Matthew Turland has a few things to say about web scraping (and recent articles covering it) on his blog today as an author of a previous article published in php|architect covering the same topic:

A friend of mine who shall remain nameless pointed a post out to me on the PHP DZone web site recently. Noting that the article's content was misinformed at best and downright ignorant at worst, even when examining it sheerly from the author's knowledge of PHP as a language, this friend asked that I set the author straight.

He mentions his comments on the post correcting the author on some points as well as a more "clued in" post on the xml.lt website talking about using PHP's DOM functionality instead.