Why are serious bugs in software considered "okay"?

This blogger has up a thought that has been bothering me of late, especially as I work on commercial systems where companies are paying thousands-to-millions of dollars for software.

I mean, look at what they've managed to do: computer users now expect new software to be buggy. That is, we fully expect a newly purchased product, straight out of the box, to be broken. Something that for any other product category would be an outrage is just shrugged off as the normal course of the business. If you bought a toaster that didn't toast, you'd take it back to the store and demand a refund, or at least a working toaster. But somehow, because computers are involved, if a new video game crashes, we just say "Oh, well, that's what I get for buying it before the patches come out."

The author applies this to the Opera browser and iTunes, but honestly I think it applies much more to large commercial vendor-supported software, where the attitude of the vendor is "It is our job to produce the software. It is your job to test it." Why do we accept this?

Honestly, though, I know the answer. It is because we are not willing to wait. If Apple holds off the newest release of iTunes by a month, people are all over the rumor sites demanding to know what corporate bigwig denied them the ability to view movies on their Mac. And if a commercial vendor delays shipping because of a critical bug, a potential customer will just go to their competitor who ships buggy, incomplete software faster. It makes no sense, but it is how we act in the 2000s.