Patent Office Statistics

[Beady]

It bugs me, how Randi keeps taking swipes at the Patent Office. I really wonder how much he actually knows about it; he certainly seems to buy into the popular misconception that the issuance of a patent is a kind of government endorsement. This strikes me as odd, given his stated preference for research, and for finding out what's really happening.

Anyway, I decided to do a little research of my own :eek: , and here's a highly-condensed version of what I found for 2005:

Number of examiners 4,258
Applications filed 409,532
Applications allowed 182,254
Under examination 807,379
Abandoned by applicant 116,564
Total pending 970,056
Patents issued 165,485
Average pendancy 21.9 months

That's all off the USPTO web site. What amazed me was the unexpectedly low number of examiners. Mention "government agency" and I think in terms of tens of thousands.

If things work the way they do in my own office:

"Under examination" means that the file has been issued to an examiner, who has yet to reach a decision.

"Abandoned" means that the applicant failed to respond to a question asked by the examiner.

Anyhow, with 4,000 examiners, and 410,000 applications in and 182,000 out, and bearing in mind the *real* purpose of a patent, the image of a ditzy, rubber-stamp agency *should* be somewhat diminished.

[bjb]

Anyhow, with 4,000 examiners, and 410,000 applications in and 182,000 out, and bearing in mind the *real* purpose of a patent, the image of a ditzy, rubber-stamp agency *should* be somewhat diminished.

I saw a show about the patent office that claimed the examiners had an average of two days or so to evaluate each patent. Sorry, but I don't have a better source for this information. I don't think the patent office is ditzy, but I do think they are horribly understaffed. The way I see it, Randi wants the patent office to immediately reject patents that claim to operate by paranormal forces. I think that's a pretty good idea. Overall, I don't think the patent office is overrun with idiots, but they could use some better guidelines as well as more help.

As for abandoned patents, you get a patent, then you have to renew the patent every few years to keep it current. You pay a fee each time, and the last fee is a few thousand dollars. Many patents are held by small-timers and they abandon the patent when they get to the last level. I don't think it creates much work for the patent office when a patent is abandoned in this manner.

[GodMark2]

Ah, if only we could go back to the good old days, when patent applications were required to be accompanied by a WORKING MODEL of the invention. That alone would cut the number of patents significantly.

I saw a show about the patent office that claimed the examiners had an average of two days or so to evaluate each patent. Sorry, but I don't have a better source for this information. I don't think the patent office is ditzy, but I do think they are horribly understaffed. The way I see it, Randi wants the patent office to immediately reject patents that claim to operate by paranormal forces. I think that's a pretty good idea. Overall, I don't think the patent office is overrun with idiots, but they could use some better guidelines as well as more help.

As for abandoned patents, you get a patent, then you have to renew the patent every few years to keep it current. You pay a fee each time, and the last fee is a few thousand dollars. Many patents are held by small-timers and they abandon the patent when they get to the last level. I don't think it creates much work for the patent office when a patent is abandoned in this manner.

[fuelair]

Ah, if only we could go back to the good old days, when patent applications were required to be accompanied by a WORKING MODEL of the invention. That alone would cut the number of patents significantly.


Oh yes, please let's go back to that!!!! for ANY device! and for software, a working complete copy (which must be tested!!). Patents are a great idea in principle (as are copyrights and trademarks) but one that is being misused rather a lot.

[Beady]

Ah, if only we could go back to the good old days, when patent applications were required to be accompanied by a WORKING MODEL of the invention. That alone would cut the number of patents significantly.

There's a web site around (I lost the URL ehn I moved to a new computer) that lists "interesting" patents. "Interesting" in the Rube Goldberg sense. I don't see how there could ever have been a working model; things like a device that enabled you to tip you hat to a lady without raising your hand to your head.

[Beady]

I saw a show about the patent office that claimed the examiners had an average of two days or so to evaluate each patent. Sorry, but I don't have a better source for this information.

That pretty much maches my math, based on a 2080-hour work year per person. I think I came up with ~23 hours.

The way I see it, Randi wants the patent office to immediately reject patents that claim to operate by paranormal forces. I think that's a pretty good idea.

So, how do you decide it operates on paranormal forces before accepting and evaluating the application? How much time would be saved per person per year? *Would* time be saved? What are the repercussions if applications are rejected out of hand? What if the examiner, himself, believes in these paranormal forces?

As for abandoned patents, you get a patent, then you have to renew the patent every few years to keep it current.

Thanks. Didn't know that.

[GodMark2]

There's a web site around (I lost the URL ehn I moved to a new computer) that lists "interesting" patents. "Interesting" in the Rube Goldberg sense. I don't see how there could ever have been a working model; things like a device that enabled you to tip you hat to a lady without raising your hand to your head.

That device submitted a working model.

It consisted of a band fitted about the head, with a cantilevered section for the hat to sit on. A string from the non-hat end of the lever ran through a coat's sleeve, to the hand. A pull on the string caused the lever to vault, tipping the hat.

It was on display at the Smithsonian Institute (along with several other odd inventions) when I last visited.

[geni]

Ah, if only we could go back to the good old days, when patent applications were required to be accompanied by a WORKING MODEL of the invention. That alone would cut the number of patents significantly.

It would also kill off pharmaceutical researc in very short order.

[GodMark2]

It would also kill off pharmaceutical researc in very short order.

No, it would kill off the ability to patent a chemical, but would still allow for the methods and machenery used to produce that chemical to be patented, ensuring the inventor a preiod of market advantage. Research would still have the ability to grand market advantage.

[eowyn]

It bugs me, how Randi keeps taking swipes at the Patent Office. I really wonder how much he actually knows about it; he certainly seems to buy into the popular misconception that the issuance of a patent is a kind of government endorsement. This strikes me as odd, given his stated preference for research, and for finding out what's really happening.



I understood Randi's argument to be that because there is (in general) a public perception that a patent IS a government endorsement, then the patent office has a responsibility to not provide the patent for the worst offendors.

If you want to change the situation there are two options:
- educate the public in what a patent actually is
- change the rules on granting patents.

[paiute]

It would also kill off pharmaceutical researc in very short order.

If I had to submit a sample of a chemical with the patent application, I could do it. What it would do is cut down on the wide claims a chemical patent contains. If I prepare for patent one molecule, the attorney writes the patent to try and cover a whole panoply of possible derivatives one "skilled in the art" could imagine making from that one molecule. I may have prepared one structure and my patent claims ownership of one thousand I never made.

[bjb]

I've been involved in patents for aerospace applications. One patent was a safety mod for a space vehicle so to build a working model would have taken many years and cost many tens of millions of dollars.

These kinds of patents are used to give my company an advantage over our competetion by protecting our ideas. We do this because the government is notorious for giving our ideas to our competitors and letting us fight between ourselves for new business. I know this is true because some of our best ideas came from our competition!

[Beady]

I understood Randi's argument to be that because there is (in general) a public perception that a patent IS a government endorsement, then the patent office has a responsibility to not provide the patent for the worst offendors.

I'm trying to wrap my mind around this. Patent examiners should refuse to grant patents, even when something qualifies under patent law, because of public ignorance? That is, federal officers have a moral responsibility to ignore the law because of public misperception of the law.

My, that's interesting. Shall we apply this principle to other types of government officials, and to other areas of law?

[bjb]

That's a very good point, but the patent office does not necessarily need to ignore existing laws:

Utility patents may be granted to anyone who invents or discovers any new and useful process, machine, article of manufacture, or compositions of matters, or any new useful improvement thereof.

The patent office just needs to reject patents that are not 'useful'. Since paranormal forces are not proven to exist, how can anything based on the paranormal be called useful?

The patent office can grant a design patent for a completely useless object, though. For examle, I could get design patents for a new type of dowsing rod even though it wouldn't work. The public would see the design patent as a government endorsement of my dowsing rod. I really don't know if there's a good way around this without changing the law. Maybe design patents need a new name?

[eowyn]

In Australia, we have Registered Designs, not design patents. In some ways, this is similar to copyright. It covers things like 3 dimensional designs, but the object does not have to be useful.

Here, the tests for a patent are:
1 is it a patentable subject (software is included)
2 is it novel
3 is it inventive
4 is it useful
5 has it been secretly used (needs a no answer)

The test for 'useful', is that it does what it claims to do. I gather the tests for a patent in the US are different

(The above is based on my knowledge from one lecture in legal subject at university) (edited to add: by that i mean I don't claim to know much)

[eowyn]

Beady, I admit the faulty logic on my part.

[Beady]

Beady, I admit the faulty logic on my part.

:eek: Stop that!!! This is the JREF forum, and any member here will tell you that you are supposed to state a position or proposition, then defend it to the death. Logic, reason and data are all irrelevent; all that matters is that you prevail.

Now, go back and try again.

;)

[Beady]

The patent office just needs to reject patents that are not 'useful'. Since paranormal forces are not proven to exist, how can anything based on the paranormal be called useful?

There is nothing in the law that says the invention has to operate according to principles and mechanisms that are already known to science and industry. Indeed, such a provision would pretty much invalidate the ultimate purpose of patents in general.

I really don't know if there's a good way around this without changing the law.

This is my central point, and is the very one that Randi appears unable or unwilling to recognize. Government officers (should) operate according to the law, and the law is written by Congress, not the Patent Office. If there are holes in patent law, it is not the patent examiner's role to patch it up or to gloss it over. If an application qualifies under the law as written, then it should get the patent.

Among (many) other considerations, it's much easier for the examiner to defend his decision in case of a legal challenge.

ETA "(many)"

[geni]

No, it would kill off the ability to patent a chemical, but would still allow for the methods and machenery used to produce that chemical to be patented, ensuring the inventor a preiod of market advantage. Research would still have the ability to grand market advantage.

By the time we get to the bulk scale the methods and machinery of bulk manufactor are pretty standard.

[geni]

If I had to submit a sample of a chemical with the patent application, I could do it. What it would do is cut down on the wide claims a chemical patent contains. If I prepare for patent one molecule, the attorney writes the patent to try and cover a whole panoply of possible derivatives one "skilled in the art" could imagine making from that one molecule. I may have prepared one structure and my patent claims ownership of one thousand I never made.

With current techniques it is not a problem to make 1000 different molecules. The patent office is not going to please with haveing to identify serveral million molecules per year.

[bjb]

There is nothing in the law that says the invention has to operate according to principles and mechanisms that are already known to science and industry.

No, but the invention needs to be useful, which implies that it needs to do what it claims to do. The patent office already rejects patent applications for perpetual motion machines and this doesn't seem to be a violation of law. They just need to expand their list of inventions that get immediately rejected, that's all. Of course, now the question is: Who gets to make this list? I'm in favor of Randi's goal but it isn't so simple to get there. Besides, even if we can stop utility patents for useless inventions, there are still design patents, and an invention doesn't need to be useful to qualify for a design patent.


Indeed, such a provision would pretty much invalidate the ultimate purpose of patents in general.

I'm not sure that I agree. The primary purpose of a patent is to allow an inventor to profit from their invention. A patent is intended to protect the invention from being used by others. However, the patent doesn't actually stop anyone from infringing upon the invention. It only gives the inventor the basis for a lawsuit against someone who infringes upon the invention.

The general public seems to have another mistaken idea as to the purpose of patents, which is that the government is trying to encourage new ideas by providing patents for them. In certain cases, this couldn't be less true because anything inventions developed under government funding can't be patented! The reason for this policy is that the government knows patents actually impede the growth of technology by limiting who gets to use it. When the government funds a project, anything that comes from it belongs to them first and the company second. All of that wonderful technology developed by NASA-funded space programs was not protected by patents. A company is only allowed proprietary rights to the technology and has the first shot at using it, though. Some small companies don't want to do government work for this reason. Whatever they develop will get used by their larger competitors and the little guys will get driven out of business by someone using their own ideas against them.

[Beady]

The primary purpose of a patent is to allow an inventor to profit from their invention.

And that, in turn, is to encourage innovation. Innovation is, by definition, something new. There is no inherent requirement that the innovation operate according to principles that are already known.

The general public seems to have another mistaken idea as to the purpose of patents, which is that the government is trying to encourage new ideas by providing patents for them.

You can blame the Founding Fathers for that "mistaken idea":

"To promote the progress of science and the useful arts, by securing for limited times to authors and inventors the exclusive right to their respective writings and discoveries." (Article 1, Section 8 of the United States Constitution)

I anticipate with intense interest your argument that promoting the progress of science does not involve the encouragement of new ideas, or that new ideas bear no relation to discoveries, or that neither bears a relation to the heretofor unknown.

Now you're beginning to sound like Randi.

[Horatius]

There is no inherent requirement that the innovation operate according to principles that are already known.

You can blame the Founding Fathers for that "mistaken idea":

"To promote the progress of science and the useful arts, by securing for limited times to authors and inventors the exclusive right to their respective writings and discoveries." (Article 1, Section 8 of the United States Constitution)


The argument that patents actually stiffle invention is a common one these days, but the intent of patent law has always been to encourage innovation by promoting the publication of all the details needed to implement an invention, in return for a limited monopoly on the patented subject matter. The idea is that the published information may stimulate ideas in others, leading to a generally constant advance in technology. This may or may not work in practice, but that's hard to prove one way or the other.

I actually work for the Canadian Patent Office, so I deal with these sorts of issues fairly regularly. What people need to realize is that there are some serious legal restrictions (established by both the law, and the juriprudence surrounding the law) on what an examiner can, cannot, and must do in a lot of cases. There are times when I'm legally required to accept statements that I personally think are bunk, but which I can't prove are bunk. This can get quite frustrating at times.

As one example, an applicant for a patent (at least in Canada) cannot be held to their "theory" of how their apparatus works, so long as they can demonstrate that it does work. So if an applicant says their new power source is "quantumly entangled quarkite scuttlebutt" or some such thing, I can't reject it solely on that basis. I could require a model (but not, interestingly, a "working" model), but we simply don't have the time or resources to test all such models, even if they did "work".

There's also the problem that not all bogus applications are the same. Some, such as the teleportation apparatus in this week's commentary, are filed by people who are clearly delusional. These are actually pretty easy to get rid of. The problems are mostly with those who deliberately set out to get a bad patent (such as for cold fusion). These applicants spend considerable effort to write an application that exploits the weaknesses in the examination process. They can quite often produce a document that I cannot legally reject. The best I can hope for is to delay it, and drive up their costs of doing business.

[Beady]

Horatius, please send a copy of your post to Randi.

[Horatius]

I've e-mailed Randi on this topic previously. I'm putting together a paper on patent issues that I intend to submit for the next TAM (once the call for papers is published), which he seemed interested in. I'll let you know how that turns out, if it's accepted.

[Beady]

I've e-mailed Randi on this topic previously. I'm putting together a paper on patent issues that I intend to submit for the next TAM (once the call for papers is published), which he seemed interested in. I'll let you know how that turns out, if it's accepted.

Thanks. Unfortunately, I have no plans to be at TAM5.