I'm sorry I used the clpm example -- a CPAN directory or personal web site with lots of "real" project code is a much better indicator. I still like to see community participation because Perl is Open Source. Does the candidate understand the community? Contribute? Leverage on-line resources? Waste a lot of time on rec.games? ;)
BTW, the cheating problem is the same. Just because google doesn't turn up hits for a suspicious paragraph doesn't mean it wasn't copied verbatim -- it may be copied from a library book for example. These papers just take more time and effort to grade. My wife might need to review previous work, talk to other teachers, query the student, etc.
Same thing when a candidate does not turn up in a google search. These people just take more time and effort to understand. When screening a few dozen candidates, the first cut must happen quickly otherwise we lose the most qualified people.