[00:12:23] halfak: do you have time to write some conclusions? [00:12:27] what else needed [00:12:44] Also I'm waiting for DarTar to take a pass on this too [00:15:23] Yeah. I think we're close to read. [00:15:25] *ready [00:15:32] I'm still working thought the related works section. [00:15:59] OK. I do some edits in limations [00:21:23] Just finished a pass in related works. [00:21:37] Still need to prose-ify the wikidata quality and below. [00:21:41] I'll leave that to you. [00:21:49] I'm working on the evaluation section in methods. [00:22:12] BTW, I have a hard stop in 40 mins. [00:22:22] Then I pull DarTar in for a pass and let you handle the rest. [00:22:30] * halfak puts as much into this 40 mins as he can. [00:23:33] ok [00:23:35] thanks [00:36:01] halfak: I finished prose-ifing but please have a pass on it [00:37:41] OK. Just finishing conclusiion. [00:37:52] amazing [00:44:40] Working on related work [00:44:58] thanks [00:45:07] I do an editing pass on the conclusion [00:46:54] "As we showed it doesn't apply to Wikidata. The difference in the concept of classification in their project and ours can be the reason." [00:46:56] What? [00:49:13] the Tan were into correctness of edits [00:49:21] while we are looking for vandalism [00:50:22] they showed user reputation in making correct/incorrect edit doesn't determine future of corretness of edits [00:51:19] Oh! Might need some word smithing. [00:51:32] you are the master [00:51:36] of word smithing [00:51:37] Make sure to make a clear statement about how correctness/incorrectness != vandalism/non-vabndalism [00:51:48] Just about to save some edits. [00:51:53] TO related works [00:52:03] awesome [00:53:12] if you have some time, do an editing pass overall [00:53:27] just fixing typos, etc. [00:54:38] if you don't have time, just take a pass on features table [00:56:17] Regretfully, my time is up. [00:56:34] BUT, I'll come back to take another pass tonight (CST) [00:56:37] * halfak checks calendar [00:56:58] I think I can do a pass at 0400 UTC. [00:57:08] I'll send you updates via telegram. [00:57:16] Now, I'm going to ping Dario to have him take a full pass. [00:57:40] awesome thanks [00:58:22] still "pick_a_vandalism_paper" [00:59:34] Oh yeah. Most of them start out with a "7% of edits to Wikipedia are vandalism" [00:59:40] Just pick one that does and cite that there [00:59:49] sure [01:01:27] OK. DarTar is going to take his pass soon. I told him to edit boldly and leave inline or talk-page notes. [01:01:42] I also told him he can find you in #wikimedia-research [01:02:35] OK [03:02:38] Hello all. [03:03:01] YuviPanda ^ [03:04:30] Will there be sth with the paws? Last time we chatted you mentioned "it's down for a few days" but now it's almost a week that passed. [03:04:46] yeah, unfortunately I got pulled away on vacation :( [03:04:50] so it'll be down until next week [03:05:00] Phew. [03:05:33] I managed to solve this intermitently but I don't like the soltion :/ [03:05:38] yeah [03:05:43] PAWS should be back solidly sometime next week [03:05:45] the timing was bad [03:06:28] Having a virtual machine is eating up the memory and I need it since it's not just me who uses the machine. [03:09:18] Do you have suggestion on which distro to obtain that will have a minimal memory and hard disk footprint - I will run only the Jupyter and a simple terminal? I guess sth with LXDE or the likes. [03:11:16] Besides this new Plasma 5 is buggy - I had troubles when changin the SDDM. [03:12:31] Anyway feel free to answer whenever possible, I'll be checking out the chatlogs. Just ping my nick so I'd know what to search for. [03:13:00] Gtg now. Bye [03:14:21] pipvioj: I don't think your choice of distros matter, since they are just installable via pip [05:01:57] o/ amir1 [05:02:13] hey halfak [05:02:23] How's it going? [05:02:25] I'm adding figures and list of features [05:02:33] Everything else is done [05:02:35] How's our length? [05:02:44] 6.3 [05:03:03] Woah! Cool. They might give us a hard time for having a short paper! [05:03:28] by adding figures and features we would have close to ten [05:03:39] Oh! perfect then :) [05:03:50] halfak: I send you the tex file [05:04:02] Is it up on github? [05:04:15] can you help me adding figures while I work on the features? [05:04:19] no [05:04:24] it's not ready yet [05:04:41] It's OK. Will be easier to work together. OK if I put it on github? [05:05:58] sure [05:06:00] why not [05:06:59] halfak: Also Dario is working on the methods section afterwards, I will apply his changes once he is done. We have no time for that [05:07:11] *for waiting until he finishes [05:07:18] we have less then 3 hours [05:07:27] halfak: sent [05:18:03] halfak: what do you think? [05:18:29] Still getting it organized. What, exactly do you want me to do right now? [05:18:56] Amir1, ^ [05:20:00] halfak: add figures [05:20:22] I'm trying to work on the features list [05:20:35] OK. i think i understand. [05:20:48] How have you been compiling? PDFlatex? [05:22:20] yes [05:31:55] halfak: I'm done with features now [05:32:03] I want to share it with you [05:33:43] OK. Will put in the doc. [05:33:58] Am working with DarTar to help him understand one of the feature definitions. [05:34:13] awesome [05:42:59] OK. figures in. Working on referencing them. [05:43:04] Did you send feature list? [05:43:13] no [05:43:24] I paste it somewhere now [05:43:35] OK. Can send what I have to you in 5 min. [05:43:49] https://www.irccloud.com/pastebin/KtUSUoRy/ [05:44:04] halfak: ^ just put them in Features section [05:52:08] Amir1, committed to the sample_subsets branch we have been working from [05:52:15] gitpull and look in the paper directory [05:53:08] kk [05:53:59] Amir1, DarTar just told me he hacked the feature descriptions heavily in the wiki [05:54:45] it's okay. I change it to list using two regexes [05:54:51] halfak: ^ [05:55:03] Great :) [05:55:54] Make sure you review the proportion of qid changed [05:56:16] sure [05:57:09] halfak: can you email me the tex file? the git pull takes long time and raw content in github is blocked now(!) [05:57:37] OK. [05:58:55] from DarTar: "in the table with the fitness of the model, the features group names don't match those in the feature description table" [05:59:01] Seems likely. I took some liberty [05:59:07] Change to your judgement [06:00:07] in the tex file or meta? [06:00:25] meta [06:00:38] ok [06:01:45] Probably the tex file from now on. [06:01:52] These liberties, I took a long time ago [06:01:57] When building the fitness tables. [06:03:03] it's okay [06:09:34] I applied Dario's hacks on features [06:09:40] Still other parts needed [06:21:04] halfak: done [06:21:16] All of DarTar CEs applied [06:22:00] only one thing left: the link to scikit learn doesn't stay at the proper place because it doesn't have space [06:23:01] also it would be great if you review the pdf [06:23:57] halfak: just sent the tex and pdf [06:24:18] I have only 1:30 min. [06:24:34] http://http://scikit-learn.org/ [06:24:36] http://scikit-learn.org/ [06:24:38] Wooops [06:25:38] yup, everything is done now [06:42:10] OK. Sent the tex back [06:42:15] That's all I've got for tonight. [06:44:00] I was applying Dario's the whole time [06:44:03] :((((((((( [06:44:16] halfak: can you say what was your changes [06:44:39] Oh! didn't you just send me the tex for review? [06:44:46] I'll make a diff [06:44:53] also references to figures doesn't work, it references to sections not figures [06:45:10] thanks [06:47:29] diff http://pastebin.com/tfev0Wxz [06:48:15] Oh! move the \label{} after the \caption{} in the figures! [06:48:18] Amir1, ^ [06:48:22] That should fix the references [06:48:29] thanks [06:57:29] Dario told me that he left some notes in his edit comments [06:58:33] Nevermind. Looks like those items have been covered. [07:00:14] cool [07:00:23] the only thing left is applying your changes [07:02:19] Sorry that the diff isn't great [07:03:09] I used wdiff [07:03:18] it's awesome [07:09:18] wdiff: http://pastebin.com/kcyhk2zk [07:13:06] I applied everything [07:13:25] halfak: the note you added to the footnote make it really ugly [07:13:36] for scikit learn [07:14:44] Happy to have you change it back :) [07:23:29] http://www.kdd.org/kdd2016/calls [07:23:55] halfak: ^ which one "Deployed", "Discovery" or the other one [07:24:54] Deployed [07:26:12] As I thought [07:33:57] I think I submitted it [07:34:04] \o/ [07:34:10] in last thirty minutes [07:34:27] there is an interesting question here halfak: it asks if the first author is a full time student. I said yes [07:34:41] +1 [07:34:46] it might be helpful in review or vice versa [07:34:56] Maybe. It's the truth. [07:35:14] ok [07:35:35] I haven't slept for ~30 hours and worked all the time [07:35:52] I'm going to sleep for * at least * 24 hours [07:36:19] Damn it! I told you to sleep :P [07:36:21] Enjoy it! [07:36:26] A job well done :) [07:36:33] See you when you're rested [07:36:35] o/ [07:36:50] o/ [07:36:53] you too [07:37:01] I think it's pretty late back there [08:42:07] hello all [08:43:59] YuviPanda, I installed a Linux virtual machine in Win7 and in it cython,scipy, etc. so IUC it does matter which distro I install as a VM, yes? [08:44:07] Gtg now, bye [09:32:18] pipivoj nope, doesn't matter :) I would recommend ubuntu or debian :) [13:35:37] Hi all [13:36:39] I c niw that I wasn't clarify enough [13:37:27] I'm running a win7 box snd in/on it a VM [13:38:24] So I think it dies mstter whuch distro to install ss vm [13:39:16] So methinks it do matter which distro to have as vm [13:39:57] YuviPanda, am I wrong? [13:40:17] Going now, bye [15:01:49] o/ [15:18:21] hello halfak|Mobile [15:18:28] no meeting today? [15:19:11] Hey. Late night last night submitting kdd paper. I hoped in irc to see if anyone would show up [15:19:24] Can got in call of you give my 5 min [15:19:35] no worries [15:19:39] I can wait if you want me to [15:20:13] Na. Could do irc quick. I think the main sync questing I have is re the report and any new campaigns. [15:20:56] not much progress for those, I intend to have some progress today [15:21:04] I'm a bit behind on getting arwiki's reverted model deployed [15:21:06] MY talk request entered the top 50 of 42 [15:21:37] Gotcha. Amir1's did too. Mine was rejected. [15:22:22] Did you see the wikimania-l thread? [15:22:38] *wikimedia-l [15:24:12] not quite [15:25:28] I recommend it. Good read. Made me feel better. My reviews were super low quality. Turns out, it was the same for everyone. [15:26:08] A reviewer told Lydia she didn't know anything about "Wiki data" [15:26:21] hmm [15:26:26] I got positive reviews [15:26:35] they are saying I can present if a few people dont show up [15:29:17] halfak|Mobile we ought to think of an alternate method of presentaiton [15:29:19] Yeah. My reviews were generally positive except the one that just said "6" and nothing else [15:29:22] those alternate tracks [15:29:43] Meh. I think I'm not going to wikimania this year [15:29:49] aw [15:29:57] You should. :) [15:30:05] But the hackathon, I'll be there [15:30:17] I am unsure if I can get to the hackathon in israel :/ [15:30:41] Yeah. I don't like the political situation this sets up. [15:30:50] Wish it was in Austria [15:31:02] That was the other big bid [15:31:30] I dont care too much of the politics, its visas and all. It tends to make me travel to a different country just to apply for countries like US and UK [15:31:32] Nothing against Israel, but a lot if people can't travel there [15:32:06] it would cost me around 250-300 euros just to get to paris and back probably [15:32:26] given one day of hotel stay is probably needed too since you need to queue up early [15:32:49] anyways [15:33:19] I would encourage for wikimania. it would probably be easier for amir to get there too [15:38:03] There's a limit to the number who can attend from the WMF. Even if I tried to go, it would be unlikely. And I have a conflict with something to do in Minnesota, so I think I'll just do that. [15:38:44] this is too bad [15:38:59] maybe I wont attend either :/ [15:39:24] I even got an italian plug :p [15:46:01] o/ amir1 [15:46:16] Hey, halfak|Mobile [15:46:31] I just woke up and realized we have a meeting [15:46:52] I'm at home but I can be at university in ten min. then we start the meeting [15:46:58] halfak|Mobile: ^ [15:47:27] Just syncing via irc today [15:47:39] Long night last night 😀 [15:48:07] Talked to ToAruShiroiNeko about wikimania and the report [15:48:36] My next work is going to be with setting up arwiki's reverted model [15:49:07] ya [15:49:14] One thing I'm still troubled by is switch away from the svc models [15:49:40] Higher fitness, but we're going to need our toll devs to set new thresholds [15:49:47] Tool! [15:50:02] It's going to be a trick to communicate that. [15:50:05] it's ok [15:50:13] we can fix them easily [15:50:21] if they didn't respond in time [15:53:18] halfak|Mobile: we are good to deploy the extension. Only Ops needed [15:53:41] as I told you. We merged one of the biggest patches last night [15:54:43] Gotcha. I'll try to get ops to comment this coming week. Yuvi has been out of town [15:58:52] awesome. I think that is a huge victory [16:02:26] halfak|Mobile: what should we do for the hack session? [16:03:23] Honestly, I'm looking for a day off. I can stay on irc if anyone needs me though. [16:04:02] maybe I'm still at the runners' high [16:04:05] :D [16:11:48] Amir1: something for now it later... We should move the current version if the outside back onto the wiki so that we can continue work on it openly. [16:12:03] * now or later. No rush [16:12:45] oh :( [16:17:18] hah you are feeling like me last week :p