| --- Log opened Sun May 04 00:00:10 2014 |
| 00:25 | | HotShot[ipad] [HotShot@Nightstar-v7se27.try.wideopenwest.com] has joined #code |
| 00:37 | | Derakon[AFK] is now known as Derakon |
| 00:37 | | mode/#code [+ao Derakon Derakon] by ChanServ |
| 00:53 | | Typherix [Typherix@Nightstar-n91qrf.lnngmi.sbcglobal.net] has quit [Ping timeout: 121 seconds] |
| 00:58 | | Typherix [Typherix@Nightstar-n91qrf.lnngmi.sbcglobal.net] has joined #code |
| 00:59 | | * McMartin looks at the FAQ for the Bumblebee project. |
| 00:59 | <&McMartin> | Q. Where did the name "Bumblebee" come from? |
| 00:59 | <&McMartin> | A. There is more to it than meets the eye. |
| 01:00 | <&McMartin> | They go on to actually explain the joke (that is, it's a stopgap Linux system for interoperating with nVidia Optimus technology), but there is no better way to start that answer. |
| 01:04 | | JackKnife [Z@Nightstar-484uip.cust.comxnet.dk] has quit [Ping timeout: 121 seconds] |
| 01:07 | <@Azash> | https://www.ruby-forum.com/topic/4570470 |
| 01:07 | <@Azash> | "Even though the language is extremely awesome, the implementation cannot advance as it should, heavily in part due to this Japanese culture." |
| 01:07 | <@Azash> | ??? |
| 01:10 | <@Azash> | (it should be noted that the person mentioned in the topic is, IIRC, the sole maintainer of DateTime) |
| 01:25 | | celticminstrel [celticminst@Nightstar-ha4ihv.dsl.bell.ca] has quit [[NS] Quit: KABOOM! It seems that I have exploded. Please wait while I reinstall the universe.] |
| 01:25 | | celticminstrel [celticminst@Nightstar-ha4ihv.dsl.bell.ca] has joined #code |
| 01:25 | | mode/#code [+o celticminstrel] by ChanServ |
| 01:34 | | * McMartin Buildcops to the Iji soundtrack. |
| 01:36 | <&McMartin> | These days buildcop duty has more to do with restraining our unruly robots than keeping the organics in line. :/ |
| 02:35 | | macdjord [macdjord@Nightstar-7rac1r.mc.videotron.ca] has quit [Ping timeout: 121 seconds] |
| 02:39 | | macdjord [macdjord@Nightstar-7rac1r.mc.videotron.ca] has joined #code |
| 02:39 | | mode/#code [+o macdjord] by ChanServ |
| 02:51 | | HotShot[ipad] [HotShot@Nightstar-v7se27.try.wideopenwest.com] has quit [Ping timeout: 121 seconds] |
| 03:05 | <&ToxicFrog> | Holy shit |
| 03:05 | <&ToxicFrog> | Timbre has the prettiest stack traces |
| 03:07 | <&McMartin> | ? |
| 03:09 | <&ToxicFrog> | Logging library for clojure |
| 03:09 | <&ToxicFrog> | If you pass a Throwable to one of the logging functions as the first argument it extracts a stack trace from it and displays that along with the message. |
| 03:10 | <&ToxicFrog> | It looks like this: http://i.imgur.com/bY3vKe2.png |
| 03:12 | <&ToxicFrog> | I mean, the actual content of the stack trace is godawful, but :clojure: |
| 03:14 | <&ToxicFrog> | Also, new ifirc release. Minor bugfixes, pretty much. |
| 03:16 | <~Vornicus> | Those are pretty pretty |
| 03:23 | <~Vornicus> | McM: does buildcop ride with axecop? |
| 03:29 | | HotShot[ipad] [HotShot@Nightstar-t3672h.sfldmi.sbcglobal.net] has joined #code |
| 03:31 | | HotShot[ipad] [HotShot@Nightstar-t3672h.sfldmi.sbcglobal.net] has quit [[NS] Quit: Leaving] |
| 03:35 | <&McMartin> | Sadly no |
| 03:47 | | himi [fow035@Nightstar-v37cpe.internode.on.net] has quit [Ping timeout: 121 seconds] |
| 04:03 | | himi [fow035@Nightstar-v37cpe.internode.on.net] has joined #code |
| 04:03 | | mode/#code [+o himi] by ChanServ |
| 04:11 | | VirusJTG [VirusJTG@Nightstar-6i5vf7.sta.comporium.net] has quit [[NS] Quit: Program Shutting down] |
| 04:26 | <&McMartin> | http://bumbershootsoft.wordpress.com/2014/05/04/cycle-exact-delays-on-the-6502/ |
| 04:32 | <&McMartin> | Also, I can't see a series of NOP NOP NOP commands without also thinking of http://img.pandawhale.com/82387-badger-NOPE-fuck-it-gif-Imgur-oKPf.gif |
| 04:36 | <&ToxicFrog> | Thank god for macros |
| 04:56 | | Datawalke [Datawalke@Nightstar-ks2sek.pa.comcast.net] has joined #code |
| 04:57 | | Datawalke [Datawalke@Nightstar-ks2sek.pa.comcast.net] has quit [[NS] Quit: ] |
| 05:00 | | NSGuest22852 is now known as Datawalke |
| 05:04 | | Harlow [harlow@Nightstar-9hnfdm.il.comcast.net] has joined #code |
| 05:05 | < Harlow> | When did OSs start protecting their ram from other ram on the system? |
| 05:06 | <&McMartin> | In general, or in consumer OSes? |
| 05:06 | <&McMartin> | Windows NT generally (so, Win2k or XP in consumer space) for Windows. |
| 05:06 | <&McMartin> | OSX 10.0 for apples. |
| 05:06 | <&Derakon> | I remember a cheat program for my (pre-OSX) Mac. |
| 05:06 | <&McMartin> | I want to say CTSS for the world at large, clear back in the 1960s. |
| 05:06 | <&Derakon> | It scanned the specified program's RAM for values you wanted. |
| 05:07 | <&Derakon> | A few iterations of the search would narrow things down and you could then set the value to whatever you wanted. |
| 05:07 | <&McMartin> | Well, you can still do that if you're root or the equivalent. I'm taking this question as meaning "a wild pointer doesn't scribble over potentially EVERY PROGRAM RUNNING EVERYWHERE" |
| 05:07 | < Harlow> | I feel so happy when I forget something in my code and Xcode just stops running the program |
| 05:07 | < Harlow> | I am some what relived knowing that its not BREAKING my os whilst running |
| 05:07 | <&McMartin> | Win95/98/ME actually shared the top half of RAM across every program ever |
| 05:07 | <&McMartin> | Including the entirety of the core OS data structures. -_- |
| 05:07 | <&McMartin> | gj |
| 05:08 | < Harlow> | What |
| 05:08 | < Harlow> | lel |
| 05:08 | < Harlow> | Mcmartin is that really true? |
| 05:08 | <&McMartin> | Yep |
| 05:08 | <&McMartin> | It was a case study in my OS class. |
| 05:09 | <&McMartin> | (Remember, Win95 was built on top of DOS, which was pretty much a "I'm exposing all the hardware at once" kind of thing. Hardware-level memory protection/task switching didn't really come into its own until Literally Everyone had at least a 386.) |
| 05:09 | < Harlow> | when did os 10 come out? |
| 05:10 | <&McMartin> | 2001 |
| 05:10 | <@Alek> | wait, really? |
| 05:10 | <@Alek> | OSX? |
| 05:10 | <&McMartin> | Yeah |
| 05:10 | <&Derakon> | Yep, it was the New Thing when I first entered college. |
| 05:10 | <@Alek> | oh wow, it's been THAT long? |
| 05:10 | <&Derakon> | For awhile I was dual-booting because the Classic environment wasn't quite up to snuff. |
| 05:10 | <&Derakon> | (Classic within OSX, that is) |
| 05:11 | <&McMartin> | Of note because OSX was where MacOS finally also got pre-emptive multitasking. |
| 05:11 | <@Alek> | weren't they still selling iFruit then? the candy cases? |
| 05:11 | <&Derakon> | Yeah, though that was towards the end of that line, I think. |
| 05:11 | <&Derakon> | They moved to the "monitor on a swivel arm" iMac after that. |
| 05:11 | <&McMartin> | Yeah, the brushed-steel effects didn't come into play until near the Intel swapover |
| 05:11 | | RchrdB [RichardB@Nightstar-c6u.vd5.170.83.IP] has quit [[NS] Quit: Gone.] |
| 05:12 | < Harlow> | What is "pre-emptive multitasking" |
| 05:12 | | * macdjord still has a pre-OSX iMac on his desk back in his parents house. Keeps it around mostly to read old Mac CDs and to play Ambrosia games. |
| 05:12 | | Derakon is now known as Derakon[AFK] |
| 05:12 | <&McMartin> | If you've got a bunch of programs running at once, pre-emptive multitasking means that the OS is responsible for divvying up time between them, and can basically make them stop after they run out of their share of time. |
| 05:12 | <&McMartin> | (a "timeslice") |
| 05:13 | < Harlow> | so is this why even under full load Mac's seem to let the user work? |
| 05:13 | <&McMartin> | Classic MacOS and Win3.1 had "cooperative multitasking", where a process needs to explicitly give up the processor so someone else can run. |
| 05:13 | <@macdjord> | Harlow: "Okay, program X, you stop for a minute and program Y can run" as opposed to "Okay, program X, you can run - and remember to stop running in 0.1 second so program Y can have a chance" |
| 05:13 | <&McMartin> | MS has actually had pre-emptive multitasking since Windows 95, by, IIRC, hooking the MS-DOS timer interrupt. |
| 05:15 | | RchrdB [RichardB@Nightstar-c6u.vd5.170.83.IP] has joined #code |
| 05:16 | <~Vornicus> | The long story short is that in preemptive multitasking, the OS tells the program when it can run; in cooperative multitasking, the program tells the OS when other programs can run. |
| 05:17 | <~Vornicus> | Preemptive also typically comes with a thing where you say "I don't need anything right now; wake me up when there's an event to be handled" |
| 05:18 | < Harlow> | so is there a reason that when a Mac locks up its just the program that locks? (or so it seems) |
| 05:18 | <&McMartin> | Kinda |
| 05:18 | <&McMartin> | Windows gets this too but it tends to be more upfront about it when it's the app and not the OS locking. |
| 05:19 | <&McMartin> | If the OS locks, you're hosed regardless |
| 05:19 | <&McMartin> | But as a rule OS X, being Unix-based, is a lot better at that than Windows. |
| 05:19 | <@macdjord> | Harlow: Interesting thing - due to cooperative multitasking and a couple Interesting Bugs, you could lock up not just your own Mac but /every Mac on the same network/ by holding down shift at the right time. |
| 05:20 | < Harlow> | macdjord, was this back in 2000? |
| 05:21 | <&McMartin> | Pre-2000 |
| 05:21 | <@macdjord> | Not sure, I read about it a while ago and I don't remeber the details. |
| 05:21 | < Harlow> | weird, sounds like a bad bug. |
| 05:22 | <&McMartin> | Yeah, it's at "design flaw" levels and it's one of the first places where MS got on the ball before Apple did. |
| 05:23 | < Harlow> | what exactly is "Kernel panic" and why can it Never (to my knowledge) be recovered from? |
| 05:23 | <&McMartin> | That's the Unix equivalent of BSoD - a program crash inside the core OS. |
| 05:24 | <&McMartin> | There's nothing left to rescue it, so the only hope is to restart the entire system. |
| 05:26 | < Harlow> | What exactly happens that there's nothing |
| 05:26 | <&McMartin> | Hrm. |
| 05:26 | < Harlow> | "nothing left to rescue it" |
| 05:26 | < Harlow> | * |
| 05:26 | <&McMartin> | So, the hardware has some way of noticing that something extremely bad has happened |
| 05:27 | <&McMartin> | divide by zero, reference to memory that isn't actually hooked up, etc. |
| 05:27 | <&McMartin> | It handles this, basically, by checking a special part of memory whose job it is to handle that, and this points to OS code |
| 05:27 | <&McMartin> | That code generally shuts down the program, and on Windows, puts up the "this program has performed an illegal operation &c". |
| 05:28 | <&McMartin> | If the code *responsible for that kind of thing* goes wrong, it ends up in an infinite loop, conceptually |
| 05:28 | <&McMartin> | (In practice, you can nest this a few times, but some handler is The Last Handler.) |
| 05:28 | < Harlow> | hmmm and there are no double redundancies when it comes to os handling? |
| 05:28 | <&McMartin> | And if the Last Handler fails, or if it otherwise knows it can't recover (common for device driver failures), it just pops up a screen that basically says ":-(" and leaves it at that. |
| 05:29 | <&McMartin> | Oh, the redundancies go down to like five or six levels, easily. |
| 05:29 | <&McMartin> | But it's not hard for all six to fail. |
| 05:29 | <&McMartin> | And the closer to the metal you get, the fewer there are. |
| 05:29 | <&McMartin> | If something goes horribly wrong while reading from, like, a PCI bus or something, you're pretty much screwed. |
| 05:30 | <&McMartin> | "I can't talk to PCI, I can't do anything useful" |
| 05:30 | <&McMartin> | Older systems tend to just go catatonic without warning. |
| 05:30 | <&McMartin> | But the way you're supposed to get extra redundancy for things like kernel panics is to *not do things in the kernel*, more or less. |
| 05:31 | <&McMartin> | A malfunctioning kernel module might be able to be split out, but it might deliberately crash itself on the grounds the malfunctioning module might have subtly corrupted something. |
| 05:31 | < Harlow> | yikes. |
| 05:32 | <&McMartin> | Cartridge-based stuff often set the "something has gone terribly wrong" jump point to the cartridge initialization routine, so a buggy or fatally glitched-out cartridge game would generally act like you pushed RESET instead of giving an error screen. |
| 05:32 | <&McMartin> | I think the GameCube is the first console I know of where that wasn't true. |
| 05:32 | < Harlow> | can there be problems with the kernel on a bios chip even? |
| 05:32 | <@macdjord> | Harlow: Yes. |
| 05:32 | <@macdjord> | There Is Always One More Bug. |
| 05:32 | < Harlow> | :( that would be sad. |
| 05:32 | <&McMartin> | Sure; flash your BIOS with a fucked-up firmware and you can brick your system but good. |
| 05:32 | <@macdjord> | Unless you're NASA. |
| 05:33 | <&McMartin> | That tends to get you back in the old world of "system just sits there doing nothing" |
| 05:33 | <&McMartin> | That said, BIOSes tend to be among the most thoroughly tested of software because doing *anything else* with a computer of that era involves putting the BIOS through its paces. |
| 05:34 | <&McMartin> | UEFI, well... |
| 05:34 | <&McMartin> | UEFI looks a Hell of a lot like they're implementing a smallish OS in firmware to me. |
| 05:34 | <&McMartin> | (Compare BIOS, which is "on start, read the first 512 bytes of the disk and jump to it. Good luck!") |
| 05:35 | <&McMartin> | That said, "kernel" *usually* means a piece of the OS that is in the center of it. How big it is depends on the OS. |
| 05:36 | < Harlow> | 512 bytes? Is that arbitrary or really how small the first read is? |
| 05:37 | <&McMartin> | That's really how small the first read is. Those 512 bytes are responsible for getting more general disk-reading code in to read the rest of the bootloader, which ultimately loads proper drivers, and then the OS, and so on and so forth until you are off to the races. |
| 05:37 | < Harlow> | thats terrifying. |
| 05:37 | <&McMartin> | It's also, for Intel/AMD systems, mandated to be in the 16-bit assembler mode, so those 512 bytes probably also set you up to go into 32- or 64-bit mode too. |
| 05:37 | <&McMartin> | And hey, 512 bytes is four times as much memory than the Atari 2600 had RAM! |
| 05:38 | <&McMartin> | (That was small even for the time; full-scale systems of that era tended towards tens to hundreds of KB) |
| 05:38 | < Harlow> | wow |
| 05:38 | <@celticminstrel> | I think OSX doesn't even bother to pop up a screen on kernel panic. |
| 05:39 | < Harlow> | No, I've seen it. |
| 05:39 | <&McMartin> | Classic MacOS had the X_X page, right? |
| 05:39 | < Harlow> | Not cool. |
| 05:39 | < Harlow> | yeah, if it happens now its just a bunch of text in a bunch of languages telling you to turn of your computer with with an IO in the background. |
| 05:40 | <@celticminstrel> | I've seen the kernel panic screen on my laptop. I'm pretty sure I've had a couple of kernel panics on this computer, but the screen just locked up and stopped updating. |
| 05:40 | <@celticminstrel> | I suppose I could've been wrong about them being kernel panics. |
| 05:40 | <&McMartin> | If your window manager hard-locks you're basically out of luck even if the core OS is fine, yeah. |
| 05:40 | <&McMartin> | (This is where Windows tends to have its biggest stability problems these days. Vista helped some, but not as much as one would like.) |
| 05:41 | <@celticminstrel> | Are you suggesting that's what happened to me> |
| 05:41 | <@celticminstrel> | ^? |
| 05:41 | <&McMartin> | I'm suggesting it's possible. I have no idea from a distance, of course. |
| 05:41 | <&McMartin> | I would like very much to know how Firefox manages to make all my other windows nonresponsive on Windows without triggering Windows's "application is not responding, apply axe to face Y/N" dialog |
| 05:42 | | himi [fow035@Nightstar-v37cpe.internode.on.net] has quit [Ping timeout: 121 seconds] |
| 05:42 | <@celticminstrel> | Ah. |
| 05:44 | <&McMartin> | ... oh hey, a copy of Hammurabi |
| 05:44 | <@macdjord> | McMartin: ... you found a giant stone pillar with rather brutal laws on it? |
| 05:45 | <&McMartin> | One of the earliest management sims. |
| 05:46 | <&McMartin> | Notable to me because the way to get the highest ratings was to happen to be hit by major plagues, thus resulting in both very low starvation rates and *dramatic* increases in per-capita land ownership~ |
| 06:00 | | himi [fow035@Nightstar-v37cpe.internode.on.net] has joined #code |
| 06:00 | | mode/#code [+o himi] by ChanServ |
| 06:30 | | himi [fow035@Nightstar-v37cpe.internode.on.net] has quit [Ping timeout: 121 seconds] |
| 06:39 | < Harlow> | I'm having some trouble with types and it iterators. If I'm creating a template class why does it not like an iterator with a definition of class type "Type". |
| 06:40 | < Harlow> | could there be a reason for that? if anyone wants to take a look its line 23 http://pastebin.com/mZsPfPRA |
| 06:43 | | himi [fow035@Nightstar-v37cpe.internode.on.net] has joined #code |
| 06:43 | | mode/#code [+o himi] by ChanServ |
| 06:45 | <@celticminstrel> | ... you haven't defined an iterator type in stackTClass... |
| 06:45 | <@celticminstrel> | ...wait, is it normal to declare your constructor like that? |
| 06:47 | < Harlow> | Well this constructor was given to me |
| 06:48 | <@celticminstrel> | I see. |
| 06:49 | <@celticminstrel> | Anyway, line 23 doesn't work because there's no member type iterator in stackTClass. |
| 06:50 | < Harlow> | right its just type no? |
| 06:50 | <@celticminstrel> | I'm not really sure what this is supposed to be, but I suppose a "typedef Type* iterator" would do the trick. Or, if C++11. "using iterator = Type*". |
| 06:50 | <@celticminstrel> | Unless you're doing something strange here. |
| 06:53 | < Harlow> | not really just want to compare two of the same classes with type type apparently. I would just say they're going to be strings, but that could be incorrect. |
| 06:54 | <@celticminstrel> | So what you really want is Type::iterator? |
| 06:54 | <@celticminstrel> | Though, I notice the "itor" variable isn't even used... |
| 06:54 | < Harlow> | did i really leave that code out.... :/ |
| 06:56 | < Harlow> | http://pastebin.com/ZTrysaWh |
| 06:56 | < Harlow> | sorry thats where its being used. |
| 06:58 | <@celticminstrel> | ...but stackTClass doesn't have a begin() function... |
| 07:00 | < Harlow> | http://pastebin.com/3hLrRWNd third time's the charm |
| 07:00 | <@celticminstrel> | What are you trying to do here? Is this supposed to be a stack? If so, is there some reason not to just include <stack>? |
| 07:01 | <@celticminstrel> | Oh, well, that looks kinda better. |
| 07:01 | <@celticminstrel> | Have you tried compiling this yet? |
| 07:02 | <@celticminstrel> | But, if it's supposed to be a stack, what's wrong with std::stack? |
| 07:02 | < Harlow> | yes, but i cannot because begin is not a thing so don't pay attention to that. I need to switch how they compare, think my main problem was just getting an iterator |
| 07:03 | <@celticminstrel> | Confused, |
| 07:03 | <@celticminstrel> | Begin is not a thing? |
| 07:03 | < Harlow> | no. it is not. not in this class, for stacks in general yes. |
| 07:04 | <@celticminstrel> | I still don't understand why you're doing this. |
| 07:06 | < Harlow> | its just a lab for school, nothing else. i just needed to know the iterator work around. Its rather pointless if you ask me. |
| 07:09 | <@celticminstrel> | Um, okay... |
| 07:17 | < Harlow> | Thanks for the help, Im going to need to take another look at this anyways, not using the standard stack definition is stupid. |
| 07:17 | | Harlow [harlow@Nightstar-9hnfdm.il.comcast.net] has quit [[NS] Quit: BED] |
| 07:32 | | himi [fow035@Nightstar-v37cpe.internode.on.net] has quit [Ping timeout: 121 seconds] |
| 07:45 | | himi [fow035@Nightstar-v37cpe.internode.on.net] has joined #code |
| 07:45 | | mode/#code [+o himi] by ChanServ |
| 07:46 | | celticminstrel [celticminst@Nightstar-ha4ihv.dsl.bell.ca] has quit [[NS] Quit: And lo! The computer falls into a deep sleep, to awake again some other day!] |
| 08:10 | | JustBob [justbob@ServerAdministrator.Nightstar.Net] has quit [[NS] Quit: ] |
| 08:13 | | JustBob [justbob@ServerAdministrator.Nightstar.Net] has joined #code |
| 08:13 | | mode/#code [+o JustBob] by ChanServ |
| 08:28 | | himi [fow035@Nightstar-v37cpe.internode.on.net] has quit [Ping timeout: 121 seconds] |
| 08:30 | | JackKnife [Z@Nightstar-484uip.cust.comxnet.dk] has joined #code |
| 08:30 | | mode/#code [+o JackKnife] by ChanServ |
| 08:40 | | himi [fow035@Nightstar-v37cpe.internode.on.net] has joined #code |
| 08:40 | | mode/#code [+o himi] by ChanServ |
| 09:55 | | Orthia [orthianz@Nightstar-3tp.juj.184.203.IP] has quit [Ping timeout: 121 seconds] |
| 09:58 | | AverageJoe [evil1@Nightstar-fb1kt4.ph.cox.net] has joined #code |
| 10:00 | | luke [flukiluke@Nightstar-q2b7ep.blinkenshell.org] has joined #code |
| 10:02 | | Orthia [orthianz@Nightstar-3tp.juj.184.203.IP] has joined #code |
| 10:02 | | mode/#code [+o Orthia] by ChanServ |
| 10:16 | | macdjord is now known as macdjord|slep |
| 10:34 | | Orthia [orthianz@Nightstar-3tp.juj.184.203.IP] has quit [Ping timeout: 121 seconds] |
| 10:57 | | Syka [the@Nightstar-t98.7kd.131.1.IP] has joined #code |
| 10:58 | | Syka is now known as NSGuest32259 |
| 11:00 | | Orthia [orthianz@Nightstar-3tp.juj.184.203.IP] has joined #code |
| 11:00 | | mode/#code [+o Orthia] by ChanServ |
| 11:04 | | AverageJoe [evil1@Nightstar-fb1kt4.ph.cox.net] has quit [[NS] Quit: Leaving] |
| 11:10 | | Orthia [orthianz@Nightstar-3tp.juj.184.203.IP] has quit [Ping timeout: 121 seconds] |
| 11:28 | | NSGuest32259 is now known as Syk |
| 11:34 | | Orthia [orthianz@Nightstar-3tp.juj.184.203.IP] has joined #code |
| 11:34 | | mode/#code [+o Orthia] by ChanServ |
| 12:01 | | VirusJTG [VirusJTG@Nightstar-6i5vf7.sta.comporium.net] has joined #code |
| 12:02 | | himi [fow035@Nightstar-v37cpe.internode.on.net] has quit [Ping timeout: 121 seconds] |
| 12:15 | | himi [fow035@Nightstar-v37cpe.internode.on.net] has joined #code |
| 12:15 | | mode/#code [+o himi] by ChanServ |
| 12:54 | | Vornicus [Vorn@ServerAdministrator.Nightstar.Net] has quit [Connection reset by peer] |
| 13:30 | <@ErikMesoy> | The terms "context-free grammar" and "recursive descent parser" sound like very strange names for parts of the thing I'm supposed to be writing for this programming class: a grammar checker which shall accept sentences like "I know the actor whom she likes" and reject sentences like "she likes he". |
| 13:31 | <@Azash> | They do make a lot of sense |
| 13:31 | <@Azash> | CFG is a good way to represent grammar |
| 13:31 | <@Azash> | Er, CFL |
| 13:31 | <@Azash> | (Context-free language) |
| 13:31 | <@Azash> | ErikMesoy: I recommend you get Sipser's Introduction to the Theory of Computation from some Free Online Legal Library Service or similar |
| 13:32 | <@ErikMesoy> | Oh, I'm not saying the parts are bad. I'm saying they have names I can't get used to. |
| 13:32 | <@Azash> | It covers them |
| 13:32 | <@Azash> | Ah |
| 13:32 | <@ErikMesoy> | http://covers.oreilly.com/images/9780596516499/cat.gif <-- What I've got for reference at present |
| 13:33 | <@Tamber> | A cat is fine too. |
| 13:34 | <@ErikMesoy> | Tamber: I can't tell if you're joking about the filename or if that redirects to a cat pic to avoid scrapers for external viewers. |
| 13:34 | < luke> | The link is fine. |
| 13:34 | <@ErikMesoy> | For me it shows a red-and-white cover of "Natural Language Processing with Python" with weird whales (which I believe is the usual O'Reilly style of letting the books be identified by animals). |
| 13:35 | <@Tamber> | Erik: The former. :) |
| 13:36 | <@Tamber> | Though the latter would be mildly amusing. |
| 13:38 | | himi [fow035@Nightstar-v37cpe.internode.on.net] has quit [Ping timeout: 121 seconds] |
| 13:44 | <@ErikMesoy> | This subject has a high number of fast changes from "I have no idea what I'm doing" to "Oh this is simple, just an annoying amount of implementation details and edge cases" for me. |
| 13:45 | <@Azash> | Anyway Erik while you might not find the terms very natural, remember that they are the names for the generic formal methods and you're just using a specific CFL that you're parsing |
| 13:46 | <@ErikMesoy> | Even the term "context-free language" strikes me as weird. What would its counterpart be? A language where one conjugates by whether the speaker is indoors or outdoors? |
| 13:47 | <@Azash> | Well, lo and behold, its counterpart is the context-sensitive language |
| 13:48 | <@Tamber> | Erik: A language where the meaning of each word depends on the ones surrounding it, presumably? |
| 13:48 | <@Azash> | Tamber: Context-free basically means that adding possible characters/strings/whatever is just a simple replace |
| 13:49 | <@Azash> | eg. A -> AB | null, B -> foo | null |
| 13:49 | <@ErikMesoy> | Tamber: But English isn't context-free by that standard! :V |
| 13:49 | <@Azash> | So the CFL in question produces (foo)* |
| 13:50 | <@Azash> | While a context-sensitive language can have more complicated replaces |
| 13:50 | <@Tamber> | Azash, ah! |
| 13:50 | <@Azash> | Like BAB -> BAAB or whatever |
| 13:50 | <@Tamber> | Erik: Exactly, it isn't. ;) |
| 13:50 | | JackKnife [Z@Nightstar-484uip.cust.comxnet.dk] has quit [Ping timeout: 121 seconds] |
| 13:50 | <@ErikMesoy> | Then explain how come I find myself writing a context-free grammar to parse English sentences. :p |
| 13:50 | <@Azash> | Basically the context-(free|sensitive) determines whether you can specify a context for the item that will be replaced |
| 13:51 | | himi [fow035@Nightstar-v37cpe.internode.on.net] has joined #code |
| 13:51 | | mode/#code [+o himi] by ChanServ |
| 13:51 | <@Azash> | ErikMesoy: Because you don't need a context-sensitive grammar to do English grammar structure |
| 13:52 | <@Azash> | There's a difference to be made between the meanings of context in the formal language method and the idea of context in the English grammar itself |
| 13:52 | <@Azash> | http://www.pling.org.uk/cs/com6791img/phrasestructuretree.png |
| 13:52 | <&ToxicFrog> | ErikMesoy: NLP? Have fun. |
| 13:53 | <@ErikMesoy> | Azash: It looks to me as though a context-sensitive grammar can be emulated by a context-free grammar with sufficiently complicated replace rules. Is this wrong? |
| 13:54 | <@Azash> | ToxicFrog: Don't think it's processing as much as parsing |
| 13:54 | <@Azash> | ErikMesoy: I can't remember, tbh |
| 13:54 | <@Azash> | But I don't think so |
| 13:54 | <&ToxicFrog> | ErikMesoy: you are wrong. |
| 13:55 | <&ToxicFrog> | (have you covered the Chomsky hierarchy yet?) |
| 13:56 | <@ErikMesoy> | ToxicFrog: No. |
| 13:56 | <@Azash> | ErikMesoy: Sipseeerrrrrrrrrrr |
| 13:56 | <@ErikMesoy> | So far there has been one mention of Chomsky, and that was in Further Reading. |
| 13:59 | <&ToxicFrog> | So, the Chomsky hierarchy orders languages based on what kind of formal grammar is necessary to describe them (or, equivalently, what kind of machine is needed to recognize them -- or it orders grammars based on what kinds of languages they can describe, it's all kind of the same thing) |
| 14:01 | <&ToxicFrog> | The smallest category is the regular languages, corresponding to regexes and DFAs; a superset of those are the context-free languages, corresponding to CFGs and PDAs. |
| 14:02 | <&ToxicFrog> | A superset of those is the context-sensitive languages, corresponding to CSGs and LBAs. Any context-free language can be described by a CSG, but the converse is not true. |
| 14:04 | <&ToxicFrog> | And then at the top you have type 0 grammars, the unrestricted grammars, consisting of anything that can be recognized by a Turing machine. |
| 14:04 | <&ToxicFrog> | And yeah, Sipser's Intro to the Theory of Computation is pretty excellent. |
| 14:05 | <@ErikMesoy> | Does Gödel get in on this anywhere with grammars that only describe language which are complete and consistent? :p |
| 14:11 | <&ToxicFrog> | It's been a while since I looked at Gödel, but my understanding is that axiom systems are not the same thing as grammars or the languages they describe :P |
| 14:26 | | JackKnife [Z@Nightstar-ro94ms.balk.dk] has joined #code |
| 14:26 | | mode/#code [+o JackKnife] by ChanServ |
| 14:32 | <@ErikMesoy> | Well, I was thinking there might be some variant of "Sentence S: Turing machine T cannot recognize sentence S" involved. |
| 14:43 | <@Tarinaky> | http://i.imgur.com/MnnHR3f.png << Kerbal Space Program |
| 14:45 | <@Tarinaky> | Caption: Jeb kerman's Mon's vibrator is here |
| 14:45 | <@Tarinaky> | *mom's |
| 14:46 | < [R]> | <McMartin> I would like very much to know how Firefox manages to make all my other windows nonresponsive on Windows without triggering Windows's "application is not responding, apply axe to face Y/N" dialog <-- probably by making everythig have to get shit from swap. IME if Windows is hitting swap, it's going to be slow as sin. |
| 15:06 | <&ToxicFrog> | ErikMesoy: Chomsky says nothing about semantic meaning, though - only recognition. |
| 15:17 | <@ErikMesoy> | All right. |
| 15:45 | < [R]> | <Lugia> part of the problem is java ruining the ability to write good code |
| 16:45 | | Reiv [NSwebIRC@Nightstar-q8avec.kinect.net.nz] has quit [Ping timeout: 121 seconds] |
| 16:47 | | Syk [the@Nightstar-t98.7kd.131.1.IP] has quit [Connection closed] |
| 17:29 | | Typherix [Typherix@Nightstar-n91qrf.lnngmi.sbcglobal.net] has quit [Connection closed] |
| 17:29 | | Typherix_ [Typherix@Nightstar-n91qrf.lnngmi.sbcglobal.net] has joined #code |
| 17:34 | | celticminstrel [celticminst@Nightstar-ha4ihv.dsl.bell.ca] has joined #code |
| 17:34 | | mode/#code [+o celticminstrel] by ChanServ |
| 18:31 | | macdjord|slep is now known as macdjord |
| 19:28 | | mode/#code [+o RchrdB] by ChanServ |
| 20:21 | | [R] [rstamer@genoce.org] has quit [Operation timed out] |
| 20:31 | | [R] [rstamer@Nightstar-d7h8ki.org] has joined #code |
| 22:02 | | JackKnife [Z@Nightstar-ro94ms.balk.dk] has quit [Ping timeout: 121 seconds] |
| 22:02 | | Vornicus [Vorn@ServerAdministrator.Nightstar.Net] has joined #code |
| 22:03 | | mode/#code [+qo Vornicus Vornicus] by ChanServ |
| 23:34 | | Reiv [NSwebIRC@Nightstar-q8avec.kinect.net.nz] has joined #code |
| 23:34 | | mode/#code [+o Reiv] by ChanServ |
| 23:36 | | himi [fow035@Nightstar-v37cpe.internode.on.net] has quit [Ping timeout: 121 seconds] |
| 23:49 | | himi [fow035@Nightstar-v37cpe.internode.on.net] has joined #code |
| 23:49 | | mode/#code [+o himi] by ChanServ |
| --- Log closed Mon May 05 00:00:26 2014 |