[00:01] Hostgator just suspended me. :) [00:01] They want account verification. [00:01] I assume this is because I bought a 3 year business account. [00:01] epic [00:08] jeez, that was fast. [00:31] I'm refunding hostgator and switching to webfaction. [00:33] <3 [00:34] SketchCow: let me know if you have any questions w/ the webfaction setup [00:34] SketchCow: the app/domain mapping is a bit unique to them [00:36] I want to run a mediawiki on a domain. [00:36] It's not that hard [00:36] If it IS hard, something's wrong [00:39] He;;p [00:39] Hello* [00:40] Hi [00:40] Is there ssl here? I cant seem to connect to port 6697 [00:40] Oh, it's connecting now, sorry [00:41] Thank you for signing up with us. [00:41] I just bought a 5 year account at webfaction. [00:42] Is that the dealbreaker problem you were talking about on twitter? [00:43] The dealbreaker problem was hostgator suspending my account while I was working to move files to it. [00:43] And then demanding ID, and then taking an hour or two. [00:43] Sorry, so fired [00:44] Either demand my ID as a final step, or don't demand it. [00:44] SketchCow: i've heard good things about rackspace [00:44] Put it up, make it work, THEN jam it down, ESPECIALLY when I just gave you $350 bucks, sorry, you're shit and I'm done. [00:44] If webfaction shits the bed, rackspace is next. [00:44] have you tried namecheap? [00:45] I use them for domains, not hosting. [00:45] Ah [00:46] SketchCow: they actually have a one-click install for mediawiki [00:46] v1.8.2 [00:47] Good. [00:48] SketchCow: https://panel.webfaction.com/app/create?script_url=https%3A%2F%2Fwiki.webfaction.com%2Fwiki%2FMediaWiki-1.8.2Install%3Fformat%3Dtxt [01:18] Wow, that's one fucked up interface. [01:18] I wish you had told me it would be something by geeks to provide to nerds to make them all feel like sexless robots with infinite processing space and time [01:19] I want to hear the AWESOME metric this approach fulfills [01:19] I want it so you go to archiveteam.org and it shows the wiki. Looks like this thing tries to shove you into webapps. [01:19] SketchCow: i told you to ask :) [01:20] SketchCow: create mediawiki app, create archiveteam website, add architeteam domain to said website [01:20] and you'd golden [01:20] *you're [01:21] I like that. [01:21] 1. Turn lead to gold [01:21] haha [01:21] 2. Forge gold into statue [01:21] 3. Tell golem to emulate gold statue [01:21] 4. Instant sandwich! [01:21] I just want a sandwich [01:21] it's setup so you can have, say, wordpress running under a certian path on a domain [01:22] like mediawiki for /, wordpress for /blog/ [01:22] Yeah, it's on crack. [01:22] Anyway, you're no help [01:22] I obviously have to go through some sort of cult transition to think in this retarded geek environment [01:22] it takes 3 seconds to setup [01:22] So dropping spoilers that mediawiki is a sled isn't going to help [01:22] Yes, it takes me 3 seconds to find out the processing of a machine too [01:22] Unless I'm dropped into Unix like a foundling [01:23] Allow me some time to go read this crazy moonbat environment up [01:24] SketchCow: http://docs.webfaction.com/user-guide/websites.html [01:24] Too late, I'm working over here [01:28] gotta be fast to keep up with the cow [01:29] Error: Traceback (most recent call last): [01:29] File "/var/tmp/api_system.28582", line 96, in ? [01:29] func(account, app_name, autostart, extra_info, password, server, session_id, username) [01:29] File "/var/tmp/api_system.28582", line 58, in create [01:29] server.create_db(session_id, db_name, 'mysql', password) [01:29] File "/usr/local/lib/python2.4/xmlrpclib.py", line 1096, in __call__ [01:29] return self.__send(self.__name, args) [01:29] File "/usr/local/lib/python2.4/xmlrpclib.py", line 1383, in __request [01:29] verbose=self.__verbose [01:29] File "/usr/local/lib/python2.4/xmlrpclib.py", line 1147, in request [01:30] return self._parse_response(h.getfile(), sock) [01:30] File "/usr/local/lib/python2.4/xmlrpclib.py", line 1286, in _parse_response [01:30] return u.close() [01:30] File "/usr/local/lib/python2.4/xmlrpclib.py", line 744, in close [01:30] raise Fault(**self._stack[0]) [01:30] xmlrpclib.Fault: [01:30] Yeah, there's some aweseom [01:30] Well, thanks for dropping me into a true level of hell [01:30] It's like driving a car through a playground of babies and steering with PERL [01:30] I mean, a serious, level of hell [01:30] 16 characters or less. [01:30] Let's do surgery with algrebra. [01:30] welcome to 1983. [01:31] Shut up, once we used 17 characters in a filename and a man DIED. [01:31] He DIED. [01:31] was it CP/M? [01:31] cause serves him right if it was. [01:32] I heard he was trying to use sed and awk into his perl compost because he was cooking a muffin [01:32] Luckily his C++ preprocessor was fibbling the foorogotz and we didn't blow up a nearby church [01:32] This is all in the script [01:33] is the mediawiki in the debian/ubuntu repos not an option for some reason? [01:33] for all i know it's really old or no one uses it or it isn't there [01:33] Your account will be canceled in the next 24 hours (usually much sooner). [01:33] but apt installing something is easy [01:33] Goodbye, Webfaction [01:33] Onto rackspace [01:33] Good for me to have $700 out in the cloud [01:34] you are guaranteed to get $700 back, but it may not all be your $700. [01:34] Well, actually [01:34] You know what? [01:34] Fuck it [01:34] One more fucking day of fucking viruses on the fucking mediawiki. [01:35] Because fuck it, I don't need this shit. [01:35] I can handle the giggling bucktooth nerd feedback that we're "hacked" for 24 hours until I cool down [01:36] Rackspace doesn't look like a good fit. [01:38] SketchCow: sorry for the hassle. I seriously did not mean to provoke you into trying to fix this right away, I was just talking about it this afternoon because I was stymied that I seemed to be the only person afflicted. [01:39] what exactly happened [01:40] I run mediawiki too and I'm scared [01:41] do we just want a wiki? [01:41] if so, github has a great wiki service [01:42] powered by git, so we're not relying on them either [01:51] Yeah, no. [01:51] I'm speaking with the host for textfiles.com. He's a fan, he's good, he helps me. [01:51] We'll straighten this shit right out [01:52] What's up SketchCow [02:08] Three sniper bullets short of a high school shooting [02:08] hang in there, brother [02:08] if there's anything I can help with, let me know [02:09] I'm just going to go with the guy, I'm sure. [02:09] I like The Guy [02:10] Also, I am poor these days. [02:10] $300 for 5 years makes sense. [02:10] $240 a year? Not so much. [02:24] Cheered up. [02:25] Going with the textfiles.com host, he's been there for me over the years. [02:25] Also, the Musee' Mechanique just gave me permission to film in there. [02:25] No lights, no wires on floor, one guy, one camera. [02:25] ah, fun place [02:25] Yes! [02:25] I knew they likely got a lot of shit-ass crews. [02:25] that should be fun [02:25] liked it better when it was on the cliffs [02:25] that should be really awesome [02:26] I went there once and it was an awesome place [02:26] > Please cancel my "archiveteam" account effective immediately. I would like to take advantage of your 60 day money back guarantee and request a full refund. [02:26] I confirm that your account has been closed and a full refund has been sent. [02:26] > Your site is for nerds telling geeks how to act like robotos. [02:26] Can you tell us what exactly did you find hard/difficult in our panel? [02:26] [02:26] Did you consult our docs at docs.webfaction.com or try to contact support about it? [02:30] Note to self: never recommend anything to SketchCow. Ever. [02:32] many have tried / few have not died [02:32] I have this hilarious exchange [02:33] Where Tim Berners Lee is talking to people about whether he has the old webserver code [02:33] And he says not, and mails us at the archive for a copy [02:35] and the archive says... [02:35] Also, it's my fault, I should have known that the guy who has sex with Heroku servers and made it his full-time job would probably look at a web host that lauds the experience of steering a hay truck with bash scripts and think "oh, he's gotta get in on that delicious buffet" [02:36] Like, if you're not into spicy food and you ask the guy whose cube smells like you got punched in the face by Vindaloo where's a good place to get grub [02:36] The archive is at a loss, but thanks to the links, we'll be in better shape [02:40] * SketchCow kennethre NO GETTING BITTER [02:41] My quest in life is now to start a user-content company that gets massively popular and suddenly kill it without warning. [02:41] No, no, it doesn't taste right if you plan ahead [02:42] It's gotta be a surprise [02:42] haha [02:42] You have to think it's all for the good of humanity and speak at SXSW and in interviews [02:42] Then when you sit at your desk, head down, while they haul out the beanbags, you look up with seething eyes [02:42] changing the world [02:42] Soulant: cloud spirituality [02:42] I'm laughing so hard right now [02:43] I will crush your dreams as mine have been crushed [02:43] also sex with Heroku servers could be awesome [02:43] like [02:43] need more? [02:43] HEROKU SCALE [02:43] yeah [02:43] yipdw: it is indeed [02:43] They surround you with their little mouths [02:43] All want some shared you [02:44] It goes to 11. [02:44] that said, I deployed a CouchDB-based app to heroku + cloudant today [02:44] and after trying to hook up the app to the Cloudant DB I saw the "You're Doing It Wrong" talk that specifically mentions why couchrest_model is bad [02:44] guess what I'm using [02:44] so uh I don't know what the point of that was, except thanks Heroku, I guess [02:45] <3 [02:51] need more? [02:51] HEROKU SCALE [02:51] hahaha [02:51] if you're a pimp, you can even use heroku ps to keep track of all the bitches you've choked [02:52] Here, cheering up emo kennethre [02:52] http://www.archive.org/~jake/free-couch/ [02:52] Free Couches on Craigslist. [02:52] haha awesome [02:52] :D [02:52] wtf? [02:52] yes, but when can you deliver the couch? I live on the 4th floor. [02:53] not bad: http://www.archive.org/~jake/free-couch/5Id5L75W43K23F53J4c1ob627dceda7c51c7d.jpg [02:53] crazybeans. [02:58] http://i.qkme.me/36ckpl.jpg [03:05] OK, I'm going to call it a day [03:05] I got some stuff done, but I'm behind [03:05] Tomorrow, tomorrow things get done. [03:05] I'm moving URLTE.AM to my textfiles.com machine, it's a fucking static site and people have been very patient with me about it. [03:09] OK. [03:09] Over the next day or two, urlte.am should go to textfiles.com. When it's looking right, I'll plunk our replacement back. [03:11] DNS servers: [03:11] ns1.easydns.com [03:11] ns2.easydns.com [03:11] remote1.easydns.com [03:11] remote2.easydns.com [03:11] ns3.easydns.org [03:11] ns6.easydns.net [03:11] Well, that was fast. [03:13] OK, it'll be good shortly. [03:13] Good, something off the fucking plate [03:13] Now, to go see SF [03:13] If kennethre only LIVED here, we'd be having DINNER [03:13] enjoy your evening, brother [03:13] on a fucking plate [03:13] SketchCow: :( [03:14] SketchCow: I'll be in san jose next thursday [03:14] er, santa clara [03:19] some of you may have seen this already, but it's a really awesome archive thing, so I'd like to share it again: http://infovore.org/archives/2012/02/26/a-year-of-links/ [03:20] that is a good one [04:06] "Do you want to type them in by hand?" [04:07] shit... i wish i still had that "The Internet Yellow Pages" book [04:08] perhaps I still do have it in a box somewhere [04:09] metadata: apparently always my enemy [04:09] even such simple things as "what books are in this box?" [04:11] oh god, the Internet Yellow Pages [04:11] Michael Wolff, right? [04:14] i don't remember [04:17] looks like there were several books by that name [04:18] iirc, the cover looked like this: http://ecx.images-amazon.com/images/I/414R8NBC1XL._SL500_AA300_.gif [04:24] SketchCow: If you're still accepting images for your next talk, Mine is on the wiki: http://archiveteam.org/index.php?title=User:Scumola [04:27] Also, tweets mentioning 'oscar' during last night's broadcast (I've got to do something with all of that data): https://skitch.com/scumola/8fmex/steves-twitter-trend-grapher [04:28] "oscar 17 crystal 1"? [04:28] assuming billy crystal [04:29] but what is the number? [04:29] That just happens to be the values on the graph where my mouse was. Ignore that. :) [04:31] Also, I've been unwinding all urls from url shorteners that have appeared on the twitter spritzer feed since Oct, 2011. I've got about 23M unwound urls now and counting. [04:33] fuck yeah [04:33] urlteam++ [04:34] Yea, I haven't forgotten about urlteam. :) [04:37] I've got the whole twitter thing down to a science now. I have nagios watching the stream, so I get paged when the stream goes down or I've lost more than a couple minutes of data. I archive (bzip2) the data in minute-buckets and then go through each minute bucket and unwind all of the urls recursively and insert into a DB. I keep one week's worth of the full tweets in the DB also for [04:37] twitter-based searches if I need it. [04:37] I've got a small sqlite database of goo.gl short urls that showed up in a torrent listing site rss feed [04:38] how large do the minute buckets generally get? [04:38] Coderjoe: That's cool. I was initially scraping the url-shorteners too, but I figured that I wanted the ones that people were actually using. [04:39] well, mine is because I wanted to use the feed, but some people submit the torrentinfo page link on another site, while pretty much everything else is a link to download the .torrent file [04:39] each minute is only like 5MB or so uncompressed. Tweets are only about 2k each and the spritzer feed is like 2% of the full firehose. [04:40] but it's enough to get a good sample size and most urls are re-tweeted anyway, so I am probably capturing most of the 'important' urls. [04:40] each hour is about 350MB of data. [04:41] and some of them are behind goo.gl urls. so I wrote a php script that pulls the feed, unshortens goo.gl urls (and saves the results in a database for later lookup, so I don't hammer google each time), and then convert the torrentinfo links to download links [04:41] tweets are 2k? [04:41] a day is about 6.9GB uncompressed, 700MB compressed [04:41] Yea, thereabouts. It varies, but 2k is a good ballpark figure. [04:41] for a 140-char message? [04:42] that's a lot of overhead [04:42] yea, there's a bunch of metadata in there. Lat/Lon, user profile stuff, ... [04:42] 17:54:58 +0000 2011","profile_sidebar_fill_color":"ffffff","id":306331380,"utc_offset":-21600,"favourites_count":7,"url":null},"in_reply_to_screen_name":null,"truncated":false,"id":174353887023730688,"created_at":"Tue Feb 28 04:43:00 +0000 2012"} [04:42] i0.twimg.com\/profile_background_images\/414657846\/tumblr_lyqhl7NzZL1r972d6o1_500.png","profile_background_color":"EBEBEB","protected":false,"id_str":"306331380","profile_background_tile":false,"profile_image_url_https":"https:\/\/si0.twimg.com\/profile_images\/1847372264\/sel_normal.png","name":"AA\u2665","default_profile_image":false,"follow_request_sent":null,"created_at":"Fri May 27 [04:42] ription":"","is_translator":false,"show_all_inline_media":true,"following":null,"profile_background_image_url":"http:\/\/a2.twimg.com\/profile_background_images\/414657846\/tumblr_lyqhl7NzZL1r972d6o1_500.png","default_profile":false,"profile_link_color":"b6ffb6","time_zone":"Central Time (US & Canada)","verified":false,"geo_enabled":false,"profile_background_image_url_https":"https:\/\/s [04:42] ser":{"notifications":null,"friends_count":235,"profile_sidebar_border_color":"DFDFDF","screen_name":"andy_arceo","contributors_enabled":false,"lang":"es","statuses_count":28438,"profile_use_background_image":true,"location":"","listed_count":0,"profile_text_color":"000000","followers_count":203,"profile_image_url":"http:\/\/a0.twimg.com\/profile_images\/1847372264\/sel_normal.png","desc [04:42] {"contributors":null,"text":"y pulido me aplico la flechita OTRA VEZ","entities":{"urls":[],"user_mentions":[],"hashtags":[]},"in_reply_to_user_id":null,"place":null,"retweeted":false,"coordinates":null,"retweet_count":0,"source":"web","in_reply_to_status_id_str":null,"geo":null,"in_reply_to_status_id":null,"favorited":false,"in_reply_to_user_id_str":null,"id_str":"174353887023730688","u [04:43] That's one tweet. [04:43] yum [04:43] so the full feed is how much? [04:43] the full feed costs money, but is about 70Mbps uncompressed. [04:43] whee [04:43] the spritzer feed (the one that I get) is free and it's only 1.5Mbps uncompressed. [04:44] I'd love to get my hands on the other social media feeds like foursquare and others. [04:44] i bet we could get twitter to send us dumps [04:45] i know some people in ops over there [04:45] they charge money for the data - big-time. [04:45] only the sprtizer feed is free. [04:45] yeah but we're archivists, it's different [04:45] plus, the full twitter feed of everything would be huge, I'm guessing. [04:45] compressed, it shouldn't be SO bad [04:46] statuses/firehose [04:46] Returns all public statuses. The Firehose is not a generally available resource. Few applications require this level of access. Creative use of a combination of other resources and various access levels can satisfy nearly every application use case. [04:46] heavily compressed [04:46] oh nice [04:47] Returns a random sample of all public statuses. The default access level, âSpritzerâ provides a small proportion of the Firehose, very roughly, 1% of all public statuses. The âGardenhoseâ access level provides a proportion more suitable for data mining and research applications that desire a larger proportion to be statistically significant sample. Currently Gardenhose returns, very roughly, 10% of all public st [04:47] statuses/sample [04:47] atuses. Note that these proportions are subject to unannounced adjustment as traffic volume varies. [04:48] Just the spritzer feed (2% of the full firehose) is (on my system) ~90GB compressed for everything since Oct 2011. [04:49] i wonder what level the randomization happens [04:50] like, would different connections get different results? [04:51] I think that each "hose" is just a modulus of the tweet id. So, the spritzer is $id mod 100 = 1 or something like that. [04:51] so everyone getting the spritzer feed would get the same tweets. [05:19] damn, 90GB of tweets [05:20] I hope someone has wordclouded that shit [05:20] yipdw, from who? [05:21] Zuu-: swebb2's copy of the spritzer feed [05:21] ah [05:46] hai, is there a channel for mobileme? [05:46] #memac [07:40] Pardon my ignorance, but is archiving Twitter necessary, given the LoC archive? [07:41] swebb2: loc? [07:41] oh nvm duh [07:41] Library of Congress [07:41] yeah [07:44] Is it to provide a public dump vs. request only? [08:21] * joepie91 is running a few boxes on the mobileme thing [08:21] er [08:21] * joepie91 is running the mobileme thing on a few boxes * [08:22] Jump over to #memac, that's where the mobileme party is going [08:22] o [08:23] ^_^ [08:40] Hi, [08:40] Psst. [08:40] I've talked to people. [08:40] Inside people. [08:41] The LoC tweet archive? [08:41] I wouldn't buy a lot of stock in it. [08:41] Wouldn't bet the farm, you know. [08:43] Wouldn't bet the farm, you know.Juuuust sayin' [08:44] figures [08:47] Yeah. [08:47] So do whatever we can [08:47] Oh, that new wikimedia project. Daddy likes. [09:02] FUCKING FINALLY. [09:02] URLTE.AM IS BACK [09:02] SketchCow: thank you! [09:02] oh, not me. [09:02] Now, let's not have THAT happen again. [09:02] What are you up to, radii? [09:03] I've been saying "Free the edit stream!" to every wikipedian I come across for years, so when I saw the WM Commons project I figured I should get online [09:03] Ah, OK. [09:03] Welcome aboard [09:03] URLTE.AM finally back. [09:04] but alas I have a timezone and dayjob challenge, so I'm just going to idle here for the next long while. [09:04] We're exactly like Wikipedia is in how we run things [09:04] what, no, deletionists are not welcome here. [09:05] but with less drama, less deletes and a lot more rape [09:05] no deletes. [09:05] no is less! [09:05] * chronomex shakes head [09:08] * radii deletes my awakeness [09:08] * ersi injects a coffee machine into radii [09:08] ersi: in 8 hours I will agree with that. [09:11] http://web.me.com/stressakustik [09:11] awesome page [09:20] Tomorrow, we see about the wiki. [09:21] Oh, I will love THAT overlap. [09:38] damn.. while working 12 hour days kicks ass on payday, the rest of the time it sucks [09:48] dcmorton: I was just reading a blog post about that: http://thecodist.com/article/why_i_don_39_t_do_unpaid_overtime_and_neither_should_you [10:00] worth a read: http://www.stevepavlina.com/blog/2006/07/10-reasons-you-should-never-get-a-job/ [10:08] http://qaa.ath.cx/LoseThos.html [10:09] ^ i thought the post was jumping the shark, but if you take the time to watch the videos and look at his twitter stream, I think the other's quite correct [10:14] joepie91: that post is a bit rediculous in parts [10:14] kennethre: it's hyperbole, obviously :P [10:15] joepie91: this guy must have had a really soul-crushing gig :) [10:15] nah [10:15] imo hyperbole is necessary there [10:15] for the masses I'd agree [10:15] it throws people out of their normal line of thinking and makes them reconsider :) [10:16] yes, exactly [10:16] those that don't need it will recognize it as being hyperbole, usually :) [10:16] I used to do hyperbole posts [10:16] I feel like I grew out of it though [10:16] also, lookie at http://www.ramhost.us/?page=status-core [10:16] if you didn't read #memac yet [10:16] :P [10:16] the bottom two graphs [10:16] geeze [10:16] haha [10:16] * joepie91 enjoys seeing those graphs pealk [10:16] peak* [10:16] I just briefly hit 156mbit total [10:17] before that I was on 130mbit total for a while [10:17] 60mbit down and 70mbit up or so [10:17] I do have the idea I'm going to be running out of bandwidth very soon though.. ._. [10:17] or traffic*, rather [10:17] maybe they only limit incoming traffic? [10:18] 88.211 GB of 600 GB (14.7%) [10:18] oops [10:18] :x [10:18] that was on 3GB a few hours ago [10:18] per box or per account? [10:18] that's that one server [10:19] I have 5 other servers, but I can afford for this one to be cut off due to bandwidth exceeding [10:19] kill it and start a new one when you're doing :) [10:19] *you're done [10:19] :P [10:19] it's just a storage server [10:19] oh man [10:19] doesn't really do much [10:19] i think i used to use these guys [10:19] I did! [10:19] ramhost? [10:19] they're <3 [10:19] no they were horrible [10:19] worst support ever [10:19] you sure you're talking about ramhost? [10:19] pretty sure [10:20] i don't know of any others that offer both openvz and kvm [10:20] buyvm does [10:20] though it was openvz and xen at the time [10:20] and so does some .nz provider [10:20] mabye it was someone else [10:20] nah [10:20] ramhost never did xen [10:20] hmm [10:20] * kennethre searches the logs [10:20] ah [10:20] vpslink.com [10:20] so, so terrible [10:21] ramhost always scores really high on support, the one issue some people have with them is that they're rather straightforward and to the point [10:21] :P [10:21] and some people experience that as 'unfriendly' [10:21] mm [10:21]