| --- Log opened Fri Sep 25 00:00:43 2009 |
| 00:01 | | Orthia [Orthianz@Nightstar-2a20c8f3.xnet.co.nz] has quit [Connection reset by peer] |
| 00:09 | | Orthia [Orthianz@Nightstar-2a20c8f3.xnet.co.nz] has joined #Code |
| 00:10 | | * Rhamphoryncus plots to upset every other programmer in the world by using 1-based numbering rather than 0-based |
| 00:13 | <@MyCatVerbs> | Rhamphoryncus: be fair. Compromise. |
| 00:13 | <@MyCatVerbs> | 0.5 |
| 00:13 | < Rhamphoryncus> | heh |
| 00:14 | <@MyCatVerbs> | Which is apparently actually the case when writing shaders... |
| 00:14 | <@MyCatVerbs> | (0.5,0.5) being exactly the center of the first pixel. :D |
| 00:15 | < Rhamphoryncus> | heh |
| 00:15 | < Rhamphoryncus> | Because it's referring to a range, rather than a unit |
| 00:16 | <@MyCatVerbs> | Fair duds. |
| 00:16 | < Rhamphoryncus> | first pixel == 1st pixel == pixel #1 == *offset* of 0 from the first pixel |
| 00:26 | < Rhamphoryncus> | oi. I'm actually tempted to write a list subclass that uses 1-based indexing. Mostly because the objects contained use 1-based, so there's a mismatch |
| 00:29 | < gnolam> | Depends on the shader. |
| 00:29 | < gnolam> | And range issues are, well, an issue with /every/ graphics API. |
| 00:37 | | gnolam [lenin@Nightstar-38637aa0.priv.bahnhof.se] has quit [[NS] Quit: Z?] |
| 00:41 | | You're now known as TheWatcher[T-2] |
| 00:44 | | You're now known as TheWatcher[zZzZ] |
| 00:49 | | Orthia is now known as Reivthia |
| 01:00 | | AnnoDomini [farkoff@Nightstar-11c116b1.adsl.tpnet.pl] has quit [[NS] Quit: I am Magellan-R-MGE. I see your group is missing a Hygiene officer.] |
| 01:08 | <@McMartin> | Yeah, I don't think it's possible to design a range system that doesn't break somebody's problem horribly, making its solutiont terrible. |
| 01:25 | | Derakon[AFK] is now known as Derakon |
| 02:19 | | Reivthia [Orthianz@Nightstar-2a20c8f3.xnet.co.nz] has quit [Connection reset by peer] |
| 02:19 | | dmlandrum [dmlandrum__@Nightstar-54510c8f.sfldmi.ameritech.net] has joined #Code |
| 02:27 | | Orthia [Orthianz@Nightstar-2a20c8f3.xnet.co.nz] has joined #Code |
| 02:43 | | Attilla [The.Attilla@FBC920.642D35.0878D5.AE151E] has quit [[NS] Quit: ] |
| 03:36 | | Thaqui [Thaqui@27B34E.D54D49.F53FA1.6A113C] has joined #Code |
| 04:32 | | thalass is now known as Thalass|Domesticated |
| 05:42 | | Orthia [Orthianz@Nightstar-2a20c8f3.xnet.co.nz] has quit [Connection reset by peer] |
| 05:43 | | Reiv [82d94c4d@Nightstar-b4529b0c.zen.co.uk] has joined #Code |
| 05:51 | | Orthia [Orthianz@Nightstar-2a20c8f3.xnet.co.nz] has joined #Code |
| 06:35 | < Rhamphoryncus> | am I the only one who thinks Dykstra never saw Sesame Street? |
| 06:36 | < Rhamphoryncus> | He's managed to convince the computing world that 0 is 1 |
| 06:40 | <@Kazriko> | Counting from 0 is really convenient in many ways... |
| 06:41 | <@Kazriko> | far less in the way of adding +1 to things for comparisons and such. :) |
| 06:41 | | * Kazriko is an ex-pascal programmer that used to love 1-counting, but has since learned how nice 0 counting is. :) |
| 06:42 | < Rhamphoryncus> | I accept that is convenient in some ways. I don't accept the anecdotes of it being more convenient than the alternative. Nor have I read any direct justification for redefining 1 as 0 |
| 06:43 | <@Derakon> | #define ZERO 1 |
| 06:43 | < Namegduf> | Arrrrgh. |
| 06:43 | <@Derakon> | Pretty sure I saw that on The Daily WTF once~ |
| 06:43 | < Rhamphoryncus> | Derakon: probably |
| 06:43 | | * Kazriko shrugs, goes back to coding. |
| 06:44 | <@Kazriko> | Rhamphoryncus, Eh, ok, let me point this one out. for x=0;x<length;x++. With 1 counting, it's for x=1; x<length+1;x++. When doing list slicing, you'd have to add 1 to all your lengths to get the slices right. it's just a mess. |
| 06:45 | < Rhamphoryncus> | I found one person mention that base-10 uses n*10**0 as smallest digit. They didn't seem to think of 10**-1 :P |
| 06:45 | | * Derakon notes that "< foo + 1" is equivalent to "<= foo". |
| 06:45 | <@Kazriko> | Derakon, yeah, but I'm trying not to give him ammo. :) |
| 06:45 | <@Derakon> | Learn 2 debate, noob~ |
| 06:45 | <@Kazriko> | *sigh* |
| 06:45 | <@Derakon> | :) |
| 06:45 | | * Kazriko goes back to writing actual code. |
| 06:45 | < Rhamphoryncus> | Kazriko: yeah, that's a good rationale for supporting "slice of length x" directly in the language. Sorry, that wasn't your argument ;) |
| 06:46 | <@Kazriko> | Feel free to use whatever you want, it's not as if I care. I was just trying to explain the rationale. |
| 06:46 | <@Kazriko> | Why does everything have to be a debate instead of an explanation... |
| 06:46 | <@Derakon> | Because geeks tend to be pedantic and will argue over anything if they think it can prove their superiority? |
| 06:46 | < Rhamphoryncus> | I've seen that rationale, and a few related ones. They're good points. They also completely ignore the opposition |
| 06:47 | < Rhamphoryncus> | "You should get a blue car." "Why not a red one?" ".... you should get a blue car." |
| 06:47 | <@Kazriko> | And in my opinion, it's a stupid argument, use what is most convenient in your language. In every language I use, 0 is more convenient, so I'll use it. *shrug* |
| 06:48 | <@Kazriko> | if I were coding something in Pascal, i'd count from 1. I'm not though. |
| 06:48 | < Rhamphoryncus> | what brought this up today is that I was counting objects, presenting the number to the user (ie me), but then I want to go use that for a unit test.. well that number doesn't match the list indexes |
| 06:49 | | * Kazriko complains that Modbus stupidly counts from 1 on registers when the hardware it ties to counts from 0. :p |
| 06:49 | | * Rhamphoryncus seconds that |
| 06:49 | <@Derakon> | Yyyyeah, whatever system you use, it should be consistent. ¬.¬ |
| 06:50 | <@Kazriko> | They either need to label the hardware AI1 through AI12, or label the modbus registers 0 through 11... |
| 06:50 | | * Derakon heads to bed. |
| 06:50 | | Derakon is now known as Derakon[AFK] |
| 06:50 | <@Kazriko> | since you can't use 0 on modbus, they need to label the inputs starting at 1... |
| 06:50 | < Rhamphoryncus> | IMO, 0-based has a deeper justification if you have a limited number of bits |
| 06:51 | < Rhamphoryncus> | 3 bits, 8 values, naturally has a range of 0..7 |
| 06:52 | <@Kazriko> | the silliest thing in modbus is that over the wire, it's 0 based, but everyone who uses it says coil addresses start at 1, digital ins at 10001, analog ins at 30001, and holding registers at 40001. you subtract that from the register before sending it over the line. |
| 06:53 | < Rhamphoryncus> | heh |
| 06:55 | < Rhamphoryncus> | I wonder if you could justify using modulus instead. ie, register 8 is register 0 |
| 07:02 | | * Kazriko needs to buy a t-shirt that says "No, I don't care to debate what is better, counting from 0 or counting from 1." |
| 07:05 | | gnolam [lenin@Nightstar-38637aa0.priv.bahnhof.se] has joined #Code |
| 07:10 | | Kazriko is now known as Kazriknaway |
| 07:34 | | Reiv [82d94c4d@Nightstar-b4529b0c.zen.co.uk] has quit [[NS] Quit: CGI:IRC (EOF)] |
| 08:07 | < Rhamphoryncus> | I've come to the conclusion that base-0 isn't base-0. The underlying problem is with the positional numeric system (can't write 10 in a single digit), so it uses a slightly different system. It eliminates the value of 0. What's stored as 0 in our normal positional system literally means "one" in the base-0 system |
| 08:11 | | * gnolam ponders stateless PRNGs. |
| 08:13 | < Rhamphoryncus> | wha? |
| 08:14 | | * Rhamphoryncus will be back in about 20 minutes |
| 08:14 | < gnolam> | Well, not /truly/ stateless. But decent 1D/1D noise functions. |
| 08:14 | | AnnoDomini [farkoff@Nightstar-11c116b1.adsl.tpnet.pl] has joined #Code |
| 08:14 | | mode/#code [+o AnnoDomini] by Reiver |
| 08:29 | < gnolam> | Also: cha-ching! |
| 08:31 | < Rhamphoryncus> | back |
| 08:31 | < Rhamphoryncus> | 1D? |
| 08:37 | < gnolam> | The dimensionality. |
| 08:39 | <@Vornicus> | Complexity theory is awesome. |
| 08:40 | <@Vornicus> | I just replaced a factorial-time algorithm with a factorial-squared-time algorithm... and I improved the speed of the algorithm by... factorial time. |
| 08:42 | < Rhamphoryncus> | heh |
| 08:43 | | Vornicus is now known as Vornicus-Latens |
| 08:50 | <@Vornicus-Latens> | ...actually I improved it by exponential time, now that I look at it. |
| 08:51 | <@Vornicus-Latens> | (the central binomial coefficient approaches 4^x) |
| 08:59 | | Rhamphoryncus [rhamph@Nightstar-a62bd960.abhsia.telus.net] has quit [Client exited] |
| 09:14 | | You're now known as TheWatcher |
| 10:24 | | Attilla [The.Attilla@FBC920.642D35.0878D5.AE151E] has joined #Code |
| 10:24 | | mode/#code [+o Attilla] by Reiver |
| 11:14 | | Thaqui [Thaqui@27B34E.D54D49.F53FA1.6A113C] has quit [Client closed the connection] |
| 11:27 | | Attilla [The.Attilla@FBC920.642D35.0878D5.AE151E] has quit [Client closed the connection] |
| 11:28 | | Attilla [The.Attilla@FBC920.642D35.0878D5.AE151E] has joined #Code |
| 11:28 | | mode/#code [+o Attilla] by Reiver |
| 11:41 | < gnolam> | http://igoro.com/archive/human-heart-is-a-turing-machine-research-on-xbox-360-sh ows-wait-what/ |
| 12:03 | | * TheWatcher vaguely pokes Reiver |
| 12:32 | | crem [moo@Nightstar-8ca3eea7.adsl.mgts.by] has quit [Ping timeout: 121 seconds] |
| 12:34 | | * TheWatcher pours vast and untiluted burning hatred over the developers of Flash |
| 14:03 | < gnolam> | Why this time? |
| 14:06 | <@TheWatcher> | Oh, input boxes that give out html despite being told not to in the UI (manually setting <instance>.html = false; in the frame seems to work), Selection.setFocus(<instance>) doing fuck all (it should make the specified input box get focus, it does nothing) |
| 14:06 | <@TheWatcher> | and other similar crap like that |
| 14:07 | <@TheWatcher> | Whole damned thing is a pile of bugs and kludges with a horrible UI shoved on the front. |
| 16:00 | | Alek [omegaboot@Nightstar-ac83c982.emhril.sbcglobal.net] has quit [Ping timeout: 121 seconds] |
| 16:05 | | Alek [omegaboot@Nightstar-3d8f2590.emhril.sbcglobal.net] has joined #Code |
| 16:30 | | Kazriknaway is now known as Kazriko |
| 17:00 | | Rhamphoryncus [rhamph@Nightstar-a62bd960.abhsia.telus.net] has joined #Code |
| 18:03 | | SmithKurosaki [Smith@Nightstar-82ad4434.dsl.teksavvy.com] has quit [Connection closed] |
| 18:05 | | SmithKurosaki [Smith@Nightstar-82ad4434.dsl.teksavvy.com] has joined #Code |
| 19:41 | | gnolam [lenin@Nightstar-38637aa0.priv.bahnhof.se] has quit [Ping timeout: 121 seconds] |
| 19:42 | | gnolam [lenin@Nightstar-38637aa0.priv.bahnhof.se] has joined #Code |
| 20:00 | | Finale [c0cb88fe@Nightstar-14e5d099.mibbit.com] has joined #Code |
| 20:19 | | You're now known as TheWatcher[afk] |
| 20:30 | | Derakon[work] [Derakon@Nightstar-d44d635e.ucsf.edu] has joined #Code |
| 20:32 | | * Derakon[work] crosses his fingers.h |
| 20:32 | < Derakon[work]> | s/h// |
| 20:33 | < Derakon[work]> | I've added a one-second sleep after moving the microscope stage, and increased the delay between photos from 23ms to 300ms. And it's been running for...about three hours now without crashing. |
| 20:34 | < Finale> | nice |
| 20:45 | | AbuDhabi [farkoff@Nightstar-380ba2c3.adsl.tpnet.pl] has joined #Code |
| 20:47 | | AnnoDomini [farkoff@Nightstar-11c116b1.adsl.tpnet.pl] has quit [Ping timeout: 121 seconds] |
| 20:48 | | Finale [c0cb88fe@Nightstar-14e5d099.mibbit.com] has quit [[NS] Quit: http://www.mibbit.com ajax IRC Client] |
| 20:49 | < Rhamphoryncus> | yay for horrible bodges! |
| 20:50 | < Derakon[work]> | It doesn't count if the software I'm modifying is itself a horrible bodge! |
| 20:52 | < AbuDhabi> | If it works it isn't stupid. |
| 20:52 | < Derakon[work]> | False! |
| 20:52 | < Derakon[work]> | This software works, mostly, and I can assure you it's stupid in many places. |
| 20:52 | < Derakon[work]> | That doesn't prevent it from being useful, mind you. |
| 21:02 | < Rhamphoryncus> | I agree with that |
| 21:03 | < Derakon[work]> | It's the classic tradeoff between "do it right, slowly" and "do it wrong, quickly". Both solve the problem the user has, both make useful software. One is fast but makes future developer slower and more error-prone. The other is slow and makes future development easier. |
| 21:04 | < Derakon[work]> | Of course, typically this choice isn't made consciously. The guy who wrote this software originally was just getting it done the way he knew how. |
| 21:05 | | Finale [c0cb88fe@Nightstar-14e5d099.mibbit.com] has joined #Code |
| 21:11 | < Rhamphoryncus> | Did you ever verify that the pure python version had the problem? |
| 21:24 | < Derakon[work]> | No, I haven't moved in that direction since the pure python version is also horribly out of date. |
| 21:24 | < Derakon[work]> | Updating it to work properly is going to be a major chore. |
| 21:24 | < Derakon[work]> | Also: it crashed. :( |
| 21:26 | < Derakon[work]> | Current test: excising the code that is directly around the crash. |
| 21:27 | < Derakon[work]> | Next test: getting rid of moving the microscope stage altogether, since that's the major difference I see between the crashing experiments and the not-crashed-in-five-years experiments. |
| 21:27 | < Derakon[work]> | (By "directly around" I mean "where the logs say we stop working, even though since this looks like memory corruption that probably has nothing to do with the fault") |
| 21:52 | | * Vornicus-Latens returns home, ready to add improvements to his RPS finder. |
| 21:54 | < Derakon[work]> | Rock Paper Scissors finder? |
| 21:54 | <@Vornicus-Latens> | Yes. |
| 21:55 | <@Vornicus-Latens> | I have written a Python script that will find all possible fair RPS rulesets with n symbols. |
| 21:55 | < Derakon[work]> | Ahh. |
| 21:56 | <@Vornicus-Latens> | (there's 3 for n = 7) |
| 21:56 | < Derakon[work]> | Are there 2 for n=3, or 1? |
| 21:56 | <@Vornicus-Latens> | 1. |
| 21:56 | < Derakon[work]> | Righto. |
| 22:00 | | * Vornicus-Latens destroys magic numbers! |
| 22:02 | | * Vornicus-Latens drops an entire exponential factor out of the number of rulesets to try! |
| 22:02 | <@Vornicus-Latens> | (Unfortunately it's /still/ factorial time) |
| 22:04 | <@Vornicus-Latens> | ---Actually I lie; that section was exponential in the first place. |
| 22:04 | <@Vornicus-Latens> | Now it's just less so. |
| 22:05 | | * Vornicus-Latens now does battle with another exponential factor. |
| 22:35 | | * Derakon[work] eyes a comment that includes the datestring "20091718". |
| 22:35 | < Finale> | ... |
| 22:36 | < Finale> | the only SENSIBLE interpretation is the 20th of September, 1718. |
| 22:36 | <@jerith> | ... and Finale beat me to it. |
| 22:36 | | * AbuDhabi hands jerith the Bag of Shame. |
| 22:37 | <@jerith> | I was too busy obsessively hitting refresh on graphs of happiness related to work stuff. :-/ |
| 22:38 | | Finale [c0cb88fe@Nightstar-14e5d099.mibbit.com] has quit [[NS] Quit: back soon, otherwise gone for the weekend] |
| 22:54 | | * Vornicus-Latens gets it right the first try. |
| 22:54 | < Derakon[work]> | Woot. |
| 22:54 | <@Vornicus-Latens> | All right. Now that I've added in my improvements, I should be able to hit RPS9 quickly. |
| 22:54 | < Derakon[work]> | You should add a feature to pick random nouns for each of the actors and random verbs to relate each of the noun pairs. |
| 22:55 | | Kazriko [kaz@teela.arkaic.com] has quit [Ping timeout: 121 seconds] |
| 22:56 | <@Vornicus-Latens> | Pfff. Possibly after the processor is done. |
| 22:56 | <@Vornicus-Latens> | Oh. That does remind me though, I need to rejigger the printer so it prints unambiguously for 2-digit values. |
| 22:58 | < Derakon[work]> | <Fire hydrant> defenestrates <Plato's allegory of the cave> |
| 23:00 | <@McMartin> | <Helicopter> [invigorates] <Earthworm> |
| 23:02 | < Derakon[work]> | <Earthworm> [filibusters] <quasar> |
| 23:02 | <@Vornicus-Latens> | <Highlighter> [aggrandizes] <The Cathedral of St. John the Divine> |
| 23:03 | | Kazriko [kaz@teela.arkaic.com] has joined #Code |
| 23:03 | | mode/#code [+o Kazriko] by Reiver |
| 23:04 | <@jerith> | <Bohemian pretentiousness> [unequivocates] <utter disregard for sense of self> |
| 23:04 | < Derakon[work]> | Wait, I'm not up on my postmodern philosophy. Does that actually make sense? |
| 23:05 | <@Vornicus-Latens> | I think it actually might. |
| 23:05 | | Vornicus-Latens is now known as Vornicus |
| 23:05 | | You're now known as TheWatcher |
| 23:05 | <@jerith> | I hope not. |
| 23:05 | <@McMartin> | Someone could try. |
| 23:06 | <@McMartin> | "unequivocates" is one of those words that shouldn't mean anything, and which is by that very fact full of subtle nuance that the speaker has, naturally, mastered. |
| 23:06 | <@McMartin> | Therefore, the underlying truth that authors do not in fact have telepathic projection powers is a shocking new revelation that will turn philosophy on its head. |
| 23:09 | | * Vornicus sets his thing to 9, thinks he needs to add parallelization support! |
| 23:09 | <@Vornicus> | 'cause this is gonna take a while. |
| 23:14 | <@jerith> | Vorn: Is this thing a means to and end or an end in itself? |
| 23:15 | <@Vornicus> | An end in itself. |
| 23:16 | <@jerith> | If it isn't alread a ProjectEuler problem, it might be worth suggesting it... |
| 23:17 | <@Vornicus> | Possibly; it's a lot of thought though to even come up with these algorithms. |
| 23:17 | <@Vornicus> | And I don't know how I'd make it an Euler problem. |
| 23:19 | <@Vornicus> | As far as 9 goes, it takes about 0.44 seconds per candidate to discover its canonization. |
| 23:20 | | Derakon[work] [Derakon@Nightstar-d44d635e.ucsf.edu] has quit [Ping timeout: 121 seconds] |
| 23:33 | <@Vornicus> | at 1000 seconds, it's covered about 2300 candidates, and ten canonical graphs. It discovered the last four at around 240 seconds. |
| 23:35 | <@Vornicus> | 1174 seconds: 11 and 12! |
| 23:35 | <@Vornicus> | 14... |
| 23:40 | | AbuDhabi [farkoff@Nightstar-380ba2c3.adsl.tpnet.pl] has quit [[NS] Quit: I Make Stuff Up. From the moon. Made by mooninites. They wear moonboots.] |
| --- Log closed Sat Sep 26 00:00:58 2009 |