Chapter 21. Web Page Tastefulness
The Web is slowly but surely turning into a cyber-stripmall, complete with flashing neon signs, tasteless ads, and outlet stores. Snazzy graphics crowd out textual information, giving both myself and my low-bandwidth modem a headache. A page purporting to be the definitive guide to some subject turns out to consist of lists of links, most of which are dead. Frames proliferate like weeds, crowding out the page content with scrollbars within scrollbars within scrollbars. Meanwhile, hopping, jittering, flashing, bleeping, and morphing applets dance in and out of my visual field, jerking my attention away from whatever mindless promotional copy I was trying to wade through.
What to do about the proliferation of web junk? Will the Internet actually collapse under its own weight as technology pundit Bob Metcalfe predicted way back in 1995?
Perl to the rescue. You don’t have to wade through 19 pages of trash to find the gem buried in the twentieth. You can have a Perl agent do the wading for you. In a column that I wrote for WebTechniques in May 1997, I suggested a series of indexes to measure the tastefulness of a web page. Some of the indexes were serious, such as the ratio of words in hyperlinks to total words on the page (pages with sparsely scattered links are more likely to contain real information than pages consisting almost entirely of links), or the number of potential advertisements on the page. Others were tongue-in-cheek, such ...