Yahoo and T-Mobile preserve an ancient tradition: inaccessible Captchas
This post is cross-posted to Same Difference, and you can comment there or here.
Captchas are a method websites use to tell whether a visitor is a person or a computer. This is typically used to prevent automated use of their system, and in the case of forums and blog comment boxes, this is used to stop spam. This often takes the form of letters contained in a picture, which a person can see and read but a computer can’t understand because, if they even download the image file, they won’t (unless they have special character recognition software) be able to tell what letters are in it. Captchas (which stands for “completely automated public Turing test to tell computers and humans apart”) exploded in popularity in the mid-2000s, because blogs got snowed under with adverts for indecent material and gambling.
Most websites, including a lot of sites run by people on tiny budgets using cheap shared hosting and free software content management systems like WordPress, use a captcha with an auditory alternative, i.e. you can listen to the letters instead of read them. There is some criticism of this as often the audio is garbled and the letters aren’t that clear. It took a while for this ‘accessible’ form of Captcha to catch on but the vast majority now use one. One of them even claims to offer words from real books that are being digitised by a big public library, although the fact that many of the words are gobbledegook and exist in no known language - indeed, often couldn’t possibly be words - suggests that not all of these images are from real books. But either way, it’s a long time since I saw an inaccessible one. I thought they’d died out.
Apparently, though, two major companies still use them. Last weekend, I tried joining T-Mobile’s forums to ask a question about a problem upgrading my Android phone. I also joined the CyanogenMod forum and XDA Developers, both technical forums and neither run by a major multinational telecommunications company. But it was only T-Mobile that required me to fill out a Captcha from an inaccessible image. I could read it, but my friend Kimberley couldn’t when she tried to join a disabled people’s chat room on Yahoo that she had frequented many years ago.
Captchas are not even all that secure - a few years ago the Guardian’s technology editor, Charles Arthur, noticed that spam was getting through on some websites after the Captcha had been filled in, so either spammers were using optical character recognition (OCR) or getting workers in overseas call centres to fill them in by hand. Either way, the system got cracked and there should be no further excuse to put this barrier in front of blind and visually impaired internet users. Since every blog run on a shoestring budget has better ways of controlling spam, there is no excuse for companies the size of T-Mobile and Yahoo not to.
Possibly Related Posts:
- Brianna Ghey and stable-door logic
- Rosie Jones: Am I a Ret*rd, reviewed
- Threads: Strangled at Birth
- What is ‘snitching’? When is it bad?
- Age like wine, age like milk?
