Suppose for a moment that you invent something, but don't tell anyone. Then someone else can obviously patent it.

Suppose you publish something on your blog. The patent examiner doesn't read blogs, so someone else can obviously patent it.

Suppose you publish it in a well-respected journal of computer science. The patent examiner doesn't read journals either, so someone can obviously patent it.

Now obviously, none of these patents are valid, but that is not how the patent system works.

Or suppose someone else is trying to patent the exact same invention that's already patented by you. The patent examiner makes a mistake and issues a patent.

Oh, you mean that patent on the stick as a dog toy? Yeah, that one sucked. I can't believe it took four years, a reexamination by the USPTO Director and referencing no less than two prior patents as prior art to invalidate all of its claims:

http://www.scribd.com/doc/49521/Worlds-Craziest-Patent

Is it? The way I see it, the patent may not hold up to appeal, but that's not the same thing as not being patented by someone else.

And appealing a patent? That isn't cheap. :(

There is nothing preventing someone else from patenting something you have invented but did not patent. They may not be able to stop you from using your invention, but patents are first-come, first served.

Please, if you are going to make an assertion like that, you need to explain exactly how it is an incorrect statement. Maybe you misunderstood me, or maybe you misunderstand the state of the law. Without an explanation, you leave no opportunity to respond to the specific issue you have, forcing the person who have called out to take a stab in the dark and try to guess why you say this. I feel this is disingenuous.

The incorrectness of your statement is explained in every definition of patent requirements. For example, it's in the 2nd paragraph of Wikipedia article called "Patent". I'll even give you a direct link to this requirement: http://en.wikipedia.org/wiki/Novelty_(patent)

In this case, then, the parent comment is correct, and you are wrong.

If I invent something, and neither sell it or make it known to the public (thus it continues to meet the novelty requirement), then there is nothing preventing someone else from patenting said invention.

No, the statement is "things have to be patented...". It's incorrect because you can either patent it, or publish without acquiring a patent. (I'm focusing on this part, because the statement was made to justify patenting. It's not, you don't have to patent anything.)

Like I said, if you're going to make such an assertion, please provide an explanation how it is incorrect. Linking to wikipedia as you did is fallacious for more than one reason, including argument from authority, argument from false authority, and on the basic fact that you did not actually make a claim, you just implied someone else made a claim, and left it to me to guess where that claim was made and what part of the second paragraph showed an error in my claim.

The article you linked to actually confirms my assertion by going over the situations where one inventor will get a patent on something another inventor also created. Therefore, one is incentivized to patent something to obtain protection, which is the claim I was making in the first place.

So, not only are you flat out wrong, you have twice been given the opportunity to provide an explanation for your assertion and twice failed to do so. Meaning you are not worth taking seriously because you are apparently more interested in saying I'm wrong than making an argument, which means I think you're trying to argue to the person rather than the point. Downvoting only convinces me of this.

"The incorrectness of your statement is explained in every definition of patent requirements. For example, it's in the 2nd paragraph of Wikipedia article called "Patent". I'll even give you a direct link to this requirement: http://en.wikipedia.org/wiki/Novelty_(patent)