Several reasons, really. It depends on a lot on the subject.

In Physics and Math courses, the textbook problems are often solved by the author's students, and are often wrong. Every math course I took (Calculus 1-3, Differential Equations, Engineering Math 1&2) had several cases where one or more homework problems would be incorrect. Same for Physics. For example, pulley system problems where the person doing the illustration accidentally swapped the locations of two forces. Again, these are usually cases where the professor either (1) was in a hurry or (2) most of the work was done by his students. In my Intro to EE course, the problem set was so horribly screwed up (even the sample problems had errors) that our instructor provided everyone with a solutions manual. Why did he use it? Because the choice of textbook was decided by the department, not the professor.

One of my favorite examples was back in high school. We were preparing for a certain standardized test for a comp sci elective I was taking, and we received in advance a copy of the library we would have to work with during the test. While familiarizing ourselves with it, we discovered numerous logic errors, and even a few syntax errors. We corrected them, but when we got around to the test, we discovered the errors had not been corrected by the test provider. So we were taking an exam (for college credit) referencing a library we *knew* was broken, and had to just pretend that it was a black box that actually functioned.

There *are* that many variants of key components. I remember several cases in the past few years where a game failed to function properly on a *single* model of video card because of a bug in the card's programming, or an update to the graphics subsystem of a certain MMO that broke the display on a handful of cards. You can't test on a P4 and say it works on every Intel chipset, you can't test one Radeon and say it works on all ATI cards. Factor in possible issues when combining hardware, and the list grows exponentially.

What ends up happening with the qualified list is that as time goes on, the list gets smaller and smaller as companies work to minimize their liability for returns. We have an app controlling our RFID card reader at work that has, among other things, the following requirements (it will *not* work on anything else): no multi-core processors, 2000 or XP (but only if you don't install SP3, otherwise you have to hack the registry to get it working again). How did they respond when we contacted them for support? "We don't support any setups other than what we listed, so either install it on a machine that meets the specs or pay $8K for an upgrade to a new version." This is how I see the future if we go to a "supported" list. If you don't fit the artificially narrow list, the company will be more than happy to let you pay again for an upgrade that supports current_spec_list+1. Otherwise, you can feel free to sod off.

Ultimately, support lists help the seller, not the buyer, because the seller can decide how narrowly he wants to define the list.

*Raises hand* I do exactly this. All my real work is done in a Linux enviro. My primary machine dual-boots XP exclusively for gaming. Granted, I realize I am an exception to the rule, but be careful with those "no one does X" phrases.

Sellers *choose* to allow returns for buyer's remorse because it is good PR. They are not legally obligated to do so. Similarly, I can *choose* to give you a refund if you don't like your game, but I don't *have* to. If the product/game is defective, then yes, there are legal issues to force me to refund your money. But I'm not aware of any laws in the US forcing me to accept returns for any random reason.

Not really. Getting an illness from his food would be like my software accidentally erasing your hard drive. Those are serious issues that justify a strong response.

A gameplay bug would be more like your hotdog coming with mustard that had no flavor, or a bun that had a hole in it. Annoying, but mostly harmless. And as a small business, I wouldn't have time to personally test every batch of mustard or examine every bun.

I would love to see this filter of his. I've never seen a filter for *anything* that could truly be said to be 100% accurate.

I like this idea.

Where I grew up, we had a children's science museum (Great Explorations: http://www.greatex.org/index.php ) that I used to visit during summers (they did 1-2 week summer camps). The most popular exhibit (most of them were rotated in and out) was always the Touch Tunnel. Totally dark inside, with corridors, ramps, etc. You had to feel your way through to the end. Kids loved it, even when the lights were on. The idea was really simple: giving kids the experience of relying on something other than their sight, and it was really effective.

It is great to see the author's company contributing to a kid's museum. I still remember some of the things I learned at those summer camps (like the letters of the alphabet in ASL). I always loved learning, but it was those camps that really sparked my interest in the sciences.

I took my adopted sisters there once a few years ago (they were adopted at 5/6 years of age when I was 19). I think I had more fun with the exhibits than they did *grin*

You do know that pretty much every ebook reader ever sold supports plain .txt/rtf formats, right?

OTOH, ASCII has limitations. For example, no formatting, no images, no MathML. ASCII is fine if all you care about is the latest Dirk Pitt novel, but it is woefully inadequate for everything else.

Blowing up your own tanker that cost hundreds of millions of dollars to build and carries another 100+ million in cargo is better than paying 2 million in ransom? Yeah, that'll teach them to screw with us. Now they have the crew and no prize. Guess what's getting ransomed now. And hey, then we can start dropping bombs, and instead of ransoming the crew, they'll just execute them instead.

Use the stick too much, and the animal bites back.

This is pennies on the dollar. To a business, this is no different than greasing the local zoning board before getting permits for a new building complex.

If only I had mod points.

The people complaining about this are the same people who would bitch about the lack of scientific rigor in "What kind of geek are you?" quizzes.

Sounds like my method for dealing with political activists. I always find out who they are campaigning for and tailor my response to fit, with the goal of choosing the least compatible option. For example, I respond to Republicans with "Anarchist". Other fun responses are Green Party and "Can't vote, I'm a convicted felon".

No, no, you're looking at it all wrong. What he really means is that securing windows is much easier when you use a blunt stick. Like, say, a sledgehammer.

Install Office 97 and load up the flight sim?
Disconnect all the fans and use it as a space heater on cold nights?

By now, though, I suspect this is the only possible outcome:
http://xkcd.com/350/

You still see them in the medical community (ER docs, nurses, etc), but that's about the only place I ever seen them nowadays.

I've always had the same reaction. As a kid, I had lots of issues. Rheumatic fever, asthma for many years, chicken pox, plenty of sick days.

As an adult, nothing. In the 8 years since I left home, I've had *one* sick day. I had three roommates at the time all flattened out for a week by a stomach flu. Three sick people, two bathrooms to throw up in. Toward the end, I had one queasy night, threw up once, and was out golfing the next morning.

I've always attributed it to the fact that I was exposed to so much stuff as a kid. Even the stuff that caused me asthma as a kid (talking "rush me to the hospital" asthma), nowadays it just makes my eyes dry (cat dander).

I've never been obsessed about germs. I hardly ever wash my hands, I don't shy away from people who sneeze and cough, etc. The body needs, to quote George Carlin, "germs to practice on".