This is exactly what I do. If I'm doing any questionable browsing I pop open Opera. Most browsers have a delete/clear/private browsing option but I don't trust them to be completely thorough.

I've recently discovered though that Flash stories an equivalent to cookies across browsers that don't get cleared out or filtered. There's always something.

http://www.macromedia.com/support/documentation/en/flashplayer/help/settings_manager07.html

I am colorblind. I would not have been able to register for this site without enlisting the help of a friend or coworker. That is a very broken system.

The problem with semantic CAPTCHA replacements is that the problem space is extremely finite. For your boat example, you'd have a set of boat images you'd have to find or create and a subset of them would be steam boats. It would be a labor intensive process to generate all the images. A bot/script to reload the sign up form could quickly aggregate all the images and questions that would come up by refreshing the page a few thousand times. A human could answer all questions for all images once, and have a bot with a 100% success rate.

CAPTCHA works because the problem space is theoretically infinite. Bots that break them don't get through every time, just most of the time. Your proposed system would be defeated 100% of the time by the above bot.

The logical extension to this problem would be to blur or obfuscate the images of boats so that the bot couldn't just easily identify them, but then you're just re-creating CAPTCHA except with blurry boats...

There are a few flaws with this idea. Primarily that it blocks colorblind individuals from registering for the site, and there are much more colorblind internet users than visually and hearing impaired.

This is also not very difficult to break. Assuming that the letters and numbers aren't obfuscated the same way CAPTCHA images are (if they are then this is just another CAPTCHA), a bot would be able to parse the characters out of the image. It could then classify the characters into groups of colors, pick one group randomly, and guess. There couldn't be more than four or five colors in the image since asking to differentiate between aqua/navy/royal/pale blue is unreasonable for a human (but interestingly enough, not difficult for a computer). That would give you a bot with a ~20-25% accuracy rate.

Beyond that, you could parse the question as well, looking for the words red, blue, green, black, etc. and classify ranges of hex colors into associated color names. That would greatly increase success rate of guesses.

This is not a reliable CAPTCHA replacement and in fact seems not very difficult to break.

"You have been tagged in 1 photo"

Well I have an R4. It's amazing, but there are a few games that don't run or lock up because of speed issues. The two above mentioned have apparently fixed that.

It's really by far the most elegant piracy/hacking/homebrew/etc. tool that I have ever used.

Lots of info on Wikipedia: http://en.wikipedia.org/wiki/Nintendo_DS_storage_devices

Either the CycloDS or Acekard seem to be best. Your best bet would be an Acekard 2i, which will work on the new DSi hardware. Not having to carry around carts pretty much makes the console for me. There's no good reason to for everyone to be carrying around the little plastic boxes other than copy protection. Not to mention being able to audit games. There is an overwhelming amount of terrible games for the DS, even with first party and reliable franchises.

There. Fixed that for you.

You have a very loose definition of "information".