News for hackers. Stuff that matters.
Re:Facebook's legal team bans developer of F.B. Purity from ... by lessnonymous on Wednesday December 19, 2012 @05:37PM (#4945209)
As someone who's done a lot of data mining, I'm always surprised to find that sites with lots of data don't do this. Your output generator just needs to randomize some class names and throw a couple of pointless elements around the place to make it MUCH more difficult.

There's one site I've scraped that has code like this: <div class="name"><span class="first_name">...</span> <span class="last_name">...</span></div><div class="street_address">...</div>

Re:Facebook's legal team bans developer of F.B. Purity from ... by r00fus on Thursday December 20, 2012 @04:00AM (#4946840)
How is this marginally difficult to parse?

A CSS3 selector could easily just get this - in jQuery:

$('div span').attr('class')

will easily give you the expression that returns all the classnames - easily mapped to whatever output you want.

Re:Facebook's legal team bans developer of F.B. Purity from ... by ucho on Thursday December 20, 2012 @06:30AM (#4947155)
He was actually giving example of page that is ridiculous easy to mine.