[00:09] snap-l: i found 3 good bands due to club metal #4. i'm especially surprised that i like http://unwoman.bandcamp.com/album/unremembered [00:39] brousch: Yeah, that's an awesome album. [00:40] Bought that on disc because I liked it so much [00:51] i like so few female singers [00:59] What were the other bands you found? [01:08] neurotech, cryogen second [01:10] old Hanzel Und Gretyl is great. [01:15] Uber Alles is awesome [01:16] brousch: Yeah, I was really happy to have found Neurotech [01:16] that pretty much sealed me putting out the episode. :) [01:16] I want more of that sort of music [01:20] http://www.youtube.com/watch?v=FOtjiMYUJ_s [01:20] unfortunatley they lost their domain [01:40] <_stink_> heh, that python fiddler thing looks really broken. [01:41] <_stink_> importing json and trying to dumps() a list causing a library traceback, and the gmail oauth isn't set up right. [02:10] heh, sure it's got issues [11:58] morning [11:58] ++ [12:01] morjning [12:01] brousch: http://pyrseas.wordpress.com/2011/08/29/dueling-frameworks/ [12:01] he's ranting on your flask, attack! [12:03] hehe [12:04] i think 0.8 due next week has the class-based thing [12:05] and i have trouble with the database passing [12:05] passing? [12:05] the per request session management? [12:05] well, passing the db variable around [12:05] import it [12:06] you're not doing db stuff in your controller actions are you :P [12:06] the db connection needs the app [12:06] and then the app needs the db [12:06] hmm, maybe one day I'll check it out. I keep meaning to find an excuse to do something in flask [12:06] I've had to figure it out in pylons and pyramid and command line scripts so usually able to get a handle on the stuff these adys [12:07] flask-sqlalchemy needs the app to make the base in the models [12:07] but not looked at how that flask plugin/addon works [12:07] but then to use the db in the app, you have to import models [12:07] it doesn't seem right to me [12:07] oh, circular dep issue? [12:07] yeah [12:08] if it's all in one file, like a very simple app, then it's fine. but when i break out models it gets ugly [12:08] well normally what you find is what pylons/pyramid does [12:08] the models originally are attached to a base with no db on it [12:09] then there's an "init_models(engine)" [12:09] that accepts a build db engine and assigns it to the base [12:09] that way on initial import the models can load, but just don't talk to anything, and then at the end of your app "boot" phase, it grabs the engine data from your config and runs init_models [12:10] https://github.com/mitechie/Bookie/blob/master/bookie/models/__init__.py#L44 [12:10] for example in bookie [12:12] that is the biggest init.py i've ever seen [12:12] yea, I'm bad, I should move the bmark stuff all out [12:12] but it starts as a single models.py file [12:12] and I moved it into a module and didn't clean it all up [12:22] good morning [12:22] in my little cherrypy python test i was about to run into the circular issue again [12:22] the main app has all of the config, and the model needs to access it to get the db config [12:23] but then app needs the model to do db stuff [12:24] so i was going to have the model access the config without going through the app, but then the config is being parsed twice, or there are two config files [12:25] booo [12:26] just delay, you don't need the model stuff until the ned of the app code anyway. [12:26] that's what that initialize_sql thing is doing [12:26] get an app up somewhere brousch and I'll edit it and show you what I mewan [12:26] i think i see it now [12:26] have a function in the models.py that someone else can call to create the engine [12:27] right [12:27] then the model has no need to know the config, the app can call the create_engine function with the right dburi [12:27] call that after imports and all that work is deon from your app [12:27] right [12:27] exactly [12:27] much better [12:27] and you don't end up with models dependant on a config file [12:28] you can still load them via ipython, external script and just feed it an engine [12:28] i like that [13:13] http://bazaar.launchpad.net/~barry/flufl.enum/trunk/files/head:/flufl/enum/__init__.py?file_id=flufl-20100424151503-h6gekpiiqsaprkiu-1 [13:13] does that load for anyone else? [13:14] nope [13:16] k, glad it's not just me then [13:54] that was kind of not unpleasant [13:54] tv is cancelled, officially a cord cutter [13:54] finding the number took some work, but not real pressure/hard time cancelling [13:54] they know you'll be back [13:55] lol [13:55] yea, actually don't think I've not had cable/etc since I can remember [13:55] so will be a new trial for sure [14:03] rick_h_: I am sans cable down here [14:03] jcastro: yea, with roku + ota hd I'm hoping it's enough for the wife's needs [14:03] going to buy a second roky for upstairs I think [14:03] I have an hdhomerun but I need an antenna, just haven't bothered [14:03] roku, damn I can't type today [14:03] yea, finally got an antenna and put it up in the attic this weekend [14:03] works great, couple of channels look really nice [14:05] crazy to think for the cost of tv service could buy a roku a month :/ [14:08] i just use the $15/mo cable service. all local channels nice and clear [14:08] $99 antenna, no need for $15/mo [14:09] does that still require a box brousch ? [14:09] no [14:15] rick_h_: bah, my Flask circular dependency is my own stupidity. i am using an application-specific path inside of my model. i can easily fix that [14:15] ah, ok === jjesse_ is now known as jjesse [14:42] i refactored out the circular dependency and eliminated a few lines of code in the process [14:43] why is it when i do something the right way i end up with less code? [14:46] heh. :) [14:46] It's the way of it. :) [14:55] it's a very true sign [14:55] whenever I'm forcing a bad design I have all kinds of crap code trying to make things happen [14:55] and later on, it hits me, and code gets cut in 1/3 [15:44] Can someone riddle me why I keep running ruby on my machine? [15:44] because there's something cool on there? [15:44] and you have an on again/off again relationship with tracks? [15:44] yes [15:44] what does Ruby give you that Java or C can't? [15:45] And for some reason I keep following the development branch that goes from working to not working [15:45] gamerchick02: tracks [15:45] cool libraries/packages :) [15:45] is it love? coffee in the morning? [15:45] * gamerchick02 is being silly again [15:45] * rick_h_ parses snap-l's line there and wonders if that was rhetorical or not [15:45] rick_h_: Which line? [15:46] "And for some reason I keep following the development branch that goes from working to not working [15:47] " [15:47] sorry, monday, I'm not coherent today [15:47] then why run the dev branch? [15:47] gamerchick02: sssshhhhh [15:47] Convert tracks to use Bundler for gem resolution [15:47] lol [15:47] i run the dev-daily of chromium, but if it borks, i go back to firefox. no big hairy deal. [15:47] woo! ruby, where there's no standard tool to do anything, but 20 ways to do it differently [15:48] rick_h_: I like being silly sometimes [15:48] and happy Monday to all. [15:48] Which apparently is like saying "lets use pip, but not the one that works" [15:48] snap-l: you hit up devin on CHC? I'm going to be late and wonder if we can work out the gift card purchasing situation. [15:48] https://github.com/carlhuda/bundler/issues/865 [15:49] Neck: fuck me in [15:49] rick_h_: That's the plan, or I'm screwed to make it out again. [15:49] k [15:49] I need to go find food...grrrrr [15:50] Apparently when you do "gem install bundler" it does not put a command "bundle" ahywhere that is usable [15:51] ah, because ruby on Ubuntu installs libraries in /var [15:51] how fucking brilliant [15:54] yea, I had to manually patch paths for that in arch as well [15:54] puts stuff in /opt [15:54] why? [15:54] unless you're Sun, nothing goes in /opt [15:55] /var/lib/gems/1.8/gems/bundler-1.0.18/lib/bundler/index.rb:110: [BUG] Segmentation fault [15:55] I think I need to seriously write a python implementation of Tracks [15:55] this is stupid. [15:56] <_stink_> o/ [15:59] Installing ZenTest (4.6.1) /usr/lib/ruby/1.8/rubygems/installer.rb:170:in `install': ZenTest requires RubyGems version ~> 1.8. Try 'gem update --system' to update RubyGems itself. (Gem::InstallError) [16:41] uh oh, snap-l having fun? [16:41] time to figure out rvm or something me thinks [16:41] forget that system level ruby stuff [16:41] I'm done. [16:41] lol [16:41] I'm rewriting this thing in Python [16:41] virtuanenv --no-site-packages tacks_not_ruby [16:42] I'll need some help with the interface to get what I want, but frankly I'm sick of feeling like my system is overrun by idiots. [16:42] you could call it 'pins' [16:42] man, if only you knew anyone that did web development, some uis, and in something not ruby :P [16:42] and flavor of the month club shit [16:42] or 'pyns' [16:43] rick_h_: Yeah, and if only there was a framework that could tie all of the models and views and controllers together [16:43] ouch, that's a toughy [16:43] Blazeix: Half tempted to call it Amish Rake Fight [16:43] (except there's already a band called that) [16:43] to google it is [16:45] Regardless, I need to migrate my stuff again to todo.sh so I at least have something to look at [16:45] But this is it. I'm done with Ruby. [16:45] have you ever played around with emacs org mode? [16:46] I have a friend who organizes his entire life using that [16:46] there's a vim port, before you get the idea of launching emacs in a virtualmachine [16:46] Blazeix: Have you ever stuck your finger in a light socket? [16:46] not a fan? [16:46] Not particularly [16:46] It specifically calls out GTD as something it handles :) [16:46] both of emacs and of org mode [16:47] maybe I haven't given it enough time, but most of the text based ones I've seen use project identifiers that look more like a Scott Adams text adventure [16:47] GET BANK [16:47] CLEAN GARAGE [16:47] PLAY FIDDLE [16:47] I don't work that way. [16:47] you play the fiddle? [16:48] gamerchick02: and I dance with the devil in the pale moonlight [16:48] ooooh, cool. :) [16:48] all i can think of is hyperbole and a half when you posted that: GET BANK, CLEAN GARAGE, etc [16:50] What are you, a wizard? [16:51] (note: no, I don't play the fiddle, unless you count making cat-in-pain noises). [16:52] the snap-l went down to Georgia? No? [16:53] I can play that song... on the radio. [16:53] :-) [16:53] :-D [16:54] speaking of snap-l and radio... I was listening to Pandora on my commute today, and I "liked" Nickelback and Metallica within minutes of each other. I contemplated the possibility this might make Pandora explode... [16:55] Wolfger: No, just shows you have no taste. ;) [16:55] I should have not liked Metallica? [16:56] Depends... was it pre-load or post-load? [16:56] I have no idea what that even means [16:57] What song was it that you liked? [17:00] Sad But True [17:01] pre-load [17:01] though I tend to referrer to metallica as pre-black and post-block [17:01] the load era is just something that's better forgotten [17:02] Yeah [17:03] created this station off a single song suggestion a couple weeks ago for Alcohaulin' Ass by Hellyeah. So far, it's been pretty good for me. [17:07] Nice, wolfger. but why Nickelback? [17:08] *sigh* Not you too... [17:08] no, i'm asking why? i like a couple of songs by them [17:08] i don't openly admit it, lest my hipster cred be tarnished... [17:08] lol [17:08] :-D [17:08] i've got the black rimmed glasses and the chucks. [17:09] I like Nickelback. The songs that don't hit the top 40 more than those that do. [17:09] i don't fit skinny jeans, but hey, 2 out of 3 ain't bad. [17:09] ah. :) [17:09] well, except Rock Star. That one's cool *and* popular [17:09] :) [17:51] snap-l: where on your TODO list does the Pyramid-based task organizer go? [17:51] It goes into todo.sh. ;) [17:52] before you can add todo.py to your tasks list, you must first create todo.py [17:55] your mission is to write a minimal task list today, and have it always working as you enhance and build it. this way you will only add features you need as they become more important than other tasks on your list [17:55] remember, api first, then app :) [17:56] hmm, so is he rewriting tracks or todo.sh? [17:56] i have no idea what the difference is [17:56] tracks is a web based app, todo.sh is a command line plain text deal [17:56] i'm just trolling to procrastinate doing real work [17:57] yea, I'm done with real work for the day [17:57] \o/ [17:57] http://joeybeninghove.com/2011/08/16/backbone-screencast-introduction-views/ watching this [17:57] but man is he hard to watch/listen to [17:57] getting some yummy ideas though [17:58] I'm rewriting tracks [17:58] cool [17:58] At least that's the plan. :) [17:58] from ruby to python you should be able to almost just port it vs rewrite it [18:02] i demand a native android client [18:03] * Wolfger just heard brousch volunteer [18:03] yes, i volunteer to demand a native android client [18:05] speaking of which, has anyone played around with phonegap at all? [18:05] brousch: was playing around with it [18:05] i made 2 tiny apps with phonegap [18:05] thoughts? It seems like a good way to make webapps with app store support. [18:06] do they provide JS bridges to native APIs? [18:07] yes [18:07] some APIs [18:07] http://docs.phonegap.com/ [18:07] yea, not all [18:07] it is basically a wrapper around webkit with some native API access [18:08] so your app is javascript/html/css [18:14] hah! http://net.tutsplus.com/tutorials/javascript-ajax/getting-started-with-backbone-js/ [18:15] backbone collections are just my xxxMgr objects [18:15] awesome [18:27] <_stink_> snap-l: i wrote pydidit as a sqlalchemy/curses todo app, but i confess it was my first attempt and is pretty broken. and it's not necessarily GTD. :) but let me know if you get something going, i'd love to look at it/help out. [18:28] for those of you with the HP tablet, looks like CyanogenMod now works on it [18:28] http://techcrunch.com/2011/08/29/cyanogenmod-team-gets-android-working-on-hp-touchpad/ [18:32] jjesse: works is relative... no touchscreen support yet I thought [18:36] touchpad - touch screen = brick? [18:36] if bt works you can keyboard [18:36] think they had a bt keyboard for the pre [18:36] it's a good step, I had heard the graphics chip was diff than most android devices, so if they got a gui, that's a good step [18:36] _stink_: I'd really love some of your help with this [18:37] <_stink_> snap-l: if you create a project or googlegroup or anything, please include me [18:37] I'm going to set up some models and hopefully push up some code real-soon now. [18:37] _stink_: at CHC? [18:37] It'll likely be on github. [18:37] <_stink_> awesome. [18:37] rick_h_: yeah it's an adreno 220 [18:37] :P [18:37] <_stink_> rick_h_: someday :P [18:38] <_stink_> this Wednesday is the end of the zoo summer concerts [18:38] hmm, have to squeeze in some time before hockey starts :P [18:38] how's the boy doing? [18:38] <_stink_> he's good. seems to have gotten past a few of the devil child things he picked up a few weeks ago [18:39] <_stink_> but pretty typical goofy toddler atm [18:39] <_stink_> yours? [18:39] hah! let me know how you get those out. toddler-hood is going to test my new found adult patience [18:39] oh he's great as long as I'm biking him around or mowing for him to watch, or something else that involves dad working his @#$#@ off [18:40] <_stink_> we just try to be relentless in correcting him [18:40] <_stink_> hah [18:40] _stink_: yea, steady wins the race you hope [18:40] <_stink_> rick_h_: hah, so he's making you work. [18:40] <_stink_> i think there are some potential health benefits to fatherhood [18:40] I credit him with half my weight loss. He's a slave driver [18:40] <_stink_> i can lift more than i could beforehand [18:40] heh, that's for sure [18:52] Three is the worst age so far [18:53] potty training is going to kill me [18:53] so you've got that to look forward to [18:54] yea, just ordered a potty seat [18:54] and stool, so getting to start that [18:54] <_stink_> we got a potty, but it's just sitting for now [18:54] <_stink_> not really thinking about that yet [18:55] the day care room he's in makes him do it before he moves on [18:55] <_stink_> damn [18:55] <_stink_> pressure. :) [18:55] but yea, it's more of an early stage, trying to get him interested, "do you need a new diaper" etc [18:56] sweet, my python namespaced packages are working [18:56] time to start up my own python namespace [18:57] from mitechie.something import awesomesauce [18:57] <_stink_> hah [18:57] I wonder if my boss would let me post our tools up [20:59] https://github.com/craigmaloney/PyTracks [21:01] right at 5pm? seriously? [21:01] :P [21:01] ;) [21:02] It's just a pyramid shell right now [21:02] <_stink_> o/ [21:06] I FUCKING HATE CAPTCHAS [21:22] http://identi.ca/conversation/79356145#notice-82009866 <- seriously? [21:46] I hesitate to say this, but I'm with psquid [21:46] * Blazeix ducks [21:54] also, I read that thread as psquid prefacing all his comments with "oh snap!" [21:55] http://identi.ca/conversation/79356145 [21:55] Blazeix: captchas are not a solution. They're a stop-gap [21:56] http://identi.ca/notice/82013382 <- Blazeix [22:00] snap-l: are you against captchas in their current form, or captchas in general? [22:01] by "in general" I mean taking the literal definition of an automated way of telling comps and humans apart. [22:02] maybe every 50th captcha I run into I have trouble deciphering, but there's a refresh button to bypass it. [22:03] I'm against them because they're user-hostile [22:03] you've yet to answer any of psquids questions with a serious answer. [22:03] so what's your ideal solution? [22:03] Awesome, he blocked me. [22:03] I'll agree that captchas aren't the ideal solution, but they are a solution. [22:04] my point is that they're not effective because humans can route around them to get spam accounts [22:04] and the problems with normal users outweigh the benefits. [22:06] snap-l: you mean mechanical-turk style? [22:06] Yes. [22:06] yeah, that is an issue, and I don't see a way around it. [22:07] but it's a difference of rates [22:07] a computer can spam 100s (or 1000s) per minute. [22:07] or at least, way faster than a human can. [22:07] Understood, but there are better ways of dealing with that that don't inconvenience users. [22:08] like what? [22:08] moderation, throttling [22:08] The one thing that comes to mind is an empty field that's hidden [22:08] throttling wouldn't be useful, due to bot net spam. [22:09] moderation would be insane without some other way of limiting spam. [22:09] There's also algorithmic pattern matching (akismet) [22:09] In each of these, it's transparent to the user. [22:11] akismet is pretty awesome, but it isn't perfect, and you have false positives. [22:11] I'm very passionate about making the user experience better, and CAPTCHAs inhibit that. [22:11] Blazeix: You get false positives with CAPTCHAS [22:12] true, but I'd wager less than akismet. though i don't have any stats to back that up [22:12] I've had well-known sites eat postings that I just said "fuck it" because the CAPTCHA killed my post [22:12] I don't either [22:12] yeah, you raised that on the ident thread [22:12] that's an issue with crappy sites, rather than CAPTCHAs [22:12] and that's what really pissed me off about that conversation was me venting my spleen turned into having to defend why I think they're stupid [22:12] any site worth its salt will use some ajaxy thing [22:13] I suppose I'm not helping by raising it in IRC, then [22:13] I have yet to see a captcha that made me thrilled to use it [22:13] Blazeix: I feel more comfortable talking to you about it because I know you [22:13] You're not going to block me because I'm an ass. [22:15] I guess very few CAPTCHAs induce happiness (short of the find a puppy captchas), but a site not being overrun by spam does. at least for me. [22:16] True, but they're a false sense of security [22:18] I don't believe there's a solution that can give you true security. I'm not arguing it's anywhere close to perfect, but it does help, and I wouldn't want captchas to disappear tomorrow. [22:18] I definitely wouldn't call them "ineffective." But we can continue this at CHC :) [22:18] if you want [22:24] No, I'm good. :) [22:26] http://identi.ca/notice/82016008 [22:31] http://seldo.tumblr.com/post/9549775746/this-is-genuinely-microsofts-idea-of-a [22:36] oh jesus that's ugly [22:37] the only good part is that the ribbon is hideable, so you can get rid of all that crap and use keyboard shortcuts. [22:37] the way god intended. [22:38] I love the design differences between MS and Apple [22:38] MS makes the buttons bigger [22:38] Apple does away with the buttons [22:45] am I right in thinking that tmpfs should't fill up? [22:45] tmpfs /tmp tmpfs nodev,nosuid,mode=1777 0 0 [22:45] tmpfs 4097040 4097040 0 100% /tmp [22:45] Filesystem 1K-blocks Used Available Use% Mounted on [22:48] This is really fucked up. [22:48] Going to remove tmpfs and reboot. [22:48] brb. [22:53] OK, let's see how that works. :) [23:49] http://online.wsj.com/article/SB10001424053111904787404576535211589514334.html [23:49] man [23:49] this is awesome [23:52] jcastro: That's awesome. [23:53] I'd forgotten how truly dysfunctional HP became. [23:55] wow, I didn't realize their CEO was from SAP. That's doom right there.