4 ms·

This is easy: just put your algorithm behind a web API. When I'm solving my NP-complete problems instantaneously, for free, I'll believe you.

• stevejohnson 16y ago
  "Instantaneously, for free" is nowhere near what this paper claims (O(n^4m)).
  • jrockway 16y ago
    But P is good and NP is bad, and fast is good and slow is bad, so P must be fast!
    • khafra 16y ago
      "Proof by halo bias;" this could be a valuable new technique to math in the social sciences.
• dangrossman 16y ago
  Why would a polynomial time algorithm become instantaneous if you stick it behind an API?
  • ezalor 16y ago
    The magic of the cloud! Imagine running that on Heroku (for instance): it would be costly (price) but possible.
    • swolchok 16y ago
      Just parallelize your algorithm and use n^5 machines ;)
      • wlievens 16y ago
        Is any polynomial algorithm parallelizable?
• bdr 16y ago
  Yeah this is dumb, should have read the OP.
  • natep 16y ago
    Should have taken time to think. You don't even need to read the OP to know what you said is nonsense.