| --- Log opened Fri Nov 13 00:00:06 2009 |
| --- Day changed Fri Nov 13 2009 |
| 00:00 | <@Vornicus> | I'm lost. |
| 00:00 | < Derakon[work]> | Also, for the most part I hope to leave the interactions with other computers untouched for now, just because this is already a complicated project. |
| 00:00 | < Derakon[work]> | Vorn: okay, say I have a function that does "for i in range(100): yield i". |
| 00:00 | <@Vornicus> | Right |
| 00:01 | < Derakon[work]> | I do "foo = function; foo(); foo.next(); foo.next();" etc. |
| 00:01 | <@Vornicus> | Python fucks with that internally to create an object. |
| 00:01 | < Derakon[work]> | Between each call to next() I can interact with the UI, check for status updates, etc. |
| 00:02 | < Derakon[work]> | In general, assuming I'm reading this right, the function is actually checking a global state variable to see if the thing it's waiting for is done yet; that variable is updated by another thread. |
| 00:03 | <@Vornicus> | ugh |
| 00:03 | < Derakon[work]> | For example, the function spins off a MoveStageThread class instance; that thread speaks to another computer to move the stage, and then sets the global variable when it's done. |
| 00:03 | < Derakon[work]> | Yeah, even if the underlying concepts stay, that design's gonna get the boot. |
| 00:04 | < Derakon[work]> | I could easily rework it to have the function return the name of the flag it's waiting to be set, and have the caller do the blocking. |
| 00:05 | < Derakon[work]> | (Oh man, I am going to refactor this code so hard...) |
| 00:06 | < Derakon[work]> | (I bet I can get the 357-line sebRun.py file down to under a hundred lines) |
| 00:13 | < Finale> | that's what SHE said! |
| 00:13 | <@Vornicus> | Idunno. Way I'd do it is... geh. setting a w... wait, you /can/ do that, if you tell it to try to set a property... |
| 00:14 | < Derakon[work]> | Hmmm? |
| 00:16 | <@Vornicus> | I was going to say set a watch |
| 00:17 | < Derakon[work]> | (Note: Googling for "Python watch" is unlikely to be helpful) |
| 00:17 | <@Vornicus> | ANd then I realized that what you can do is set up a property that can register callbacks... |
| 00:17 | <@Vornicus> | Like a debug variable watch |
| 00:18 | < Derakon[work]> | I'm not envisioning how this would work. Where is this property? On the Experiment class? Or on the thing that I'm waiting on? |
| 00:19 | < Derakon[work]> | (Side note: waiting for things is currently busywaiting. Fun!) |
| 00:22 | <@MyCatVerbs> | Derakon[work]: the traditional way to do this (maintain a pile of state but not have to do it manually) is coroutines, to which generators are a half-decent approximation. |
| 00:23 | <@MyCatVerbs> | The approach that you are using may even be sane. |
| 00:24 | | * Derakon[work] tries to talk to the other programmer on his team, who immediately gets sidetracked by the question of if there's error handling if the hardware fails to do its job. Not the point, Mac. >.< |
| 00:24 | | You're now known as TheWatcher[T-2] |
| 00:24 | < Derakon[work]> | MCV: I can guarantee you that the approach currently being used is not entirely sane, because I could rewrite it to be cleaner without changing the basic principles. |
| 00:27 | | You're now known as TheWatcher[zZzZ] |
| 00:33 | | Derakon[work] [Derakon@Nightstar-d44d635e.ucsf.edu] has quit [[NS] Quit: Leaving] |
| 00:34 | | Orthia [Orthianz@Nightstar-407ad43a.xnet.co.nz] has joined #code |
| 00:44 | < Finale> | hrm. |
| 00:44 | < Finale> | anyone up on X-Men lore? |
| 00:52 | < gnolam> | Why? |
| 00:54 | < Finale> | wondering if anyone knew of a male mutant with one blue eye and one an ugly brownish green? |
| 00:58 | < gnolam> | ... David Bowie? |
| 00:58 | | * gnolam hides. |
| 01:13 | | Derakon[AFK] is now known as Derakon |
| 01:19 | < gnolam> | And another vulnerability in Flash. Shocker. |
| 01:20 | < Finale> | O_o |
| 01:24 | | Finale [c0cb88fe@Nightstar-14e5d099.mibbit.com] has quit [[NS] Quit: http://www.mibbit.com ajax IRC Client] |
| 01:29 | < simon`> | ugh |
| 01:29 | < simon`> | "epistaticity" |
| 01:29 | < simon`> | that's the word of the day. |
| 01:51 | | * gnolam slaps simon` around a bit with a heteroskedastic stochastic process. |
| 01:52 | | * simon` is participating in a study group on metaheuristics while taking a statistics course, feeling the combination is incidentally very good. |
| 01:53 | | SmithKurosaki [Smith@Nightstar-6450f657.dsl.teksavvy.com] has quit [Connection closed] |
| 01:53 | < simon`> | although words like "epistaticity" make me think there's a whole bunch of things one should know before this makes any sense ;) |
| 01:53 | | SmithKurosaki [Smith@Nightstar-6450f657.dsl.teksavvy.com] has joined #code |
| 01:53 | < SmithKurosaki> | join #fleer |
| 02:18 | | * McMartin figures out how to bend Go's encoding/binary package to his will, which works on floats despite claiming not to. |
| 02:22 | < simon`> | is it pronounced as the boardgame? |
| 02:23 | <@Derakon> | I'm rather curious if they're going to continue to call it Go, seeing as there's already a programming language named "Go!". |
| 02:23 | < simon`> | they can always pay. |
| 02:24 | <@Derakon> | For backstory: http://code.google.com/p/go/issues/detail?id=9 |
| 02:24 | <@Derakon> | I heard someone suggesting they should rename it to "Issue Nine", which I find amusing. |
| 02:25 | <@McMartin> | Unless it's trademarked, there's no real way to protect two letters. |
| 02:25 | <@McMartin> | Otherwise Thompson could go after everything with the letter C in it~ |
| 02:26 | <@Derakon> | Yeah, this is more an issue of politeness. |
| 02:26 | <@Derakon> | Another suggestion: "go2". |
| 02:26 | < gnolam> | I suggest they rename it "Thunderbirds". |
| 02:29 | <@McMartin> | Modula-9~ |
| 02:29 | <@McMartin> | I just realized that the weird variable syntax is in fact precisely Pascal/Modula-3's, without 80% of the punctuation. |
| 02:31 | <@Vornicus> | "googlish" |
| 02:31 | <@McMartin> | var a [10]int; |
| 02:31 | <@McMartin> | var a:array[1..10] of Integer; |
| 02:31 | <@Derakon> | "Umlang". |
| 02:31 | <@McMartin> | The latter is Pascal |
| 02:32 | <@Derakon> | (Then we just need someone to make "Uhlang"~) |
| 02:32 | < gnolam> | Or just make it a bit more metal: Gö. |
| 02:33 | <@Derakon> | Gø, surely. |
| 02:34 | < gnolam> | Pfft. Umlauts are metal. Slashes aren't. Even if they represent the same letter. |
| 02:36 | <@McMartin> | http://hkn.eecs.berkeley.edu/~mcmartin/u4scan.go |
| 02:36 | <@McMartin> | Needs better formatting, but that comes next. |
| 02:37 | <@McMartin> | (output formatting, that is) |
| 02:59 | | Orthia [Orthianz@Nightstar-407ad43a.xnet.co.nz] has quit [Client closed the connection] |
| 03:14 | | Attilla [The.Attilla@FBC920.480E8C.3E679A.E258A8] has quit [Client closed the connection] |
| 03:38 | < gnolam> | Damn you, Stephenie Meyer. |
| 03:38 | < gnolam> | Every time I see the "Twilight" bookmark I have I feel dirty. |
| 03:40 | < gnolam> | (I need to tweak the light level of the scene during sunrise and sunset.) |
| 03:40 | <@McMartin> | Clearly, when "Twilight" is checked, you need to add gratuitous sparkles. |
| 03:42 | | Orthia [Orthianz@Nightstar-9fa3bf3f.xnet.co.nz] has joined #code |
| 03:43 | <@McMartin> | ... wait, why does Spiff not have emacs *or* vi installed |
| 04:09 | < simon`> | who/what is Spiff? |
| 04:09 | < simon`> | is there ed or nano? lacking vi is not very nice. |
| 04:16 | <@McMartin> | Spiff is my linux machine. It does have nano. |
| 04:18 | <@Reiver> | I thought vi came pretty much standard? |
| 04:19 | < simon`> | Reiver, then Linux became popular, and all the newbies complained, and now nano is often found as the standard editor. |
| 04:20 | <@Reiver> | I mostly thought that nano and vi were the standard editors. |
| 04:20 | <@McMartin> | ed is the standard editor. |
| 04:20 | < simon`> | we have a song at my dept. with a phrase that translates into "You don't learn anything without making errors, so make the learning curve steeeep." |
| 04:20 | <@Reiver> | ... what is ed? |
| 04:20 | < simon`> | Reiver, the standard editor. |
| 04:20 | < simon`> | ;-) |
| 04:21 | <@McMartin> | A prankish holdover from the teletype days. |
| 04:21 | | * Reiver throws simon` off the earth. |
| 04:21 | | * simon` bumps back |
| 04:21 | <@Vornicus> | http://www.gnu.org/fun/jokes/ed.msg.html |
| 04:21 | <@McMartin> | Basically, vi without the concession of an actual view of the text. |
| 04:21 | <@McMartin> | You must know. |
| 04:22 | <@Reiver> | O.o |
| 04:22 | <@McMartin> | It's from the days when teletypes were hooked to actual printers, so the text created was echoed directly to the output. |
| 04:24 | < simon`> | hmmm... |
| 04:25 | <@Derakon> | I kinda wonder how often ed gets used interactively these days. |
| 04:25 | < simon`> | I've got two and a half hours left. should I try to read and summarize a text about genetic programming or read a chapter in a basic statistics book and perform the exercises? |
| 04:25 | <@Derakon> | 150 minutes until what? |
| 04:25 | < simon`> | until lecture starts. |
| 04:26 | <@Derakon> | Hm. Genetic programming is likely to be more interesting, but statistics is likely to be more useful. Your call. |
| 04:26 | < simon`> | (that is, the statistics lecture, heh) |
| 04:26 | <@Reiver> | Read the stats. |
| 04:29 | | * simon` wants to overcome his tendency to procrastinate excessively |
| 04:29 | <@Derakon> | First step: log off of IRC. |
| 04:29 | < simon`> | ;-) |
| 04:29 | < simon`> | I did! several times! ;-) |
| 04:29 | <@Derakon> | And yet you're still here. |
| 04:30 | < simon`> | (it's like smokers that say they quit after every smoke) |
| 04:33 | | * Derakon idly eyes http://wordwarvi.sourceforge.net/ |
| 05:02 | <@Reiver> | Amusing, pity it's linux only :p |
| 05:03 | | Syloqs-AFH [Syloq@NetworkAdministrator.Nightstar.Net] has quit [Connection reset by peer] |
| 05:16 | | Orthia [Orthianz@Nightstar-9fa3bf3f.xnet.co.nz] has quit [Client closed the connection] |
| 05:16 | | Alek [omegaboot@Nightstar-744f51af.emhril.sbcglobal.net] has joined #code |
| 05:19 | | gnolam [lenin@Nightstar-38637aa0.priv.bahnhof.se] has quit [[NS] Quit: Z?] |
| 05:24 | | Orthia [Orthianz@Nightstar-9fa3bf3f.xnet.co.nz] has joined #code |
| 05:38 | <@McMartin> | Better formatting is in. http://hkn.eecs.berkeley.edu/~mcmartin/go/u4scan.go |
| 05:54 | | * Reiver pokes ToxicFrog experimentally. |
| 05:57 | | * ToxicFrog leaks numbers |
| 05:57 | | * Reiver steals them for his own nefarious purposes! |
| 05:58 | <@Reiver> | Specifically, I was wondering if you were still using the computer you set up to play Supreme Commander on, or if you'd upgraded since? |
| 06:00 | | * McMartin posts his preliminary thoughts on Go: http://mcmartin.livejournal.com/124794.html |
| 06:01 | <@ToxicFrog> | Reiver: I've added another gigabyte of memory, IIRC, but that's it. Why? |
| 06:01 | <@Reiver> | Well, I was gifted a lovely copy of Borderlands. |
| 06:02 | <@ToxicFrog> | Borderlands? |
| 06:02 | <@Reiver> | Take a Diablo action-rpg, dry-hump Fallout 3 with it, and make it look like a graphic novel. |
| 06:02 | < SmithKurosaki> | It's the FPS that just came out for 360/pc |
| 06:03 | <@Reiver> | Lots of fun. ^.^ |
| 06:03 | <@Reiver> | Unfortunately, this was the point in which I realised that my machine might be slightly aging, as it cannot run with sufficient stability beyond 640x480, non-antialiased. |
| 06:03 | <@Reiver> | This, uh, suggests that my Raedon 1650 Pro is, perhaps, a mite outdated...~ |
| 06:04 | <@Reiver> | I was mostly wondering, given your machine was approximately the same vintage as mine: Is it worth it to find a replacement card, or should I be ripping out the guts and installing new ones entirely? |
| 06:05 | <@ToxicFrog> | Assuming it's PCIe, you have a nice upgrade path for graphics |
| 06:05 | <@ToxicFrog> | Durandal is using a X1950 XT, which is of the same generation but a few tiers higher up. |
| 06:08 | <@Reiver> | Yeah, it was PCIe |
| 06:08 | <@Reiver> | Do you know how modern a card the machines of that generation should be able to take? My worry is that it's hard to find vintage cards for less than a low-end one in the store. |
| 06:11 | <@ToxicFrog> | Arbitrarily modern, PCIe is still the latest. |
| 06:11 | <@ToxicFrog> | The main worry is not "how recent a card can it take" but "is the bottleneck in fact the card and not the CPU" |
| 06:11 | <@Reiver> | PCIe doesn't have a speed throttle like AGP did with the x2 x3 x8 etc? |
| 06:11 | <@Reiver> | Oh, that one I am fairly certain of |
| 06:12 | <@ToxicFrog> | Um. It kind of does, but it's inherent to the port; a PCIe bus consists of a number of channels, which can be combined into wider connectors with more bandwidth |
| 06:12 | <@ToxicFrog> | All PCIe video cards use the widest port, PCIe x16, which combines 16 channels. |
| 06:12 | <@Reiver> | As the game will run quite smoothly in 640x480, but starts to get rather jagged and glitchy when I want to run it in my native resolution of 1280x1040. :p |
| 06:13 | < SmithKurosaki> | That sounds like a really weird res, what is the scale on thaht |
| 06:13 | <@Reiver> | er, 1024, but you know what I mean |
| 06:13 | < SmithKurosaki> | Ahh kk |
| 06:13 | <@Reiver> | Unless there's a significant chance that boosting the screen resolution would be stressing the CPU, TF? |
| 06:16 | < SmithKurosaki> | I know that when I wanted to play spellforce, I had to turn down the screen res to get it to run nicely |
| 06:16 | < SmithKurosaki> | (And i don |
| 06:17 | < SmithKurosaki> | 't think turning down the screen resolution actually affects your view |
| 06:17 | <@ToxicFrog> | SmithKurosaki: it doesn't affect your frustrum size (the amount of game-world you can see), at least in most 3d games. |
| 06:18 | <@ToxicFrog> | Reiver: unlikely. |
| 06:18 | <@ToxicFrog> | Also, how did 1280x1024 get to be a popular resolution anyways? |
| 06:18 | <@ToxicFrog> | It's 5:4! |
| 06:18 | < SmithKurosaki> | That's what frustrum means! |
| 06:18 | <@Reiver> | What does 1024x768 factor down to? |
| 06:19 | < SmithKurosaki> | 3:4 |
| 06:19 | <@ToxicFrog> | 4:3 |
| 06:19 | < SmithKurosaki> | tHAT |
| 06:19 | <@Reiver> | Hn |
| 06:19 | <@ToxicFrog> | Like every other pre-widescreen resolution |
| 06:19 | <@Reiver> | In that case I can blame soley LCD monitors, which are lousy for having it. |
| 06:19 | <@Reiver> | Is it half a widescreen scale, maybe? |
| 06:19 | | * Reiver is clutching at straws, but hey |
| 06:19 | <@ToxicFrog> | Nope |
| 06:19 | <@ToxicFrog> | Widescreen is 16:10 |
| 06:19 | <@ToxicFrog> | Or 16:9 for the cinema |
| 06:20 | <@Reiver> | Go fig, then. |
| 06:20 | <@Reiver> | Though I do find the smidge more vertical height is useful all the same. |
| 06:20 | <@Reiver> | (For browsing, etc, where you rarely need the full width.) |
| 06:21 | < SmithKurosaki> | That is true |
| 06:21 | <@Vornicus> | In some games, higher resolution means that it tries to render farther away (because more distant things will show up as more pixels) |
| 06:22 | <@Vornicus> | And, frustrum means "a segment of a cone/pyramid that does not include the tip" |
| 06:22 | <@ToxicFrog> | I use it here as a common abbreviation for "viewing frustrum" |
| 06:23 | <@Vornicus> | The view frustrum, in a 3d game, is the pyramid segment that is within view of the screen; usually it does not include things that are really close to the camera, for a variety of reasons. |
| 06:31 | <@Vornicus> | Also: I wonder why there are so few "tall" monitors |
| 06:32 | <@Vornicus> | I mean, most of the time we care more about tall than wide; toolbars tend to be along the top and bottom, documents tend to be tall; web pages tend to be tall. |
| 06:32 | < SmithKurosaki> | wide > tall |
| 06:32 | <@Derakon> | Vorn: because you can buy wide monitors with swivels and rotate them 90°. |
| 06:32 | <@Vornicus> | I haven't met any you can do that with. |
| 06:32 | <@Reiver> | TF: Sooo... if you were looking at all this, would you bother with a gfx card upgrade, or swap out the whole machine? |
| 06:33 | <@Reiver> | It's currently a Raedon 1650 Pro, AMD 4200 Duo, 2GB RAM. |
| 06:33 | <@Derakon> | Vorn: http://www.newegg.com/Product/ProductList.aspx?Submit=ENE&N=2010190020+129701960 8&QksAutoSuggestion=&ShowDeactivatedMark=False&Configurator=&Subcategory=20&desc ription=&Ntk=&CFG=&SpeTabStoreType=&srchInDesc= |
| 06:33 | <@Vornicus> | SmithKurosaki: /why/, though? the only application I can think of where wide is better is video. |
| 06:33 | < SmithKurosaki> | yes |
| 06:34 | <@Vornicus> | Everything else - and this is huuuuuge amounts of stuff - tall is better. |
| 06:34 | <@ToxicFrog> | Reiver: pretty sure the CPU is actually better than Durandal's; I'd put in a new video card and maybe upgrade to a Phenom if you have an AM2+ socket |
| 06:34 | <@Reiver> | (Yeah, Chalain bought it for me as a birthday present.) |
| 06:34 | <@ToxicFrog> | Vornicus: actually, for gaming I find that I tend to want more view to the sides over more view up/down |
| 06:36 | <@Vornicus> | Yes, that's video |
| 06:38 | <@Reiver> | I suspect, Vorn, it's a matter of trendy. |
| 06:38 | <@Vornicus> | But then you have: text (tall), spreadsheets (usually tall), web (tall), PDFs (tall)... |
| 06:38 | <@Reiver> | Which basically means: Video is the flashiest stuff you can do with a monitor, so it's what most monitors get built towards. |
| 06:39 | <@Reiver> | And a wide monitor just looks sexier than a tall one on a desk, which is itself horizontal. |
| 06:39 | <@ToxicFrog> | Or you can just rotate the monitor 90%. |
| 06:39 | <@ToxicFrog> | Er, 90? |
| 06:39 | <@Reiver> | (If it can be rotated.) |
| 06:39 | <@ToxicFrog> | 90? |
| 06:39 | <@ToxicFrog> | dammit |
| 06:39 | <@Derakon> | 90°? |
| 06:39 | <@Vornicus> | TF: and hope to god your monitor and your OS support it. |
| 06:39 | <@Reiver> | TF forgot his alt codes~ |
| 06:39 | <@Derakon> | Vorn: Windows and Macs should both inherently support rotated displays. |
| 06:40 | <@Derakon> | I'm sure Linux has a half-dozen config codes for it. |
| 06:40 | <@Reiver> | As for the horizontal usefulness, a lot of folks have started to realise a lot of the width is wasted, and started to throw handy dandy tools into the sides. |
| 06:40 | <@Reiver> | Still waiting for Word 2010 to come out with the ribbon along the side~ |
| 06:40 | <@Derakon> | One of the first things I do when I get a new Mac (not often, mind) is move the dock to the side. |
| 06:41 | <@ToxicFrog> | Reiver: U-codes, actually :P |
| 06:41 | <@Vornicus> | Yeah, my mac has dock on the side. I could technically do the same with the Windows taskbar but it doesn't really work well in that direction. |
| 06:41 | <@Reiver> | Your technical clarification does not refute my statements intent~ |
| 06:43 | <@Vornicus> | Also, regarding MS Office: the two primary UI improvements I can think of are "ribbon to the side" and "color-coded by application" |
| 06:44 | <@Reiver> | So word is yellow, excel is green, etc? |
| 06:44 | <@Vornicus> | Word is blue, excel is green, powerpoint is orange, access is maroon, outlook is yellow... |
| 06:44 | <@Vornicus> | Oh, and they really need to ribbon up Visio. <3 Visio |
| 06:45 | <@Reiver> | You're thinking color-coding so that way you can instantly recognise which app you're in just by the color scheme? |
| 06:46 | <@Vornicus> | Yes |
| 06:46 | <@Reiver> | That's... actually, that's a bloody good idea. |
| 06:46 | <@Vornicus> | It's really minor and kinda silly, but it makes it easier to find your window. |
| 06:47 | <@Reiver> | I was going to say 'cute, but redundant', then I realised how often you're going by -- yeah |
| 06:47 | <@Vornicus> | And now that I think about it, ribboning Visio doesn't really make much sense. |
| 06:47 | <@Vornicus> | It doesn't have the same kind of workflow that other apps do. |
| 06:49 | <@ToxicFrog> | Oh yeah, that reminds me (via a kind of convoluted chain of thought) |
| 06:49 | <@ToxicFrog> | Does windows 7 have multiple desktops? |
| 06:49 | <@Vornicus> | Man I hope so. |
| 06:50 | <@Reiver> | IIRC, yes |
| 06:50 | <@Reiver> | But I've not toyed with it much. |
| 06:50 | <@Reiver> | I'll tell you when I get it next week and install it, shall I?~ |
| 06:50 | <@ToxicFrog> | Sure~ |
| 06:50 | <@Reiver> | (Hey, they're offering it for free!) |
| 06:51 | <@ToxicFrog> | ...although it occurs to me that if Server 2008(?) is feature-equivalent to W7, I can just rdp in Lockhart and check. |
| 06:51 | <@Reiver> | (So unlike Vista, it actually has value for money.~) |
| 06:51 | <@ToxicFrog> | *into |
| 06:51 | <@Reiver> | Not entirely quite, but you can look. |
| 06:57 | <@Vornicus> | (the Ribbon concept encourages in some ways a particular workflow. It's clearest in Excel: you start on the Home tab, enter a bit of data, go to the Formulas tab, get your formulas going, go to the Data tab, add gloms of data and get it organized, go to the Insert tab to add graphs and stuff, then Page Layout to deal with what in general you want it to look like when you print it... |
| 06:58 | <@Vornicus> | back to Home to add proper formatting, then Review to check for stupid errors, and then after that you spend most of your time on Data to look at your stuff filtered and so forth. |
| 06:59 | <@Vornicus> | In each step there's one particular tab that has nearly everything you want to deal with. |
| 06:59 | <@Reiver> | I clearly work differently; I tend to find I wish certain tabs were universal, or in different places. |
| 07:00 | <@Vornicus> | Well, universal tabs, there's the "quick access toolbar" |
| 07:00 | <@Vornicus> | Right click a ribbon button and "add to quick access toolbar" |
| 07:03 | <@Vornicus> | Also tell me more |
| 07:03 | | AnnoDomini [farkoff@Nightstar-3e109cd7.adsl.tpnet.pl] has joined #code |
| 07:04 | | mode/#code [+o AnnoDomini] by Reiver |
| 07:04 | <@Reiver> | Tell you more what? |
| 07:04 | <@Vornicus> | What things do you wish were in different places? |
| 07:04 | <@Reiver> | Oh. That's a tricky one, I've not touched in a couple weeks |
| 07:05 | | Derakon is now known as Derakon[AFK] |
| 07:05 | <@Reiver> | But I suspect the 'home' tab inconviniences me when all I want to do is see the shape of a simple graph for number comparisons - such a function is simple enough I'm inclined to want to do it just to make sure my number entry is correct with a visual inspection. |
| 07:06 | <@Reiver> | Aside: I really, really wish there was a way to easily flip horizontal rows into vertical columns, or at the least, that it was properly labeled. |
| 07:06 | <@Vornicus> | "transpose" |
| 07:07 | <@Vornicus> | Copy, paste special, transpose |
| 07:07 | <@Vornicus> | Has always been there. |
| 07:07 | <@Reiver> | Okay I did not know that's what Transpose did. |
| 07:08 | <@Reiver> | Always? |
| 07:08 | | * Reiver fires up Excel. |
| 07:08 | <@Vornicus> | I've been using Microsoft Excel since 1990, and it was there then. |
| 07:08 | <@Reiver> | Huh. |
| 07:08 | <@Vornicus> | So if not always, then at least nearly always. |
| 07:08 | <@Reiver> | Okay, that's nifty. |
| 07:08 | <@Reiver> | But kind of a convoluted place to find it. |
| 07:09 | | * Reiver has been looking for that button for over a decade~ |
| 07:10 | <@Vornicus> | There is also the "transpose" array function; you use it with the magic array function modifier key... |
| 07:10 | <@Vornicus> | What you do is you select the area you want the array formula to work on, type in your formula, then press ctrl-shift-enter |
| 07:10 | <@Reiver> | whut |
| 07:11 | <@Reiver> | Okay see I was like, looking for a button on a menu or something |
| 07:11 | <@Reiver> | One or two versions had it in the "Data" dropdown or something, but I don't think it was universal. |
| 07:11 | <@Vornicus> | :D |
| 07:39 | <@Reiver> | Hey, anyone know/be able to check how a Nvidia 9800 compares to a Raedon 1650 Pro? |
| 07:39 | <@Reiver> | Graphic card comparisons make baby Reivles cry. |
| 07:48 | <@Reiver> | 8400, too. |
| 07:49 | <@Vornicus> | You too? |
| 07:51 | <@Reiver> | They hurt mein brain. ;_; |
| 08:23 | | Vornicus is now known as Vornicus-Latens |
| 08:53 | | * AnnoDomini mutters. |
| 08:54 | <@AnnoDomini> | Could anyone point me at a noob-friendly site where I can learn how to install Apache and MySQL on Debian from console? I need to set this damn server up. |
| 08:57 | <@McMartin> | Hm. Try googling "aptitude tutorial" |
| 08:57 | <@McMartin> | "aptitude" is the Debian package manager |
| 08:58 | <@AnnoDomini> | Thanks. |
| 08:58 | < Namegduf> | AnnoDomini: Enter "aptitude" as root. |
| 08:58 | < Namegduf> | You should be able to figure it out. |
| 08:59 | <@McMartin> | The status markers in aptitude are Not Helpful without some grounding. |
| 08:59 | | Rhamphoryncus [rhamph@Nightstar-a62bd960.abhsia.telus.net] has quit [Client exited] |
| 08:59 | < Namegduf> | Yeah; it has built-in help for that, though. |
| 08:59 | <@McMartin> | True |
| 08:59 | <@McMartin> | That said, I was *much* happier with aptitude once I realized it had a usable commandline mode |
| 09:00 | <@McMartin> | The curses mode was, for me, full of, well, curses |
| 09:00 | < Namegduf> | Huh. |
| 09:00 | <@McMartin> | This probably has to do with yum sort of being my default workflow from my Fedora days; aptitude in cmdline mode is quite close to that. |
| 09:04 | <@McMartin> | (In particular, package search with aptitude search | less I found much more handy than doing the interactive search where the whole screen changes and you have to hit 'n' a billion times to get to where you want) |
| 09:04 | < Namegduf> | Well, I know regular expressions. |
| 09:04 | < Namegduf> | And the interactive earch accepts tehm. |
| 09:04 | < Namegduf> | So, yeah, I don't have the "billion times" problem. |
| 09:05 | <@McMartin> | So does the cmdline - however, my search terms tend to be too general for their taste |
| 09:05 | <@McMartin> | So I get 47 billion libfoo-curl-devel etc. things because I didn't know whether I needed to search for "libfoo" or "foo" |
| 09:06 | | Tarinaky [Tarinaky@Nightstar-4fa31ea0.adsl.virginmedia.net] has quit [Connection closed] |
| 09:07 | | You're now known as TheWatcher |
| 10:32 | | Attilla [The.Attilla@FBC920.480E8C.3E679A.E258A8] has joined #code |
| 10:32 | | mode/#code [+o Attilla] by ChanServ |
| 11:00 | | Alek [omegaboot@Nightstar-744f51af.emhril.sbcglobal.net] has quit [Client closed the connection] |
| 11:06 | | Alek [omegaboot@Nightstar-744f51af.emhril.sbcglobal.net] has joined #code |
| 11:22 | | Thaqui [Thaqui@27B34E.D54D49.F53FA1.6A113C] has joined #code |
| 11:45 | <@AnnoDomini> | Hmm. Okay, I've found some tutorial on how to install LAMP. |
| 11:45 | <@AnnoDomini> | But when I type in "apt-get install apache2 php4 libapache2-mod-php4" it says that apache2 is already there and that the two others have no installation candidates. |
| 11:46 | <@AnnoDomini> | What does that mean? |
| 12:01 | <@TheWatcher> | I'd guess your LAMP instructions are probably out of date. The packages might be php5 and libapache2-mod-php5 or just s/php5/php/ - I can't say for certain, as I don't know which distro you're using (and I personally don't tend to use debian/ubuntu/friends) |
| 12:02 | | * AnnoDomini tries s/4/5. |
| 12:02 | <@AnnoDomini> | It seems to have worked! |
| 12:03 | <@TheWatcher> | Wewt |
| 12:05 | | * AnnoDomini desperately tries to remember how one created files. |
| 12:05 | <@TheWatcher> | ? |
| 12:06 | <@AnnoDomini> | I want to create /var/www/test.php to test out if this thing is working. |
| 12:08 | <@TheWatcher> | echo "<?php phpinfo(); ?>" > /var/www/test.php |
| 12:09 | <@AnnoDomini> | I'd use that, but I've already used "nano /var/www/test.php"... how do I save the file and exit? ;_; |
| 12:10 | <@AnnoDomini> | I think I figured out how to write to disk. But how do I exit? |
| 12:10 | <@TheWatcher> | In nano - Ctrl+X |
| 12:11 | <@AnnoDomini> | Yay! |
| 12:11 | <@TheWatcher> | (the shortcuts should be shown along the bottom, in general ^ means 'ctrl+'. So ^X is 'ctrl+x') |
| 12:11 | <@AnnoDomini> | What does "XOFF ignored, mumble mumble" or "XON ignored, mumble mumble" mean? |
| 12:13 | <@TheWatcher> | XON and XOFF are control codes that terminals uses to, essentially turn on and off display of data (useful if something is outputing data faster than the terminal can display it). |
| 12:13 | <@TheWatcher> | They're kinda meaningless when you're using a terminal text editor though, so nano supresses them |
| 12:20 | | Thaqui [Thaqui@27B34E.D54D49.F53FA1.6A113C] has quit [Connection closed] |
| 12:25 | | Alek [omegaboot@Nightstar-744f51af.emhril.sbcglobal.net] has quit [Ping timeout: 121 seconds] |
| 12:31 | | Alek [omegaboot@Nightstar-0f92a982.emhril.sbcglobal.net] has joined #code |
| 12:43 | <@AnnoDomini> | How do I check what IP this thing has? |
| 12:47 | <@TheWatcher> | Internally? /sbin/ifconfig will list it (probably under eth0). Externally? http://www.whatismyip.com/ is probably fastest... |
| 12:53 | <@AnnoDomini> | There doesn't seem to be an 'ipconfig' in /sbin/. |
| 12:53 | <@TheWatcher> | ifconfig, not ipconfig |
| 12:54 | <@TheWatcher> | (and it might be /usr/sbin/ifconfig really) |
| 12:55 | <@AnnoDomini> | Awright. There it is. |
| 12:56 | <@AnnoDomini> | Yay, php works! |
| 12:56 | <@TheWatcher> | Yay |
| 13:26 | | Orthia [Orthianz@Nightstar-9fa3bf3f.xnet.co.nz] has quit [Ping timeout: 121 seconds] |
| 13:28 | <@AnnoDomini> | Hmmm. MySQL is insalled, but when I try to 'mysql', it denies me access. |
| 13:28 | | Alek [omegaboot@Nightstar-0f92a982.emhril.sbcglobal.net] has quit [Ping timeout: 121 seconds] |
| 13:31 | <@TheWatcher> | Have you tried 'mysql -u root mysql'? |
| 13:31 | <@TheWatcher> | Did you set a root password with mysqladmin? |
| 13:32 | <@AnnoDomini> | I tried 'mysql -u root' and have set up a password when prompted at installation. |
| 13:32 | | Alek [omegaboot@Nightstar-fc052118.emhril.sbcglobal.net] has joined #code |
| 13:33 | <@TheWatcher> | mysql -u root -p <databasename> |
| 13:33 | <@TheWatcher> | it'll prompt you to enter that password |
| 13:34 | <@AnnoDomini> | It works! Thanks. |
| 13:45 | | Orthia [Orthianz@Nightstar-d5f7841f.xnet.co.nz] has joined #code |
| 13:45 | <@AnnoDomini> | Whee. I has a phpmyadmin. |
| 14:00 | | Namegduf [namegduf@Nightstar-7ec84b32.bath.ac.uk] has quit [[NS] Quit: Need to take this laptop elsewhere; will be back later.] |
| 14:04 | | You're now known as TheWatcher[afk] |
| 14:36 | | AnnoDomini [farkoff@Nightstar-3e109cd7.adsl.tpnet.pl] has quit [Ping timeout: 121 seconds] |
| 14:38 | | gnolam [lenin@Nightstar-38637aa0.priv.bahnhof.se] has joined #code |
| 14:41 | | AnnoDomini [farkoff@Nightstar-6e923c5e.adsl.tpnet.pl] has joined #code |
| 14:41 | | mode/#code [+o AnnoDomini] by Reiver |
| 16:04 | | Syloqs_AFH [Syloq@NetworkAdministrator.Nightstar.Net] has joined #code |
| 16:06 | | Syloqs_AFH is now known as Syloqs-AFH |
| 16:06 | < gnolam> | "IRC. Internet Relay Chat. It's how hackers talk when they don't want to be overheard." |
| 16:06 | | You're now known as TheWatcher |
| 16:06 | < SmithKurosaki> | o.0 |
| 16:09 | < gnolam> | "Numb3rs" doing a "CSI: NY": http://www.youtube.com/watch?v=O2rGTXHvPCQ |
| 16:10 | < gnolam> | "Luckily, I speak leet." |
| 16:13 | <@TheWatcher> | Yeah, a depressing amount of Numb3rs eps that involve computers in any way are so completely incorrect it's not even funny |
| 16:14 | < Alek> | and this is new to Hollywood? |
| 16:15 | | * AnnoDomini stabs himself in the eyes. |
| 16:15 | <@TheWatcher> | No. Just, y'know, vaguely wish they'd actually do some goddamn research |
| 16:15 | < Alek> | so does everybody else. |
| 16:16 | < Alek> | my dad keeps bitching about how Soviet and Russian soldiers are always in the wrong uniforms. |
| 16:17 | < Alek> | my beef is similar, but mostly in books - people writing thrillers and adventures and stuff and using Russian words that are ALL WRONG and stuff |
| 16:17 | < Alek> | but then, they tend to be ex-military, having learned from Berlitz or some such |
| 16:18 | < Alek> | and not from an expert or native. |
| 16:18 | | SmithKurosaki [Smith@Nightstar-6450f657.dsl.teksavvy.com] has quit [Client closed the connection] |
| 16:18 | <@AnnoDomini> | Barlitz School of Aviation and Air Conditioner Repair. |
| 16:19 | | SmithKurosaki [Smith@Nightstar-6450f657.dsl.teksavvy.com] has joined #code |
| 16:19 | < Alek> | Berlitz Language Learning and Cultural Training |
| 16:24 | < gnolam> | Screenwriters and directors get everything wrong, that's a fact. |
| 16:25 | < gnolam> | But there are degrees of wrongness. |
| 16:25 | <@AnnoDomini> | Everything and EVERYthing. |
| 16:25 | < gnolam> | It's one thing to have soldiers wearing ushankas when they should have been wearing pilotkas, it's another to make a common household object - a computer - basically fill the role of /magic/. |
| 16:26 | < gnolam> | Especially since it leads to a hideously distorted view of what computers and "hackers" can do. |
| 16:52 | | Rhamphoryncus [rhamph@Nightstar-a62bd960.abhsia.telus.net] has joined #code |
| 17:14 | | Derakon[work] [Derakon@Nightstar-d44d635e.ucsf.edu] has joined #code |
| 17:21 | | * jerith prods Rhamphoryncus in the direction of the Rolo wave. |
| 18:19 | | You're now known as TheWatcher[afk] |
| 18:52 | < Derakon[work]> | Okay, here's a pseudocode description of how the program is currently handling blocking on asynchronous events: http://paste.ubuntu.com/318033/ |
| 18:52 | < Derakon[work]> | Clearly this can be improved. |
| 18:52 | < Derakon[work]> | The question is, what exactly should be the design? |
| 18:54 | <@McMartin> | Traditionally, a semaphore would be used. |
| 18:54 | <@McMartin> | Not sure how you can do those cross-machine, though. |
| 18:55 | <@McMartin> | Actually, for this, a condition variable is best. |
| 18:55 | <@McMartin> | Grab lock before call to moveStage. waitForSomething then waits on the CV associated with that lock (releasing it atomically with its falling asleep) |
| 18:56 | < Derakon[work]> | My goals are: a) cleaner than what I just described; b) able to process UI signals while waiting, with a minimum of fuss; c) able to handle more or less arbitrary call stacks (i.e. should still work even if it's f() calling g() calling h() calling moveStage, without g() or f() having to do anything special). |
| 18:56 | <@McMartin> | endMoveStage then grabs that lock, signals the CV, does any last-minute cleanup if needed, releases lock. |
| 18:56 | <@McMartin> | "while (boolean): pass" is a busywait loop that's like waiting for a CV signal except it eats 100% CPU. |
| 18:57 | < Derakon[work]> | Yeah. |
| 18:57 | <@McMartin> | So, CV. =P |
| 18:57 | < Derakon[work]> | I've stuck a "sleep(10ms)" call in there and am checking to see if that fixes the crash, incidentally. |
| 18:57 | <@McMartin> | It's most like a semaphore if you can have cross-thread ones (where it starts at zero, wait tries to lower it, and end raises it) but CVs are Generally Better, semantically |
| 18:58 | <@McMartin> | But now, I need to commute. |
| 18:58 | < Derakon[work]> | Okay, seeya. Thanks. |
| 19:13 | < Derakon[work]> | Okay, if I understand condition variables correctly, I think this describes what you were proposing: http://paste.ubuntu.com/318045/ |
| 19:13 | < Derakon[work]> | It's been a long time since I did any concurrency work though. |
| 19:19 | < Rhamphoryncus> | Derakon[work]: For UI you generally want to channel into something the gui toolkit's event loop can watch, ie a socket |
| 19:20 | < Derakon[work]> | I'm afraid my brain is operating at roughly 60% maximum efficiency right now. More detail? |
| 19:21 | < Rhamphoryncus> | replace the condition with a TCP condition. Replace the .notifyAll() with a write to the socket |
| 19:21 | < Derakon[work]> | And this gains me what? |
| 19:22 | < Rhamphoryncus> | The .wait() can be moved into the GUI's event loop. Usually that's what people want to allow the UI to continue |
| 19:22 | < Rhamphoryncus> | Although on further thought you may prefer using a separate thread.. |
| 19:22 | < Derakon[work]> | Yeah, that's what I was thinking. |
| 19:22 | < Derakon[work]> | Spin the entire experiment to its own thread. |
| 19:22 | < Rhamphoryncus> | How are you talking to the remote computer? |
| 19:22 | < Derakon[work]> | Pyro. |
| 19:23 | < Derakon[work]> | http://pyro.sourceforge.net/ |
| 19:24 | < Rhamphoryncus> | bytecode transfer o.O |
| 19:25 | < Derakon[work]> | I hadn't gotten around to investigating how it actually works. |
| 19:26 | < Rhamphoryncus> | Assuming callback remains on the client, and is called in another thread as a single operation.. |
| 19:29 | < Rhamphoryncus> | it should work as expected |
| 19:29 | < Rhamphoryncus> | okay, so a better suggestion: use an Event rather than a Lock and Condition |
| 19:32 | < Derakon[work]> | Okay, so thread 0 runs the GUI. User clicks to start an experiment. Thread 0 makes thread 1 for the experiment. Thread 1 wants to move the stage. It creates thread 2 and passes it an Event that it created, then calls wait() on the event. Thread 2 calls out to the remote computer to move the stage, then sleeps until it's done. It then calls set() on the Event, and exits. |
| 19:32 | < Rhamphoryncus> | http://paste.ubuntu.com/318060/ |
| 19:33 | < Derakon[work]> | Ah, righto. |
| 19:34 | < Rhamphoryncus> | semaphores should be taken out and shot. They're an attempt at creating a threading axiom, a fundamental primitive on which everything else can be built. They suck for practical usage |
| 19:34 | < Rhamphoryncus> | Conditions are not much better |
| 19:35 | < Rhamphoryncus> | The irony is that when you consider performance, forward progress, etc, they fail at being fundamental primitives. You often need something else |
| 19:36 | < Rhamphoryncus> | oi, proper syntax helps the syntax highlighting: http://paste.ubuntu.com/318071/ |
| 19:36 | < Derakon[work]> | Hee. |
| 19:37 | < Rhamphoryncus> | still pretty weak highlighting |
| 19:39 | < Rhamphoryncus> | Which of your 3 threads is handling pyro? |
| 19:40 | < Derakon[work]> | Man, I don't know. |
| 19:40 | < Derakon[work]> | Haven't gotten that deep into the code yet. |
| 19:41 | < Derakon[work]> | For the most part threads aren't currently being used explicitly. |
| 19:42 | < Derakon[work]> | Augh. "from Pyro.ext import remote_nons as r" |
| 19:43 | < Derakon[work]> | Please stop using goddamn single lowercase letters as module names. ;.; |
| 19:43 | < Rhamphoryncus> | agreed x_x |
| 19:43 | < Derakon[work]> | If "greppable" isn't an adjective, "ungreppable" certainly is as of this moment. |
| 19:44 | < Rhamphoryncus> | .. I have no idea when pyro copies across the network and when it proxies |
| 19:48 | < Rhamphoryncus> | mmm implicit threaded imports |
| 20:25 | | Alek [omegaboot@Nightstar-fc052118.emhril.sbcglobal.net] has quit [Ping timeout: 121 seconds] |
| 20:29 | | Alek [omegaboot@Nightstar-c2835c7a.emhril.sbcglobal.net] has joined #code |
| 20:38 | | * Derakon[work] takes a look through his HTTP access logs, discovers what looks like a spammer trying to use his photo album manager script to muck up the photo comments. |
| 20:38 | < Derakon[work]> | Unluckily for the spammer, I set my permissions properly once I'm done managing albums. |
| 20:38 | < Derakon[work]> | I am curious how they found the manager script in the first place though. It's not linked from anywhere. |
| 20:39 | < Derakon[work]> | ...oh, wait, right. Index listing. |
| 20:41 | | Alek [omegaboot@Nightstar-c2835c7a.emhril.sbcglobal.net] has quit [Ping timeout: 121 seconds] |
| 20:47 | | Alek [omegaboot@Nightstar-114b3e79.emhril.sbcglobal.net] has joined #code |
| 20:57 | | Namegduf [namegduf@Nightstar-7ec84b32.bath.ac.uk] has joined #code |
| 21:24 | | You're now known as TheWatcher |
| 21:40 | | * Derakon[work] eyes the current experiment code, tries to figure out how to adapt it to his alternate design. |
| 21:40 | < Derakon[work]> | This is ugly. :( |
| 22:03 | | Kazriko [kaz@Nightstar-e09690fa.client.bresnan.net] has quit [Ping timeout: 121 seconds] |
| 22:07 | | * gnolam stabs Steam. |
| 22:08 | <@TheWatcher> | Can I join in? |
| 22:09 | < SmithKurosaki> | Why steam? |
| 22:09 | <@TheWatcher> | Because it is made of spiders. |
| 22:09 | < gnolam> | Indeed. Please join in. |
| 22:09 | < SmithKurosaki> | :( |
| 22:10 | < gnolam> | Let's go Murder on the Orient Express on its ass. >:E |
| 22:10 | <@TheWatcher> | (actually, chunks of it are fine, but the UI designer in particular deserves to be hung by his funes) |
| 22:11 | | Kazriko [kaz@Nightstar-e09690fa.client.bresnan.net] has joined #code |
| 22:11 | | mode/#code [+o Kazriko] by Reiver |
| 22:11 | < gnolam> | And the reliability of their servers is... less than stellar. It's more 2 nines downtime than 6 nines uptime. :P |
| 22:11 | < gnolam> | And the whole client is a /hog/. |
| 22:12 | < gnolam> | There's just no reason that it should take half a minute to a minute to start up. |
| 22:13 | < SmithKurosaki> | It's usin the IE engne.. |
| 22:20 | | SmithKurosaki [Smith@Nightstar-6450f657.dsl.teksavvy.com] has quit [Client closed the connection] |
| 22:21 | | SmithKurosaki [Smith@Nightstar-6450f657.dsl.teksavvy.com] has joined #code |
| 22:32 | | Alek [omegaboot@Nightstar-114b3e79.emhril.sbcglobal.net] has quit [Client closed the connection] |
| 22:36 | <@McMartin> | IE stops being ridiculous around IE7, which is standard now, so there's no excuse :colbert: |
| 22:39 | <@AnnoDomini> | There's something I can't put a finger on that pisses me off about IE7. |
| 22:39 | < simon`> | http://www.dyberemening.dk/blog/wp-content/notatree.jpg <- this is not a tree |
| 22:39 | | Alek [omegaboot@A2BA3E.D9488C.BA9218.BA9A2C] has joined #code |
| 22:39 | <@AnnoDomini> | The only version I had neutral-ish feelings towards was 5.5. |
| 22:40 | <@McMartin> | The only one I'm willing to have any respect for is IE8, and I feel they could have done a bit more to meet their previous captives halfway |
| 22:41 | <@McMartin> | By which I mean that IE8 really shouldn't identify itself as IE, because when it does, lots of sites break their pages to work in IE6, and IE8 is actually standards-compliant and thus renders them brokenly~ |
| 22:41 | < simon`> | AnnoDomini, so it's a question of when hatred grows out of one by prolonged IE exposure. |
| 22:43 | < Namegduf> | McMartin: For this, it has compatibility mode. |
| 22:43 | | Vornicus-Latens is now known as Vornicus |
| 22:43 | < Namegduf> | McMartin: Also, anyone still serving IE6 hacks to IE8 is Doing It Wrong. |
| 22:43 | < simon`> | McMartin, yes, they totally broke their non-existing naming scheme by simply choosing the next integer. if it followed Windows' naming scheme, IE8 should've been Windows Internet Experience NE |
| 22:43 | < Namegduf> | The browsers are quite identifiable by version, and IE7 didn't need a lot of IE6's hacks. |
| 22:43 | <@McMartin> | simon`: Yeah, see, the latter would confuse them and then they'd serve proper CSS, and everyone would win; that's what I was proposing |
| 22:44 | <@McMartin> | Namegduf: Hm, I thought 7 still had massive issues with that, though 6 was comical |
| 22:44 | < simon`> | Namegduf, anyone serving WWW hacks is doing something wrong in their life. |
| 22:44 | <@McMartin> | Or using autogenerated code. |
| 22:44 | < Namegduf> | simon`: Running a website is wrong? Matter of opinion, I guess. |
| 22:44 | < simon`> | Namegduf, not at all. I'm just an opponent of wasting much time on making websites look pretty in all modern browsers. =) |
| 22:45 | <@McMartin> | Most of the Web 2.0 style people where I'm not using that as a sneer serve mechanically generated JS that handles incompatibility as if it were a particularly ugly ABI. |
| 22:45 | < Namegduf> | s/a website/a real website/ |
| 22:45 | | * McMartin hasn't directly messed with GWT, but has seen the results and been fairly happy therewith. |
| 22:45 | < Namegduf> | McMartin: 7's issues were along the lines of "one line of CSS to fix bug", as opposed to "entirely different set of layout schemes because IE6 calculates lengths entirely wrong". |
| 22:45 | <@McMartin> | Aha |
| 22:45 | < Namegduf> | They were *bad*, but two orders of magnitude less. |
| 22:45 | <@McMartin> | Go tit. |
| 22:45 | <@McMartin> | Got it, even |
| 22:46 | | * McMartin unleashes a swarm of birds |
| 22:46 | | * simon` looks in amazement |
| 22:46 | <@McMartin> | My CSS is enough to mean I can make my pages be XHTML Strict, but that's about it. |
| 22:46 | < Namegduf> | JS browser detection is an option to be carefully considered with the fact that you can do serverside browser detection |
| 22:47 | < Namegduf> | And not require JS to access the website without it looking broken. |
| 22:47 | <@McMartin> | Well, this was the non-sneering Web 2.0 thing, where "the whole point is to have a served interactive application of some kind, and not having JS means you're missing the point in the first place" |
| 22:47 | <@McMartin> | That's a subset of the stuff calling itself that. |
| 22:48 | < Namegduf> | Yeah. |
| 22:48 | < Namegduf> | Those aren't invalid approaches, either. |
| 22:49 | <@MyCatVerbs> | I'm quite fond of the "this is an internal management page; we'll just tell the users to use Safari or Firefox and we'll suspend their cookie jar priviledges if they turn JS off in their browser" approach. |
| 22:50 | < Namegduf> | Yeah; sadly, that's a rather wrong methodology if what you're maintaining is not, in fact, an internal management page. |
| 22:50 | <@McMartin> | Quite |
| 22:51 | <@McMartin> | (Our own internal pages do that and it makes me sad because I hat Mac Firefox, but so it goes) |
| 22:51 | < Namegduf> | (Annoys me because people use it to keep IE-only pages all the time) |
| 22:51 | <@MyCatVerbs> | Of course for anything public-facing, I test primarily in FF3.5 and IE6, and the policy on JS is progressive-enhancement-only. |
| 22:51 | < Namegduf> | (Still valid if you're supporting standards right) |
| 22:52 | < Namegduf> | Yeah, that's how I do it, although I try to rest other browsers. |
| 22:52 | <@McMartin> | I'd toss Chrome into the mix just to make sure WebKit isn't having a cow. |
| 22:53 | < Namegduf> | (And its Chrome, FF3.x, and IE6, because Chrome is my main browser, but Gecko is just glitchy enough to deserve attention) |
| 22:53 | <@McMartin> | (Chrome will also more or less cover Safari and Konqueror, so ++) |
| 22:53 | < Namegduf> | Yeah. |
| 22:53 | < Namegduf> | Well, I don't *try* to support or ever test Konqueror, really. |
| 22:54 | < Namegduf> | Hmm, unless KDE4 still has a varient. |
| 22:54 | < Namegduf> | I don't know, I've not heard of anyone using it. |
| 22:54 | < Namegduf> | It used to be such a horrible, horrible browser. |
| 22:54 | <@MyCatVerbs> | Arora is quite handy for this. Webkit embedded in QT, so I can run it in Linux and get a good approximation to what I'm inflicting on the poor defenceless Safari users. |
| 22:54 | <@McMartin> | I mention it only because I remember vaguely that it's where WebKit got its start |
| 22:55 | <@MyCatVerbs> | I used to quite like Konqueror/KHTML at one point. |
| 22:55 | < Namegduf> | Huh. |
| 22:56 | < Namegduf> | Even Konqueror's mother didn't like it. |
| 22:56 | <@MyCatVerbs> | Mainly because I was working within a fuckawfully tiny university quota, and Konqueror never spat ~75MB of binary shite into my homedir. Unlike Firefox at the time. |
| 22:56 | < Namegduf> | Well, yes, but its interface and options setup makes IE7 look well-designed |
| 22:56 | < Namegduf> | Or made, anyway. |
| 22:56 | <@MyCatVerbs> | You hold control and press L and type a URL and press enter and it works. |
| 22:56 | <@MyCatVerbs> | What else do you want, a pony? |
| 22:57 | <@MyCatVerbs> | I know I don't. Ponies are too much like hard work. |
| 22:58 | <@TheWatcher> | FOR PONY! |
| 22:59 | < Namegduf> | Well, currently, my requirements are... response times to UI interaction measured in ms rather than s |
| 22:59 | <@TheWatcher> | *ahem* |
| 22:59 | < Namegduf> | (So Firefox is right out) |
| 22:59 | < Namegduf> | Search engine support |
| 22:59 | < Namegduf> | Spellcheck |
| 22:59 | < Namegduf> | Tabbing, an interface which looks uncluttered and has non-hideous icons. |
| 23:00 | < Namegduf> | Generally good rendering, acceptable defaults for styles. |
| 23:00 | <@McMartin> | Sounds like Chrome to me |
| 23:00 | <@McMartin> | Is Chromium any good? |
| 23:00 | < Namegduf> | Chromium is excellent |
| 23:00 | < simon`> | omg, pwnies! |
| 23:01 | | * Namegduf has been using the Google Chrome Linux x86_64 dev channel as his primary browser for the last couple of months |
| 23:01 | <@McMartin> | Good to know. |
| 23:01 | < Namegduf> | I was using Firefox, but it kept crashing. |
| 23:01 | < Namegduf> | And doing the "response time in seconds" thing. |
| 23:01 | <@McMartin> | I got the impression that code churn was a bit... lackadaiscal after the Finest Bug Report Ever. |
| 23:01 | < gnolam> | TheWatcher: We actually used that as a battle cry during reenactment once. :) |
| 23:02 | <@TheWatcher> | gnolam: good man :) |
| 23:02 | <@McMartin> | (That bug report being the "Expected behavior: X button Actual behavior: Some dude's head" issue.) |
| 23:02 | < Namegduf> | No, not really. |
| 23:02 | < Namegduf> | Some guy played a joke, once |
| 23:03 | <@McMartin> | Good to know |
| 23:03 | <@McMartin> | And yeah, I read that as "accidentally checked in his test case and pushed it out to distro" instead of "prank" |
| 23:04 | <@McMartin> | Oy, Karmic thinks that "Chromium" is that terrible mouse-based scrolling shooter |
| 23:04 | < Namegduf> | Well, it's publicly documented. |
| 23:04 | <@McMartin> | True enough |
| 23:04 | < Namegduf> | And I think it was only in top of trunk/nightlies, not in the dev channel |
| 23:04 | < Namegduf> | Which gets basic QA |
| 23:04 | | * McMartin nods |
| 23:05 | < Namegduf> | And yeah, the package is called chromium-browser for that reason. |
| 23:05 | < Namegduf> | Basically, the guy checked it in as a joke placeholder to "encourage" the guy whose head it was to commit the real one. |
| 23:06 | < Namegduf> | (The real X button, basically) |
| 23:06 | <@McMartin> | chromium-browser looks like Ubuntu hasn't picked it up yet. |
| 23:06 | | Orthia [Orthianz@Nightstar-d5f7841f.xnet.co.nz] has quit [Client closed the connection] |
| 23:06 | < Namegduf> | No; you need to download the package from... |
| 23:06 | <@McMartin> | (aptitude search chromium would have otherwise grabbed it, but I searched for chromium-brwoser by name to make sure) |
| 23:06 | < Namegduf> | http://dev.chromium.org/getting-involved/dev-channel |
| 23:06 | <@McMartin> | Aha. |
| 23:06 | < Namegduf> | The .deb, when installed, adds Google's repo to your sources.list |
| 23:07 | <@McMartin> | (I probably won't; the machine running Karmic is due to be retired permanently in a month or so once Astatine comes online) |
| 23:07 | < Namegduf> | And thus it remains fully and proper updated, Google-maintained, and yet properly run through the distro packaging system. |
| 23:10 | <@MyCatVerbs> | Comment 56, "New buttons are the worst close\resotre buttons I've ever seen. How can I return old ones?" |
| 23:10 | <@MyCatVerbs> | Comment 57, "do you want the head back?" |
| 23:14 | < Namegduf> | Useful for that, too. |
| 23:14 | <@McMartin> | Heh |
| 23:14 | < simon`> | any of you ever read about genetic programming? |
| 23:15 | < simon`> | I wonder how program size is determined; if it's a product of mutation. I can't imagine a fixed program size is any good. |
| 23:16 | <@MyCatVerbs> | Depends how you represent programs. |
| 23:16 | <@MyCatVerbs> | If your programs are a term algebra then usually variable program sizes pretty much fall out naturally. |
| 23:17 | < simon`> | how so? |
| 23:17 | <@MyCatVerbs> | Like if your programs look like Lisp programs. |
| 23:18 | <@MyCatVerbs> | Or syntax trees, even. |
| 23:18 | < simon`> | why couldn't they vary in size then? trees can grow. |
| 23:18 | <@MyCatVerbs> | Having those trees have arbitrary sizes is quite natural. |
| 23:18 | <@MyCatVerbs> | Oh, ambiguity. |
| 23:19 | < simon`> | ah, right |
| 23:19 | <@MyCatVerbs> | When I said "fall out naturally" I mean that the feature comes about as a natural result. |
| 23:21 | <@MyCatVerbs> | Hell, if you represent programs as strings of byte code then you can happily have arbitrary-length strings. |
| 23:22 | <@MyCatVerbs> | You'd want to have mutation operators that do things like copy-and-paste and cut-and-paste and even cut-and-discard segments of code at random for that, I guess. ^^ |
| 23:22 | < simon`> | but which genetic operator would result in multiplying the number of operations? |
| 23:22 | < simon`> | ah yes, copy and paste would result in a growth. |
| 23:22 | <@MyCatVerbs> | I would say ideally none. :) |
| 23:23 | | Vornicus [vorn@ServerAdministrator.Nightstar.Net] has quit [[NS] Quit: ] |
| 23:23 | <@MyCatVerbs> | I assume that you would be better off growing programs additively rather than multiplicatively. |
| 23:23 | <@MyCatVerbs> | So that growth is roughly linear rather than exponential. :) |
| 23:23 | < simon`> | well, program size should definitely be a fitness function. I could always prevent an explosion in the selection phase. |
| 23:23 | < simon`> | s/function/criterium/ |
| 23:24 | <@MyCatVerbs> | I'd rate number of cycles over program size. :) |
| 23:24 | | Vornicus [vorn@ServerAdministrator.Nightstar.Net] has joined #code |
| 23:24 | | mode/#code [+o Vornicus] by Reiver |
| 23:24 | < simon`> | right |
| 23:25 | < simon`> | that reminds me of dormant genetic material |
| 23:26 | <@MyCatVerbs> | On the other other paw, fixing the program size might not actually be a bad idea. |
| 23:26 | <@MyCatVerbs> | The terrible secret of spa-genetic programming is that the more you constrain the search space, the less looking the damn thing has to do. ;) |
| 23:27 | < dmlandrum> | Spa-genetic? Gradually evolving the ideal jacuzzi? |
| 23:27 | <@MyCatVerbs> | dmlandrum: Do you have stairs in your house? |
| 23:27 | < simon`> | MyCatVerbs, good point. |
| 23:28 | < dmlandrum> | Umm, yes. |
| 23:28 | | * dmlandrum waits for the punchline... |
| 23:28 | | * MyCatVerbs pushes dmlandrum. |
| 23:29 | | * MyCatVerbs shoves dmlandrum. |
| 23:29 | | * simon` installed emacs and wanted to give it a chance |
| 23:29 | < simon`> | first off, its fonts are broken in my minimal ubuntu. |
| 23:29 | <@MyCatVerbs> | Wanted to? Past tense. |
| 23:29 | < dmlandrum> | MyCatVerbs: Funny thing is, I googled the term and nothing came up. |
| 23:29 | <@MyCatVerbs> | I mena, past tense? |
| 23:30 | < dmlandrum> | Just a bunch of Russion dating sites. |
| 23:30 | < dmlandrum> | Err, Russian |
| 23:30 | <@MyCatVerbs> | *mean, the Force is not with me tonight. |
| 23:30 | < simon`> | MyCatVerbs, my enthusiasm for the project is dwindling |
| 23:30 | <@MyCatVerbs> | dmlandrum: Ah, no, that's not it. That's not it at all. :) |
| 23:30 | < dmlandrum> | I'm assuming SPA is an acronym... |
| 23:30 | < simon`> | the viper mode is not too bad, although it feels like using vi 10 years ago. why didn't they make a more vim-like clone! |
| 23:31 | <@MyCatVerbs> | dmlandrum: http://www.albinoblacksheep.com/flash/terriblesecret |
| 23:31 | <@MyCatVerbs> | No, no, I was just pretending to correct myself for the sake of a lame joke. |
| 23:31 | <@MyCatVerbs> | Not even a joke, just a reference. |
| 23:31 | < dmlandrum> | Ah |
| 23:31 | < dmlandrum> | When I went and made an even lamer joke. |
| 23:32 | | Alek [omegaboot@A2BA3E.D9488C.BA9218.BA9A2C] has quit [Ping timeout: 121 seconds] |
| 23:32 | < dmlandrum> | Heh |
| 23:33 | | Alek [omegaboot@Nightstar-57e80a71.emhril.sbcglobal.net] has joined #code |
| 23:33 | < dmlandrum> | Some people have a lot of time on their hands. :) |
| 23:34 | <@MyCatVerbs> | simon`: I donut know about you, but I'm vaguely fond of plain old fashioned vi. |
| 23:34 | < simon`> | MyCatVerbs, originally it was for completing some Emacs regular expressions exercises for my compiler course. I then realised that its regexp-builder mode treats regular expressions differently than its regular regexp i-search mode. and viper's forward searching mode apparently had a bug. I spent so much time figuring that out. |
| 23:35 | < simon`> | MyCatVerbs, I don't mind old-fashioned vi when I'm on minimalist systems, but otherwise I really enjoy multiple-level undos and syntax highlighting (although vim's implementation of that is bloated) |
| 23:36 | <@MyCatVerbs> | I don't think vim's syntax highlighting is actually bloated. |
| 23:36 | <@MyCatVerbs> | It's pretty conservative, even, it's all just finite state machines. |
| 23:36 | <@MyCatVerbs> | The issue is more that it's quite inefficient anyway. :/ |
| 23:37 | < simon`> | well, okay, maybe bloated is the wrong word. I'd like to improve on it. have you ever written a syntax file? |
| 23:37 | <@MyCatVerbs> | No, too lazy. |
| 23:38 | < simon`> | it does some weird things, too, for example if you jump way down in a file, its syntax highlighting hasn't parsed whatever's above, so it might get syntax highlighting wrong for a large block unless you page up a time or two and hit ^L. |
| 23:38 | <@MyCatVerbs> | Yeah. I'm quite familiar with that failure mode, I hit it semi-regularly. |
| 23:38 | < simon`> | also, the quality of the syntax files vary greatly because some of them are really old. |
| 23:39 | <@MyCatVerbs> | I did not know that last bit. |
| 23:39 | <@MyCatVerbs> | I'd always just assumed that it was because of the limited power of the approach. |
| 23:39 | < simon`> | I made an improvement to TCL's that wasn't accepted into vim because it didn't have support for Tk (I couldn't bother making that a separate addition, since I never use Tk in TCL) |
| 23:40 | < simon`> | I think it's because the syntax is slightly difficult to work with. I'm thinking of the many ways to specify offsets (whether something should be counted in or not, eaten or not, etc.) |
| 23:41 | < simon`> | wait. someone actually improved TCL's syntax file now. that's great. :) it sucked for so long. |
| 23:41 | < simon`> | mine was like 6-7 years ago anyways. |
| 23:42 | < simon`> | (what it most significantly lacked was highlighting of [commands] inside strings) |
| 23:43 | <@MyCatVerbs> | You still ever program in TCL? |
| 23:44 | < simon`> | never |
| 23:44 | < simon`> | it's just a fond memory. |
| 23:44 | < simon`> | it was my first programming language. :) |
| 23:44 | <@MyCatVerbs> | Oh well. Glad to hear that it's at least fond. ^^ |
| 23:45 | < simon`> | I hear they got tail-recursion less than a year ago. the community is not dead! ;-) |
| 23:56 | | AnnoDomini [farkoff@Nightstar-6e923c5e.adsl.tpnet.pl] has quit [[NS] Quit: ACHTUNG! ALLES LOOKENSPEEPERS! Das Internet is nicht fuer gefingerclicken und giffengrabben. Ist easy droppenpacket der routers und overloaden der backbone mit der spammen und der me-tooen. Ist nicht fuer gewerken bei das dumpkopfen. Das mauskli] |
| --- Log closed Sat Nov 14 00:00:43 2009 |