←2016-07-13 2016-07-14 2016-07-15→ ↑2016 ↑all
00:02:39 -!- adu has joined.
00:07:14 -!- JenElizabeth8 has quit (Quit: Nettalk6 - www.ntalk.de).
00:08:33 <boily> hppavilion[1]: hppavellon[1]. why the matinal BLASPHEMY?
00:08:47 <hppavilion[1]> boily: matinal?
00:08:57 <boily> eeeeh... morning-y?
00:09:05 <boily> you BLASPHEMYed me this morning.
00:09:44 <boily> (hmm... according to Google Translate, there's no adjectival form of "morning" in English. strange.)
00:12:37 <hppavilion[1]> I really want a custom-made car where all of the controls are in the back seat (and there is no front seat)
00:12:51 <hppavilion[1]> So people do a double take when they see me driving from the back of the car
00:13:00 <hppavilion[1]> I mean, I probably couldn't legally drive it (it'd be hard to see anything)
00:13:02 <zzo38> Then what are you going to store in the front? The engine? A TV set?
00:13:10 <hppavilion[1]> And it'd be too expensive to be practical
00:13:11 <hppavilion[1]> But still
00:15:11 -!- tromp_ has joined.
00:17:32 <hppavilion[1]> On a more realistic note, I want any car I have to have a display mounted on the rear window that allows me to send messages to drivers behind me
00:17:32 -!- variable has joined.
00:18:21 -!- function has joined.
00:18:32 -!- variable has quit (Remote host closed the connection).
00:20:57 <Cale> boily: The adjective form of morning is just "morning"
00:21:11 <quintopia> helloily
00:21:15 <boily> Callo. beuh :/
00:21:20 <boily> quinthellopia!
00:21:31 <quintopia> what's the third coolest city in the U.S.
00:21:36 <quintopia> ?
00:21:39 <Cale> OED doesn't list it as such somehow, instead preferring to list a bunch of specific compounds of morning together with another noun, which seems odd.
00:22:16 <quintopia> well
00:22:19 <boily> third?
00:22:31 <quintopia> there certainly are only a handful of situations in which morning is adjectived
00:22:46 <quintopia> morning meeting, morning coffee, morning constitutional...
00:23:06 <quintopia> boily: on the assumption that NY and LA will be the first two
00:23:17 <Cale> Oh weird, in the 1989 version of OED, they had it clearly listed as an adjective as well.
00:23:25 <boily> meeting matinal, café matinal, eeeeh... constituosité matinale???
00:23:49 <boily> quintopia: oh, that kind of cool! I was stuck on weather cool.
00:25:48 <boily> I wouldn't know, really... I only remember Boston, New York and Washington DC. My family and I went to LA too, but it was a long long time ago.
00:26:14 -!- adu has quit (Quit: adu).
00:26:46 <boily> Perhaps Austin?
00:27:06 <boily> quintopia: can you vouch for Atlanta being the Third Coolest?
00:27:30 <pikhq> Weird, at least as a US English speaker "morning" in those phrases parses as an adjective, not as a compound noun.
00:27:31 <quintopia> atlanta is just a city
00:27:39 <quintopia> it does have me, so that's pretty cool.
00:27:47 <quintopia> austin is probably cooler
00:27:47 <fizzie> I think I'll switch from "account creation and anonymous editing disabled" to "ask an 'impossible' CAPTCHA question", to see if the spammers are actually passing it or just bypassing it, and also to provide an explanation.
00:28:49 <quintopia> account creation disabled?
00:29:26 <boily> but but... what if someone had a stroke of inspiration and wanted to expand the Glorious Purple Family?
00:29:27 <fizzie> Well, I had to do *something*.
00:29:50 <quintopia> what was happening?
00:29:55 <fizzie> Spam.
00:30:13 <fizzie> At a rate of about 44 new accounts and ~500 spam edits in the 15 or so minutes.
00:30:19 <quintopia> i thought spam happened constantly?
00:30:23 <quintopia> oh
00:30:26 <quintopia> yeah that's high
00:30:51 <quintopia> boily: Glorious People's Republic of Purple
00:31:22 <hppavilion[1]> Corporation's Republic of Capitalia
00:31:29 <\oren\> three strange notes have been left in our mail slot
00:31:50 <hppavilion[1]> If corporations are people too, they should get their lifespan limited to ~78 years
00:32:08 <quintopia> true
00:32:16 -!- MDude has joined.
00:32:29 <\oren\> "I love you" "You are special" and "Yor are beatiful" (sic). What are the note-leaver's intentions?
00:32:37 <quintopia> or else people should have their lifespans indefinitely expanded
00:33:10 <quintopia> \oren\: market research for future valentine's cards
00:34:07 <\oren\> these are physical, handwritten notes
00:34:33 <boily> he\\oren\. do yor feel beatiful?
00:34:35 <quintopia> boily: what is left to be done with the aubergines? any more than ubergenes would be too much and any less than purple would be too little.
00:34:54 <boily> if you can't go up nor down, go sideways?
00:35:08 <\oren\> more beautiful than beatific at any rate
00:35:09 <quintopia> hmmmmmm
00:35:25 <boily> we have ubergenes that adds to the concept. chop off what is in aubergine. what are you left with?
00:35:44 <quintopia> Écraser?
00:36:52 -!- tromp_ has quit (Remote host closed the connection).
00:37:13 <quintopia> if you chop off what is in aubergine you get uh....only negatively addressed memory locations, any variable names that aren't a and b--unless we chop out the idea of a memory array entirely?
00:37:25 <boily> hmmm....
00:37:45 <boily> what are there besides linear arrays?
00:38:05 <boily> graphs? topologies? weird mathematical stuff? Taneb?
00:38:08 <quintopia> you also lose the ability to read or change the instruction pointer, and the ability to do I/O
00:38:31 <boily> I have a sentimental attachment to IO.
00:38:45 <quintopia> perhaps it can be done differently
00:39:03 <quintopia> Courge!
00:39:11 <hppavilion[1]> Clearly, \oren\ has a confidential admirer that only the top world leaders can be informed of
00:39:21 <hppavilion[1]> (That didn't work very well)
00:39:47 <quintopia> hppavilion[1]: \oren\ and everyone else at that address
00:39:57 <hppavilion[1]> quintopia: Oh?
00:40:05 <hppavilion[1]> quintopia: I don't know where \oren\ lives
00:40:08 <hppavilion[1]> (Unfortunately)
00:40:17 <hppavilion[1]> boily: Do you have \oren\'s GPS coordinates on file?
00:40:22 <quintopia> hppavilion[1]: "our mail slot" implies multiple recipients
00:40:28 <hppavilion[1]> Ah
00:40:38 <boily> hppavilion[1]: there you go: 43.6559769, -79.4141707
00:40:53 <hppavilion[1]> I'm a bit scared it's a joke, but checking anyway
00:40:58 <boily> maybe there are multiple \orens\?
00:41:21 <quintopia> boily: oren and \\oren\\ and /oren/ and //oren//
00:41:44 <\oren\> it isn't a joke, and I'm the only one currently living here
00:42:01 <hppavilion[1]> One side of the street has...
00:42:05 <hppavilion[1]> What, a duplex?
00:42:10 <hppavilion[1]> Looks too big to be a duples
00:42:12 <hppavilion[1]> *duplex
00:42:18 <quintopia> \oren\: oh, then secret admirer seems likely!
00:42:19 <hppavilion[1]> (like, the individual sides)
00:42:26 <hppavilion[1]> Ah
00:42:27 <quintopia> \oren\: or someone who wants you to think that
00:42:28 -!- MoALTz has quit (Ping timeout: 272 seconds).
00:42:35 <hppavilion[1]> duhn... duhn... DUHN
00:42:43 <\oren\> I'm in a two story house with a pine tree in the front
00:42:58 <\oren\> it is attached to the house on both sides of it
00:43:32 <hppavilion[1]> The other side of the street has condos or apartments or something...
00:44:50 <\oren\> that's the old folks home that has a fire every few days
00:44:58 <hppavilion[1]> Target acquired
00:45:02 <hppavilion[1]> Launching
00:45:45 <boily> hppavilion[1]: hey, careful with that ICBM. a little bit off and it'd hit one of our Magnificent Crumbling Buildings.
00:46:16 <hppavilion[1]> Detonation confirmed
00:46:20 <hppavilion[1]> Summary: http://bit.ly/29Phz2C
00:46:43 <hppavilion[1]> \oren\: The first time I read that I thought it was "a few days ago"
00:46:48 <hppavilion[1]> But, every few days, really?
00:47:31 <\oren\> yeah because they won't stop smoking
00:48:07 <hppavilion[1]> Ah
00:48:21 <hppavilion[1]> (For clarity, in this scenario North Korea /really/ doesn't like \oren\ in particular)
00:49:19 -!- adu has joined.
00:55:16 -!- adu has quit (Quit: adu).
01:00:49 -!- function has quit (Quit: 1 found in /dev/zero).
01:03:17 -!- tromp_ has joined.
01:28:51 <tswett> Intercontinental Crumbling Building Missile.
01:29:31 <quintopia> i had assumed that was the joke yes
01:30:13 <tswett> I had assumed that that was a coincidence that nobody else had yet noticed.
01:30:41 <tswett> So I'm pretty sure that in my silly little ordinal notation, {{{{}:{}}:{}}:{}} is an extremely large ordinal.
01:30:57 <shachaf> Is that surreal numbers?
01:31:03 <tswett> No, that's ordinal numbers.
01:31:28 <tswett> Each notation is a dictionary with finitely many key-value pairs, where each key and each value is a notation.
01:31:42 <boily> `learn ICBM/ICBMs are Crumbling Building Missiles. The I is currently unknown.
01:31:57 <HackEgo> Learned 'icbm/icbm': ICBM/ICBMs are Crumbling Building Missiles. The I is currently unknown.
01:32:10 <shachaf> Doesn't the I stand for Intercontinentalballisticmissile?
01:32:18 <tswett> {} is 0. {0:0} is 1.
01:32:20 <shachaf> `sedlast s/unknown/classified/
01:32:27 <HackEgo> wisdom/icbm/icbm//ICBM/ICBMs are Crumbling Building Missiles. The I is currently classified.
01:32:34 <tswett> So the concise way of writing that notation above is {{1:0}:0}.
01:33:53 -!- groteworld has joined.
01:34:58 <izabera> can we talk about how well designed is argument parsing in sh?
01:35:39 <tswett> Yes, but not in Sanskrit.
01:35:45 <tswett> We can talk about that in any other language.
01:35:45 <izabera> ok
01:35:51 <izabera> let's go with english
01:36:00 <quintopia> engli sh?
01:36:03 <boily> tswellott. bad sanskrit past experiences?
01:36:20 <izabera> -o and -c take a parameter, but they take the first "free" argument as their parameter
01:36:35 <izabera> like this sh -of vi is the same as sh -f -o vi
01:37:19 <izabera> sh -oovi vi vi is the same as sh -v -i -o vi -o vi
01:38:08 <izabera> also most sh options can be enabled with + instead of - , like this sh +o vi
01:38:29 <izabera> but -c isn't one of those options, so this should be invalid sh +oc vi :
01:38:49 <izabera> yet it's valid in several shells because this shit is fucked up
01:40:13 <izabera> and you can't have sh -co : vi because for some reason this must consider : as the option name and vi as the command
01:41:05 <izabera> so in that case you need sh -co vi :
01:41:14 <izabera> even if -c comes first
01:42:08 <izabera> and bash adds an additional set of options that are enabled with -O with the same semantics as -o because it worked so well
01:42:21 <izabera> who thinks this is shit?
01:44:07 -!- groteworld has quit (Quit: ZZZzzz…).
01:44:08 <boily> izbellora. it is quaint and pittoresque. looks good from afar :D
01:44:21 <izabera> hoily
01:46:07 <tswett> This is the way it should work:
01:46:31 <boily> damn. once again I typoed the izaberaporthello...
01:46:54 <tswett> {command: sh, arguments: {f: nil, o: vi}}
01:48:08 <izabera> i'm not saying you should use json to start a shell..
01:48:09 -!- groteworld has joined.
01:49:19 <izabera> it should just parse arguments and options like every other utility -_-
01:49:41 <izabera> sh -fovi == sh -f -o vi
01:49:42 -!- groteworld has quit (Client Quit).
01:50:19 <shachaf> @tell pikhq The HackEgo thing is fixed. You can @tell oerjan that you don't need to be told to unignore it.
01:50:19 <lambdabot> Consider it noted.
01:50:37 <tswett> I'm one of those extremists who believes that commands shouldn't parse their own arguments.
01:51:05 -!- groteworld has joined.
01:51:26 <shachaf> Parsing is scow.
01:51:49 <izabera> parsing is necessary
01:52:06 <shachaf> Sort of.
01:52:35 <izabera> if they don't parse their command line arguments, you can't have a unified interface to call them
01:52:49 <shachaf> Just the opposite.
01:53:02 <izabera> what do you mean?
01:53:28 <shachaf> I assume tswett means that the shell or something will have one parser for all commands.
01:53:40 <shachaf> And the commands get their arguments in structured form rather than a bunch of bytes.
01:54:00 <tswett> Yeah, that.
01:54:55 <izabera> so your idea is to put in the shell a super complex parser that handles every possible way to call any program
01:55:01 <shachaf> Yes.
01:55:18 <shachaf> Parsers are hard and error-prone, so you might as well handle parsing well in one place.
01:55:43 <shachaf> The problem with UNIX command line programs is that they serve two functions: User interface and API.
01:56:12 <shachaf> Just about anything that tries to be both UI and API is doomed to be terrible at at least one.
01:56:47 <izabera> but that's why stuff like getopt() exist
01:56:54 <izabera> so they don't have to write their own parsers
01:57:05 <shachaf> But it's such a mess.
01:57:24 <shachaf> You have things like rm $file, where $file happens to start with a -.
01:58:00 -!- boily has quit (Quit: PURCHASING CHICKEN).
01:58:38 <tswett> Look, a quit message of boily's that's something that makes sense as a quit message.
02:00:10 <zzo38> You should then use the -- if a filename with - at first but it would have been better if the filename does not start with - anyways. I use shell script with #!/bin/bash -- for this reason.
02:01:21 <tswett> People don't consistently put in -- when they're deleting a file with an unknown name.
02:01:22 <shachaf> Yes, of course you can work around it.
02:01:34 <shachaf> But if it was structured, you wouldn't need to work around it.
02:01:39 <tswett> It's easy to accidentally get it wrong.
02:02:00 <izabera> it's easy because you're not in #bash
02:02:01 <tswett> Generally, I don't like it when a computer system makes it easy to accidentally get it wrong.
02:02:04 <shachaf> If unlink() had a different behavior when a file name started with '-', people would be mad.
02:02:32 <shachaf> izabera: Obviously it's possible to train yourself to do it.
02:02:47 <shachaf> It's not literally impossible to write correct code using UNIX utilities.
02:02:51 <shachaf> But it's not a good API for programming
02:03:04 <izabera> it's fine
02:03:36 <tswett> I also don't like it when a computer system is "fine".
02:04:00 <shachaf> spot of the stockholm syndrome, governor?
02:04:07 <tswett> If something can reasonably be improved, it should be.
02:04:08 <zzo38> I think it is working OK, although the filename ought to don't start with a minus sign anyways.
02:04:29 <shachaf> If by "it's fine" you mean that it's not worth the effort to change, I agree.
02:04:40 <shachaf> It would be a big project.
02:04:47 <pikhq> shachaf: IMO the worst part is that things fail to implement Unix utilities to match the spec, which means that writing correct code with it is even *harder* than it ought to be.
02:05:08 <shachaf> pikhq: Well, it's not as if a utilty author's life is easy.
02:05:20 <shachaf> A structured API would make life easy for both parties.
02:05:21 <izabera> except when the spec is idiotic, e.g. how sh parses options ^
02:05:38 <shachaf> utility
02:05:57 <shachaf> Anyway, I'll just say that izabera doesn't like change.
02:06:05 <izabera> no that's not true
02:06:06 <zzo38> An alternative would be for the shell to parse the switches and pass as arguments to the program so that it will not cause a problem based on what the filename is.
02:06:08 <shachaf> That's a pretty rude thing to say, actually.
02:06:09 <pikhq> For instance, POSIX argument parsing is a little less fragile than GNU argument parsing...
02:06:10 <shachaf> I'll retract it.
02:06:30 <pikhq> In POSIX, you stop parsing flags on the first string that doesn't start with -, or after "--".
02:06:40 <pikhq> In GNU, you stop parsing flags after "--".
02:06:41 <shachaf> Can you believe that the Windows shell doesn't expand *? It's the program's responsibility to do it.
02:06:42 <zzo38> I do think the POSIX argument parsing is better than the GNU argument parsing, at least.
02:07:00 <pikhq> ... Which means rm * is even *more* likely to do weird shit.
02:07:05 <shachaf> pikhq: Personally I like being able to ls / -l
02:07:22 <shachaf> I'm a gnu / agnother gnu
02:07:33 <izabera> agnother lol
02:07:44 <pikhq> IMO the worst part, again, is this is a deviation from spec and from common behavior that you basically just have to know.
02:08:18 <pikhq> Unix is bad enough without things make nominally correct code break because you extend the semantics.
02:08:56 <zzo38> Making the shell to be error message if a filename has - at first might be one option that you might want to add too
02:09:47 <pikhq> ... Also weird is the weird insistence that the interactive and scripting shells be the same, or that the interactive shell should be usable for scripting...
02:10:11 <izabera> why is that bad?
02:10:26 <zzo38> But making it different would require to program the difference and cause problem when the program is not written in this way.
02:10:38 <pikhq> I mean granted, you don't want a lot of pointless differences. But a lot of the things that an interactive shell could do to make the UI less bad would break non-interactive use.
02:10:39 <zzo38> Making it same way would be simpler I think, so therefore it is better.
02:10:40 <\oren\> pikhq: I use different shell for interaction and scripting
02:10:51 <shachaf> pikhq: That's what I was saying about UIs and APIs above.
02:11:21 <\oren\> Spefically, I use bash for scripting and midnight commander for interatiocn
02:11:34 <izabera> mc counts as a shell?
02:11:39 <pikhq> e.g. an interactive shell could see you write "rm *" and execute "rm -- *".
02:11:53 <\oren\> of course it counts
02:12:01 <izabera> cough cough rm () { command rm -- "$@"; }
02:12:47 <shachaf> But that rm doesn't support flags.
02:13:10 <zzo38> The alternative I mentioned would be that the interactive shell will display an error message if a wildcard expands to a filename that begins with a minus sign by default.
02:13:16 <tswett> rm: cannot remove ‘-rf’: No such file or directory
02:13:17 <izabera> ok i'm not writing the one that expands the glob correctly *and* supports options, in one line on irc
02:13:39 <deltab> or glob could prefix with ./
02:13:42 <pikhq> There's also no particular reason that an interactive shell has to work like the traditional Unix shell in most any way...
02:13:44 <shachaf> This is the sort of mindset that says that the way to solve SQL injection is to be extra careful.
02:13:57 <deltab> as suggested in http://www.dwheeler.com/essays/fixing-unix-linux-filenames.html#dashes
02:14:06 <shachaf> Just gotta mysql_real_escape_string everything.
02:14:09 <pikhq> Literally the only reason behind keeping the traditional Unix shell the way it is, is that you break shell scripts if you change it.
02:14:13 -!- groteworld has quit (Quit: ZZZzzz…).
02:14:17 <izabera> shachaf: what mindset?
02:14:31 <pikhq> But holy *shit* is it a rough UI experience.
02:14:42 <tswett> Now I'm trying to think why it is that PowerShell is both an interactive shell and a scripting shell.
02:15:13 <tswett> Well... I definitely want a PowerShell interactive shell, and I definitely want a PowerShell scripting shell.
02:15:23 <deltab> izabera: if you changed it so that it allowed options, it would allow *unintended* options, and you'd have the original problem again
02:15:29 <tswett> And it's easier for Microsoft to make just one thing than two.
02:15:39 <pikhq> PowerShell has the slight advantage of the preexisting scripting environment on Windows being complete shit.
02:16:15 <pikhq> ... And the larger advantage of being able to ship the brand new thing and get fairly wide adoption.
02:16:31 <zzo38> SQL has host parameters if you need to call SQL codes from another programming language and include the string data from the host.
02:16:51 <pikhq> Unix shell scripting, uh, there's several different OSes which use it.
02:16:52 <zzo38> If you are programming directly in SQL then you don't have problems with injection.
02:17:17 <pikhq> And using it fairly heavily for scripting.
02:17:54 <zzo38> String escaping in SQL is supposed to be by doubling each apostrophe that appears (I think it is the same as Pascal), and this is what SQLite implements, but other systems do escaping differently and don't work so well.
02:18:11 <izabera> btw most of what all of you are saying reads as "i want to stay ignorant of how shells work and just paste in awful code snippets i found on the internet" like all the code in http://www.dwheeler.com/essays/fixing-unix-linux-filenames.html#dashes
02:19:45 <tswett> I want tools whose behavior is simple so that I don't have to learn a lot in order to be able to use them well.
02:20:17 <izabera> it's code
02:20:25 <izabera> you want to learn how to write python before writing python
02:20:40 <pikhq> izabera: I am not saying that, I'm just saying the UI is terrible and it should not require mastery of it to avoid shooting your foot.
02:21:08 <shachaf> izabera: I know how to write correct code in bash.
02:21:23 <shachaf> I'm complaining *because* I know how.
02:21:27 <pikhq> ^
02:21:39 <izabera> `` touch ./-rf; GLOBIGNORE=-*; echo rm *
02:21:48 <HackEgo> rm !\.´ advice bin canary candide cdescs emoticons esobible etc evil factor good .hg .hg_archival.txt .hgignore hw ibin interps karma le lib ls misle out paste ply-3.8 ps quines quotes share src theorems tmflry tmp wisdom wisdom.pdf
02:22:00 <izabera> ^ see? it didn't expand to -rf
02:22:10 <shachaf> But it's not deleting the file -rf either.
02:22:13 -!- Phantom__Hoover has quit (Read error: Connection reset by peer).
02:23:00 <izabera> `rm ./-rf
02:23:05 <HackEgo> No output.
02:23:19 <tswett> Fwiw, in PowerShell, if you want to delete all files in the current directory, that's...
02:23:21 <tswett> ls | rm
02:23:23 <tswett> Or, for long:
02:23:28 <tswett> Get-ChildItem | Remove-Item
02:23:41 <shachaf> What if I want to create all the files that don't exist?
02:23:50 -!- Hoolootwo has left ("Leaving").
02:23:51 <izabera> whawt
02:23:57 <izabera> s/w//2
02:24:33 <tswett> Actually, I'm pretty sure you can just do:
02:24:33 <tswett> *
02:24:36 <tswett> Er.
02:24:38 <tswett> rm *
02:24:56 <tswett> And that will be interpreted as "remove all items which match the wildcard string '*'".
02:25:24 <tswett> PowerShell doesn't expand globs... ever, I think.
02:25:41 <izabera> so that leaves the burden on each program
02:26:06 <shachaf> I don't mind the shell expanding globs.
02:26:12 <shachaf> But it ought to be done the right way.
02:26:52 <izabera> you should just start to use something like -- whenever you expand a glob
02:26:53 <tswett> izabera: yeah, I suppose it does.
02:27:11 <tswett> I think PowerShell would also let you do this:
02:27:12 <tswett> rm (ls)
02:27:30 <tswett> List the contents of the current directory and delete that.
02:27:35 <pikhq> shachaf: Something where it understands you writing the string "rm * -rf" and executes {"rm", "-rf", "--", results of glob here, 0}?
02:28:09 <shachaf> Something where you pass more structured data between the shell and rm.
02:28:21 <shachaf> I don't know what the details should be like.
02:28:31 <pikhq> Mmkay, that's a little tricky without changing some huge details of Unix.
02:28:40 <shachaf> Yes, of course I'm talking about changing everything.
02:28:45 <pikhq> My interpretation is the closest you can do while only changing the shell.
02:28:48 <pikhq> Right.
02:28:49 <izabera> that becomes overly verbose for interactive use
02:28:50 <tswett> Remove-Item (Get-ChildItem) -Recurse -Force
02:29:03 <shachaf> Yes, it's a tricky UI/language design problem.
02:29:11 -!- Hoolootwo has joined.
02:29:42 <shachaf> The UNIX approach is at one point in a big space of potential solutions.
02:30:12 <shachaf> I don't think it's a particularly good point, though it has some important advantages that other solutions should strive for.
02:30:14 <pikhq> With the major advantage of it (which is big, mind you) is that it already is implemented and popular.
02:30:23 <zzo38> What I intend to do for my new kind of computer design is the main command line interface is just Forth.
02:30:40 <izabera> so user friendly
02:30:44 <tswett> Personally, I prefer PowerShell over Forth.
02:32:19 <shachaf> I think this whole discussion is an example of Alan Kay's answer in https://news.ycombinator.com/item?id=11940276
02:32:35 <zzo38> Normally you would not need this interface though since you can just insert the CD/DVD and then push START and it will start.
02:36:31 -!- lambda-11235 has joined.
02:37:59 <lambda-11235> Yeah, I wrote an LC compiler where parenthesis are only allowed with and required for application.
02:38:40 <lambda-11235> So (\x. x x) (\x. x x) is illegal syntax, but (\x. (x x) \x. (x x)) is legal.
02:38:51 <tswett> Hey there.
02:38:59 <tswett> I think I like it.
02:40:21 <lambda-11235> tswett: The nice thing is that using that rule precedence doesn't matter.
02:42:36 -!- jaboja has quit (Ping timeout: 246 seconds).
02:45:06 -!- hppavilion[1] has quit (Read error: Connection reset by peer).
02:52:03 -!- JenElizabeth8 has joined.
03:01:14 <izabera> hercules has cut over 200k heads and hydra is still going strong
03:06:19 <tswett> But you cannot lose!
03:06:36 <izabera> actually i can, my lifetime is limited
03:10:40 <HackEgo> [wiki] [[WhoScript]] https://esolangs.org/w/index.php?diff=48886&oldid=46034 * MCS-Kaijin * (+57)
03:12:30 -!- augur has quit (Remote host closed the connection).
03:19:38 -!- tromp_ has quit (Remote host closed the connection).
03:20:10 -!- spiette has quit (Ping timeout: 272 seconds).
03:20:57 -!- Kaynato has joined.
03:47:31 -!- adu has joined.
03:57:46 -!- augur has joined.
04:24:27 -!- tromp_ has joined.
04:32:11 <myname> i am disappointed by the "dr who theme" of whoscript
04:32:48 <myname> it's basically renaming stdin/stdout and giving jumps and if another name
04:33:14 <myname> it doesn't even feature rebirth
04:34:15 <alercah> it should make every jump go to a random instruction near to, but never exactly on, the targeted instruction
04:34:53 <myname> that would just result in nopslides everywhere
04:35:48 <myname> you may have to trick a little if all nops you can do are actually composed commands, but it's still possible
04:36:37 <alercah> nops are forbidden
04:36:53 <myname> you cannot do,that, i th7nk
04:37:35 <myname> i am not sure if a langiage without the possibility to produce nops is turing complete
04:37:49 <alercah> you might be able to produce compound nops
04:37:56 <alercah> but then starting in the middle of one won't necessarily be a nop
04:38:11 <myname> well, yeah, but you might solve that
04:38:30 <alercah> also the degree to which you are off should clearly be unlimited
04:38:44 <myname> like, having a nop composed of increments and decrements of a var, compare with another at the end of the slide, fix accordingly
04:38:46 <alercah> so that no amount of nop sliding is big enough to guarantee you don't end up stuck in an asteroid
04:39:03 <myname> or just use a var that is of no use whatsoever and just increase it
04:39:47 <Hoolootwo> yeah, there would be lots of ways to do pseudo-nops
04:40:36 <myname> i would make a language that _needs_ rebirths, but every time it does so, it is a bit unclear about its state
04:41:16 <myname> also, there always need to be a companion thread
04:42:22 <myname> dr who does remind me a bit of interactive proofs
04:48:37 <lambda-11235> What about a language that could reset its state to an earlier point in time?
04:49:34 <izabera> that's what lisp does
04:51:03 <myname> call/cc everywhere?
04:52:40 <Hoolootwo> you can also do that in C-INTERCAL with backtracking
04:53:06 <Hoolootwo> http://catb.org/esr/intercal/ick.htm#Backtracking
04:59:33 -!- augur has quit (Remote host closed the connection).
05:00:08 -!- augur has joined.
05:01:24 -!- tromp_ has quit (Remote host closed the connection).
05:05:00 -!- augur has quit (Ping timeout: 258 seconds).
05:05:11 -!- adu has quit (Quit: adu).
05:35:03 -!- Kaynato has quit (Ping timeout: 240 seconds).
06:07:51 -!- adu has joined.
06:15:17 -!- tromp_ has joined.
06:17:24 -!- lambda-11235 has quit (Quit: Bye).
06:19:33 -!- tromp_ has quit (Ping timeout: 240 seconds).
06:33:47 -!- adu has quit (Quit: adu).
06:35:53 -!- Akaibu has joined.
06:39:14 -!- oerjan has joined.
06:42:22 <oerjan> @messages-foul
06:42:22 <lambdabot> shachaf asked 9h 40m 7s ago: to remind me to /unignore HackEgo when the spam issue is fixed twh
06:42:22 <lambdabot> pikhq asked 9h 34m 27s ago: me too
06:42:22 <lambdabot> izabera asked 9h 34m 3s ago: me three
06:42:22 <lambdabot> izabera asked 9h 16m 51s ago: no need to remind me anything
06:42:22 <lambdabot> shachaf said 8h 12m 58s ago: never mind tdh
06:42:51 <oerjan> pikhq: please unignore HackEgo
06:42:54 <oerjan> (sheesh)
06:42:58 <pikhq> It is done.
06:44:02 <shachaf> oerjan: as if you don't like getting messages
06:44:15 <shachaf> if you didn't want foul ones you should've just used another command
06:45:25 <oerjan> i already knew they were foul, that's why i used it hth
06:45:43 <shachaf> self-foulfilling prophecy
06:47:23 <oerjan> sounds like it was a good time to be away, anyway.
06:47:39 <shachaf> what, you're not logreading?
06:48:04 <oerjan> yes i am. it was very good for me to be away from such a mess.
06:48:32 <shachaf> You gotta read between the lines.
06:48:50 <shachaf> Maybe download the logs and filter out the HackEgo lines.
06:48:58 <oerjan> i haven't got much beyond searching for my own nick yet.
06:49:09 <oerjan> also, i seem to have email too.
06:49:18 <oerjan> (agora has reawakened)
07:15:07 <Sgeo> oerjan, ooh?
07:17:08 <shachaf> hm
07:17:18 <shachaf> i had a good pun with the word "swat" but i've forgotten it
07:17:19 <shachaf> it's a shame
07:17:26 <shachaf> it would have earned me a proper swatting
07:26:11 -!- augur has joined.
07:30:26 -!- augur has quit (Ping timeout: 250 seconds).
07:36:31 -!- augur has joined.
07:49:00 <oerjan> Sgeo: there is activity on the mailing lists, and several people have reregistered.
07:49:12 <Sgeo> I should take a look tomorrow
07:50:24 -!- rdococ has joined.
08:08:01 -!- hppavilion[1] has joined.
08:27:53 -!- AnotherTest has joined.
08:35:16 <int-e> oerjan: Sounds like an NP problem (namely: guess the right phrase)... you should use a SWAT solver.
08:39:42 <oerjan> int-e: did you mean to ping shachaf
08:40:13 * shachaf swats int-e -#
08:49:38 <int-e> oerjan: MAYBE!
09:03:12 -!- tromp_ has joined.
09:07:06 <hppavilion[1]> I detect NP problems
09:07:51 -!- tromp_ has quit (Ping timeout: 276 seconds).
09:15:27 <shachaf> I detect solutions to NP problems.
09:17:31 <int-e> oerjan: somehow, to my mind, you emit a stronger "pun" signal than shachaf does.
09:17:55 <shachaf> oerjan has been punning since before I was born.
09:18:00 <shachaf> He even puns unintentionally.
09:18:22 <hppavilion[1]> And that's saying something; shachaf is at LEAST 81 years old
09:18:36 <shachaf> `? oerjan
09:18:49 <HackEgo> Your mysterious articled cackling zombie øverlord kommisjonær emeritus oerjan is a lazy expert in future computation. Also a Precambrian Norwegian who mildly dislikes Roald Dahl with a pasjon. Lately when he tries to remember a word, "amortized" pops up. His arch-nemesis is Betty Crocker. He sometimes puns without noticing it.
09:18:56 <shachaf> @wn precambrian
09:18:56 <hppavilion[1]> Say what you like about puns, but they can be a very rewording experience
09:18:57 <lambdabot> *** "precambrian" wn "WordNet (r) 3.0 (2006)"
09:18:57 <lambdabot> Precambrian
09:18:57 <lambdabot> n 1: the eon following the Hadean time and preceding the
09:18:57 <lambdabot> Phanerozoic eon; from about 3,800 million years ago until
09:18:57 <lambdabot> 544 million years ago [syn: {Precambrian}, {Precambrian
09:18:59 -!- augur has quit (Remote host closed the connection).
09:18:59 <lambdabot> eon}, {Precambrian aeon}, {Precambrian period}]
09:19:40 <shachaf> int-e: do you have any fancy things like goodstein's theorem for me
09:20:05 <int-e> shachaf: there was a hydra yesterday...
09:20:24 * int-e needs to catch a bus
09:21:10 <shachaf> well, that hydra was the same thing
09:21:10 * oerjan seems to have caught it http://www.madore.org/~david/math/hydra.xhtml
09:21:28 <shachaf> i just mean any fancy theorems or fast-growing functions or that sort of thing
09:23:57 <b_jonas> int-e: about the pentagon from yesterday, I said under 1e9 first four primitive steps, but that was stupid, there are much less because the compass collapses.
09:24:00 <b_jonas> I'll have to recompute it
09:24:28 <b_jonas> also, I think even with the flaw you pointed out, it might be possible to brute-force it, because I must still get _something_ useful after 4 steps
09:29:15 -!- Sgeo_ has joined.
09:32:34 -!- Sgeo has quit (Ping timeout: 272 seconds).
09:55:14 <oerjan> <shachaf> A channel op could KickEgo <-- the set of active channel ops is alas ~= the set of active wiki admins.
09:55:34 <shachaf> oerjan: yes, i lamented that fact hth
09:56:47 <b_jonas> M:tG: wait, so in Eldritch Moon they print a creature whose name ends in " Standard Bearer". Guess what ability it has? No, not the "Standard Bearer" ability. The spellshaper ability. It's not even a spellshaper by type.
09:56:58 <oerjan> huh lament is still there.
09:57:26 <oerjan> shachaf: technically lament may have been on freenode. he's now, anyway.
09:57:44 <shachaf> oerjan: whoa whoa whoa
09:57:44 <oerjan> whether he would be helpful is a different matter.
09:57:57 <shachaf> you contracted the is which would have been emphasized
09:58:06 <shachaf> the "is"
09:58:09 <oerjan> wat?
09:58:28 <shachaf> "he's now"
09:58:46 <shachaf> but he isn't. maybe you meant "not now" or something
09:58:48 <oerjan> i emphasized the now.
09:58:59 <oerjan> oh he is. just under a slightly different nick.
09:59:08 <oerjan> same account though.
09:59:18 <shachaf> true, it's not emphasized
09:59:23 <shachaf> but that's pretty bizarre
09:59:42 <shachaf> there must be some grammar rule that forbids it
10:00:43 <oerjan> couldn't've been
10:04:26 <oerjan> also, ais523 is the admin that knows how to actually fight that kind of spam afaik.
10:05:56 <oerjan> it's good fizzie showed up. i suspect i wouldn't have had enough powers or knowledge to stop it.
10:10:01 <hppavilion[1]> I just saw a few seconds of a video in 4K
10:10:09 <hppavilion[1]> In other news, ALL GLORY TO THE HYPNOTOAD
10:15:10 <shachaf> We need to give fizzie a pager.
10:16:28 -!- MoALTz has joined.
10:17:59 <fizzie> shachaf: Will I get paid extra when I'm on-call?
10:18:17 <shachaf> fizzie: You'll get paid with extra vacation from #esoteric.
10:20:26 <fizzie> A spam alerting thing might make sense though -- was pure luck I happened to glance at the channel when I did. Wonder if there'd be a reasonable way to set one up.
10:24:17 <shachaf> Are you on an oncall rotation at work?
10:24:23 <shachaf> I've forgotten your username.
10:25:31 <fizzie> Not really. I mean, we use oncall for managing some triage-style rotations, but that's not the same thing.
10:26:05 <shachaf> I think you've said it before but maybe it was someone else.
10:27:11 <fizzie> There's been talking of having a more on-call oncall so that "our" SREs have someone to talk to if they get lonely (or have a problem, I guess).
10:27:17 <fizzie> It's just my last name.
10:28:05 <shachaf> Ah, too easy.
10:28:36 <shachaf> whoa whoa whoa, fizziecoin
10:28:41 -!- Phantom_Hoover has joined.
10:29:06 <fizzie> Weren't you supposed to be an *ex*-Googler?
10:29:08 <shachaf> `? fizzie
10:29:15 <HackEgo> fizzie is not fnord with a monad but the sneaky king of #esoteric, see http://codu.org/projects/hackbot/fshg/index.cgi/raw-file/tip/src/fizziecoin.jpg
10:29:39 <izabera> i need to upgrade hercules
10:29:48 <izabera> how do i give him a machine gun or something?
10:29:58 <oerjan> fizzie: he's not supposed to a lambdabot admin either, but...
10:30:01 <oerjan> *to be
10:30:02 <izabera> 600k heads so far
10:30:25 <shachaf> fizzie is the sneaky one, according to HackEgo.
10:30:30 <oerjan> izabera: maybe not use the auto?
10:30:37 <izabera> BUT NOW I HAVE TO
10:30:52 <oerjan> shachaf: you're just so sneaky you've removed the evidence hth
10:31:27 <shachaf> Anyway this information is public.
10:31:36 <fizzie> Oh, right.
10:32:47 <fizzie> There's a Redis Pub/Sub variant of the update feed we're piping to HackEgo, maybe there'd be some low-effort way of getting that through GCM into notifications on my phone.
10:33:08 <fizzie> It's not like it's (usually) that active.
10:33:48 <fizzie> Also I could then publish whatever I use in the Play Store, and anyone could get esolangs updates into their phone.
10:34:23 <shachaf> Anyone with an Android phone.
10:34:33 <shachaf> Which is anyone who counts, I suppose.
10:34:38 <fizzie> Close enough.
10:34:41 <fizzie> I'm sure that sort of app would have an audience in the ones.
10:35:00 <shachaf> Can't you just use an RSS feed or something?
10:35:15 <fizzie> That'd be laggy.
10:35:17 <shachaf> I don't think your SLA has to be that good.
10:36:23 <fizzie> The cleanup time is linear to the time they've spent spamming, unless I manage to install one of those dodgier spam-cleaning extensions or go through the trouble of getting some of the hacky batchy things working.
10:36:55 <b_jonas> `wisdom
10:36:58 <b_jonas> `wisdome
10:37:03 <HackEgo> ​/home/hackbot/hackbot.hg/multibot_cmds/lib/limits: line 5: exec: wisdome: not found
10:37:05 <HackEgo> thwackamacallit//A thwackamacallit is like a whatchamacallit, but more painful. See mapole.
10:37:22 <b_jonas> `wisdom
10:37:26 <HackEgo> ​#esoteric//#esoteric is the only channel that doesn't exist. After monqy left it became slightly off-centër. It's about 30 m (100 ft) across. oerjan seems to be making a lawn in the northern part, but it keeps getting dug up by free ranging moons. May contain crude drawings of nuts.
10:37:28 <b_jonas> `random-card
10:37:32 <HackEgo> Hedron Matrix \ 4 \ Artifact -- Equipment \ Equipped creature gets +X/+X, where X is its converted mana cost. \ Equip {4} \ ROE-R
10:38:41 <fizzie> Although I have a vague notion Chrome on Android does push notifications now.
10:38:55 <shachaf> `` F="$(find wisdom -type f | shuf -n 1)"; sed 's#\. #\.#' "$F"
10:38:58 <HackEgo> flower.what IS a flower?
10:39:12 <shachaf> `? flower
10:39:14 <HackEgo> flower. what IS a flower?
10:40:32 <shachaf> `cwlprits flower
10:40:40 <HackEgo> int-e ais523 Bike FreeFull oerjan FreeFull shachaf shachaf nitia
10:40:59 <shachaf> `hoag wisdom/flower
10:41:03 <HackEgo> ​<int-e> revert accbc9c5c7ec \ <ais523> echo wisdom/* | shuf | head -n 10 | xargs rm \ <Bike> revert \ <FreeFull> for x in wisdom/*; do rev "$x" > "$x"a; mv "$x"a "$x"; done \ <oerjan> revert \ <FreeFull> run rm -rf wisdom \ <shachaf> revert 0 \ <shachaf> run rm -rf wisdom/* \ Initial import.
10:41:14 <shachaf> man
10:41:17 <shachaf> that's a lot of vandalism
10:42:15 <fizzie> I did find https://pushassist.com/push-notifications-mediawiki/ but that's so commercial.
10:42:23 <fizzie> "Increase engagement and drive more repeat traffic to your MediaWiki site with desktop & Android (Chrome & Firefox) push notifications."
10:42:55 <shachaf> Can't you just have it send you an email?
10:43:46 <shachaf> Or a page, if you're set up for that.
10:46:41 <fizzie> I know MediaWiki can send emails for changes to pages on a specific watch list. Not sure if I can convince it to send emails for everything.
10:47:17 <shachaf> How are changes being sent to HackEgo?
10:47:42 <shachaf> You can just write a bot that monitors HackEgo's messages in #esoteric
10:47:47 <fizzie> $wgUsersNotifiedOnAllChanges
10:47:54 <fizzie> There's a flag for that, apparently.
10:48:09 <fizzie> "Array of usernames who will be sent a notification email for every change which occurs on a wiki. The user will not be sent an email for changes that they themselves make."
10:48:15 <fizzie> Well, I could just enable that, then.
10:49:37 <fizzie> For HackEgo, we use the $wgRCFeeds mechanism, which does Redis Pub/Sub pushes and raw UDP (one packet/edit). We use the latter, plus socat UDP-RECV:8147,bind= STDOUT | stdbuf -oL cut -c 1-400 | stdbuf -oL sed -e 's/^/PRIVMSG #esoteric :[wiki] /' | socat STDIN UNIX-SENDTO:/tmp/multibot.HackEgo to smuggle them to IRC.
10:50:32 <shachaf> But then you won't get notified if you spam the wiki.
10:54:23 <b_jonas> shachaf: you do if you use two different users and add both to the list
11:17:31 -!- hppavilion[1] has quit (Ping timeout: 240 seconds).
11:33:20 -!- boily has joined.
11:34:42 <boily> `wisdom
11:35:10 <HackEgo> soviet russia//¯\(°​_o)/¯ soviet russia?
11:47:09 <b_jonas> `bardsworthlist 2016-07-14
11:47:17 <oerjan> hily
11:47:20 <HackEgo> bardsworthlist 2016-07-14: b_jonas
11:48:26 <b_jonas> `? sarlacc
11:48:35 <HackEgo> sarlacc? ¯\(°​_o)/¯
11:48:45 <shachaf> `cat bin/listlist
11:48:49 <HackEgo> ​#!/bin/sh \ set -e \ export LANG=C \ cd /hackenv/bin;exec ls -dF *[lL]ist*
11:48:53 <shachaf> hm
11:49:00 <shachaf> I want a notification list for when new lists are created.
11:49:11 <shachaf> `cat bin/mklist
11:49:13 <HackEgo> cat: bin/mklist: No such file or directory
11:49:23 <shachaf> `cat bin/makelist
11:49:24 <HackEgo> file="bin/$1"; shift; cp bin/emptylist "$file"; for n in "$@"; do echo "$n" >> "$file"; done
11:49:54 <shachaf> `makelist makelistlist
11:50:04 <HackEgo> No output.
11:50:09 <shachaf> `` echo shachaf >> bin/makelistlist
11:50:14 <HackEgo> No output.
11:51:03 <b_jonas> shachaf: then add yourself to listlist
11:51:13 <b_jonas> I'll try to remember that list next time I create a list
11:52:02 <oerjan> shachaf: hey, `makelist takes arguments now hth
11:52:12 <shachaf> oerjan: i remembered too late
11:52:18 <shachaf> didn't someone tell me about remembering too late the other day
11:52:54 <shachaf> `sled bin/makelist//s/.\{13\}/name="$1"; file="bin/$name"; makelistlist "$name"/
11:52:58 <HackEgo> sed: -e expression #1, char 32: unknown option to `s'
11:53:08 <shachaf> `sled bin/makelist//s/.\{13\}/name="$1"; file="bin\/$name"; makelistlist "$name"/
11:53:12 <HackEgo> bin/makelist//name="$1"; file="bin/$name"; makelistlist "$name"; shift; cp bin/emptylist "$file"; for n in "$@"; do echo "$n" >> "$file"; done
11:53:20 <shachaf> `makelist makelistlist shachaf
11:53:25 <HackEgo> makelistlist makelistlist shachaf: shachaf
11:53:40 <shachaf> oops
11:54:00 <shachaf> Arguments don't work with the regular running syntax.
11:54:02 <shachaf> `revert
11:54:06 <shachaf> Then what's the point?
11:54:19 <HackEgo> rm: cannot remove `/home/hackbot/hackbot.hg/multibot_cmds/env/.hg/store/data/canary.orig': Is a directory \ Done.
11:55:04 <b_jonas> `? subscribe
11:55:11 <HackEgo> subscribe? ¯\(°​_o)/¯
11:55:25 <shachaf> The list infrastructure could use some overhauling.
11:55:36 <shachaf> It should stop using self-modifying executables, for one.
11:56:08 <b_jonas> shachaf: right, but then you should figure out some easy interface to subscribe and unsubscribe lists
11:56:27 <b_jonas> feel free to overhaul it
11:58:13 <boily> hellørjan!
11:58:27 <shachaf> THe only reason for me to overhaul it would be to avoid you overhauling it in Perl that I can't read or modify.
11:58:46 <shachaf> Anyway, I'm going to sleep.
11:59:52 <b_jonas> shachaf: nah, it's culprits and noping that I want to overhaul in perl, not lists
12:00:32 <b_jonas> (well except maybe `list itself, to make myself immune to it but in a sneaky viral way that you can't easily fix because culprits re-infects it)
12:00:55 <b_jonas> (maybe I should just make `list use the overhauled culprits)
12:02:26 <int-e> b_jonas: "something useful" is not nearly specific enough, but the way I interpret it, the construction I saw has nothing very useful after 4 steps.
12:02:58 <b_jonas> int-e: I can make it more specific:
12:03:06 <int-e> (to be precise, none of the missing four vertices of the pentagon are there)
12:04:05 <int-e> (hoping that's not spoiling too much)
12:05:51 <izabera> http://duktape.org/
12:06:15 <b_jonas> after 4 steps, I want either (a) two points, each on a line of one of the five sides of the pentagon or the big circle that contains the outer intersection points of the sides; or (b) one or two circles or lines whose union contains two outer intersection points of the sides
12:06:44 <b_jonas> (and that's not always enough)
12:07:34 <b_jonas> int-e: basically, after 4 moves I can restrict the branching of the search so much that the 5th move must be one of some very specific moves, and usually completely impossible
12:08:20 <b_jonas> it's still a lot of search, but with primitives only, I think it's possible
12:08:24 <b_jonas> I'm not completely sure
12:10:50 <b_jonas> but I also have the feeling that it's probably easier to find a non-primitive 10 step solution than to do the brute force for a primitive one,
12:10:59 <b_jonas> because there's still some of the geometry I don't understand here
12:12:16 <b_jonas> hmm wait, I have an idea
12:12:42 <b_jonas> let me try some things
12:20:02 <b_jonas> ok, I got a horrible 15 move non-primitive solution, but at least it doesn't look like just an obvious pessimization of the 11 move one I've got
12:23:00 -!- oerjan has quit (Quit: Later).
12:28:37 -!- boily has quit (Quit: RESET CHICKEN).
12:28:54 <b_jonas> nice, I've got a 12 move solution
12:39:08 <b_jonas> I've got a very ugly 13 move solution
12:39:45 -!- MoALTz has quit (Ping timeout: 276 seconds).
12:41:01 <b_jonas> ok wait, I still don't understand how this geometry stuff works
13:58:35 -!- tromp_ has joined.
14:10:53 -!- Sgeo_ has quit (Read error: Connection reset by peer).
14:14:09 -!- tromp_ has quit (Remote host closed the connection).
14:14:23 -!- Sgeo has joined.
14:19:32 <prooftechnique> g
14:20:01 <prooftechnique> Sorry, ERC bugged out and wouldn't let me delete that :(
14:20:20 <b_jonas> `randomquote
14:20:28 <HackEgo> ​/home/hackbot/hackbot.hg/multibot_cmds/lib/limits: line 5: exec: randomquote: not found
14:20:31 <b_jonas> fungot, are you here?
14:20:31 <fungot> b_jonas: see? i like marle better than " princess,' the chosen time has come! he's strong and he's gonna thrash those monsters! yea! is it?
14:20:40 <int-e> . o O ( what is the European Research Council doing with prooftechnique? )
14:20:41 <b_jonas> yes
14:20:57 <prooftechnique> int-e: I would think a fair bit
14:21:02 <b_jonas> int-e: ERC is errorrorrecting code
14:22:04 -!- groteworld has joined.
14:22:05 * int-e is still waiting for an isosceles triangle to sneak up on b_jonas.
14:27:02 -!- spiette has joined.
14:35:23 -!- `^_^v has joined.
14:37:20 -!- groteworld has quit (Quit: ZZZzzz…).
14:37:26 -!- Akaibu has quit (Quit: Connection closed for inactivity).
14:38:19 -!- groteworld has joined.
14:48:12 -!- Kaynato has joined.
14:52:39 <FireFly> ^style irc
14:52:39 <fungot> Selected style: irc (IRC logs of freenode/#esoteric, freenode/#scheme and ircnet/#douglasadams)
15:03:07 -!- adu has joined.
15:14:37 -!- tromp_ has joined.
15:18:51 -!- tromp_ has quit (Ping timeout: 250 seconds).
15:35:16 -!- groteworld has quit (Ping timeout: 244 seconds).
15:51:19 <Taneb> fungot, what's new with you?
15:51:19 <fungot> Taneb: there's still nothing threatening about **********s. and hey, the entire point
15:51:36 <Taneb> That was very profound
16:04:59 <int-e> `? *
16:05:00 <HackEgo> Twinkle, twinkle, little star!
16:07:23 -!- AnotherTest has quit (Ping timeout: 250 seconds).
16:10:23 -!- AnotherTest has joined.
16:11:04 <myname> people disagreed with me over 6 times 9 being 42. i was disappoint
16:11:18 <izabera> tell em base 13
16:11:37 <myname> i could, but i told em adams
16:11:49 <tromp> use church numerals to avoid all confusion
16:12:56 <Taneb> > fix ("ONE MORE THAN " ++)
16:14:28 <Taneb> Aaargh, I still need to document COMPLEX on the wiki
16:29:09 <b_jonas> `olist 1044
16:29:10 <HackEgo> olist 1044: shachaf oerjan Sgeo FireFly boily nortti b_jonas
16:32:03 -!- LKoen has joined.
16:33:55 <Sgeo> b_jonas, ty
16:43:00 <Sgeo> ais523 is now in Facebook trending because of a speedrun
16:44:33 <Taneb> Well done him
16:47:48 -!- LKoen has quit (Remote host closed the connection).
16:53:35 -!- LKoen has joined.
16:54:06 -!- LKoen has quit (Remote host closed the connection).
16:59:56 <b_jonas> oh!
17:00:12 <b_jonas> `random-card \A..thersnatch
17:00:17 <HackEgo> AEthersnatch \ 4UU \ Instant \ Gain control of target spell. You may choose new targets for it. (If that spell becomes a permanent, it enters the battlefield under your control.) \ C15-R
17:00:40 <b_jonas> I didn't know they finally printed this
17:01:22 <b_jonas> (now I want an aura version of this too)
17:16:37 <b_jonas> (that would be much cheaper, because then you don't get the permanent)
17:17:10 -!- AnotherTest has quit (Ping timeout: 258 seconds).
17:17:17 <int-e> myname: Douglas Adams said about this: I may be a sorry case, but I don't write jokes in base 13.'
17:18:16 <myname> i know
17:21:59 <b_jonas> `random-card
17:21:59 <HackEgo> Relic Barrier \ 2 \ Artifact \ {T}: Tap target artifact. \ LE-U, 5DN-U
18:15:35 -!- MoALTz has joined.
18:16:27 -!- AnotherTest has joined.
18:19:45 <zzo38> Other version would be: Target player gains control of target spell. You may choose new targets for it.
18:30:29 <int-e> "You"?
18:31:49 <int-e> that could be awkward
18:39:40 -!- PinealGlandOptic has joined.
19:03:13 -!- augur has joined.
19:03:59 -!- bb010g has joined.
19:12:28 -!- zgrep has changed nick to zgrep_the_change.
19:13:13 -!- wob_jonas has joined.
19:15:30 -!- augur has quit (Remote host closed the connection).
19:15:49 <zzo38> I thought of idea of a new kind of poker game, that you cannot make change of poker chips. For example if you have 100 white chips and 100 red chips, and white chips are worth 1 point and red chips are worth 5 points. If later on in the game you have to pay 3 points to call but you have only red chips, then you cannot call; you must fold or raise.
19:16:49 <shachaf> Why?
19:17:19 <zzo38> Can you be more specific please?
19:17:33 <shachaf> Why do you want to disallow change?
19:18:08 <wob_jonas> couldn't you just pay two red chips then?
19:18:36 <zzo38> It is just an idea, that is all.
19:18:55 <zzo38> wob_jonas: You can but it is a raise. You cannot call because you need three white chips to call.
19:19:48 <wob_jonas> and if the other player also has only red chips after those 3 white chips, then you two can't stop until you fold or all in
19:20:00 <wob_jonas> dunno, that would be a strange game
19:20:31 <zzo38> Yes, I know it is strange, but that is just the idea I came up with. Variants of this kind of game can be made up of course.
19:23:19 <shachaf> How about a variant where you can exchange a red chip five white chips at any time?
19:23:21 <zzo38> But I do not think it would be common if you started with 100 white chips, that after that you have two player both with only red chips anyways like that, unless of course their intention is to go all in
19:23:49 <shachaf> for
19:24:02 <shachaf> Red chips are supposed to be worth five white chips, but if you can't trade them easily, then they're worth less than that.
19:24:20 <shachaf> You need liquidity to make them really be worth that much in a case like that.
19:24:23 <zzo38> shachaf: In a normal poker game you could, assuming you have enough poker chips to implement that, of course.
19:24:26 <shachaf> Enjoy being locked in your matrix of solidity.
19:25:11 <int-e> solid advice?
19:26:23 <zzo38> Another way can be: If at the beginning of the deal you have only red chips, you can exchange two red chips for ten white chips; if the ante or blind is not a multiple of five this will be necessary anyways. This also mean that white chips are slowly replenished in this way.
19:31:37 <wob_jonas> shachaf: yes, that's what happens here too, sadly, ever since 2015 automn when the Post announced they're no longer changing forint banknotes to smaller ones for free.
19:32:14 <wob_jonas> their rates are so high that I'm actually incentivized to buy a single envelope for 5 forint, repeated up to 7 times, to change a 20000 forint note to 2000 forint notes, rather than pay their normal change fees
19:32:25 <wob_jonas> that would take a lot of time of the cashier of course
19:32:35 <wob_jonas> so they might kick me up when I try it
19:33:10 <wob_jonas> hmm no, it's only up to 5 times I think
19:34:56 <wob_jonas> no, up to 4 times
19:34:59 <wob_jonas> or something, I can't count
19:39:29 -!- augur has joined.
19:49:51 <wob_jonas> `random-pokemon
19:49:51 <HackEgo> ​/home/hackbot/hackbot.hg/multibot_cmds/lib/limits: line 5: exec: random-pokemon: not found
19:50:02 <wob_jonas> `encounter
19:50:03 <HackEgo> ​/home/hackbot/hackbot.hg/multibot_cmds/lib/limits: line 5: exec: encounter: not found
19:50:08 <wob_jonas> `random-encounter
19:50:09 <HackEgo> ​/home/hackbot/hackbot.hg/multibot_cmds/lib/limits: line 5: exec: random-encounter: not found
19:53:27 -!- zgrep_the_change has changed nick to zgrep.
19:58:15 -!- adu has quit (Quit: adu).
20:01:19 -!- rdococ has quit (Quit: Leaving).
20:31:47 -!- augur has quit (Remote host closed the connection).
20:34:26 -!- wob_jonas has quit (Quit: http://www.kiwiirc.com/ - A hand crafted IRC client).
20:34:30 -!- PinealGlandOptic has quit (Quit: leaving).
20:42:44 -!- gamemanj has joined.
20:49:07 -!- adu has joined.
20:53:20 -!- adu has quit (Client Quit).
21:08:42 -!- adu has joined.
21:37:29 -!- Cale_ has joined.
21:38:07 -!- augur has joined.
21:38:53 -!- Cale has quit (Ping timeout: 250 seconds).
21:40:44 -!- Cale_ has changed nick to Cale.
22:02:58 -!- Akaibu has joined.
22:03:01 -!- augur has quit (Remote host closed the connection).
22:09:27 -!- augur has joined.
22:16:45 -!- adu has quit (Quit: adu).
22:19:59 -!- bb010g has quit (Quit: Connection closed for inactivity).
22:29:16 <fizzie> Was there something one should do in particular in Cambridge, England?
22:29:56 <gamemanj> sound posh
22:29:59 <gamemanj> drink tea
22:30:21 * gamemanj thinks for a second
22:30:33 <gamemanj> sneak into the library and read stuff at midnight?
22:32:10 -!- AnotherTest has quit (Quit: ZNC - http://znc.in).
22:32:16 <fizzie> I think we'll be doing a day trip, not overlapping with midnight.
22:32:28 -!- adu has joined.
22:32:40 <gamemanj> If it's a day trip, then you probably live in England, correct?
22:32:47 <fizzie> London, yes.
22:33:11 <gamemanj> Ok, so basically that just leaves Cambridge-specific stuff.
22:33:54 <fizzie> Yes, I thought the tea thing was a little too general.
22:34:00 <gamemanj> Maybe try dressing to blend in with the crowd... and sneak into the library to learn stuff. (I have no idea, really.)
22:34:35 <fizzie> For the record, I just stuck on the ", England" part because I also have a trip to Cambridge, MA, and didn't want to be ambiguous.
22:35:27 <gamemanj> error: too many cambridges
22:36:12 <fizzie> And I'm not event going to Cambridge, Ontario at all.
22:42:59 -!- zgrep has changed nick to zgrep_the_Slow.
22:43:36 -!- MoALTz has quit (Ping timeout: 272 seconds).
22:46:35 -!- boily has joined.
22:54:59 <quintopia> helloily
22:58:44 <izabera> another terror attack in france https://www.reddit.com/live/x99pqdwudg0l/
23:00:55 <quintopia> shit
23:01:06 <quintopia> FUCK YOU 2016
23:01:52 -!- jaboja has joined.
23:08:33 -!- tromp_ has joined.
23:11:11 -!- adu has quit (Quit: adu).
23:11:35 <boily> quinthellopia, izabellora.
23:11:37 <boily> AGAIN?
23:12:01 <izabera> looks like so
23:19:09 <quintopia> and the murderer already managed to get himself martyred. fuck the world. anyone know where i can get a rip van winkle drug?
23:20:15 -!- gamemanj has quit (Ping timeout: 264 seconds).
23:22:31 -!- tromp_ has quit (Remote host closed the connection).
23:25:03 -!- oerjan has joined.
23:33:31 <quintopia> i've been sucked into the random-space-jam-mashups vortex of the youtube remix community again--
23:33:54 <quintopia> the danciest of strange attractos
23:37:50 <boily> youtube has weird corners.
23:39:49 <quintopia> some of the corners are sharp
23:40:32 <shachaf> oerjan: Day
23:40:36 <oerjan> magandang gaboily
23:40:43 <oerjan> shachaf: wat?
23:40:55 <shachaf> Well, you usually say Nite when you leave.
23:40:59 <shachaf> But then there's no way to respond.
23:41:10 <shachaf> On the other hand it's hardly daytime for you right now.
23:41:15 * oerjan notices that his nvg shell account seems to have lost its quotas. he's not sure whether to be delighted or worried.
23:41:29 <shachaf> What are its quotas?
23:41:33 <shachaf> Something microscopic?
23:42:04 <oerjan> a few hundred Mb
23:42:27 <oerjan> by "lost" i mean quota now reports there are no quotas.
23:42:54 <boily> trathnóerjana maith duit. tagalog?
23:43:38 <oerjan> yep!
23:44:14 <boily> yé :D
23:44:38 <oerjan> that did look gaelic, which it was.
23:44:40 <quintopia> coily
23:45:13 <quintopia> (do you recognize that lang oerjan?)
23:45:48 <oerjan> quintopia: a _bit_ short...
23:46:00 <oerjan> could be ciao, which is at least italian.
23:46:04 <boily> oerjan: the Irish variant of.
23:46:13 <quintopia> oerjan: nope
23:46:19 <quintopia> boily: your guess?
23:47:17 <oerjan> i still guess it's romance or slavic. portuguese?
23:47:48 <oerjan> (just because it uses c. hm so does turkish...)
23:47:52 <boily> most probably not slavic.
23:48:12 <boily> vietnamese, possibly? eventually?
23:48:22 <oerjan> boily: i _think_ some south slavic languages borrowed ciao...
23:48:26 <boily> no!
23:48:29 <boily> found!
23:48:34 <boily> quintopia: nice choice :D
23:48:43 <oerjan> wat.
23:49:14 -!- adu has joined.
23:50:22 <oerjan> maybe it's actually québécois. that language is weird...
23:50:50 <boily> ha ha! we're still using «bonsoir».
23:50:55 <oerjan> time for some more cheating...
23:52:00 <oerjan> croatian has Ćao
23:52:30 <boily> not even close to the cigar.
23:52:59 * oerjan is obviously just reading through a list now.
23:53:23 <shachaf> An ølist?
23:53:34 <shachaf> `? oerjan
23:53:38 <shachaf> `? ørjan
23:53:58 <HackEgo> Your mysterious articled cackling zombie øverlord kommisjonær emeritus oerjan is a lazy expert in future computation. Also a Precambrian Norwegian who mildly dislikes Roald Dahl with a pasjon. Lately when he tries to remember a word, "amortized" pops up. His arch-nemesis is Betty Crocker. He sometimes puns without noticing it.
23:54:00 <HackEgo> ​Ørjan is oerjan's good twin. He's banned in the IRC RFC for being an invalid character. Sometimes he publishes papers without noticing it.
23:54:02 <boily> `? œrjan
23:54:03 <HackEgo> ​œrjan? ¯\(°​_o)/¯
23:54:18 <oerjan> hm nothing promising on that list
23:54:29 <oerjan> as in, nothing starting with "co"
23:54:53 <shachaf> `slwd ørjan//s#^#Your pal #
23:54:59 <HackEgo> wisdom/ørjan//Your pal Ørjan is oerjan's good twin. He's banned in the IRC RFC for being an invalid character. Sometimes he publishes papers without noticing it.
23:56:12 <shachaf> your pal ørjan johansen
23:57:30 <shachaf> an expert in new ro science
23:57:42 <shachaf> ro science is presumably the science of persistent data structures
23:57:54 <quintopia> *rho science
23:58:00 <shachaf> Ah, even better.
23:58:16 <shachaf> http://www.ncbi.nlm.nih.gov/pubmed/7494472
23:58:17 <shachaf> oh man
23:58:21 <shachaf> a new rho mutation
23:58:41 <oerjan> bah there, at the end of "good evening"
23:59:02 <oerjan> my only defense is i actually thought of it before finding it.
23:59:28 <oerjan> that's pretty short indeed.
23:59:58 -!- `^_^v has quit (Quit: This computer has gone to sleep).
←2016-07-13 2016-07-14 2016-07-15→ ↑2016 ↑all