Perl: the Markov chain saw | |
PerlMonks |
Re: Re: Words that suck...by YuckFoo (Abbot) |
on Jan 17, 2002 at 04:10 UTC ( [id://139394]=note: print w/replies, xml ) | Need Help?? |
Thanks for you thoughts ChemBoy. You can see I lean to
high byte constructions. It always seems to me that
conditions should come before statements (ducks). This was a ten minute hack to show a coworker how lame the passwords he makes up are, I have to use them. Not sure I agree about which string is easier, 'asdfghjkl' is a single keystroke and a swipe. But indeed some words have a momentum component that is not scored. Also the keyboard wasn't quite properly layed out, an offset is probably needed. If 'w' is column 1, shouldn't 's' be column 1.2 and 'x' be column 1.9? Also I shouldn't just add the row distance to the col distance something more like sqrt(dr*dr+dc*dc) maybe. Obviously scoring is flawed because 'social' scores harder than 'socialize'. Length should play more of a factor. Caps should be implemented with high penalty for 'Y'. Perhaps a typo probability for those pinky letters that'll trigger backspaces? You're right, so many variations I punted. clintp, if you've never used 'icicle' or 'coaxial' for one handed adult conversations, you're not doing it right. YuckFoo -who wishes his name was Fred (score=1) instead of Edward (score=2.2) -but is glad his name is not Pamela (7.6) or Claudia (6.83)
In Section
Cool Uses for Perl
|
|