00:15:30 -!- upgrayeddd has joined.
00:19:26 <Gracenotes> for the hell of it, I decided to get macademia nuts
00:20:35 <Sgeo> "You do realize this crap contains aspartame right?
00:20:35 <Sgeo> I think I'll stick to the ones without chemicals"
00:24:08 <Gracenotes> sugar-free candy usually has erythritol. or, slightly worse, maltitol or sorbitol. or, worst, aspartame.
00:24:35 <Sgeo> How about chemical-free candy?
00:24:46 <Sgeo> Also, these things that I'm addicted to aren't considered candy I think
00:26:11 <Gracenotes> everything is candy. cough drops are candy.
00:27:01 <Gracenotes> Yes, they have other purposes, but they're more than just a spoonful of sugar.
00:27:18 <Gracenotes> (unless you seek out the kinds that are less fun to eat)
00:31:40 <Sgeo> But too much of candy-like non candy things could presumably be more harmful than too much candy
00:31:47 <Sgeo> Too many vitamins is bad, for example
00:34:52 <Gracenotes> yeah, perhaps, and ideally no one should need to take vitamins (unless they have a nutritional impairment, like a thyroid issue or veganism)
00:35:26 <Gracenotes> Most things that people buy have been very heavily engineered to be palatable, though. People don't realize that.
00:37:12 <Sgeo> But I'm worried about eating too many of these Listerine strips
00:37:15 <pikhq_> Turns out our tastebuds aren't actually designed to tell us whether something's good for us or not.
00:37:54 <Sgeo> What I quoted I was more facepalming at the 'ones without chemicals' thing
00:38:21 <Gracenotes> assuming a lack of agriculture and domesticated anything
00:38:25 <pikhq_> Which is to say, things that where good for us way back when.
00:38:37 -!- mnoqy has joined.
00:39:02 <pikhq_> They don't adapt to what we need right now, they're adapted to survival in an environment where the really tasty foodstuffs are also rare.
00:39:23 <pikhq_> And you'd actually want to pig out on, say, that bison you just killed.
00:40:31 <Gracenotes> evolution really screwed up, making an organism that was superb at changing the environment around it without regard for externalities.
00:41:19 <Gracenotes> put another way, it did its job too well.
00:41:19 <Sgeo> Isn't that inevitable though when selecting based on X' that doesn't quite match the ideal X?
00:41:33 <Sgeo> bad incentives etc
00:41:48 <pikhq_> You get what you optimize for, not what you want.
00:44:29 <Gracenotes> Sgeo: you should limit yourself to 3 strips a day. put all of your energy into allocating those strips.
00:44:53 <Gracenotes> try to put it into an existing habit, such as brushing your teeth
00:45:25 <Gracenotes> oh, and if you fuck up, self-flagellate. 30 times, each side of the back. that's the most important part.
00:45:56 <Gracenotes> I know, I know, a lot of people say "But I don't want to scar my skin!" blah blah blah
00:46:36 <Gracenotes> ...yeah, fine, last two lines optional. >.>
00:51:23 <Sgeo> I already had two more since then
00:51:29 <Sgeo> I need a candy to substitute for ir
00:54:02 <Gracenotes> I also recommend not eating sweet things or refined grains at all
00:57:15 <Sgeo> Assuming that mints will taste like this wonderful taste
01:02:02 <Sgeo> I get the impression that a lot would be bad, but more like 24 all at once or something
01:03:48 <Sgeo> Although my body does seem to be treating it as food and lowering my hunger :(
01:10:47 <Gracenotes> Sgeo: you will not stay hungry for long tho
01:11:49 <Sgeo> Gracenotes: er
01:12:00 <Sgeo> I need to _be_ hungry, these things are stopping me from being hungry
01:12:08 <oerjan> "After Eights are now sold across Europe and North America, where they are kosher certified by the Orthodox Union, and one billion After Eight mints are made annually."
01:13:39 <oerjan> Sgeo: why do you need to be hungry, and can't you just eat something real
01:13:52 <Sgeo> Because if I'm not hungry, I won't eat
01:14:38 <Gracenotes> sugar and refined grains make you less satiated than whole foods like fruits, vegetables, and meats hth
01:15:18 <Sgeo> Eat more sugary snacks and refined grains. Got it.
01:16:05 * pikhq_ declares this to have been a damnable week
01:16:30 <oerjan> Sgeo: i recall something about sugary drinks not triggering the body's satiation properly, unlike sugary food
01:17:03 * Sgeo checks if ais523 is here on pikhq_'s behalf
01:17:18 <Gracenotes> I don't think sugary food does either, compared to non-sugary food
01:17:29 <oerjan> it's ok the week doesn't have a soul than can go to hell hth
01:22:04 <Sgeo> Pretty sure http://www.scrapbooksthatteach.com/files/2012/12/father-time.gif disagrees
01:25:40 <oerjan> what they changed the recipe in for after eights in 2007
01:25:50 * oerjan is not quite sure if he's tasted them since
01:32:18 <oerjan> "Listerine did not make mouthwash as much as it made halitosis."
01:32:35 <oerjan> i'm not quite sure this is candy Sgeo
01:33:09 <Sgeo> I'm not drinking liquid mouthwash
01:33:32 <Sgeo> I'm dissolving in my mouths strips of presumably a different substance that is somewhat safer to swallow
01:35:16 <oerjan> "For a short time, beginning in 1927, the Lambert Pharmaceutical Company even marketed Listerine Cigarettes."
01:57:28 -!- Bike has quit (Ping timeout: 264 seconds).
01:59:39 -!- Bike_ has joined.
02:00:41 -!- Bike_ has changed nick to Bike.
02:33:35 <tswett> Is eating sugary snacks and refined grains actually a good idea when one is underweight?
02:33:49 <tswett> It's a bad idea when you're overweight, but that doesn't make it a good idea when you're underweight.
02:34:50 <tswett> Sgeo: do you think you'd be more likely to eat better if you had unlimited access to every type of food, already prepared?
02:35:27 <tswett> I'm sure that I would. More veggies, less candy.
02:37:29 <zzo38> Can you eat other things, such as your own blood?
02:39:52 <oerjan> eating your own blood is unlikely to gain weight hth
02:41:22 <tswett> Unless the reason you haven't been gaining weight is that you've been bleeding on a regular basis.
02:42:15 <Bike> my blood's more maple-y
02:45:33 <zzo38> I suppose you won't gain weight, but do you like to eat your own blood anyways?
02:45:48 <Gracenotes> tswett: imo it's a good way to become sink your metabolism and become overweight later in life
02:48:31 <tswett> Delete everything up to the previous x character.
02:54:03 <Sgeo> tswett: possibly
02:54:19 <Sgeo> I'd certainly be more likely to eat more often
02:54:38 <tswett> Are there any reasonably healthy foods that you enjoy eating?
02:59:08 <Gracenotes> bananas are pretty healthy, although they are actually incredibly sweet.
02:59:25 <Gracenotes> they just don't taste it because of, well, they have banana taste and texture.
02:59:26 <tswett> I guess bananas are pretty good, yeah.
03:00:40 <Sgeo> Whenever I would talk about my issues to teachers in highschool, they suggested pizza
03:00:47 <Sgeo> Don't know if that counts for anything
03:03:05 <zzo38> Pizza is a bread with other stuff on it such as tomato and cheese and chess.
03:07:21 -!- copumpkin has quit (Ping timeout: 264 seconds).
03:07:58 -!- copumpkin has joined.
03:27:03 <doesthiswork> if you're lactose tolerant it is possible to live healthily off of milk and potatoes.
03:28:21 <doesthiswork> and soybeans are one of the few crops I know of that is bred for nutritional content
03:45:58 -!- sacje has quit (Excess Flood).
03:46:26 -!- sacje has joined.
03:53:52 -!- douglass has joined.
04:07:21 <tswett> So as everyone knows, one of the great problems of the world is that it's hard for a purely functional programming language to represent a cyclic data structure.
04:08:29 <tswett> And the reason for this is that in a purely functional programming language, you can't really have two separate pointers to a single mutable object.
04:10:52 <zzo38> There are a number of other things I also find it difficult to represent in Haskell, or in any other programming language, which makes it even more difficult.
04:11:09 <tswett> Not the sort of cyclic data structure I'm after.
04:12:06 <zzo38> Linear logic might help a bit, but it doesn't help everything.
04:12:15 <tswett> So here's an idea that you might use in a resource-conscious language.
04:12:54 <tswett> Have some type constructor... I have no idea what to call it, so let's call it Shoe.
04:13:51 <tswett> Given any object, you can turn it into a pair of Shoes. Given just one of the Shoes, you're allowed to borrow the object, but you have to put it back in order to get the Shoe back.
04:13:54 <tswett> Finally, you can destroy both Shoes to get the object back once and for all.
04:14:53 <tswett> Obvious problem 1: race conditions happen.
04:15:05 <doesthiswork> tswett: immutable cyclic data structures seem easy to create.
04:15:15 <tswett> Obvious problem 2: if you try to use both Shoes simultaneously, you'll deadlock.
04:15:29 <tswett> doesthiswork: true, but I'm thinking about mutable ones.
04:16:13 <zzo38> Could you make more stuff with the type system by adding a new kind of quantification?
04:19:50 <tswett> So the problem is that race conditions and deadlocks can happen if you have two pointers to the same object. Thus, pointers can't form an arbitrary graph; they have to form a tree.
04:20:24 <tswett> So what if we somehow make it so that the pointers form a tree *at any given time*, but the pointer structure can be changed to support all the deque operations in constant time?
04:20:48 <doesthiswork> are you not allowed to refer to things that haven't been fully defined yet?
04:21:00 <tswett> Recursive definitions are allowed.
04:21:48 <tswett> If we make it so that there exist "inactive pointers" which can't actually be used...
04:22:00 -!- Nisstyre has quit (Quit: Leaving).
04:23:42 <tswett> It would be perfectly easy to add deques as a primitive data type, but I want to figure out how to make it possible to actually implement them.
04:28:36 <tswett> Okay, here's another idea.
04:29:01 -!- doesthiswork has quit (Read error: Connection reset by peer).
04:29:49 <tswett> There are these things called Batons, which can be freely created. You can trade an object for a bunch of pointers to it, but all of its pointers must be associated with the same Baton. You can only access an object if you have both a pointer and the corresponding Baton.
04:30:23 <fizzie> I was under the impression that Data.Sequence was an amortized-O(1) deque. (But maybe that doesn't count given that the implementation is not at all "deque-ish".)
04:30:54 -!- doesthiswork has joined.
04:31:21 <tswett> Perhaps it is, but I'm thinking about "real" O(1), not amortized.
04:32:04 <doesthiswork> darn it, my client left out half the conversation without telling me
04:33:08 <tswett> Yes, I feel like this will work.
04:34:19 <Bike> doesthiswork: we have nice logs
04:34:27 <tswett> You can create a Baton. Given an object and a Baton, you can give up the object to create a pointer to the object associated with the Baton.
04:34:49 <doesthiswork> bike: that's how i know that there were messages I didn't see
04:35:30 <tswett> Given a pointer, you can duplicate it; given two pointers to the same object with the same Baton, you can delete one. Given a pointer, you can give up the Baton to borrow the object; upon returning the object, you get back the Baton.
04:35:47 <tswett> Given the last remaining pointer to an object, and its Baton, you can destroy the pointer. And if a Baton has no pointers, you can destroy the Baton.
04:35:52 -!- oerjan has quit (Quit: leaving).
04:37:04 <tswett> So it's like, the "real" pointer here is the Baton, and the Baton is the thing which has to be part of a tree. The "pointers" merely serve to select the pointee of the Baton.
04:39:10 -!- zzo38 has quit (Remote host closed the connection).
04:56:33 -!- sacje has quit (Ping timeout: 264 seconds).
04:57:35 -!- sacje has joined.
04:58:45 <HackEgo> olist 906: shachaf oerjan Sgeo FireFly
05:11:32 <Sgeo> I swear, I'm going to have to change my name to Rich Burlew so the things you say start making sense
05:13:17 <shachaf> Do you disagree that it's good?
05:14:31 <Sgeo> I disagree that I'm somehow responsible for it being good
05:14:59 <shachaf> Who said you were responsible?
05:15:35 <shachaf> You told me "hey, there's an update [which Rich Burlew wrote, presumably]!". I read the update. It was good. I commented to the person who told me.
05:18:11 <doesthiswork> rss readers notify you of a wide variety comics
05:18:36 <Sgeo> So do I! (If 2 counts as a wide variety)
05:18:52 <shachaf> doesthiswork: Hmm, good point.
05:19:14 <shachaf> You should hook up an RSS reader to some other comics (like supermegacomics.com) and tell me when they update.
05:20:53 <Sgeo> SMComics updates on a regular basis? Why do you need a human RSS reader?
05:21:09 <Sgeo> NOW UPDATING EVERY TUESDAY AND THURSDAY EVENING
05:21:30 <Sgeo> No idea what that is
05:21:35 <Bike> rice boy updates are easy to track
05:21:41 <shachaf> Sgeo: Well, it didn't update on Tue!
05:21:48 <shachaf> Hmm, someone told me to read Rice Boy once.
05:21:52 <shachaf> I'd forgotten. I guess I should read it.
05:22:25 <Sgeo> There's a reason I don't have lists for xkcd and Freefall
05:25:09 <Bike> god, it's crazy how much dahm's art's improved
05:33:27 <tswett> http://lpaste.net/91292 – batons, refined!
05:34:31 <tswett> A pointer really represents two things: a way of identifying an object, and a way of accessing it. This idea separates a pointer into two separate things: a "ticket" and a "coat check".
05:34:46 <tswett> A ticket is a means of identifying an object. Since it can't be used to identify the object, multiple tickets can exist for just one object.
05:35:04 <tswett> s/to identify/to access/
05:35:19 <tswett> And a coat check is the means of accessing the object. Since it can't be used to identify the object, multiple objects can belong to just one coat check.
06:01:45 -!- Taneb has joined.
06:03:46 <Taneb> I like today's FreeFall
06:04:45 <shachaf> Taneb: Today's oots is p. good but has ````spoilerz''
06:04:48 <shachaf> so don't start on this one.
06:06:24 <Taneb> shachaf, should I enter IOCCC?
06:07:39 <Gracenotes> the best IOCCC entries are golfing interesting algorithms
06:09:15 <Gracenotes> golfing, byzantifying, and rococoifying
06:19:04 <coppro> dammit, I missed the underhanded c competition
06:19:41 <coppro> I had a *fantastic* way to be underhanded, too
06:20:02 <lifthrasiir> I still don't know why IOCCC judges changed a size limitation
06:20:21 <lifthrasiir> as the total byte size limitation is same (up to 4,096 bytes)
06:21:27 <kmc> hi everyone
06:22:14 <coppro> my plan was to use a hashtable with an if statement that accidentally does an assignment
06:22:25 <coppro> and my account just so happened to hash to 0
06:22:27 <Gracenotes> I can't wait to see how underhanded people are.
06:22:37 <olsner> coppro: if they haven't finished judging, maybe you could try to submit it anyway?
06:22:46 <coppro> olsner: I haven't the time to write it
06:22:54 <Gracenotes> coppro: hm, I think accidental-= is old-hat nowadays
06:23:00 <Gracenotes> not nearly underhanded enough in this day and age
06:23:44 -!- augur has quit (Remote host closed the connection).
06:25:47 -!- augur_ has joined.
06:28:29 <coppro> Gracenotes: I think I had some other device to obfuscate that this was the bug
06:28:41 <coppro> since if you just do it with nothing else, the compiler wil catch it pretty easy
06:28:57 <coppro> the real trick was to make sure that it continued to work
06:38:12 -!- sacje has quit (Ping timeout: 260 seconds).
06:43:26 <Taneb> In C, does assignment propagate value?
06:43:55 <kmc> what do you mean?
06:44:49 <Gracenotes> a = b is an expression whose value is b and whose type is a's
06:45:52 <kallisti_> 16:41 < Taneb> kallisti's is unusual
06:48:57 <kallisti_> Taneb: if I understand correctly, C's assignment propagates bits.
06:49:57 <Taneb> So, I can do "a = b = c = 4;"
06:50:32 <olsner> as long as none of a, b and c alias each other
06:51:01 <Bike> assignment like that is reasonably common
07:01:52 <kmc> if they alias then it's UB due to multiple assignment with no sequence point?
07:06:25 <fizzie> The value of the assignment is the value the left-hand-side will have after the assignment.
07:06:50 <Gracenotes> If the instruction executes at the exact moment of high tide in northern Sri Lanka it's UB
07:07:30 <fizzie> [10:07:17] <fizzie> ,cc unsigned char b, c = 255; unsigned u1, u2; u1 = c + 1; u2 = b = c + 1;
07:07:34 <fizzie> [10:07:18] <candide> fizzie: <no output: b = 0; c = 255; u1 = 256; u2 = 0>
07:11:52 <Deewiant> "If the value being stored in an object is read from another object that overlaps in any way the storage of the first object, then the overlap shall be exact and the two objects shall have qualified or unqualified versions of a compatible type; otherwise, the behavior is undefined."
07:13:52 -!- tertu has quit (Ping timeout: 276 seconds).
07:14:40 <fizzie> The objects cover exactly the same bytes, presumably.
07:16:16 <fizzie> However, I don't think "a = b = c = 4;" involves any reads from objects b or c.
07:17:00 <Deewiant> fizzie: "The implementation is permitted to read the object to determine the value [of the left operand] but is not required to, even when the object has volatile-qualified type."
07:17:05 <fizzie> Deewiant: Yes, I just read that.
07:17:35 <fizzie> A (non-normative) footnote, how nasty to put something like that in it.
07:19:54 -!- epicmonkey has joined.
07:21:07 <Deewiant> I guess that means it can even read from a, although the value isn't used?
07:21:56 <Deewiant> The value is "discarded" but it doesn't seem like anything precludes computing it first
07:24:09 <fizzie> I guess. I mean, plain "a;" can read from a.
07:24:34 <fizzie> Or at least should intuitively be able to.
07:25:03 <Deewiant> Well, I think it could be argued that discarding the value means that it shouldn't be computed in the first place
07:25:40 <Jafet> It could be argued that C sucks
07:25:48 -!- kallisti_ has quit (Ping timeout: 260 seconds).
07:25:54 <Deewiant> That's a non-normative statement
07:34:15 -!- kallisti has joined.
07:35:27 <olsner> hmm, so *volatile_int_p = 0; might read from memory too (before or after writing?) ... that would be interesting if the location happens to have side effects on read
07:35:33 -!- douglass has quit (Ping timeout: 264 seconds).
07:35:54 <olsner> e.g. if reading launches the missiles and writing aborts the launch
07:36:04 <Jafet> Good hardware design
07:36:15 -!- epicmonkey has quit (Ping timeout: 240 seconds).
07:36:24 <olsner> I think people who make missile launchers do the best hardware design, probably
07:41:37 <olsner> > [...] which was then known still as a "Proscribed Action Link." The military leadership, however, soon realized that this term had negative connotations for the use of weapons by the officer corps (proscribed = "prohibited"), and changed the meaning of PAL to "Permissive Action Link" (permissive = "allowing / tolerating").
07:41:38 <lambdabot> <hint>:1:5: parse error on input `]'
07:46:52 -!- Jafet has quit (Quit: Leaving.).
07:47:51 -!- sacje has joined.
08:00:38 -!- heroux has quit (Ping timeout: 240 seconds).
08:02:28 -!- heroux has joined.
08:03:56 -!- conehead has quit (Quit: Computer has gone to sleep.).
08:06:07 <fizzie> olsner: "What constitutes an access to an object that has volatile-qualified type is implementation-defined."
08:06:31 <fizzie> Just briefly glancing at the pointer can be a read, too.
08:07:02 <Deewiant> Surely not? The pointer itself isn't volatile-qualified
08:07:21 -!- heroux has quit (Ping timeout: 264 seconds).
08:07:48 <fizzie> Deewiant: But it's implementation-defined what constitutes an access. I don't think that's restricted in any way to e.g. expressions involving the object.
08:08:19 -!- heroux has joined.
08:38:04 -!- augur_ has quit (Quit: Leaving...).
08:38:22 -!- augur has joined.
08:46:49 -!- Frooxius has quit (Ping timeout: 276 seconds).
09:00:06 -!- epicmonkey has joined.
09:02:12 -!- Frooxius has joined.
09:17:25 <Taneb> If a dying star starts with enough energy, can it fuse past iron?
09:32:37 <FreeFull> Taneb: It's called a supernova
09:33:09 <FreeFull> Supernovas are how all elements past iron got made, and also they are good because they spread those elements around
09:33:45 <Taneb> Taneb's problem with his science education number 36 solved!
10:47:57 -!- carado has joined.
10:56:33 -!- copumpkin has quit (Ping timeout: 264 seconds).
10:57:10 -!- copumpkin has joined.
11:10:09 -!- Jafet has joined.
11:32:11 -!- Sgeo has quit (Read error: Connection reset by peer).
11:36:37 -!- MindlessDrone has joined.
12:23:23 -!- sacje has quit (Ping timeout: 246 seconds).
12:54:12 -!- boily has joined.
12:54:12 -!- boily has quit (Client Quit).
12:54:49 -!- boily has joined.
12:57:28 <boily> good regular morning!
13:00:27 -!- metasepia has joined.
13:03:36 -!- epicmonkey has quit (Read error: Operation timed out).
13:18:03 -!- epicmonkey has joined.
13:39:52 <Taneb> I really want to do voice acting right now...
13:40:25 <fizzie> You can start a YouTube show. (Everyone can.)
13:41:55 <Taneb> I shall need an animator!
13:43:18 <fizzie> You can voice-act a still picture drawn in Paint.
13:44:09 <boily> you can do ASMR :D
13:44:40 <Taneb> What does that stand for?
13:44:57 <boily> http://www.reddit.com/r/asmr
13:48:21 <Taneb> This is going badly... Audacity seems to hate me
13:48:38 -!- Bike has quit (Ping timeout: 240 seconds).
13:50:34 -!- Bike has joined.
13:55:29 -!- CADD has quit (Quit: Lost terminal).
13:56:08 <Taneb> Okay, the quality on this mic seems much better
13:56:48 <Taneb> Now I need a script!
13:57:26 <Taneb> You know, when I first played minecraft I only used my left hand, because I was talking into this microphone and it's one of those hand-held ones
13:59:21 <boily> how'd you move around?
14:00:45 <Taneb> I could reach WASD from my laptop's touchpad
14:01:03 <Taneb> Couldn't re-angle while mining, though
14:15:44 -!- tertu has joined.
14:42:56 <boily> I'm still confused as fungot about the Cont monad. I guess today will be a NIM.
14:42:56 <fungot> boily: and there are a lot of
14:43:15 <boily> indeed. each week, a new Non-Illumination Monday.
14:49:37 <fizzie> I didn't know "confused as fungot" was a term.
14:49:37 <fungot> fizzie: but you need to resist those urges :) http://paste.pocoo.org/ show/ lisppaste/ colorize it, and because you don't have to live with
14:57:06 <boily> fungot: I don't resist you. you are far too tempting a bot for me not to love you.
14:57:06 <fungot> boily: was never named iirc, vb.net has builtin oo too ( please don't panic)
15:02:17 -!- conehead has joined.
15:28:03 -!- yorick has joined.
15:35:40 -!- Taneb has quit (Ping timeout: 276 seconds).
15:36:55 -!- iamfishhead has joined.
15:45:37 -!- jsvine has joined.
15:45:58 -!- Taneb has joined.
15:46:48 <JesseH> Hmm, I need better methods for implementing languages.
15:47:16 <JesseH> Are there some wiki pages about this or something?
15:47:21 <JesseH> Anyone have any good suggested reading?
15:50:44 <tswett> Well, what are your current methods?
15:52:13 <JesseH> Basically, I split the language up into tokens, but it can only handle simple statements
15:52:55 <JesseH> They work like functions, and can only handle a couple arguments each
15:53:26 <tswett> That'll work for simple languages, yeah. For more complicated languages, you'll probably want to use a parsing library.
15:54:17 <Vorpal> You could hand write a recursive parser as well
15:54:37 <Vorpal> I heard clang has a hand-written parser some time ago.
16:01:23 <fizzie> For almost any reasonable language, someone's written some kind of a parser-generator tool.
16:02:48 <fizzie> There's a kind of a table at http://en.wikipedia.org/wiki/Comparison_of_parser_generators
16:04:13 -!- aloril_ has quit (Read error: Operation timed out).
16:05:21 -!- jsvine has quit (Quit: Leaving.).
16:06:23 -!- Taneb has quit (Ping timeout: 245 seconds).
16:09:38 -!- Taneb has joined.
16:12:06 -!- aloril has joined.
16:21:48 -!- jsvine has joined.
16:22:47 -!- Taneb has quit (Ping timeout: 246 seconds).
16:25:14 -!- quintopia has quit (Ping timeout: 268 seconds).
16:45:11 -!- doesthiswork has quit (Quit: Leaving.).
16:55:08 -!- sebbu2 has changed nick to sebbu.
17:08:50 -!- zzo38 has joined.
17:24:56 <fizzie> Huh, the batteries in this remote seem to have explomoded.
17:25:11 <fizzie> Now the battery bay is: a mess.
17:25:34 <Lumpio-> Did you read the instruction manual!?
17:25:39 <Lumpio-> It says not to leave batteries in for a long time!
17:27:56 <fizzie> But I've been using it. Aren't you supposed to be able to do that?
17:29:41 <boily> fizzie: you're the first person I hear who had his batteries exploming.
17:30:22 <fizzie> Well, I think it must've been more of a leak than an explosion. But one side has sort of burst and there's a mess of white stuff.
17:34:23 <boily> so it's an explomesseak?
17:41:51 -!- lmt has joined.
18:02:58 -!- douglass has joined.
18:04:37 <zzo38> What assemblers can execute the programs they compile as a postprocessor?
18:04:45 -!- lmt has left.
18:06:38 <zzo38> Do you know of any, and if so, are they native code or emulated?
18:06:48 <zzo38> Can the DOS DEBUG program do this?
18:07:28 <fizzie> SPIM, kinda, if you count it as an assembler, but that's really a bit backwards way of looking at it, since it's more of an emulator with an integrated assembler. I don't think it even can write the code to file at all.
18:08:05 <zzo38> That's what it is? What computer does it emulate?
18:08:51 <fizzie> It emulates a hypothetical MIPS32 system, with a couple of "syscalls" do do input/output.
18:08:58 <fizzie> (It's designed for educational use.)
18:11:08 <zzo38> Unofficial MagicKit is a 6502 assembler with an integrated emulator. There is a special 64K bank for postprocessing, and if it has a nonzero reset vector, it will run the code in that bank before writing the output file. I think it might sometimes be useful to have in other assemblers too, but I don't know of any.
18:11:33 <zzo38> (The other banks are then accessed using memory-mapped I/O.)
18:12:04 <fizzie> I haven't heard of such a thing either.
18:46:26 -!- Taneb has joined.
18:48:43 -!- epicmonkey has quit (Ping timeout: 276 seconds).
18:53:14 -!- MindlessDrone has quit (Quit: MindlessDrone).
18:56:50 <Taneb> Having a mildly confusing conversation mixing up This Is The End with The World's End
18:57:15 <Taneb> Then someone mentioned Watson and someone else thought she was referring to the actor who played John Watson in Sherlock
18:59:35 -!- rodgort has quit (Ping timeout: 246 seconds).
19:02:18 -!- rodgort has joined.
19:06:50 -!- clog has quit (Ping timeout: 256 seconds).
19:08:44 <Bike> scare quotes eh
19:10:33 <fizzie> I don't think it counts as a proper AI until it kills people.
19:11:28 <elliott> `addquote [on the name "Watson"] <fizzie> And not the IBM "AI"? <Bike> scare quotes eh <fizzie> I don't think it counts as a proper AI until it kills people.
19:11:31 <HackEgo> 1076) [on the name "Watson"] <fizzie> And not the IBM "AI"? <Bike> scare quotes eh <fizzie> I don't think it counts as a proper AI until it kills people.
19:11:55 <Fiora> fizzie: have you seen https://dl.dropboxusercontent.com/u/167055177/nsc-isis%20maiden%20voyage.html ?
19:12:40 <fizzie> Fiora: Not before now, no.
19:12:52 -!- Bike has quit (Ping timeout: 240 seconds).
19:13:10 -!- Bike has joined.
19:13:16 <Fiora> (it's a game where you get to play as a possibly evil/rampant/terrible AI)
19:14:03 <Bike> does killing people at a trivia game not count
19:14:55 <Fiora> I'm not sure having a very fast button pressing device really counts -_-
19:15:17 <Bike> why do you hate the future Fiora....
19:15:28 <elliott> what if the button is really spiky
19:15:41 <Fiora> I was kind of confused by like that whole thing because I watched some of it and I was like
19:15:57 <Fiora> "Ummm... okay, this computer is getting like 1/4 of the questions wrong, and isn't really doing that great, it's just pressing a button really fast"
19:16:10 <elliott> I wonder how much work went into actually physically pressing the button quickly. I assume it's not really the bottleneck
19:16:15 <elliott> (I didn't watch any of it)
19:16:34 <Fiora> they claimed I think that like, they made it "fair" by delaying it randomly by 0-50ms? which. um. doesn't seem very fair
19:16:51 <Bike> doesn't seem as interesting as chess AI, overall
19:17:54 <Bike> maybe i've just inherited elliott's dislike of speech recognition.
19:20:45 <fizzie> "Watson consistently outperformed its human opponents on the game's signaling device, --"
19:22:39 <elliott> maybe the robot just pushed the button and they had an intern answer the questions.
19:22:44 <fizzie> But apparently it's been programmed not to press the buzzer before it has achieved some pre-set confidence level.
19:22:49 <fizzie> Or that's what they claim, anyway.
19:23:18 <elliott> pre-set confidence level or if it has a real good hunch
19:28:22 <zzo38> The computer was very good at it, however, one of the players made a severe error near the end considering the tournament format, and they might have had a good chance of winning if they didn't make that mistake.
19:29:02 <zzo38> Someone claimed that "WATSON" is "World Annihilating Trivia Solving Ontological Network" (I forget who told me) but probably not.
19:31:12 -!- kallisti has quit (Ping timeout: 246 seconds).
19:31:40 <fizzie> I think it should've been called "Deep Watson".
19:32:12 -!- augur has quit (Ping timeout: 246 seconds).
19:32:12 -!- mnoqy has quit (Ping timeout: 246 seconds).
19:32:13 -!- Ghoul_ has quit (Ping timeout: 246 seconds).
19:32:34 -!- augur has joined.
19:33:09 -!- conehead has quit (Ping timeout: 246 seconds).
19:33:09 -!- carado has quit (Ping timeout: 246 seconds).
19:33:11 -!- kallisti has joined.
19:33:11 -!- kallisti has quit (Changing host).
19:33:11 -!- kallisti has joined.
19:33:24 <zzo38> Winning at Jeopardy! isn't entirely about always knowing the correct responses; there are a few other things too.
19:34:16 -!- nooodl_ has joined.
19:34:16 <olsner> did watson remember to answer with a question every time?
19:34:23 -!- carado has joined.
19:34:51 <zzo38> Yes, there was practice so if there was that mistake in the program they would have corrected it immediately.
19:35:54 <fizzie> "You survived [299] days before killing your human."
19:37:17 <zzo38> Why is it in brackets?
19:37:28 <boily> it's a list of days.
19:38:07 -!- nooodl has quit (Ping timeout: 276 seconds).
19:39:08 -!- variable has quit (Ping timeout: 246 seconds).
19:39:31 -!- variable has joined.
19:41:01 <metasepia> Minecraft is a sandbox indie game originally created by Swedish programmer Markus "Notch" Persson and later developed and published by Mojang.
19:41:26 <metasepia> Year 376 was a leap year starting on Friday of the Julian calendar.
19:41:39 <fungot> olsner: there is at most one variable in the of the slot where it " points" for goading others to suicide and those points get them more power.
19:42:30 -!- sacje has joined.
19:42:36 -!- sacje has quit (Changing host).
19:42:36 -!- sacje has joined.
19:48:10 <olsner> gloomy fungot. glumgot.
19:48:11 <fungot> olsner: you keep saying, it's not considered a particularly bad swear word as far as i know what's a good book. his terms. like it, get yer lazy bum over to the writing system.
19:50:27 <boily> `learn glumgot is not a particularly bad swear word, but is still disquieting.
19:51:54 <HackEgo> boily is Canadian or something. We are not sure about Canada's existence.
19:55:36 <olsner> `pastelogs Canadian or something
19:55:52 <HackEgo> http://codu.org/projects/hackbot/fshg/index.cgi/raw-file/tip/paste/paste.20160
19:56:25 <olsner> `pastelogs seems to exist at least
19:56:39 <HackEgo> http://codu.org/projects/hackbot/fshg/index.cgi/raw-file/tip/paste/paste.11343
19:57:20 -!- zzo38 has quit (Quit: I also live at Canada.).
19:57:41 -!- epicmonkey has joined.
19:59:05 <boily> at least I'm something. I have non-zero mass!
20:45:06 -!- clog has joined.
20:46:00 -!- heroux has quit (Ping timeout: 256 seconds).
20:47:53 -!- heroux has joined.
20:49:24 <metasepia> mass definition: '''capitalized''' the liturgy of the Eucharist especially in accordance with the traditional Latin rite.
20:49:39 <boily> ~duck the other mass
20:49:43 <boily> ~duck physical mass
20:49:48 <metasepia> weight definition: the amount that a thing weighs.
20:53:46 <boily> aaaurgh. c'mon, eh?
20:54:08 <boily> ~duck irregardless
20:54:08 <metasepia> Irregardless is a word commonly used in place of regardless or irrespective, which has caused controversy since the early twentieth century, though the word appeared in print as early as 1795.
21:05:12 -!- epicmonkey has quit (Ping timeout: 268 seconds).
21:09:46 -!- copumpkin has quit (Ping timeout: 276 seconds).
21:10:34 -!- copumpkin has joined.
21:20:30 -!- nortti has quit (Ping timeout: 264 seconds).
21:20:37 -!- nortti has joined.
21:21:42 -!- oerjan has joined.
21:29:10 <boily> generic question: is it possible to SSH to a machine as user1@example.com, but you get logged in as user2?
21:29:31 <kmc> it's possible for things to be confused about what name is given to a particular UID
21:30:09 <boily> I mean, sshd is configured that way.
21:30:25 <kmc> i don't know of a particular common way that this happens
21:30:41 <boily> darn. my mischievous plans are preliminarily foiled.
21:30:50 <kmc> what are you up to...
21:31:07 <pikhq_> I suppose you could have the shell pointing to a suid program that logs in as user2?
21:31:12 <pikhq_> This would be insane of course.
21:32:28 <boily> well, I had in mind to offer a root shell to my machine to my coworkers, but a fake one with a fake shell.
21:32:33 <boily> (yes, I am bored.)
21:32:42 -!- conehead has joined.
21:32:59 <boily> pikhq_: insanity is good. it has marshmallows.
21:41:35 -!- Taneb has quit (Ping timeout: 268 seconds).
21:42:27 -!- Ghoul_ has joined.
21:49:42 <HackEgo> pbflist: shachaf Sgeo quintopia
21:50:26 <HackEgo> /home/hackbot/hackbot.hg/multibot_cmds/lib/limits: line 5: exec: egslist: not found
21:50:45 -!- Phantom_Hoover has joined.
21:53:46 <ion> http://www.reddit.com/r/IAmA/comments/1jal04/i_am_reza_aslan_scholar_of_religions_author_of/cbcu4q2?context=3
21:53:57 <Bike> the guy on fo?
21:54:15 -!- Taneb has joined.
21:55:34 -!- boily has quit (Quit: Poulet!).
21:55:49 -!- metasepia has quit (Remote host closed the connection).
22:03:02 -!- Taneb has quit (Ping timeout: 240 seconds).
22:19:51 <shachaf> ion: Did you do all the exciting exercises?
22:20:04 <ion> shachaf: I still haven’t got around to that.
22:20:12 -!- Taneb has joined.
22:21:39 <kmc> what kind?
22:21:52 <ion> LEM → callCC at least
22:22:08 <shachaf> I don't remember. Sometimes I give exercises to ion and eventually he gets around to doing them.
22:22:46 <ion> I have managed to read some pages of HoTT.
22:31:02 -!- sacje has quit (Ping timeout: 240 seconds).
22:32:14 -!- sacje has joined.
22:38:16 -!- Taneb has quit (Ping timeout: 264 seconds).
22:43:21 <kmc> Bitcoin is banned in Thailand now
22:43:22 -!- sacje has quit (Ping timeout: 276 seconds).
22:43:40 <Bike> i'm going to imagine that's an imperial edict.
22:44:19 -!- sacje has joined.
22:45:11 -!- Vorpal_ has joined.
22:47:56 -!- Vorpal has quit (Ping timeout: 260 seconds).
22:49:26 <Fiora> Bike: http://eprint.iacr.org/2004/199.pdf wow, this paper
22:49:52 <Fiora> "nope don't mind me just breaking 4 hash functions in one paper"
22:50:43 <Bike> this seems badly translated
22:50:46 <Fiora> http://link.springer.com/content/pdf/10.1007%2F11535218_2.pdf oh gosh, this guy did the SHA1 work too
22:50:49 <Bike> or uh, written
22:51:09 <Bike> straightforward though
22:51:56 <Fiora> xiaoyun wang is very much not a guy
22:52:07 <elliott> I like how the paper is just a bunch of collisions
22:52:23 <Fiora> so she apparently broke MD5, SHA-0, SHA-1, RIPEMD, and MD4
22:52:25 <Bike> elliott: straightforward!
22:54:06 <shachaf> kmc: MAP_GROWSDOWN is mostly used for mapping bird pages.
22:56:05 <kmc> i don't quite get it
22:57:05 <shachaf> imo i should've made a feather jokez instead "too late"
22:57:45 <kmc> is a growsdown a kind of bird or something
22:58:30 -!- jsvine has quit (Quit: Leaving.).
22:59:10 -!- zzo38 has joined.
22:59:10 <shachaf> https://en.wikipedia.org/wiki/Down_feather
22:59:24 <Fiora> I thought it would be useful for carrot pages because they grow down
22:59:35 <Fiora> ... shachaf that was actually amazing
22:59:38 <kmc> shachaf: :)
22:59:56 <Fiora> that was wonderful
23:01:07 <shachaf> it's not as specialized as MAP_GROWSUP, which has only ever been used by pixar
23:01:59 <Fiora> I thought it was like
23:02:28 <Fiora> puberty is madvise( person, sizeof(person), MADV_GROWSUP )
23:02:47 <shachaf> Obviously when a bird grows up it stops growing down.
23:18:04 <oerjan> puberty is when you're mad but think you are wise
23:24:19 -!- yorick has quit (Remote host closed the connection).
23:29:18 -!- sacje has quit (Ping timeout: 245 seconds).
23:32:31 -!- Sgeo has joined.
23:32:32 -!- sacje has joined.
23:32:33 -!- sacje has quit (Changing host).
23:32:34 -!- sacje has joined.
23:37:36 -!- shachaf_ has joined.
23:37:36 -!- shachaf_ has quit (Changing host).
23:37:36 -!- shachaf_ has joined.
23:37:51 <zzo38> What happens when you are wise and you think you are mad?
23:37:51 -!- shachaf has quit (Disconnected by services).
23:37:59 -!- shachaf_ has changed nick to shachaf.
23:41:14 <oerjan> zzo38: enlightenment hth
23:41:34 <shachaf> kmc: does zzo38 have the buddha nature
23:42:09 <kmc> there are slight linguistic parallels between zzo38 and Smullyan
23:42:46 <kmc> i'm not sure exactly
23:42:58 <kmc> just a few similarities in the way they write
23:43:32 <shachaf> i like smullyan's writing style
23:43:38 <kmc> me too, I think?
23:43:46 <shachaf> i also like zzo38's but in a different way
23:46:05 -!- Jafet has quit (Quit: Leaving.).
23:51:48 -!- iamfishhead has quit (Ping timeout: 245 seconds).
23:53:23 <Fiora> Bike: http://en.wikipedia.org/wiki/X-ray_pulsar_based_navigation wooow. it's like, GPS for spaceships
23:53:28 <Fiora> with pulsars instead of satellites
23:54:14 <Bike> theyre not too far away?
23:54:41 <Fiora> http://en.wikipedia.org/wiki/Pulsar_clock apparently they can do clocks too o_O
23:55:00 <Fiora> http://arxiv.org/abs/1305.4842v1.pdf yay the pulsar navigation thing has an arxiv
23:55:23 <Bike> "Digital processing of the pulsar signals is done by an FPGA device."
23:55:26 <Fiora> gosh that is actually really cool.
23:55:42 <Fiora> I'm imagining like flying my spaceship around and having a pulsar positioning system
23:55:45 <Fiora> "PPS, set course for pluto"
23:59:26 -!- madbr has joined.