[04:02] Ymgve: yes. dreamhost sucks [04:27] ugh. I hate spring [04:27] we need to get rid of pollen [05:18] I do agreGoogle is of course right. [05:19] kjdlkjdfg [05:25] Well, Balrog, if you were trying to make sure I was aware Kryoflux is a "problem", you've done it. [05:25] ok... [05:26] just so you know, the past week, which was supposed to be my spring break, turned out busier than most [05:26] so I'm rather tired in any case [05:27] and my major interest is hard drives. and disk packs. [05:27] reading those things is gonna be fun. [05:28] these are things similar enough, and in some cases identical, to the Cray 1 disk pack that was dumped [05:32] Excellent. [05:33] That whole thing went well too, once I jumped out of it. [05:33] well the difference with these is that the on-disk format is at least well documented [05:33] DEC was extremely good at that part. [05:34] among other things, we're currently working on RLL decoding for hard drives [05:34] SketchCow: how far did the on-disk-format decoding go? [05:34] The cray thing? They got very far. [05:34] A friend of mine still thinks that one of the tracks wasn't fully dumped but I have to ask him for details [05:35] did they manage to decode the filesystem? [05:35] yes, the cray thing [05:35] Chris would have to say [05:35] the hardware that Cray used was relatively common [05:35] we have to have at least a dozen such disk packs [05:35] the big 4 or 5 platter type [05:36] http://chrisfenton.com/cray-1-digital-archeology/ [05:36] I saw all that :) [05:36] I really liked how he used a custom FPGA-based device to sample it, and his own stepper [05:37] that gave me the idea, that it might be possible to make a servowriter for the RL-01/02 drives (though that's not preservation related) [05:37] anyway he does say "Head_4_data.zip (It looks like the first 20 or so tracks are missing. These may be recaptured later.)" [05:37] was this ever done? [05:39] anyway my main interest in the disk packs we have here though, is that they may contain ancient historical software and information. [05:40] Many of them are original DEC software packages. Many are backups from years and years ago. [05:41] where is "here" [05:41] a university in philadelphia pa [05:42] awesome. [05:42] dnova: are you nearby? [05:42] I'm in buffalo so not really [05:42] ahh :/ [05:42] I wouldn't be surprised if this is one of the larger DEC hoards in the area [05:42] would love to se that [05:43] a little hardware, mostly software and docs. [05:43] and when it comes to docs … it's a little ridiculous [05:43] I went out and got a scanner specifically for that but I'm afraid even that's gonna be tough [05:43] :/ [05:44] at least I do have a hard drive now so I can dump stuff as I work on it [05:44] SketchCow: when it comes to uploading archived data, what is the best way? [05:44] yum, data helicopters [05:44] raw scans at 40mb per page may be a problem :< [05:44] hooray, spinning rust [05:45] balrog: lossless (deflated) tiffs help a lot [05:45] these ARE deflated. [05:45] down to 40-60% of original usually [05:45] o [05:45] the reason for the size is my insistence on using 600dpi for technical drawings to preserve edge detail [05:45] wander over to your local uni library reading room and plug in? [05:45] aye [05:45] usually I postprocess with -levels and then knock down to 4bit which cuts sizes to more like 4mb per page [05:46] but bleh manual labor [05:46] I'm scanning 300dpi/256gray [05:46] what kind of documents? [05:46] for text, that's fine [05:46] for technical drawings with fine line-art detail, not so good [05:47] text, halftoned photos ( :( ), ok drawings [05:47] halftone is fun [05:47] my scanner though, it's able to scan halftone without moire pattenrs [05:47] patterns* [05:47] most of my stuff is 2gen photoscopy anyway [05:47] it's also got an awesome depth of field [05:47] hmmm [05:47] stick a circuit board on the glass, scan, nearly completely in focus [05:47] I also have 20 pallets of paper total [05:48] tried that with a cheap scanner. might as well use a camera there [05:48] yeah [05:48] paper to scan? [05:48] mhm [05:48] well, I don't quite have that much I don't think, but it's gotta be close; [05:48] and most of it is 11x17" [05:48] heh [05:48] have a sheetfeeder? [05:48] of course. [05:48] phew [05:48] what kind? [05:48] this is an Epson GT-15000 with feeder [05:48] got it for a steal on ebay [05:49] the only kind that the 15000 has [05:49] a large commercial-quality belt-feed [05:49] nice [05:49] I only hbave rollers [05:49] wtf why am I on irc [05:49] still, an 11x17" scan takes 90sec [05:49] why, late? [05:50] im drinkikng on the sidewalk at austin tx [05:50] lolol [05:50] should ogle the hipster boys or something [05:50] haha [05:51] SketchCow: if you have any input in regards to uploading, I'd appreciate it. I can't afford to send my own hard drives out (even temporarily) atm :-( [05:51] lame :\ [05:51] this 3TB cost me about $240 [05:51] :< [05:51] or was it $190 [05:51] still annoyingly expensive [05:53] I do have access to 100Mbit internet [05:53] still, 40mb per page is a lot even with that [05:55] I have some 40mb QIC tapes; you could put one page on each tape (this will take around an hour) and then mail them to SketchCow who will then copy the images off and upload to them to the archive. [05:59] hahahahaha [05:59] that reminds me [05:59] we have stuff on DDS tapes too [05:59] at least for that, the drives are readily available [05:59] dectape and dectape II and dec magtape and the other magtape formats… I sorta don't want to think about it [06:00] :< [06:13] In case anyone cares, these nicks: goldielox (<- in here right now) EstaTiC npt_ carl_ faalhaas are most most likely bots of some kind [06:14] balrog: Raiding Drexel? [06:14] temple [06:14] :[p [06:14] :p * [06:14] They idle in a lot of very random channels, which rotate a lot [06:14] shaqfu: you're nearby? [06:14] I'm surprised; I would've figured the engineering school would've had it [06:14] balrog: Jersey, yeah [06:14] not engineering [06:14] comp sci [06:14] Never say anything, never answer when highlighted [06:14] True [06:14] I didn't even know Temple had a CS program [06:15] they have CS and IST and ECE [06:15] among others [06:15] back in the day they have a ton of DEC hardware [06:15] almost unbelievable [06:16] Yeah [06:16] and since the sysadmin hates trashing stuff… [06:17] Just stuff they had sitting in the sysadmin's office, or old manuals out of their library? [06:18] uhm what do you mean? [06:18] I like your sysadmin. [06:18] balrog: Where were they? [06:18] shaqfu: a few hundred disk packs with all kinds of stuff. [06:18] Ah, gotcha [06:18] throughout the comp sci department [06:18] shaqfu: a room full of documentation, packed tight [06:18] those disk packs are very valuable to collectors [06:19] maybe it's best if you keep that information from the sysadmin [06:19] dnova: what do you mean? [06:19] what I want to do is preserve them properly [06:19] people pay a lot of money for those. [06:19] I'm not looking for money. [06:19] and he isn't either. [06:19] right [06:19] nevermind. [06:19] now would I sell them, if I'm sure they're dumped right? /maybe/ [06:20] but that's besides the point [06:20] anyway [06:21] the vast majority of the disk packs are RK-05, RL-01, and RL-02. [06:21] I'm pretty sure there are some RK-06 as well, and there are also the larger multiplatter disk-packs — RM-03, RM-05 [06:21] R-05 is similar to that Cray disk pack. [06:21] there are also non-DEC branded disk packs, including CDC ones [06:22] RM-05** [06:22] how many are there in total [06:23] disks? [06:23] packs [06:23] that's what I meand [06:24] MANY RK-05 (probably well over a hundred), somewhere around 50 RL-01, around 20 or so RL-02, and several of the larger ones and other types [06:24] I wanted to try to get the easy stuff first [06:24] but none of it is easy, lol [06:24] Correction with regard to my list above, npt_ is NOT a bot [06:25] balrog: that collection could be worth around $5,000 as a quick sale in a lot. [06:25] dnova: at least [06:25] NOT that I am saying you should sell it. [06:25] just, ehm... call me if you plan to sell. [06:25] that's not counting the nearly intact straight-8 [06:25] which I want to fully restore [06:26] in its heyday, there were three VAX 11/780s, a VAX 11/785, two 11/750s, and a variety of PDP-8s, 9s, 10s, 11s, 12s, and 15s [06:27] and even at least one LINC-8 [06:27] so chances are high there's software for some of the rarer ones in here. [06:31] anyway as I said, I'm not interested in selling. [06:31] if I do sell things, it will be piece by piece, after data is dumped/preserved [06:31] dnova: ^ [06:31] yes [06:31] call me when that happens [06:32] however, I would rather donate to CHM, whatever they want. [06:33] well I mean, just don't be surprised when me and my mercenaries hijack the truck full of your donation [06:33] I'll send along a portion of it to the CHM for you though. I'm not entirely heartless. [06:34] I'd prefer to deal with Al directly first [06:35] I know they're in need of certain particular things [13:43] Man, SXSW.... [13:43] I tell you, there's no less than 24 tracks. TWENTY FOUR SIMULTANEOUS TALKS AT ONCE [13:43] Do they record them for history? Maybe. [13:50] I spent some time going over the numbers. [13:50] Seriously, the answer is hard drives, not tape. [13:57] How often is these SXSWXESES events really? [13:57] Feels like I hear about them aaaall the time [14:03] Does the music festival happen in parallel also? [14:10] Basically, Friday (9th) through about next Tuesday is SXSWi (Interactive) [14:11] At the same time, overlapping, is film, both films being screened as well as film panels. [14:11] With some going as far as thursday, friday [14:11] Next Thursday, music begins, that goes for another week. [14:11] Now, to give you a size of this clusterfuck. [14:11] Interactive has 2,661 panelists. PANELISTS. [14:43] Hosting for new archiveteam.org paid up. [14:45] By no means READY yet, but up. [14:48] A medium sized step ^^_ [14:49] Shout if you need some help migrating that [14:49] I will DEFINITELY need help migrating it. [14:49] Mr joggle five thousand tasks a day [14:50] one knows you're up to a lot of stuff when you don't got time to write in here haha [14:51] It's an excellent indicator. [14:52] Hear hear [16:27] OK, the archive team DB has been dumped. [16:54] Jason, I'm interested in building a new more-featured downloader for MobileMe in a language that isn't Effing Bash [16:55] I know that's not where your head is at right this minute [16:55] but get back to me when it's a good time for you [17:00] hybernaut: Which features does it miss? [17:00] (Other than 'not being bash'. :) [17:01] my rsync gets interrupted, and I have to rerun it manually, then notify the tracker manually [17:01] so I've been hacking away at it to do things like resume [17:01] specify download and upload bw limits [17:02] minor stuff like that [17:02] so of course, I don't mean "your code is shit because I prefer some other language" [17:02] that would be unintentionally rude 8^) [17:02] Heh. [17:04] Doesn't the upload script already do resuming? [17:04] I guess mostly the tool's used on a server, and I'm trying to use it on my laptop [17:05] the seesaw script doesn't seem to resume well--maybe I should read the uploader more closely [17:06] Maybe it's better not to use the seesaw script, then, on your laptop. You can also run dld-client and upload-finished. [17:06] yea, that would be much smarter, thanks [17:06] ok, can I start over? [17:06] If you hack the download script a little, it should be possible to set the wget bandwith limits. [17:07] Hey guys, I'm not sure I'm doing this right--what's the best way to run this on my laptop? [17:07] Start over? I think you can just continue. [17:07] I'd run one or more ./dld-client.sh 's. [17:07] I looked at the scripts a little and I know a little, but I should ask for help instead of deciding to rewrite the whole thing! [17:07] (I meant restart the conversation) [17:07] Ah, okay. :) [17:08] IRC amplifies my natural stupidity [17:08] Rewriting something that already exists isn't always the most efficient solution. There may be other projects worth working on. [17:08] :) [17:09] If you edit the dld-me-com.sh script you can add --limit-rate=RATE to the various wget calls. [17:09] yea, I have done that [17:09] thanks [17:10] And then run upload-finished.sh, in a loop, if you want. [17:11] If you're looking for a programming project to work on: Google Knol needs saving. [17:12] (But it's really annoying to download.) [17:12] is there a project started yet? [17:12] #knol [17:12] #klol [17:12] sorry [17:13] http://www.archiveteam.org/index.php?title=Knol [17:14] thx [17:15] There's no download script yet. I've started a little bit, a while ago, but it's really, really annoying. [17:15] Archiveteam Wiki db dumped, now imported into new DB [17:15] hybernaut: Another project, there may be interesting public documents on iwork.com. [17:16] I saw the iWork closing announcement yesterday [17:24] Research it, hybernaut [17:24] This is your chance to shine [17:24] * SketchCow throws a knife between hybernaut and alard [17:24] Make it fast [17:25] SketchCow: which one? [17:25] either [17:29] Looks like there's nothing public on iWork.com that we could fetch for it's users [17:29] apparently this dsl modem/router is terrible [17:29] and the announcement says that a lot of the data is now in iCloud or have been for a while [17:30] lies [17:30] I'm running four copies of the fortune city downloader and now it drops connections a lot [17:32] chronomex: perhaps, but I don't find any public facing iWork data [17:32] mmm [17:32] https://encrypted.google.com/search?hl=en&q=site%3Apublic.iwork.com [17:32] sucko [17:32] https://encrypted.google.com/search?hl=en&q=site%3Apublic.iwork.com#hl=en&sclient=psy-ab&q=http:%2F%2Fpublic.iwork.com&oq=http:%2F%2Fpublic.iwork.com&aq=f&aqi=g-l1g-vC1&aql=&gs_sm=3&gs_upl=5925l6776l0l6992l7l7l0l0l0l2l141l570l5.2l7l0&gs_l=serp.3..0i13j0i15i33.5925l6776l0l6993l7l7l0l0l0l2l141l570l5j2l7l0&pbx=1&bav=on.2,or.r_gc.r_pw.r_qf.,cf.osb&fp=6cfeb4bc8f72d1c6&biw=1287&bih=1034 [17:32] Hm! [17:33] It will probably take a bit of work to track down the links, though. [17:34] And anyway, it may be even fun to say "we saved all of the public iWork data, here it is, 100MB". [17:34] :) [17:37] that would be funny [17:39] Now to play "how do I get mediawiki on the new shared server" [17:45] hah, mediawiki. fun. [18:07] I have yet to find a iWork public doc that also exists on iCloud--anyone found one yet? [18:35] db48x: Is there a wiki page about fortunecity? [18:38] db48x: Is there a wiki page about fortunecity? [18:38] soultcer: yes [18:38] http://archiveteam.org/index.php?title=FortuneCity [18:39] Now I feel stupid for not finding that. When you type Fortune in the search bar, it only finds some twitter stuff [18:39] yea :( [18:39] tracker is at http://focity.heroku.com/, and the irc channel is #fortuneshitty [18:45] hey guys [18:45] i found out a polish file hosting site, przeklej.pl, is turning off in 10 days (March 20th 2012) [18:45] http://przeklej.pl/ [18:45] sadly it doesn't even support public files anymore [20:39] Good news: The wiki is up [20:41] it is :) [20:42] I mean the new wiki [20:42] The no-spyware special. [20:43] oh? [20:43] hah [20:44] wow, how this happened? dreamhost? [20:46] I dumped dreamhost. [20:47] The dumping is taking a little while, for something like that. [20:53] Some nicks in here are some form of bot [20:54] They join channels, idle and never respond when you talk to them [20:54] Different IP addresses/hosts, but same client [20:54] Exactly: bitchx-75p2: : osmosis : Keep it to yourself! [20:54] Not sure if they logging bots, evil bots or something else [20:55] But in another channel I'm in I've taken the habit of banning them on sight [20:56] Some of the nicks used for this: goldielox EstaTiC carl_ faalhaas [20:59] Oh and another thing, these nicks cycle through exactly 9 channels. Part from one channel and immediately join a different one. [21:50] nitro2k01: meh, care if they're bots or not [21:50] This is all logged and archived anyway [22:44] was kind of miserable and snowy on friday [22:44] tomrrow 54, wednesday 66 [22:44] shit's whack [23:28] Hehe, we filled up fos [23:28] SketchCow: ^ [23:30] Hurrah! [23:32] looks like fos if full [23:32] is [23:32] Yupp, it is [23:33] must not have had a lot of space, if less than 280 GB filled it [23:33] Well, it does get bits of mobileme. [23:36] db48x: less than 200GB? You're not alone on fos [23:43] ersi: yea, yea :)