[Developers] [Data-modeling] Data load issues

Tom Morris tfmorris at gmail.com
Fri May 29 23:48:21 UTC 2009


One other thing I forget to mention, these are flooding the Genderizer
queue too.  Another useful short term patch would be to exclude these
new topics from the Genderizer queue -- at a minimum, those that have
just initials instead of givens names.  I might know that KK
Ramkrishnan is a guy because of my work with speech synthesis, but I
had to punt on a hundred or more others (not to mention all the Jans,
Pats, Lynns, etc which would be just guesses because there's no
article like there'd be with a Wikipedia based topic).

Tom


More information about the Developers mailing list