[00:01:19] I guess we'll get either very large values or very small values with few observations. [00:01:42] Intuitively, it feels like it is too aggressive to drive towards confidence, but I'm not sure. [00:03:00] halfak: choice of prior plays a bigger role with fewer observations [00:03:33] halfak: are you using Uniform(0,1) for p(desirable) and p(undesirable)? [00:04:10] Nope. We used .80/.20 which matches the real data, but it behaves like a single observation. [00:04:22] We could have boosted that so it would have normalized. [00:16:08] halfak: out of curiosity, what's p(score = 0.25|desirable) and p(score = 0.25|undesirable)? what are those values for score = 0.75? [01:00:48] bearloga, what? [01:00:55] Oh! Looks like I missed it. [01:01:12] Oh! I can answer that. :) [01:01:17] Later, I guess. [01:01:28] Anyway, I'm off for the night. Have a good one, folks! [16:17:33] o/ [16:19:33] o/ [16:41:23] o/ [16:41:26] aww [16:41:33] I need a shorter name to complete the visual pattern [16:41:42] harej: we need you! [16:42:04] :D [16:42:19] o/ [17:18:03] halfak: Did you manage to rerun mwcites by any chance? [17:19:22] tarrow, yes, but it's still going, regretfully. Either, the server is has been loaded with something else (I ran this at low priority) or my estimate was off. [17:19:55] I don't think it would have been off by more than 24 hours though. I have run this job before after all :) [17:20:37] halfak: cool, no worries! It isn't super urgent. I'll have a look again tomorrow :) [17:20:53] OK. Thanks for your patience :) [17:21:19] Hehe, you're the one doing me the favour... [17:28:22] sooo. [17:28:30] There's a major comcast outage in Minneapolis [17:28:38] But my phone can tether :) [18:34:49] It is 1234 o-clock in CST [18:34:50] :D [18:35:05] * YuviPanda vaguely wakes up and feels terrible [18:35:31] halfak: 12:34:56? [18:35:36] :D [18:36:04] in the year 789 [18:36:54] 7th day of the 8th month of stardate 91011 [18:37:25] 'cause stardates probably have months [18:37:28] :D [18:37:36] depends on which startdate [18:37:50] TOS continues in being shitty in having no set setup and people just made it up [18:38:01] TOS? [18:38:06] The Original Series [18:38:17] * Emufarmers arranges a star-date for halfak and YuviPanda. [18:38:23] lol [18:38:38] I'm not sure consistency was high on the writers' minds [18:38:47] halfak: http://trekguide.com/Stardates.htm is one of my favorite pages [18:40:08] lolwhy [18:40:18] This should be posted on r/Theinternetisbeautiful [18:40:41] https://www.reddit.com/r/InternetIsBeautiful [18:40:48] No "The" [18:41:09] I had the original TNG writer's manual [18:41:13] unfortunately it's in Chennai :( [18:41:35] I was going to gift it to a friend but then he got into an arranged marriage with a random person and so I stopped talking to him instead [18:41:38] oh well [18:59:43] * halfak uses Snuggle to find example of Teahouse invitations [18:59:45] mwahahaha [19:12:51] @seen leila [19:13:40] lzia: Hey [19:13:50] I have some stuff to talk [19:13:59] tell me when you can [19:14:31] hi Amir1. Shall we schedule something for tomorrow? [19:15:01] sure thing :) [19:15:07] thanks [20:12:09] J-Mo, FYI: https://github.com/halfak/teahouse_experiment/commit/e4d79f9b4152255046000515523ff3dc0596b669 [20:13:26] Just waiting on server time to kick it off. [20:13:35] Once my current run of mwcites is done, I'll start it up. [20:14:16] watched and starred, halfak! Add jtmorgan as a collaborator? I may have scripts to commit (I could commit copies of the current versions of the Teahouse invite scripts, or at least a README with links to same). What do you think? [20:14:57] Whatever you do, do it BOLDly :) [20:15:00] {{done}} [20:15:13] Boldly Go [20:15:14] We can always refactor later :) [20:15:21] perfect [20:24:10] * halfak finally gets some time for productivity analysis! [20:24:11] WOOOOO [20:34:30] halfak: Productively analyzing productivity? [20:34:46] Hopefully ;)