EFF takes a look at the 2011 Patent Reform Act | Hacker News Reader

EFF takes a look at the 2011 Patent Reform Act

eff.org

69 pointsgrellas14 comments

Finally, we are disappointed that the bill would take the teeth out of the Best Mode defense, which currently allows an alleged infringer to invalidate a patent when that patent fails to teach the "best way" to make the invention. Under the new bill, the patent would only have to show "a way" to make the invention, removing an important part of the bargain under which patents are granted – the disclosure of the best mode.

Wouldn't this kill any innovation when trying to "build a better mousetrap"?

The point of granting a patent is that we're supposed to gain something in return, specifically the invention disclosed on the patent application.

If we don't even get the best idea they have for how to build it, the public comes out on the losing end of the social bargain.

Which, depending on how cynical you are, might be the point.

Perhaps I'm reading this wrong, but the implication seems to be that the EFF supports first-to-file.

I understand that first-to-file clarifies things once an applicant receives a patent — they don't have to worry about pior art at that point.

However, it seems worse for everybody else: If I publish something new and patentable, but don't file for a patent, it seems like somebody else could then file for a patent on the same concept and I would end up out of luck.

To be clear, I understand first-to-file to mean that a patent can't be challenged on the basis of prior art once it's issued.

What am I missing?

"Prior art" means the state of the art at the time the patent was filed. If the patent is really an advencement on the state of the art, and it's not obvious, then first-to-file makes sense. The trouble with not having first-to-file is, that someone could keep their invention secret until a competitor files before revealing their advancements. The point of the patent system in the US is to get people to publish their inventions, so it would be counter-productive to encourage secrecy. By denying protection to un-filed inventions, the patent system encourages openness.

In a sense, that's true, but the protection for unfiled inventions is not infinite. The at-sale bar goes a long ways to helping inventors get it right without unduly delaying filing for patent protection

I've heard it both ways. First-to-file either means:

- A patent cannot be challenged on the basis of prior art once issued

or:

- If two people independently invent something, and both apply for the patent, the first one to get to the patent office gets the patent, rather than the first one to document the idea. That documentation would still be prior art, though.

People who think it's the latter tend to like first-to-file. People who think it's the former tend to not. I haven't actually heard from a patent lawyer about it, though.

I tend to think its the latter. If you look at the actual text of the bill...

(a) Novelty; Prior Art- A person shall be entitled to a patent unless--

(1) the claimed invention was patented, described in a printed publication, or in public use, on sale, or otherwise available to the public before the effective filing date of the claimed invention;

... it seems to indicate that prior art is still applicable when considering the validity of the patent. However, it's still not clear to me if prior art can be applied after the fact, or just during that 9 month period.

Ah, I knew somebody else would RTFB ;)

That seems to imply that if unpatented prior art by author A is discovered after author B applies for a patent, and the patent is invalidated, then author A cannot apply for patent. Interesting.

Or, perhaps first-to-file only applies when the reduction to practice is the act of filing a patent, rather than the act of publishing. In other words, it only applies in the special case that two people apply for patents on work that neither of them have published.

That would seem to imply that rather than applying for a patent, one should just publish.

I started to poke through the bill and the patent code to figure this out, but it's beyond me. Maybe I need more than 2 hours of sleep.

That I guess, has to deal with people not filing patents until actual use, so people can't just sit on their inventions forever.

Things that are patentable could be made not patentable simply by publishing it.

I'm not happy about that "in a printed publication" bit, though.

Yeah, did they intend it only to apply to things printed on paper or something?

Currently US patents give priority to the first to invent. So, if you invented a widget and spent six months perfecting that when your next door neighbor filed a patent one day after he "invented" it, your ability to prove you were first to invent (through evidence like your research notebooks, etc.), the USPTO could refuse or invalidate your neighbor's patent and award you the patent. That goes away with first to file, which means that the people with the deepest pockets who can afford to file patent applications quickly will wind up winning the race. Most of the rest of the world uses first-to-file.

From what the EFF has to say, this seems to be an encouraging direction for the patent system. I wonder if some non profit could be funded to monitor and oppose obvious software patents in that initial 9 month window.

Reply on news.ycombinator.com