
Tim Bray has four words for Microsoft: "Litigate or shut up". (Context) ![]()
The woe of misconfigured filters ![]()
My college has recently implemented some really poorly designed web filters. I am finishing on Wednesday, so I'm not making a big fuss about it, but let me just give you a list of the websites that are no longer accessible on the college's wifi network:
Grazr, Reason Magazine, Boing Boing, TechCrunch, CrunchNotes, Digg, Slashdot, WordPress, Creating Passionate Users, DailyKos, Instapundit, Aral Balkan, Guardian (woah!), Telegraph (double woah!), Independent (triple woah!), New York Times (quadruple woah!) Xenu.net, Vecosys, Amazon UK, Open Rights Group, RDFa.info, Wikitravel, .
Some more: 43 Folders, Cubicgarden, the W3C's XHTML2 pages, the W3C Markup Validator, the 'basic HTML' view of Gmail, NBC, ABC News, TailRank, The Sun, ITV, Wall Street Journal, the Boston Globe, the LA times, CNN, FOX News, Daring Fireball, Londonist, Adobe, Remember the Milk and, most ironically, the University of London Union (my college is part of the UoL)... the list could go on and on.
This is why web censorship is doomed to fail. I bet that within ten minutes, I could find some hardcore pornography. But I can't read the goddamn New York Times thanks to a misconfigured filter. This is what the filtering advocates want to give children. It doesn't affect me. if I seriously want to read something on a website, I will find a way. But there are many people who can't do that.
The RSS feeds for a great number of these sites work fine, by the way. The filter seems to be reading the HTML pages for their content and then deciding whether I can see them on that basis. It doesn't seem to be working for XHTML content (any pages I've visited that are delivering back properly MIME-typed XHTML are working fine) - Jyte (XHTML 1.1), my blog and Adactio are working fine. I'm sure someone could do MIME type spoofing in order to get around this filtering system. Except that when I went to make a Jyte claim "Attempts to filter or block Internet resources are doomed to failure", it conked out (probably the word 'filter' or 'block' in the URI being caught by a URI-based filter).
Just remember, British public, the University of London is a public body. Your taxes pay for it, but misconfigured filters are limiting the academic freedom of scholars and students alike. This is something that nobody ever seems to get concerned about.