Hacker News new | comments | show | ask | jobs | submit login
The best patent ever (uspto.gov)
137 points by btilly on Aug 16, 2010 | hide | past | web | favorite | 42 comments

From the patent application's detailed description:

"[0012] The inventor and the assignee of this patent [Halliburton Energy Services, Inc.] have no intention of applying the techniques described herein offensively but instead intend to use the patent defensively to discourage patent trolls and the like from extortionist practices."

(Disclosure: The attorney of record is a friend of mine.)

I'm happy to see the word trolls embedded in a legal document.

But the claims remind me why patents tend to be useless to actual innovation "The method of claim 63 where filing the claim with the patent office includes:filing the claim with the United States Patent and Trademark Office. " -- not only is that obvious, but it's repeated several times. (I'm sure these guys write better patents than I do, I just hate the metric that we use for "better")

Well, if you have years of effort put into something which can be copied easily, as is the case with mechanical inventions, drugs, or cryptography/compression algos (yes I think those actually make sense as patentable), that's one thing. Most software is the opposite -- copying it and applying to your business would often be more effort than inventing your own. So the patent system is exactly backwards for us.

Bruce Schneier has written that patents are useless for cryptography. You can't trust an algorithm until a lot of cryptographers have beaten on it, and they don't bother with patented algorithms. Patenting your algorithm just dooms it to obscurity.

Later claims add specifics. The patent office is general and can clearly refer to any patent office (WIPO, EPO one of the OAPI offices, etc.) whilst this claim is specifically for inventions before the USPTO.

Why do that? Well if there's prior art that knock out the antecedent claims but doesn't include this specificity then you still get a patent. Also for infringement purposes I think more damages can be extracted for things which are specifically claimed as opposed to generally claimed (depends on jurisdiction). Finally it helps prevent someone else from getting an addon to your invention patented - "our invention is the anti-patent invention but with the innovation of doing this before the USPTO, were awesome like that".

I don't see how claim structure like this has any bearing on innovation?

Let's not forget the patent application of patent trolling:

Generating royalty revenue using intellectual property


Looking at the independent claims, it seems like this is a patent on reverse engineering (with a computer!) something, then getting a patent on it, and finally licensing that patent to the owner of the system that was reverse engineered.

I wonder if this could be used as a Gödel patent. If you had a patent on patent trolling, it could be a weapon for patent reform: go around suing patent trolls. Either you win, in which case trolling ceases to be a viable business model, or you lose, setting a precedent for the argument that patents on business methods aren't valid. Either way, you tie up patent trolls in unproductive litigation, force them to argue against the current state of the patent system, and put a spotlight on the ridiculousness of the status quo.

> Gödel patent

I think you just coined a term.

Better copyright it!

Trademark, really.

Why not patent the trademarking of copyrights?

I think the problem with that is prior art has already been established :-/

I dunno - doesn't prior art have to be published?

Prior art has to be extent in the public domain before the priority date. Court records count, they have well established dates which makes them good. Disclosure of an invention in public, outside of the allowed uses in the US, will also invalidate a patent.

Do court records count?

If you had a patent on patent trolling

Funny you should say that. Elsewhere in this thread: http://news.ycombinator.com/item?id=1608556

I always thought "Method of exercising a cat" was the best patent ever


This situation is an example of why the USPTO is swamped.

It's not a granted patent -- it's an application; in fact, it's been rejected twice. After each rejection, the attorneys have called and spoken with the examiner, then requested a re-examination. The current state is that the agents are awaiting a response to their last request for re-examination (click Image File Wrapper):


That is a patent application. It has not been granted.

How do you tell from the page if a patent has been granted?

A patent application says "patent application." A granted patent says "patent."

Listed midway down the page is a "Serial Number" field. In the case of this application, it is 741429. At the top of the page is a "home" button. Click the home button and then select "Advanced search" underneath "PATFT: Issued Patents". In the "query" box, enter "apn/" and the serial number, i.e., "apn/741429". Click search. Five titles appear, none of which is anything close to this patent, so it looks to not have been granted.

If you really want to be sure, go to the main homepage: "www.uspto.gov", and pick "2 search" under "Patents". Then pick "Patent application information retreiveal (PAIR)". Then pick "Public PAIR". Answer the capacha, and then put the series and serial numbers into the lookup box (i.e. 11741429), leaving the radio button set for "application number". Click search. If it had issued, on the next screen under "Patent number" and "Issue Date of Patent" you would find both the number and issue date. In this case, no values, so not issued. If you want to know more you can pick the "Image file wrapper" tab and look through the documents that make up the patent application. This one is still being prosecuted, but has not issued as a patent.

Now that's a user-friendly website.

Try Espacenet or Google Patents instead. USPTO has been the worst of the patent search sites for a long time.

Go to http://ep.espacenet.com/ type in "us20080270152" click on the correct number, http://v3.espacenet.com/publicationDetails/biblio?DB=EPODOC&... .

Note that A1 tells you it's a first publication. B patents are granted.

http://www.google.com/patents doesn't appear to have this application yet.

The Slashdot joke "I patented the act of patent trolling" ceased to be funny a decade ago.

Actually trying it makes it funny again.

And succeeding really makes it a thigh-slapper.

You can apply for a patent on anything you can put into text. The barrier for printing the application (A1 docs) is exceedingly low. Indeed for a time in the UK you could apply for free IIRC, not anymore.

Interestingly I could not find this patent application at http://www.google.com/patents

That is a patent __application__, not a patent. The distinction is quite critical.

Can you even patent the process of fighting someone for patent infringement?

By Halliburton even... LOL

When did Halliburton become Richard Stallman?

When a company is that big, people can push their own richard stallman mini projects through like that.

The left hand knoweth not what the right hand doeth :)

Holy cow, didn't notice that - I thought it was some wise ass seeing what he could get away with as a joke. That this is meant as a serious patent and was accepted, oh boy.

where is the prior artwork?

Whoa, meta.

We should do a patent like this every day.

It's patents all the way down.

summary: applying for a patent on patents

bonus points: assignee is Halliburton


Applications are open for YC Summer 2018

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact