This is the website of Abulsme Noibatno Itramne (also known as Sam Minter). Posts here are rare these days. For current stuff, follow me on Mastodon

Categories

Calendar

@abulsme tweets from 2014-05-10 (UTC)

  • Alex saw a rainbow and spontaneously broke into song… What a Wonderful World. Sang the whole song. Made my day. Thanks Alex! 02:16:56
  • I will let YOU have that particular fun. :-) RT @BYUfan: @abulsme that would be so much fun! 02:49:02
  • Reading – I've got a little list (@intelwire) http://t.co/2G05SRcC0i 03:21:04
  • Let it go Bruce! Let it go! RT @BYUfan: The movie tonight is Frozen. It seems like I'm the last person in the world to see it. 03:24:35
  • Reading – U.S. Officers Kill Armed Civilians in Yemen Capital (Eric Schmitt) http://t.co/OTK7gC1aJw 04:07:45
  • Breitbart mentions my dad -> Hillary Clinton Protested Designation of Boko Haram as Terrorist Group (Warner Huston) http://t.co/fDiLmZcKbu 05:34:28
  • Made one correction and added five polls I had missed thanks to data from @bonncaruso. Thanks! http://t.co/qTS8nbucG3 06:08:35
  • Reading – How to make a simple electronic musical instrument (@Frauenfelder) http://t.co/UZrCCx00Ub 06:40:17
  • Another bug fix thanks to @bonncaruso. When I flipped left/right recently, I missed flipping category totals. Fixed. http://t.co/qTS8nbucG3 12:39:03
  • Reading – Still Paying for the Civil War (Michael M. Phillips) http://t.co/In5kBr3vwJ 13:48:49
  • No graphs yet. But I now have (very inefficient) historical trends calculations working in raw form: http://t.co/6rEmW1j0If 20:48:19
  • Next up, automatic recalculation (currently I manual trigger that), then actual graphs, then I’ll probably need to tackle efficiency. 20:49:53
  • The current way I am calculating is horribly inefficient. I have it set as a back end process so won’t trigger on web views, but still bad. 20:50:37
  • It recomputes all candidate combos, even if there are no new relevant polls, and recomputes all the historical stuff, even if not changed. 20:51:30
  • Also reexamines all polls and recalculates all the state averages for each data point, even though I have that some of that info elsewhere. 20:53:40
  • Takes over 30 minutes to calculate everything, even with limited poll data. Will get much worse in 2016. So I will have to make it smarter. 20:54:55
  • There are plenty of things I can think of to make it smarter & faster. But I wanted to get something quick&dirty done first, optimize later. 20:56:26
  • But wow is it doing a lot of repetitive calculations that it would be better off only doing once and then using the stored results… 20:57:05
  • Anyway, I spent about 6 or 7 hours on this today. I’ve got some garage cleaning work calling my name now. More poll stuff some other time. 20:58:56