| --- Log opened Fri Jan 20 00:00:19 2012 |
| 00:02 | | You're now known as TheWatcher[T-2] |
| 00:05 | | You're now known as TheWatcher[zZzZ] |
| 00:44 | | Derakon[AFK] is now known as Derakon |
| 00:49 | | * Derakon watches the Wat talk now that he's at a computer with Flash, is amused. |
| 00:56 | | eckse [eckse@Nightstar-a6cdd6ed.dsl.sentex.ca] has joined #code |
| 01:17 | | AD[Shell] is now known as ShellNinja |
| 01:24 | | maoranma [maoranma@Nightstar-1b3452f6.pools.spcsdns.net] has quit [Client closed the connection] |
| 01:32 | < celticminstrel> | ToxicFrog: I see "[Jan 19@18:02:09] <maoranma>" followed by two blank lines, several times in a row. |
| 01:33 | <@ToxicFrog> | celticminstrel: yes. It is as I explained. |
| 01:33 | <@ToxicFrog> | Try actually selecting it and you'll see each one is a line of hundreds of spaces. |
| 01:33 | <@ToxicFrog> | Being too long for the screen, it wraps to multiple lines. |
| 01:33 | < celticminstrel> | I thought spaces normally don't wrap... |
| 01:34 | <@ToxicFrog> | Wha? Spaces are a natural wrap point, most editors and chat clients will preferentially wrap at spaces if they can because then they aren't breaking up words. |
| 01:34 | < celticminstrel> | I'm sure at least some programs treat multiple spaces like a single space when it's at a wrap point... |
| 01:36 | <&McMartin> TF: I'm on irssi, on an 80-col screen, and it was two lines for me too |
| 01:36 | <&McMartin> Very odd. |
| 01:37 | <@ToxicFrog> | 4.5 lines here. |
| 01:38 | <@Tamber> | celticminstrel, right; but IRC isn't usually one of those. |
| 01:38 | < celticminstrel> | Shells in my experience wrap only at the column width though, so that seems less odd to me. |
| 01:38 | <@ToxicFrog> | And my screen is also at around 80col. |
| 01:48 | | Rhamphoryncus [rhamph@Nightstar-5697f7e2.abhsia.telus.net] has quit [Client exited] |
| 01:51 | < Eri> | Oh boy. The megaupload indictment is available on scribd, broken piece of crap that it is |
| 01:51 | < Eri> | Two pages work. I guess I wasn't interested in reading the rest |
| 01:53 | <&McMartin> One of the other channels speculates based on the nature of the charges - many of which have already been thrown out in similar cases - that it was written 2 years ago in the assumption that the CFSA would pass and hasn't been touched since. |
| 01:55 | < Eri> | Probably shelved, waiting for a good time to roll it out. I'd put money on this being damage control |
| 01:55 | <&McMartin> Nice timing, dudes. |
| 01:55 | < Eri> | "Yes, there was a huge blackout in opposition to SOPA/PIPA, but look, there's clearly pirates there" |
| 01:56 | <&McMartin> "And the existing legal system got them, so where exactly is the problem" |
| 01:56 | <&McMartin> Alternately, "And you fucked it up beyond all recognition, so why should we listen to you about anything ever" |
| 01:56 | <&McMartin> Two lines, no waiting, failure on both sides. |
| 01:57 | < Eri> | The problem is that megaupload is one among many. Then, they'll talk about how it takes years of work to take down a single website |
| 01:57 | < Namegduf> | They don't get their nice chilling effect against user-generated content sites, which was a stupid battle they'd lose. |
| 01:57 | < Eri> | I'm pretty sure it's an attempt to drum up sympathy for the MPAA |
| 01:57 | < Namegduf> | They want those sites to control content enough to stop people posting pirated stuff, and that just isn't ever going to happen. But introducing legal liability was the first step. |
| 01:57 | <&McMartin> Eri: Yeah, but their actual targets in their hurfery and durfery have been Amazon and Google. |
| 01:58 | < Namegduf> | Amazon? |
| 01:58 | <&McMartin> Amazon's music service got sued because they used dedup tech, which is explicitly in this complaint as The Evil Piracy |
| 01:58 | <&McMartin> This has been "yeah, uh, no"ed by the courts since. |
| 01:58 | < Namegduf> | Haha, wow |
| 01:59 | <&McMartin> But it was, by normal readings of what counts as being a distributor instead of a host, a grey area. |
| 01:59 | <&McMartin> Because it's no longer "user uploads stuff, we put it somewhere, they get it back later" |
| 01:59 | <&McMartin> It's "users show they have rights to a block of data, get the same data from the same point" |
| 02:00 | <&McMartin> However, in this case, the way they show it is by uploading it in its entirety so it's TMK been officially classified as a storage efficiency hack, not a change in what's really happening |
| 02:01 | <&McMartin> (If it were something like "give us the hash of the CD and you can download its contents as MP3s" you've stopped being Amazon Cloud and started being a ridiculously insecure version of Steam.) |
| 02:03 | < Namegduf> | Ah. |
| 02:03 | < Namegduf> | I thought they were accepting their loss there. |
| 02:03 | < Namegduf> | Silly me. |
| 02:06 | <&McMartin> Well |
| 02:06 | <&McMartin> At this point they probably get to get kicked in the face with precedent |
| 02:06 | <&McMartin> Like I said before, this whole thing reads like it was written 2 years ago and they've just filed it unchanged. |
| 02:36 | | maoranma [maoranma@Nightstar-1b3452f6.pools.spcsdns.net] has joined #code |
| 02:43 | | Derakon is now known as Derakon[AC2] |
| 02:51 | < maoranma> | Hey, somoene linked a video about email code that sucks, I can't find it again |
| 03:38 | | Attilla [Obsolete@Nightstar-85c2e515.as43234.net] has quit [Ping timeout: 121 seconds] |
| 03:56 | | Derakon[AC2] is now known as Derakon |
| 04:09 | | Derakon is now known as Derakon[AC2] |
| 04:12 | | Kindamoody[zZz] is now known as Kindamoody |
| 04:27 | | Eri [Eri@Nightstar-3e5deec3.gv.shawcable.net] has quit [Ping timeout: 121 seconds] |
| 04:39 | | Eri [Eri@Nightstar-3e5deec3.gv.shawcable.net] has joined #code |
| 04:42 | <@ToxicFrog> | McMartin: my wall fantasy here is SCO-level entrail feasting |
| 04:42 | <@ToxicFrog> | But there's no way it's going to happen |
| 04:42 | <&McMartin> It's not. |
| 04:43 | <&McMartin> And it won't for SOPA either, because SOPA is dual-purpose and one of those purposes they'll be able to get everyone in the country with a net-worth over $50k on board with |
| 04:47 | <@ToxicFrog> | What's the other purpose? |
| 04:49 | | Stalker [Z@Nightstar-3602cf5a.cust.comxnet.dk] has joined #code |
| 04:49 | <&McMartin> Anti-counterfeiting measures. |
| 04:50 | <@ToxicFrog> | Aah |
| 04:50 | <@ToxicFrog> | I seriously think there should be some sort of rule about cramming basically unrelated things into legislation |
| 04:50 | <&McMartin> More specifically, "online retaileres purporting to sell products from Respected Company X which are nothing of the sort" |
| 04:51 | <&McMartin> The easy solutions have unpleasant hacks |
| 04:51 | <&McMartin> Since it's fundamentally equivalent to a binding vote-alliance pact on unrelated bills. |
| 04:51 | <@ToxicFrog> | Cut down on the number of Ban Live Kitten Burnings AND ALSO Cut Funding For Planned Parenthood bills |
| 04:52 | < Eri> | Yup. It's a clever tactic, though |
| 04:52 | <&McMartin> The main problem is that when you remove the ability to do "AND ALSO direct federal contracting work to Uninterested But Bribable Asshole From District X" you can no longer get a majority for anything at all. |
| 04:52 | < Eri> | I was about to say that, yeah. |
| 04:53 | <&McMartin> Given that banning the bribe aspect was one of the few things the Tea Party caucus tried to stick to, we now have a data point! |
| 04:53 | <&McMartin> 2011 was the least productive Congress in American history. |
| 04:58 | <&McMartin> https://gist.github.com/1641705 |
| 05:06 | | Stalker [Z@Nightstar-3602cf5a.cust.comxnet.dk] has quit [[NS] Quit: Into the hole again, we hurried along our way, into a once-glorious garden now seeped in dark decay.] |
| 05:26 | | iospace [Alexandria@Nightstar-635d16fc.org] has quit [Ping timeout: 121 seconds] |
| 05:26 | < Eri> | Does Linux have really strict filesystem format standards compliance? |
| 05:26 | < Eri> | Like, I've got an SD card in my android phone |
| 05:27 | <@ToxicFrog> | Er |
| 05:27 | <@ToxicFrog> | What do you mean by "really strict filesystem format standards compliance"? |
| 05:27 | < Eri> | I was gettign there |
| 05:28 | < Eri> | 'Kay. So, I've got this card in my phone, formatted FAT32. It works in the phone. When I plug it into the windows laptop, I can mount it from the phone. When I plug it into the desktop running Mint, it doesn't work |
| 05:28 | < Eri> | I tracked down the error a while ago, and it's a corrupted filesystem |
| 05:29 | < Eri> | So, I reformatted the sd card, and then it would work, right up until I restarted the phone, at which point the format gets changed, somehow, and now Mint says it's no good, again |
| 05:29 | < Eri> | The whole time, it still works in the phone, and in Windows |
| 05:29 | < Eri> | It's reproducible, but only seems to affect me. |
| 05:30 | < Eri> | It's also reproducible across SD cards |
| 05:30 | < Eri> | Weird, no? |
| 05:30 | < Eri> | So, I was wondering if Linux, or Mint in particular, adheres to a really strict standard for FAT32, while Windows is a bit more permissive |
| 05:33 | < maoranma> | I've never had that issue. what phone? |
| 05:34 | < Eri> | HTC Desire HD |
| 05:34 | < Eri> | I've checked out a lot of support forums, no one else has the same problem, from what I've seen |
| 05:34 | < maoranma> | Odd. My EVO never did that |
| 05:37 | <@ToxicFrog> | Eri: what happens if you fsck the filesystem from linux and then try mounting it? |
| 05:38 | <@ToxicFrog> | Also, I'd be kind of interested to see the delta between images of the card taken before and after a phone restart |
| 05:39 | < maoranma> | Also, are you formating the card with the phone itself, from windows, or from linux? |
| 05:39 | | Derakon[AC2] is now known as Derakon |
| 05:39 | < Eri> | Maybe I'll work on that over the weekend. Trying to get an application package together for a workterm at the moment. How can I get just the changes between versions? |
| 05:39 | < Eri> | Formatted it once from each |
| 05:40 | < maoranma> | Hmm |
| 05:41 | | himi [fow035@D741F1.243F35.CADC30.81D435] has quit [Ping timeout: 121 seconds] |
| 05:41 | <@ToxicFrog> | Eri: try the fsck first, and see what happens |
| 05:42 | <@ToxicFrog> | As for the changes - you'd take an image of the card, reboot the phone so that it stops working, take another image, and then use a binary diff program |
| 05:42 | <@ToxicFrog> | This is not a fast operation and requires a fair amount of hard drive space |
| 05:43 | < Eri> | 8 GB card, so at least 16 GB |
| 05:43 | < Eri> | Hmm |
| 05:43 | < Eri> | I've got room for that, somewhere |
| 05:44 | <@ToxicFrog> | This is why I suggest the fsck first. |
| 05:45 | <@ToxicFrog> | It's entirely possible that on the reboot the filesystem is getting marked unclean in some way that the linux driver cares about but the windows and phone ones don't. |
| 05:45 | <@ToxicFrog> | Although I wouldn't really expect that with vfat. |
| 05:45 | <@ToxicFrog> | (NTFS yes; it's a reverse-engineered format, so the linux NTFS driver is extremely paranoid about dirty filesystems) |
| 05:46 | <@ToxicFrog> | So, yeah. I'd say, first step, see if the output from dmesg and mount is at all useful; then run it through fsck and see what that says and if it helps; and try the binary diff. |
| 06:40 | | Derakon is now known as Derakon[AFK] |
| 06:42 | | eckse [eckse@Nightstar-a6cdd6ed.dsl.sentex.ca] has quit [Client closed the connection] |
| 06:48 | | celticminstrel [celticminst@Nightstar-5d22ab1d.cable.rogers.com] has quit [[NS] Quit: Kablammo! Celticminstrel is now unavailable for comment.] |
| 08:32 | | * jerith won at git! |
| 08:32 | <@jerith> | I had a branch containing two rewrites of my code. |
| 08:32 | <@jerith> | The first was an in-place rewrite that was better than the original, but not good enough. |
| 08:33 | <@jerith> | The second was a fresh start in a new subpackage that used the first rewrite as reference and to copy/paste from where appropriate. |
| 08:34 | <@jerith> | I wanted a fresh branch containing only the second rewrite and the original code. |
| 08:34 | <@jerith> | git rebase --interactive to the rescue! |
| 09:20 | | Rhamphoryncus [rhamph@Nightstar-5697f7e2.abhsia.telus.net] has joined #code |
| 10:28 | | You're now known as TheWatcher |
| 11:04 | <@TheWatcher> | Ugh. I need more tea. Just tried to write `my $logh = $self -> {"dbh"} -> prepare("SELECT * FROM ".$self -> {"dbh"} -> prepare("SELECT * FROM ".$self -> {"dbh"} -...` |
| 11:06 | <@Tamber> | Manually recursive? |
| 11:08 | <@TheWatcher> | Apparently the part of my brain that should have been working out which bit came after the FROM ". had gone back to sleep, so the part doing the typing was just repeating, or something ¬¬ |
| 11:10 | | Attilla [Obsolete@Nightstar-7b84d1fa.as43234.net] has joined #code |
| 11:21 | | Kindamoody is now known as Kindamoody|out |
| 13:33 | | gnolam [lenin@Nightstar-202a5047.priv.bahnhof.se] has joined #code |
| 14:35 | | * Vornicus fiddles with string representation of polynomial objects, discovers that there is a shitton of special cases here. |
| 14:42 | | iospace [Alexandria@Nightstar-635d16fc.org] has joined #code |
| 14:43 | | mode/#code [+o iospace] by ChanServ |
| 15:18 | <~Vornicus> (the big remaining ones now are, it shouldn't place a + sign on the highest order term, and it may need to wrap the coefficients in parentheses, but I can't figure out how to do the latter at all.) |
| 15:24 | <~Vornicus> Okay, now that polynomial and interval are tested to my satisfaction, I can actually do piecewise functions. |
| 15:25 | | Stalker [Z@Nightstar-5aa18eaf.balk.dk] has joined #code |
| 15:25 | <~Vornicus> (they're not extensively tested, but I have covered the bases as far as I can see.) |
| 15:40 | <@jerith> | What about the exponents?~ |
| 15:41 | < Eri> | Ha |
| 15:42 | < Eri> | That's good |
| 15:42 | < Eri> | Witr |
| 15:42 | < Eri> | *Wit |
| 15:43 | <~Vornicus> exponents? |
| 15:44 | <~Vornicus> Like, how exponents of 0 and 1 should special case? Yeah, got those covered. |
| 15:47 | <~Vornicus> ... |
| 15:47 | | * Vornicus missed the obvious joke there. |
| 15:47 | < Eri> | Yeah |
| 15:48 | <@jerith> | I was trying to come up with another one, but couldn't. |
| 15:48 | <@jerith> | Although a binary reference would have worked. |
| 15:58 | | * Vornicus gnaws on it. How to print piecewise functions... |
| 15:59 | <@jerith> | Vornicus: Don't print them. Just display them on the screen. Treekiller. |
| 15:59 | | * Vornicus patpats jerith |
| 16:00 | <~Vornicus> Also I don't think I'll be actually satisfied with this code until I get rational functions in here, but that I suspect is a little beyond my current skills: at that point I would /definitely/ need to be able to make this thing spew a different type when creating objects. |
| 16:01 | <@jerith> | Vornicus: Metaclasses? |
| 16:01 | <@jerith> | type()? |
| 16:01 | | * jerith has some test code that creates classes on the fly and then instantiates them. |
| 16:02 | <~Vornicus> The big deal is that rational functions that can be represented by polynomials, should be. |
| 16:02 | <~Vornicus> But that's beyond the scope of the current system, actually. |
| 16:03 | <~Vornicus> (I'm doing some really silly things in here; when something non-callable comes in to the piecewise function I actually turn it into a Polynomial, on the assumption that it's a number.) |
| 16:10 | | * Vornicus pokes at piecewise, tries to figure out what to test. |
| 16:13 | <@jerith> | ALL the things. |
| 16:17 | <~Vornicus> RIght, but how to test them is befuddling me for a moment. |
| 16:22 | <@jerith> | Test them by taking the derivative of a spline. ;-) |
| 16:22 | <~Vornicus> I haven't written derivative yet. |
| 16:23 | <~Vornicus> (or spline) |
| 16:37 | | Attilla_ [Obsolete@Nightstar-7b84d1fa.as43234.net] has joined #code |
| 16:39 | | * Vornicus gets the basis function for the b-spline written. Yay |
| 16:40 | | Attilla [Obsolete@Nightstar-7b84d1fa.as43234.net] has quit [Ping timeout: 121 seconds] |
| 16:41 | | Attilla_ [Obsolete@Nightstar-7b84d1fa.as43234.net] has quit [Ping timeout: 121 seconds] |
| 16:41 | | Attilla [Obsolete@Nightstar-b0d89b18.threembb.co.uk] has joined #code |
| 16:55 | | Attilla_ [Obsolete@Nightstar-7b84d1fa.as43234.net] has joined #code |
| 16:56 | | Attilla [Obsolete@Nightstar-b0d89b18.threembb.co.uk] has quit [NickServ (GHOST command used by Attilla_)] |
| 16:56 | | Attilla_ is now known as Attilla |
| 17:10 | < maoranma> | Wow |
| 17:10 | < maoranma> | I never knew that |
| 17:10 | < maoranma> | The hamster dance song |
| 17:11 | < maoranma> | Comes from the beginning of Disney's Robinhood |
| 17:12 | < maoranma> | Which I suppose is Whistle Stop |
| 17:18 | <~Vornicus> oh man. |
| 17:18 | <~Vornicus> Man, I never knew that either. |
| 17:20 | < maoranma> | And Dubstep rule 34 holds true: http://www.youtube.com/watch?v=iJn7UrDTraY |
| 18:07 | | maoranma [maoranma@Nightstar-1b3452f6.pools.spcsdns.net] has quit [Ping timeout: 121 seconds] |
| 18:08 | | maoranma [maoranma@Nightstar-1b3452f6.pools.spcsdns.net] has joined #code |
| 18:14 | | maoranma [maoranma@Nightstar-1b3452f6.pools.spcsdns.net] has quit [Ping timeout: 121 seconds] |
| 18:14 | | maoranma [maoranma@Nightstar-1b3452f6.pools.spcsdns.net] has joined #code |
| 18:38 | < maoranma> | I should experiment with making some classes in python |
| 18:38 | < maoranma> | But the official tutorial is difficult to read |
| 18:50 | | Kindamoody|out is now known as Kindamoody |
| 19:54 | | Kindamoody is now known as Kindamoody[zZz] |
| 20:16 | <~Vornicus> WHat other languages have you used? |
| 20:17 | < maoranma> | Well, that are similar to python, only mIRC scripting do I know very well |
| 20:17 | <~Vornicus> ANd dissimilar? |
| 20:19 | < maoranma> | Inform 7, a natural language programming for interactive fiction. Redcode assembly for the core wars game. HTML, some PHP and perl, but very little |
| 20:20 | < maoranma> | Very little C/C++ from doing minor edits to MUD codebases |
| 20:20 | < maoranma> | Some Tk/Tcl when dealing with eggdrop bots |
| 20:20 | < maoranma> | Basic, which really isn't relevant anymore, haha |
| 20:21 | <~Vornicus> Okay. |
| 20:25 | | Vash [Vash@Nightstar-8697fea9.wlfrct.sbcglobal.net] has joined #code |
| 20:25 | | mode/#code [+o Vash] by ChanServ |
| 20:26 | <~Vornicus> Basically all you need for classes for most things are the __init__ function (which is your constructor), and how to talk about the instance |
| 20:27 | <~Vornicus> There's some other goodies - the "special method names" are names for methods used internally by the runtime, like __add__ for implementing + |
| 20:27 | <~Vornicus> and decorators and properties, but honestly most of that is not used very often. |
| 20:28 | <@jerith> | I use decorators all the time. |
| 20:29 | <@jerith> | But you don't /really/ need them. |
| 20:30 | < maoranma> | Okay, from what I understand of classes from an object oriented stand point, is that they represent a set of values and functions for dealing with those values, or a set of similar functions, things like that |
| 20:31 | <~Vornicus> Correct. |
| 20:31 | <@jerith> | Objects are data-with-behaviour. |
| 20:31 | <~Vornicus> Well, okay. ALmost. Objects are that; classes define, um. classes of objects. |
| 20:31 | <@jerith> | Classes are templates for objects. |
| 20:31 | < maoranma> | Right |
| 20:32 | < maoranma> | It's like, I can make a human class |
| 20:32 | <@jerith> | Please don't. |
| 20:32 | <@jerith> | We've just shut down the last Terminator... |
| 20:32 | | * jerith flees. |
| 20:32 | < maoranma> | Then I can make a cooker class, that inhierents the human class |
| 20:33 | < maoranma> | then I can instances bob as a cooker |
| 20:33 | <@jerith> | *instantiate. |
| 20:33 | < maoranma> | uh, yes, typing will be poor atm since one hand is balancing laptop on foot |
| 20:33 | <~Vornicus> Sure. |
| 20:34 | <~Vornicus> THough actually that's not a very good model for that. |
| 20:34 | < maoranma> | perhaps |
| 20:35 | <~Vornicus> (you'd actually create a human class, and a "skill" class that has the cooking methods, and then give some humans that skill) |
| 20:35 | <@jerith> | \o/ |
| 20:35 | < maoranma> | Yea, that would make more sense |
| 20:35 | <@jerith> | Composition over inheritance! |
| 20:35 | <@jerith> | That's some quality wisdom right there. |
| 20:35 | <@ToxicFrog> | Incidentally, "let's make classes and objects that model real-world things" is a very common trap to fall into |
| 20:36 | < maoranma> | hehe, but they're easier to grasp |
| 20:36 | <@ToxicFrog> | The point of classes is to make the program easier to organize, not to slavishly imitate reality, and the latter has resulted in a lot of really, really terrible tutorials on OO programming >.< |
| 20:37 | < maoranma> | Well sure, I've seen it done with geometry too |
| 20:37 | < maoranma> | making a class of shape, with functions to alter the shape |
| 20:38 | < maoranma> | Someone needs to relate it to pokemans, then I'll get it, rofl |
| 20:39 | <&McMartin> The problem here is that subclassing in OO is the exact opposite operation from ontology most of the time. |
| 20:39 | < maoranma> | And superclassing? |
| 20:40 | <@ToxicFrog> | maoranma: doing it with shapes is in fact the Classic Terrible Example |
| 20:40 | <@ToxicFrog> | Or perhaps the other CTE, with Car being the original. |
| 20:40 | < maoranma> | hah, I think I've seen the car one in a C# book |
| 20:41 | < maoranma> | I've seen animals in a C++ one too |
| 20:41 | < maoranma> | describe a cat |
| 20:41 | < maoranma> | Make a cat, make the cat meow |
| 20:41 | <@ToxicFrog> | That's probably #3~ |
| 20:41 | <@ToxicFrog> | http://lists.canonical.org/pipermail/kragen-tol/2011-August/000937.html - Kragen Javier Sitaker, "Goodbye, shitty Car extends Vehicle object-orientation tutorial" |
| 20:44 | <~Vornicus> there's a couple of different things you use classes for. One is creating the basics -- and a common interface -- for a group of objects that behave differently in themselves but have commonalities in how you interact with them. |
| 20:44 | < maoranma> | <blank stare> |
| 20:44 | < Namegduf> | "You can't fake the ability to turn a duck into a penguin by moving its |
| 20:44 | < Namegduf> | duckness into an animal of some other species that can be replaced at |
| 20:44 | < Namegduf> | runtime." |
| 20:44 | < Namegduf> | There are people who consider this "good practice"? |
| 20:45 | <~Vornicus> For instance, you could make an -- i don't know -- chess piece class, and then your individual piece types implement functions that list off moves and attacks it can make. |
| 20:46 | <@ToxicFrog> | In fact I have a concrete example of this in Felt. |
| 20:46 | <~Vornicus> And then you get fairy chess just by specifying the rules for each of your fairy pieces, using the common interface. |
| 20:46 | <&McMartin> Namegduf: For a non-OO example of this... |
| 20:46 | <@ToxicFrog> | There's a Token class, which is basically "anything that you can pick up and move around" - it has common properties for reacting to mouse clicks by being picked up, and suchlike. |
| 20:46 | < maoranma> | Actually, that was something close to what I have in mind to program. There was a game called Dungeon Dice Monsters that never got a lot of play, and I want to make a python implementation of it |
| 20:47 | <&McMartin> Namegduf: In ML, when doing operations on lists, it deduces what the type must be. That type can have variables in it still if the function Doesn't Care. |
| 20:47 | <@ToxicFrog> | This is subclassed by ChessPiece, which has general rules for chess (eg, snap-to-grid when you drop it on the board, capturing other pieces when you drop it on them), and an interface for legal moves |
| 20:47 | <&McMartin> The classic there being length() on lists. |
| 20:47 | <@ToxicFrog> | And then individual piece types (such as Pawn or Knight) implement the legal-moves method so that they can hilight the board properly when picked up. |
| 20:48 | < Namegduf> | McMartin: That doesn't sound very much like "moving the <functionality of a class> into another class at runtime. |
| 20:50 | <&McMartin> Oh. |
| 20:50 | <&McMartin> Sorry, I hooked to ohard on "duck" and need lunch. |
| 20:50 | <&McMartin> The buzzwords for this are "Decorator" and "Chain of Responsibility". |
| 20:51 | <&McMartin> moving a class's functionality into another is a way of implementing the same things decorators do. |
| 20:51 | < Namegduf> | That's what they're trying to get at? Eeesh. |
| 20:52 | < Namegduf> | I took it as monkey patching. |
| 20:52 | < Namegduf> | Okay, objection withdrawn, composition is cool. |
| 20:53 | <&McMartin> Well, it *is* monkey patching, under the hood, but the whole point about HLLs is that you don't care about what goes on under the hood as long as you get your guaranteed results. |
| 20:53 | < maoranma> | "I don't know why it works, but it works"? |
| 20:54 | <&McMartin> "I do know why this works, and it's ugly when humans do it, but that's what robots are for" |
| 20:54 | <~Vornicus> (now, this said: Duck inherits Animal may be a good idea if you're building a game with ducks and many other animals in it, and then just implement the behavior of the duck. May be a good idea.) |
| 20:54 | <~Vornicus> (May not be.) |
| 20:55 | <~Vornicus> (Depends on how you need to think about these things.) |
| 20:55 | <&McMartin> More seriously, "monkey patching" is less, uh, monkey when your object is fundamentally "hash table full of lambdas" as opposed to "series of cunningly designed structs of lambdas designed to work when cast to one another" |
| 20:55 | <&McMartin> The latter being the C++ approach which requires careful structure of inheritance, the former of which is smalltalk/ObjC/Python which has a lot more leeway. |
| 21:00 | <@jerith> | Monkey patch inherits from primate diff? |
| 21:01 | < maoranma> | monkey.throw() |
| 21:12 | | * Vornicus fiddles with b-splines, works on writing the knot insertion algorithm. |
| 22:00 | <~Vornicus> er, ...no. |
| 22:01 | | * Vornicus somehow gets the order backwards. |
| 22:25 | <@jerith> | Removing knots? |
| 22:27 | <~Vornicus> No, actually I got over to working on bezier stuff, specifically de casteljau. I ended up traversing the array backwards somehow. |
| 22:28 | <~Vornicus> Which was kind of cool. |
| 22:28 | <~Vornicus> Getting something that looks exactly right but happens to go exactly backwards. |
| 22:31 | < maoranma> | Why is emacs hugeeeeee~ |
| 22:32 | <@Tamber> | ...you're in ED IS THE STANDARD! mode? |
| 22:33 | <@jerith> | maoranma: Emacs has everything. |
| 22:34 | <~Vornicus> emacs is a really cool tool that lets you do, well, anything. |
| 22:34 | < maoranma> | Bake pie? |
| 22:34 | <~Vornicus> THe problem then is that everybody makes stuff for it, so now it also does everything. |
| 22:36 | < maoranma> | I'm... |
| 22:36 | < maoranma> | STILL |
| 22:36 | < maoranma> | Uncompressing... |
| 22:37 | <@Tamber> | Well, if you didn't insist on using that coal-fired machine... |
| 22:41 | < maoranma> | ...it has a calendar |
| 22:41 | < maoranma> | wtf did I just download? |
| 22:41 | < maoranma> | Someone's attempt to make an OS in notepad? |
| 22:42 | <@jerith> | maoranma: M-x psychoanalyze-pinhead |
| 22:43 | < maoranma> | I hate you |
| 22:43 | < maoranma> | Also, I have SEEN the CONSING!! |
| 22:44 | < gnolam> | Recurse, sinners! The endp is nigh! |
| 22:46 | < gnolam> | ToxicFrog: the wording of that could have been... better: "Aside from having snorted coke through my nose over "nearly tautologic diagrams"" |
| 22:49 | | iospace is now known as io|HOCKEY |
| 23:02 | <&McMartin> This is why your classier distros have a smaller core Emacs package and then 100 extra packages for the various other things |
| 23:02 | <&McMartin> However, psychoanalyze-pinhead is now in the mental standard because people get sad if it's not there. |
| 23:03 | | eckse [eckse@Nightstar-55f7bb2e.dsl.sentex.ca] has joined #code |
| 23:04 | < maoranma> | Is there a gopher client? I see I can read news, mail and RSS |
| 23:04 | <@Tamber> | Shouldn't be that hard to make one, if one doesn't already exist (Which'd surprise me). Gopher's nice and simple. |
| 23:05 | <@TheWatcher> | there is a gopher mode, yes |
| 23:05 | <@TheWatcher> | and http/https |
| 23:05 | <&McMartin> maoranma: Also, it's less "make an OS in notepad" and more "put a text editor on top of a LISP machine emulator" |
| 23:05 | <@TheWatcher> | (irc, too) |
| 23:12 | <@ToxicFrog> | What's the old joke? |
| 23:12 | <@ToxicFrog> | "Emacs is a decent OS, all it's missing is a good text editor"? |
| 23:12 | <&McMartin> Yeah, this has it backwards, IMO. |
| 23:13 | <&McMartin> Emacs is a pretty shitty OS because jumping windows is clunky and once you have more than two the "command other window" commands lead only to madness. |
| 23:20 | < maoranma> | Yeeeeeeaaaaah |
| 23:21 | < maoranma> | gonna stick with notepad++, it doesn't hurt my head |
| 23:21 | <@Alek> | thassagoodone |
| 23:22 | <@Alek> | I also like NoteTab, or at least the Lite (free) version. :P |
| 23:22 | < maoranma> | Anyone tried Ninja-IDE? |
| 23:25 | | * Vornicus gets some other random stuff working. THe b-spline stuff is spectacularly hard to follow on that page. |
| 23:26 | <&McMartin> If you're on Windows, NP++ is the best of breed by a huge margin. |
| 23:26 | | * McMartin only uses Emacs on Linux and Mac. |
| 23:33 | <@ToxicFrog> | McMartin: better than JEdit? |
| 23:35 | <&McMartin> For reasons I cannot precisely define, JEdit never clicked for me. |
| 23:35 | <&McMartin> But I'll admit it might not be a huge margin there, in objective terms. |
| 23:39 | < maoranma> | Hmm, I guess I'll need to learn how to deal with XML inside python |
| 23:39 | <&McMartin> minidom package, standard library |
| 23:40 | <&McMartin> I can post some code that's the app I use to turn my XML writeups into webpages... |
| 23:42 | <@ToxicFrog> | I tried both some years ago, and my conclusion was basically "NP++ doesn't look ugly, but JEdit actually runs on all of my computers and also has the advantage of not using Scintilla, so it wins" |
| 23:43 | <@ToxicFrog> | (the underlying objection to Scintilla being "customizing or adding syntax hilighting modes requires recompiling the entire editor") |
| 23:43 | <&McMartin> Yeah, all three of those requirements you state aren't on the metrics I'm using, so~ |
| 23:46 | <@ToxicFrog> | I will concede that if you're only on windows, and you only work in languages NP++ supports out of the box, it's better. |
| 23:48 | <&McMartin> Right, and it supports natively even OCaml, Haskell, and NSIS, so I'm generally sitting pretty tehre. |
| 23:48 | <&McMartin> *there |
| 23:50 | <@ToxicFrog> | Last time I used it (which was, granted, some years ago), it didn't support Scala, it supported Lua poorly, and the syntax hilighting for C was unspeakably ugly. And since those are my three most used languages... |
| 23:52 | | Janus [NSwebIRC@Nightstar-27138e8c.res.rr.com] has joined #code |
| 23:53 | < Janus> | I have a question! |
| 23:53 | <&McMartin> Go for it |
| 23:53 | <&McMartin> In the meantime, ToxicFrog: What kind of problem scale do you use Scala in? |
| 23:53 | <&McMartin> I've had huge problems in the small scale with it |
| 23:53 | < Janus> | The depixeling algorithm Jer and Vorn've been working with. Has anyone used it to scale a sprite down? |
| 23:57 | <@ToxicFrog> | McMartin: medium-scale. I think the biggest thing I've written it in was a static analysis tool for CPL programs (a language used internally in Bluecoat). |
| 23:57 | <@ToxicFrog> | Tiny stuff tends to get done in lua or bash, depending. |
| 23:58 | <@ToxicFrog> | It would arguably have been a better choice for Felt than Lua, if not for the fact that I already had a huge Lua codebase for it by the time I learned Scala. |
| 23:59 | | * McMartin nods |
| --- Log closed Sat Jan 21 00:00:07 2012 |