News for hackers. Stuff that matters.
Re:Pronounceable Anagrams by gknoy on Sunday May 24, 2015 @04:20AM (#9595677)
The author does some neat things. When they're comparing Really Small Numbers multiplied together, to avoid underflow, they instead calculate the logs of the pieces and add them. Very nice. :)
Re:Pronounceable Anagrams by samlewis on Sunday May 24, 2015 @06:47AM (#9595879)
It's a fairly common (but cool nonetheless) trick that often turns up in stats :)
Re:Pronounceable Anagrams by bnegreve on Sunday May 24, 2015 @06:12AM (#9595834)
Cool! One little thing is that you would expect that the is_anagram relation is a symmetrical one. I.e. that "neo" is an anagram of "one" then "one" is also an anagram of "neo", which doesn't seem to be the case with you system. Why is that?
Re:Pronounceable Anagrams by cplease on Sunday May 24, 2015 @08:16AM (#9596018)
Probably because neo is not in the source corpus.
Re:Pronounceable Anagrams by maaaats on Sunday May 24, 2015 @08:23AM (#9596031)
Couldn't find I am Lord Voldemort / Tom Marvolo Riddle. Tried upping the allowed string length, but it became struck.
Re:Pronounceable Anagrams by iheartramen on Sunday May 24, 2015 @01:22PM (#9596848)
Yup as I mentioned in the blog post, the running time is factorial with the number of characters, so unfortunately Lord Voldemort won't work :(
Re:Pronounceable Anagrams by relaxitup on Sunday May 24, 2015 @02:16PM (#9597021)
Nice. A result of "good morning" is "goon mording." This could get entertaining..