Profile_bird

Hey there! bjoernknafla is using Twitter.

Twitter is a free service that lets you keep in touch with people through the exchange of quick, frequent answers to one simple question: What's happening? Join today to start receiving bjoernknafla's tweets.

Already using Twitter
from your phone? Click here.

bjoernknafla

  1. @a_williams Please re-tweet interesting answers to that question :-) 34 minutes ago from Tweetie in reply to a_williams
  2. @xDirtyPunkx I will cite you on that! Great quote. about 4 hours ago from Tweetie in reply to xDirtyPunkx
  3. @bbrathwaite You mean an array whose elements have enum-type? about 4 hours ago from Tweetie
  4. Hawkins writes that all collect data about our brain but he states that to understand we need to create models/hypotheses and check them. about 4 hours ago from Tweetie
  5. :-) RT @xDirtyPunkx: @bjoernknafla Good parallel code always follows the ghostbusters rule; don't cross the streams! about 4 hours ago from Tweetie
  6. @groby Good vibes from me! All the best! about 4 hours ago from Tweetie in reply to groby
  7. RT @dvyukov: What tools like STM do is just elevate the level at which you are making mistakes http://bit.ly/8zwd7o about 5 hours ago from Tweetie
  8. RT @MikkoMononen: Nice digested list of local navigation methods. http://bit.ly/5ibvIN about 5 hours ago from Tweetie
  9. @meshula That made my day :-) about 5 hours ago from Tweetie in reply to meshula
  10. @meshula And: congratulations! :-) about 5 hours ago from Tweetie in reply to meshula
  11. @meshula You mean, that the entrance gallery of carbonite-frozen-10-yearers never tingled your suspicion? ;-) about 5 hours ago from Tweetie in reply to meshula
  12. RT @bbrathwaite: Hiring in Bay Area: Game Designers mid to senior level. Can't say company. Please send resumes to brendaresumes@gmail.com. about 5 hours ago from Tweetie
  13. @meshula I don't see the separation between them ;-) about 6 hours ago from Tweetie in reply to meshula
  14. @meshula Two other thoughts: a) prog langs should model how we think (cognition) to ease dev, b) we need to think differently for paralleism about 15 hours ago from Tweetie in reply to meshula
  15. @meshula Two thoughts: a) we need to treat the machines how they work (aka flight/bird wings/planes), b) should we build different machines? about 15 hours ago from Tweetie in reply to meshula
  16. @zeuxcg Thanks for sharing! Very interesting approach. about 15 hours ago from Tweetie in reply to zeuxcg
  17. @ivanassen Hehe :-) about 15 hours ago from Tweetie in reply to ivanassen
  18. @fac9 However, I doubt that high-level semantics enable peak high performance from shared mem or message passing systems. Diff langs needed. about 15 hours ago from Tweetie in reply to fac9
  19. @fac9 If high-level semantics means data-parallel patterns I concur. Strong lang constraints mainly needed for "mere mortal" devs (like me). about 15 hours ago from Tweetie in reply to fac9
  20. @TimothFarrar With the massive number of CTAs this seems to cry for a work-stealing approach + bounded queues to fit CUDA well. about 16 hours ago from Tweetie in reply to TimothFarrar