Hacker News Reader

Comment by monochromatic | Hacker News Reader

The USPTO is far from perfect, and examiners make bad calls sometimes... but it is by no means trivial to get a trivial patent issued. And the idea of a presumption of invalidity (as opposed to, say, no presumption one way or the other) is so ludicrous that it doesn't bear response.

View on HN

> it is by no means trivial to get a trivial patent issued

Actually, it's pretty easy. Here's how you do it:

1. File the application.

2. Wait for it to be rejected. The rejection will have a point-by-point explanation of why the patent was rejected. (i.e. the patent office is required by law to do your homework for you.)

3. Tweak the application so that the objections raised by the examiner are no longer valid, and re-file.

4. Repeat as necessary.

I have personally used this method successfully three times, and it only took two iterations all three times. The third time I used it on an intentionally bogus patent (http://www.google.com/patents/US20030133714) just to see if it would work. It did.

Like I said, examiners do make mistakes sometimes. For what it's worth though, your claims look pretty lengthy.

> examiners do make mistakes sometimes

That is true, but it misses the point. Apart from the occasional honest mistake, there is a system of perverse incentives that results in the reliable and predictable issuing of bogus patents even when examiners do their jobs well.

> your claims look pretty lengthy.

You must not have read many patents. I have one independent claim and six dependent claims, each with a single minor tweak on the independent claim. That's well below average.

You've got one lengthy means plus function independent claim. That is typically a poor claim. Means plus function severely limit claims despite sounding very broad.

Yes, of course it's a poor claim. The whole patent is bogus. That's the whole point.

I mean poor in terms of how good it is to assert against someone. You can easily get a really narrow useless patent, but that doesn't cause any problems.

That's true. But you can also get a really broad bogus patent. I happened not to do that, but many others have, and that does cause problems.

The point is just that the actual bar for getting a patent -- any patent -- is very low if you know how to play the game.

I mean your independent is lengthy. I don't care how many dependents you have if we're talking about the breadth of the patent.

It's about 220 words. I don't have time to do a survey, but I think that's not unusual.

But you're still missing the point: this is an intentionally bogus patent. It's essentially a patent on a perpetual motion machine that I prosecuted simply to test the theory that one can patent anything by using the method I described. The length of the claim is irrelevant.

It's not irrelevant. It's not the examiner's job to figure out if your invention works (although if it clearly doesn't, they might notice). It's his job to find out if it's new and non-obvious relative to the prior art.

Yes, you can hide the ball from the examiner. But if your claim is so lengthy that it'll never be infringed (not to say yours is quite that bad) then what's the harm in allowing it to issue?

> It's not the examiner's job to figure out if your invention works

Yes, it is:

http://www.uspto.gov/web/offices/pac/mpep/s2107.html#d0e1986...

See section II "WHOLLY INOPERATIVE INVENTIONS; “INCREDIBLE” UTILITY"

> Yes, you can hide the ball from the examiner.

My ball was not very well hidden. If my invention worked it would let you communicate faster than the speed of light. I would expect any competent patent examiner to be familiar with the basic laws of physics.

Some examiners are easier to hide the ball from than others. But if your claim can never be infringed under the laws of physics, who cares if you have that claim issued in a patent?

I care, because it is indicative of the lack of care that the PTO takes when making these decisions. I have four other patents which, in retrospect, also should not have been issued, and God only knows how many bogus patents have been issued to other people. Even if 99% of them are harmless that doesn't undo the real damage done by the rest.

In a perfect world, there would be enough funding for examiners to spent a month examining every application that comes in. In the real world, I worry a lot more about overbroad claims being issued than silly, but extremely narrow claims being issued.

I completely agree. I'm not holding up my patent as an example of the problem, merely as a data point to support my claim that getting a patent that does not meet statutory requirements (of utility and enablement in my case) is easy.

From my experience it's quite easy to get a patent issued if you have enough lawyers on staff. I have been at brainstorming sessions at a big company where we came up with around 10 ideas within an hour and often at least one of them ended up being a patent.

Of course it is not trivial to get a patent application threw, but that is the reason, big corporations have highly paid lawyers for that. The whole thing is a shadow-industry of it's own (starting at the patent office, that earns money by issuing patents).

The patent office doesn't get to keep lots of the fees it takes in.

It seems reasonable for courts to avoid injunctions or punitive damages for periods prior to a patent being tested in court. Calling this invalid or no presumption is more or less the same concept.

No, it isn't. The presumption goes to how hard it is to assert a patent in court.

That's a good thing in this case. Start with 100 junk patents to the USPTO > one gets though.

Now take that one and threaten a company. Company says the is a BS patent. Take company to court demand injunction + damages. Company responds it's invalid!

Q: Does the court force an injunction anyway?

If it does there is huge incentives to settle even if it’s a BS patent.

Reply on news.ycombinator.com