not quite sure about this
17:25:25 <b_jonas> test if you really have to know
17:25:40 <Melvar> b_jonas: Then what do you mean by “the other party has to do it as well”?
17:26:13 <oren> assuming you are both typing code samples
17:26:13 <b_jonas> Melvar: as in, if I send a message saying "you should use the Data::Dumper module" and the other party doesn't set the option, then he will see a smiley face
17:26:31 <b_jonas> no, even if only one of you is typing
17:26:35 <b_jonas> because you see your own messages too
17:26:56 <b_jonas> I think mibbit (the web irc client) used to work similarly before it got banned from freenode completely
17:27:29 <b_jonas> I think qwebirc and kiwi irc don't do such autoreplacements, but I'm not so sure
17:27:45 <oren> shared google doc works surprisdingly well as a chat client
17:28:16 <Melvar> It seems at least two clients so far misparse mirc color codes such that my bot’s output looks wrong.
17:29:00 <b_jonas> I wonder if I should get some selected colleagues to use it rather than skype
17:29:41 <b_jonas> Though it might be difficult because they already run skype for voice chat and/or video chat
17:30:22 <oren> voice chat is terrible if you have an open plan office
17:30:33 <zzo38> I also prefer using IRC too, with just text chat
17:30:44 <zzo38> For voice you just use the telephone
17:30:59 <b_jonas> zzo38: we use telephone a lot too, sure
17:31:17 <oren> at my old job we only used gmail IM thingy
17:31:19 <b_jonas> we even have some company phones with a plan such that they can call each other for free
17:32:07 <b_jonas> one problem with that is that the phone hw generally used for them is REALLY bad, unusable even compared to cheap smartphones in general
17:32:21 <b_jonas> I've temporarily put the sim card in my phone at least once because of this
17:33:29 <b_jonas> like, they're smartphones with so bad a touchscreen it ignores most touches, so I can't terminate calls or initiate them, and I can't easily figure out how to use them with a headphone because none of the two possible interfaces seem to work with them
17:33:48 <b_jonas> and at least one sometimes ignores incoming calls
17:34:12 <fizzie> Google Hangouts for everyone.
17:34:28 <fizzie> I'm not contractually obligated to say that. As far as you know.
17:34:41 <zzo38> No I mean use a landline telephone
17:35:14 <b_jonas> so temporarily moving the sim card is definitely worth the hassle, but that's hard if someone passes me the phone with someone already on the line and the speaker is set to so silent I can barely hear them, and I can't put in a headphone nor increase the volume
17:35:41 <b_jonas> the phones are so bad that using skype voice call instead can be worth, which is saying something
17:36:38 <b_jonas> zzo38: we have very few landline phones in the company. possibly none. they're going out of fashion rapidly everywhere.
17:37:02 <b_jonas> no surprise of course, because the mobile phones are better in almost everything.
17:37:46 <b_jonas> I wonder if eventually public phone boots will use mobile phone technology as well
17:38:10 <zzo38> I think the landlines are better though
17:38:38 <oren> Landlines have the advantage of reliability
17:38:41 <b_jonas> mind you, landline phones _will_ be supported for long still, basically because they can get them as a free extra to houses where they take the internet
17:39:01 <b_jonas> so they can install even new lines because of internet and cable tv
17:39:31 <b_jonas> oren: that's the traditional wisdom, but it won't be true for long as landline gets less popular and less maintained by the companies
17:39:51 <oren> Landline phones in particular work during a blackout
17:40:16 <oren> While the internet doesn't, typically.
17:40:18 <b_jonas> oren: already here in our house they're installing landline phones that go through ip, with a box outside the house that converts between landline in the house and internet outside, so they'll be only as reliable as the internet connection
17:40:28 <zzo38> Yes, those are some advantages
17:40:35 <zzo38> Also landlines use a simple protocol
17:40:45 <oren> b_jonas: that'snot a ral landline then
17:41:01 <b_jonas> so, it won't work when the internet is broken (though perhaps sometimes they'll have the internet only partly broken so the landline will still work)
17:41:39 -!- zzo38 has quit (Remote host closed the connection).
17:41:40 <b_jonas> oren: sort of. the interface towards us will be unchanged, it will behave as a landline, except for slightly cheaper price
17:41:49 <b_jonas> but it won't be as reliable as landline indeed
17:42:03 <b_jonas> it might still be a bit more reliable than both the internet and power
17:43:12 <b_jonas> also, it's a brilliant invention from the provider's side: if the internet is down, you can't complain to their customer service on telephone because the telephone is down too
17:43:27 <b_jonas> that alone will save them a ton of money
17:43:43 <b_jonas> really, customer service costs a lot
17:43:52 <b_jonas> of course, people might still complain on mobile phone or from other places
17:44:07 <Melvar> oren: In that case, wide swathes of country don’t have real landlines available anymore.
17:44:18 <Melvar> The providers don’t offer it.
17:44:42 <Melvar> Only IP-based telephony.
17:45:35 <b_jonas> also, calling the customer service is free only if you call from telephone with the same provider, so if you call from mobile, you have to pay for at least the call. that still doesn't cover most of their costs of handling the call, but it will discourage people so they'll call less.
17:45:51 <b_jonas> the more I think about it, the more brilliant it is
17:45:54 <Melvar> Mind you it still goes through a separate channel from your usual internet generally, but it isn’t analog anymore and thus won’t work on minimal power which comes from batteries in the nearest terminator.
17:46:31 <b_jonas> if they're really smart, they'll also make sure the short number of the customer service (4 digit long, which they advertise the most) only works from the same provider, and few people will know the normal length phone number (which you can call from abroad too)
17:47:20 <oren> Hmm? I only have a long number for our customer support
17:48:16 -!- hilquias has quit (Remote host closed the connection).
17:48:25 <b_jonas> oren: we have short numbers. sometimes multiple (eg. one for pre-paid mobile, one for post-paid mobile, one for business customers), and if you call the wrong number, they just send you to the other number, and sometimes they send you in circles
17:49:22 <b_jonas> heck, so many things get short numbers here (all starting by 1 now) that there are short numbers of length 3, 4, 5, and 6. the 6 digit long ones are hard to call short numbers, though those are rare.
17:50:01 <oren> I remember when numbers wer usually 7 digits
17:50:20 <oren> Now you have to dial 10
17:50:53 <b_jonas> the emergency services and a few other stuff get 3 digit numbers, telephone and internet provider service lines and stuff like that get 4 digit numbers, charity donation numbers are 4 or 5 digit long,
17:51:42 <b_jonas> commercial services that are publically useful, like car trailing service, get 4 digit long if they pay a lot, or longer if they dont; commercial services that aren't viewed as publically useful, like voting for tv reality shows, can get 5 or 6 digit long numbers if they pay a lot.
17:52:18 <oren> huh. In canada AFAIK, we only have 3 digit and 10 digit numbers
17:52:53 <b_jonas> the problem is that they keep changing the short numbers so often you can't follow them, because they're valuable assets and in number space shared by multiple providers,
17:53:26 <b_jonas> and it's very difficult to find out about them, because there doesn't seem to be any central directory listing them all, you only find out about them in advertisments of particular services.
17:53:48 <Melvar> Here the ordinary numbers are six digits. Or were, I think all the new ones are seven but the six-digit ones are still around.
17:54:46 <b_jonas> even the emergency service numbers have changed twice in my lifetime: they used to be 04, 05, 07, later those got deprecated and changed to 104, 105, 107, now those are deprecated too and they recommend using only 112, though I _hope_ the deprecated numbers are still available for very long in that case.
17:54:55 <b_jonas> other short numbers change much more often.
17:55:17 <Melvar> Here they’ve been 110 and 112 forever.
17:56:39 <b_jonas> here, ordinary numbers for mobile phones are 2+7 digits (used to be 2+7 before), landlines for budapest are 1+7 digits (used to be 1+6 before), other landlines and special area codes (like toll-free and premium) are 2+6 long.
17:57:35 <b_jonas> oh, and _all_ old numbers in budapest were changed twice: once when they were changed from 1+6 digit to 1+7 digit by adding a prefix 1, and once when all the numbers starting with 1 were obliterated so they're used for short numbers only.
17:57:47 <oren> In toronto the special numbers I know are 911 (emergency) 311 (city services) 411 (phone directory service)
17:58:29 <b_jonas> (previously short numbers used to start with 0, then with 0 and 9, then with 0 and 9 and 1, then later with only 1)
17:58:35 <Melvar> Huh. Here the six-digit local numbers have stuck around even though they’re handing out seven-digits ones now.
17:58:38 <b_jonas> (mobile phone numbers were changed only once, by prefixing a 9)
17:59:06 <b_jonas> Melvar: you can't do that in Budapest, which is a _really_ large city
17:59:12 <Melvar> And indeed, it depends on the area code; smaller areas have shorter local numbers (but usually a longer area code).
18:00:11 <b_jonas> This system works only because Budapest is big and all other towns are small.
18:00:19 <Melvar> Not sure what the city size has to do with it.
18:00:36 <b_jonas> well if you want a single area code for Budapest, you need 7 digit numbers
18:01:03 <b_jonas> there are only 800000 available 6-digit numbers, which is way too few
18:01:17 <Melvar> Oh, you probably mean the six-digit ones could clash with seven-digit ones?
18:01:36 <b_jonas> but every other town is small, so the other approx. 40 landline area codes can easily be assigned such that 6 digit is enough
18:01:45 <b_jonas> because there's no terminator in landlines
18:01:54 <Melvar> I never realized this.
18:02:15 <b_jonas> that's also why you can't have landline numbers starting with 0 or 1
18:02:17 <oren> almost 7 million people live in toronto, so there are multiple area codes (3 digit prefixes) which contain each 10000000 7 digit numbers
18:02:21 <Melvar> I thought they started connecting once one paused for sufficiently long.
18:03:11 <b_jonas> Melvar: I think they did that temporarily for a very short time at the incompatible numbering plan change for Budapest
18:03:15 <Melvar> Nowadays, they start when one pushes the green button.
18:03:27 <b_jonas> landline phones still don't
18:03:29 <fizzie> Finland had something before the EU-wide 112. I think it might've been 999.
18:03:50 <b_jonas> landline phones dial with either pulse or dtmf tone mode, neither of which has a terminator
18:04:13 <b_jonas> fizzie: I think Austria has or had 122, 133, 144 but I'm not sure
18:04:45 <fizzie> Wikipedia "List of emergency telephone numbers" to the rescue.
18:04:56 <b_jonas> 112 has been working for very long here, definitely before mobile phones got popular
18:05:20 <b_jonas> only they weren't convinced it's worth to use a unified emergency number instead of three separate ones, so they kept the separate ones
18:05:24 <b_jonas> (I'm not convinced either)
18:05:59 <fizzie> We had "10022" for the police, but that got merged into 112.
18:06:19 <b_jonas> it's easy to merge _one_ emergency service
18:06:25 <fizzie> Oh, it was 000 and not 999 before 112. And the changeover was in 1993 in Finland.
18:06:25 <b_jonas> merging all three is the hard part
18:06:40 <b_jonas> 000? that takes like ages to dial with pulse
18:06:58 <b_jonas> (not that 07 was much better)
18:07:31 <b_jonas> but a unified number certainly has some advantages
18:07:40 <b_jonas> when there's an emergency, I don't want to have to think of which number to dial
18:07:49 <b_jonas> always dialling 112 is the easiest
18:08:23 <fizzie> Some of the numbers in http://en.wikipedia.org/wiki/List_of_emergency_telephone_numbers are pretty specific.
18:08:31 <fizzie> Like the 114 for racial discrimination in Morocco.
18:09:54 <fizzie> Or the alleged 113 for "reporting spies" in (South) Korea. Although these generally don't have any citations.
18:10:09 <b_jonas> and unification everywhere in Europe is definitely useful ever since so many people have mobile phones and so many people can travel easily
18:10:47 <b_jonas> fizzie: no citations is because they're not listed properly anywhere
18:10:54 <b_jonas> not even on the homepage of providers anymore
18:11:13 <b_jonas> and printed phonebooks get less and less accessible, these days even post offices usually don't have them
18:11:50 <b_jonas> I'm not sure if they even print phonebooks these days (they still print smaller yellow-pages like business directories, and thin booklets for local services in a district)
18:12:32 <b_jonas> The terms and conditions list _some_ short numbers because it has to give their pricing, but it doesn't tell what each of them means, and is hard to use
18:12:51 <b_jonas> why can't they just have it as an appendix to online phonebooks like it used to appear in printed phonebooks (which also didn't list all short numbers btw)
18:13:28 <fizzie> The map in the phone book (for the Helsinki metropolitan region) used to be something people cut off from last year's version and use as a general map.
18:13:49 <fizzie> It was maybe 50 pages or so, with a street name -> page-and-grid-reference mapping.
18:14:35 <fizzie> I think they stopped distributing the printed copy to every household at some point.
18:14:49 <fizzie> It used to be this three-volume 15 cm stack.
18:15:04 <fizzie> (#1 for people, #2 for companies and #3 for the yellow pages.)
18:15:25 <b_jonas> I don't think we've ever had three-volume. We've had two-volume and one-volume versions for Budapest, and thinner one volume versions for each megye.
18:16:03 <b_jonas> though if you count yellow pages separately then maybe
18:17:04 <fizzie> Apparently these days you can pick up a copy from many post offices, during a one-month window, but they don't home-deliver.
18:17:11 <fizzie> It's gotten a lot thinner, though.
18:18:06 <b_jonas> that used to exist for a day. I don't know if it still exists
18:18:11 <b_jonas> also, there used to be a CD version for a while.
18:18:53 <b_jonas> there's definitely still an internet-based directory, but it's very hard to use and often broken, just like everything on the telephone provider's website
18:19:12 <b_jonas> and there's of course the directory accessible through voice calls which has slightly different content
18:20:17 <fizzie> The main Finnish one is quite okay. There's the website lookup, and the usual set of mobile apps.
18:20:54 <fizzie> The "digital" versions used to cost extra, but now it's gone down to a 0.25 EUR SMS to open an account, and no other costs.
18:21:00 <b_jonas> there might be mobile apps, I wouldn't know about it
18:21:18 <b_jonas> there's probably an SMS version, yes, I wouldn't know about that either
18:23:36 <b_jonas> fungot, do you use landline phone?
18:23:36 <fungot> b_jonas: true elegance... who is to measure recursion! :(
18:25:45 <fizzie> Thanks to the way DSL things are generally set up here, we technically have a phone-enabled landline fungot could tap into, although nothing's connected to it.
18:25:46 <fungot> fizzie: parsing of left and right sides are identified, so if i work on have browser interfaces oriented towards ie. so the important questions don't entirely respect the characterizations. my favorite ever was possibly pingpong, but i
18:26:11 <fizzie> Maybe I could hook up a modem and use the ring signal as a side channel for something.
18:26:45 <b_jonas> fizzie: why would that be useful? there's lots of other internet-based side channels you could use
18:27:01 <fizzie> If the Internet is down.
18:27:47 <b_jonas> so if you ring it, it dials a landline modem internet connection as backup, and uses expensive internet with 56 kilobit/sec speed there?
18:28:22 <fizzie> We had a free dialup at the university campus -- an elegant system, for a more civilized age.
18:28:30 <fizzie> Nobody used it, but it hadn't gotten dismantled.
18:28:45 <b_jonas> actually these days the telephone provider provides a free dialup for landlines
18:29:08 <b_jonas> it costs less than a normal voice phones, and is probably still cheap for the provider to run
18:29:16 <fizzie> Does the call still cost something? This was entirely free. (The local data, TV and phone networks were all built by the students.)
18:29:31 <b_jonas> (or at least, this had existed about two years ago)
18:29:37 <b_jonas> fizzie: yes, the call costs something
18:30:02 <b_jonas> but less than a normal local voice call from the same landline would likely cost, unless you have a plan with unlimited local calls
18:30:17 <fizzie> You could call the "internal" (campus and university) numbers for free there.
18:30:31 <fizzie> And the university provided a for-students dialup that was included.
18:30:32 -!- variable has joined.
18:30:32 <b_jonas> oh sure, if you're within the university it can be free
18:30:42 <fizzie> I think I routed our home interwebs over it once, when the Ethernet side was down for some hours.
18:30:45 <b_jonas> but this works from all landlines
18:30:55 -!- variable has quit (Remote host closed the connection).
18:34:55 <fizzie> The Helsinki telephone company used to have this pessimized-for-dialup-users pricing, where -- for 5pm-to-8am on weekdays, and all day on weekends -- there was a fixed (small) per-call cost for any call up to 30 minutes, and then the per-minute charges started ticking. Those evening-and-weekend times used to have just the per-call cost, before they realized they started getting dialup users ...
18:35:01 <fizzie> ... doing one "call" per day and still using up their capacity.
18:35:26 <fizzie> You had to hang up and redial every half an hour or so for the cheapest stay-online experience.
18:35:56 <fizzie> (If you did it right, and had a static IP, you didn't even lose active TCP connections, so no dropping from IRC.)
18:37:51 <boily> sudo aplay /dev/mem.
18:48:26 -!- notfowl has changed nick to unclefowl.
18:51:39 -!- unclefowl has quit (Excess Flood).
18:52:25 -!- notfowl has joined.
18:52:39 -!- G33kDude has joined.
18:53:46 <b_jonas> Question about Moore's law stuff. For how many years have most CPUs been having a constant 32 kilobytes of 4-way associative L1 cache? Should we expect this to continue?
18:54:24 <b_jonas> I understand why the clock speed can't be increase further above 2 or 4 gigahertze, but I don't know enough about microelectronics and cpu design about how expensive it would be to grow the L1 cache.
18:56:13 -!- GeekDude has quit (Ping timeout: 272 seconds).
18:56:17 -!- G33kDude has changed nick to GeekDude.
18:57:08 <b_jonas> (L1 data cache to be precise)
19:07:40 -!- boily has quit (Quit: ELEVATOR CHICKEN).
19:17:56 <oren> Horrible idea i just had. Some people put comments in code like
19:18:19 <oren> x = y + 4; //set x to y plus 4
19:18:32 <b_jonas> oren: yes, that's especially common in assembler code
19:18:33 <oren> Cn be generate such comments automatically?
19:18:46 <b_jonas> oren: people generate _api documentation stuff_ automatically too
19:19:45 <b_jonas> like, they write a function named void SetLength(int v) in the class Foo and automatically generate api documentation saying “Sets the length of the foo to the integer given in the parameter.” and even more fancy stuff deduced from code.
19:21:28 <b_jonas> like, they automatically add meaningless six line long comments in the code saying /*! \brief Default constructor. Creates a Foo object. */ Foo();
19:21:39 <b_jonas> because of some stupid conding standard that asks them to document every function or some such shit
19:21:53 <b_jonas> api documentation can be useful, but this kind of api docs is worse than nothing.
19:22:13 <notfowl> Pretty annoying when youre typing extra shit that just goes to doc formatting
19:23:23 <b_jonas> those comments should document only what's not immediately obvious to anyone reading the function name and type
19:25:40 <fizzie> There's an "english" command in candide, the ##c bot.
19:26:07 <fizzie> 20:26 <fizzie> ,english x = y + 4;
19:26:07 <fizzie> 20:26 <candide> fizzie: Assign to `x` the result of the expression `y` plus 4.
19:26:10 <fizzie> 20:26 <fizzie> ,english if (x()) y();
19:26:13 <fizzie> 20:26 <candide> fizzie: If the result of the function `x` is nonzero then call the function `y`.
19:26:16 <fizzie> You could just run that on each statement, and add it as a comment.
19:27:09 <b_jonas> ,english x = a*(b+c)*d; y = a*b+c*d;
19:27:29 <fizzie> Assign to `x` the result of the expression `a` times (`b` plus `c`) times `d` and then assign to `y` the result of the expression `a` times `b` plus `c` times `d`.
19:28:03 <fizzie> It recognizes some rather obscure special cases, but I don't remember what they are.
19:28:04 <b_jonas> yes, that's what it said in privmsg
19:29:00 <fizzie> void bar(char foo[static 41]) {} => "Let `bar` be a function taking `foo` as an array with optimization hint to provide access to the first element of 41 elements of char and returning void. When called, the function will do nothing."
19:29:49 -!- zadock has joined.
19:30:26 <b_jonas> notfowl: you can private message candide
19:30:55 <fizzie> It also special-recognizes the "!! operator".
19:30:59 -!- bb010g has joined.
19:31:06 <fizzie> 20:31 <fizzie> ,english !!x
19:31:06 <fizzie> 20:31 <candide> fizzie: The normalized boolean value of `x`.
19:31:54 <oren> on a related note, My mostrecent version of scrip7 generates the documentation for commands by sedding the comments and the case : values
19:33:28 -!- impomatic_ has quit (Ping timeout: 264 seconds).
19:41:12 <oren> because the sed commands are specific to the particular code, the code can simply look like this: http://ctrlv.in/572297
19:42:24 <oren> and the docs can look like: http://ctrlv.in/572298
19:42:57 <oren> ooh sequential urls
19:50:16 -!- notfowl has quit (Changing host).
19:50:16 -!- notfowl has joined.
19:50:16 -!- notfowl has quit (Changing host).
19:50:16 -!- notfowl has joined.
19:56:11 <oren> So iow, you can see what peoplearound the globe are doing by entring in numbers near the current one
20:14:32 <oren> also http://ctrlv.in/1
20:29:29 <fizzie> Apparently the answer is "mostly watching porn or playing minecraft". (Based of a N=8 sample.)
20:31:36 <Sgeo> oren, please parameterize queries
20:32:20 <Sgeo> Hmm, those values don't look particularly untrusted. Still a good habit to get into
20:34:27 <oren> which queries?
20:47:52 <Phantom_Hoover> fizzie, i'm seeing a lot of photos of an indian woman for some reason
20:48:23 <Phantom_Hoover> i suppose that's about what you'd expect for a random sample from across the globe
21:01:38 <oren> Among people who searched upload image in english especially
21:20:09 -!- hilquias has joined.
21:23:16 -!- oren has quit (Read error: Connection reset by peer).
21:26:23 -!- orin has joined.
21:30:14 -!- zadock has quit (Quit: Leaving).
21:36:36 <Sgeo> I O it's magic, you know
22:08:16 -!- PinealGlandOptic has quit (Ping timeout: 264 seconds).
22:57:45 -!- Patashu has joined.
23:02:39 -!- drdanmaku has joined.
23:27:48 -!- drdanmaku has quit.
23:28:58 -!- oerjan has joined.
23:39:09 -!- Frooxius has joined.
23:39:33 -!- mitchs_ has joined.
23:39:54 -!- Froox has joined.
23:42:03 -!- mitchs has quit (Ping timeout: 240 seconds).
23:43:53 -!- Frooxius has quit (Ping timeout: 272 seconds).
23:52:20 <FireFly> Interesting, those ctrlv.in images have OCR'd text as alt text
23:52:34 <FireFly> so you could (ab)use it as an OCR service?
23:53:19 <orin> but it takes some time to show up, it's not there right away
23:58:56 -!- doesthiswork has joined.