Talk:SE250:lab-5:stsa027

From Marks Wiki
Revision as of 10:43, 3 November 2008 by Mark (Sọ̀rọ̀ | contribs) (1 revision(s))
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to navigation Jump to search

Comments from John Hamer

  • "Chosen because I thought it was adequately large." is not good enough. You need a better reason that that!
  • buzhash RC 7500/2500: llps_i = 132 — this number is spectacularly higher than the other llps values you report for buzhash. Didn't you think that was odd, and worth mentioning?
  • All your llps values for buzhashn are exceedingly high. I rather suspect you have done something wrong. However, I can't check because you have not shown the code you used.
  • ava_Integer 7500/2500: llps = 5, expecting 10.6686 — now a value much lower than expected. Again, worth noting.
  • Java_Integer typical 7500/2500: llps = 796, expecting 10.6686 — wow, Java Integer jumps all over the place.
  • Java_Object 7500/2500: llps = 7500 — this isn't hashing: the function is returning the same result every time.
  • Why are there three sets of results for each of these hash functions? There are only two input sources. I'm confused.
  • Theoretical ranking: you ran a number of statistical tests, but only reported Entropy. How come?

Good to see a summary of the results. However, there were many anomalies that deserved investigating, rather than dismissing with a brief note.

A fair effort.