Then what you want isn't a netbook. You want an ultra-portable-PC, which have existed for a while and will keep continuing to exist.

What I want (and the masses want) is cheap, reasonably powerful laptops. Quite honestly I want something $200 that can browse the full web, have a reasonably decent keyboard, etc. I want components that are enough to multi-task, play music, movies, etc. and also to play some games. Not the newest releases, but be able to play most games reasonably.

The leading bit on signed numbers is the actual sign. The leading bit on unsigned numbers is part of the number. Even if you store them both as twos complement, you have an extra bit which is interpreted differently depending on whether or not it is signed. How can you say that makes them exactly the same? How can they be stored exactly the same, yet give different ranges of values for the same number of bytes? I just didn't want to explain the entire thing figuring a lot of people didn't really need the explanation.

``Why would you use a signed integer for a value like this?''

Probably because you would be programming in a language that makes it easier to use signed integers than, say, a type that could actually represent arbitrarily large values, as long as there were enough bits of memory available. Had the program been written in, say, Common Lisp, Ruby, or Python, using such an unconstrained numeric would actually have been the easiest thing to do.

I haven't had a TV in my house since 2001. I watched on my computer from the indie video store, later netflix, and now the net. Incidentally, since hulu et al., my TV viewership has gone way up. I do have a big monitor in front of the couch, now.

I honestly hadn't watched any TV until Battlestar Galactaca and Lost sucked me in.

Now there's Glee, Community, Dexter...the list goes on.

p.s. it means I'm cooler since I've been without a TV longer, right?

If the test isn't worth enough to update, then it probably isn't worth using. How many million did it cost to validate the original 10? According to the Wikipedia page the blots were made in 1921 with 400 subjects. That doesn't scream millions to me.

For a few million you might even be able to write software that would generate "inkblots", thereby solving the perceived weakness in the test.