[btrettel]

> How was it possible to patent this in 2018?

Easy. I'm a former patent examiner and in my view, the root cause of most bad patents is obviously the lack of time patent examiners get.

Most people here don't understand how patents are granted. Patent examiners don't get a lot of time. If I'm an examiner, and I can't find it in the time provided, and the application doesn't have some other issues (101, 112, etc.), the application is likely to be granted. This is not a matter of caring about the quality of the work. I'm confident that few critics of the USPTO would do a better job than current examiners under the same time constraints.

The most effective way to eliminate bad patents would be to increase the amount of time patent examiners get. The amount of time is based on some IBM study from the 1960s from what I know. Some adjustments to the time have been made, but it's nowhere near enough. Yes, we now have better search technologies, but we also have at least two orders of magnitude more documents to search.

I'm told that the amount of time examiners get probably won't increase without congressional intervention. Right now the USPTO only makes money through user fees. The USPTO receives no tax revenue at all. This situation is actually worse, as the Department of Commerce diverts some of the USPTO's revenues for other projects. At the very least the USPTO should be given control over their own money, and they should also receive tax revenue. Then examiners can be given more time and do a better job.

Here's why some other approaches won't work:

- Punishing examiners for making bad decisions will just make an already stressful job more stressful. This seems to be the current focus at the USPTO. Fortunately I haven't heard anyone being fired due to poor quality, so I think it's mostly talk.

- Adding more ways for companies to kill bad patents after they've been granted favors large corporations who can afford to kill bad patents. Small corporations and individuals are still powerless against bad patents.

[chmod775]

That's ridiculous.

Having lots of such work is no excuse to do it worse. It is however an excuse to have a growing backlog.

When it takes 10 years to have a patent granted governments will do something about it, but they won't if you "make do".

Appointments for certain government offices in my municipality are booked for 3 months in advance right now. It made the news and the local government is increasing staff.

This would have looked very differently if someone just decided to cut the allotted time for appointments in half.

Degraded service will be tolerated for a long long time. Broken or nonexistent service less so. Imagine the outcry if people and companies can't get patents anymore.

Thinking of it, patents should probably only be given to natural people and at most one every ten years per person (unless replacing an earlier patent), and who can only sign away up to 50% to a non-natural entity. That'll cut down on the bullshit as well. The notion that one person among billions can come up with multiple patentable ideas in such a timespan is patently ridiculous and need not be entertained. Patentable ideas should take research or domain knowledge accumulated over years and not be a five minute shower-thought.

[webmaven]

> Having lots of such work is no excuse to do it worse. It is however an excuse to have a growing backlog.

Unfortunately, examiners are evaluated based upon the number of applications they process.

Furthermore, when an examiner denies a patent or a claim, the patent application can be amended and refiled. Over and over again. Until the examiner grants the patent. Which still only counts as one patent toward the examiner's quota.

So, that backlog that you're imagining sitting there passively waiting actually represents an ever growing workload for the examiners, while their career-limiting KPIs get worse and worse.

[mulmen]

> Unfortunately, examiners are evaluated based upon the number of applications they process.

Well that sounds like it could be the entire problem.

[webmaven]

Well, not the entire problem, but yeah, a major chunk of it.

The problem of how to evaluate examiners' productivity is pretty similar to evaluating software developers'.

Unfortunately, the measure that is being used is akin to counting PRs merged.

It isn't hard to see how the PTO arrived at this method of evaluation: patent examiner time is their most constrained resource and they aren't provided the funds to hire more examiners, so of course they are focused on making the most efficient use of examiners' time to evaluate as many patent applications as they can.

Even so, patents are examined pretty thoroughly (just not thoroughly enough to prevent any bogus ones from slipping through) and it takes almost two years for a patent to be granted.

[mulmen]

Yeah I debated using that word. Probably not entirely accurate. Something like “if you change that all the other problems are insignificant” might be better.

I think your explanation of how this can happen is plausible. But it’s also the problem with treating government services like businesses.

USPTO incentives should be aligned with the public good, not the bottom line.

This differs from a service like the USPS or Amtrak that address a market failure.

[webmaven]

> This differs from a service like the USPS or Amtrak that address a market failure.

I think you may not understand the underlying economic forces.

Patents are explicitly addressing an existing market failure (that of suboptimal funding of innovation due to free rider disincentives). The mechanism is a grant by government fiat of an exclusive right for a limited time, enforced by the courts, and these grants (and the products covered by them) then become subject to ordinary market forces.

[webmaven]

Relevant post: https://blog.google/outreach-initiatives/public-policy/refor...

[car_analogy]

While I agree the USPTO should be better funded and examiners should have more time, what I find problematic is the attitude of the current approach, and of your solutions. That there is a torrent of bad patents, and the USPTO needs enough resources to fight them off, or they'll break through.

No! The USPTO is not a defending army trying to keep out barbarians. It grants patents. If it does not have time, or resources, then it does not grant patents. If the patent examiner feels they don't have enough time, the patent is denied - too bad. You can re-file, paying a fee for extended examination.

I feel any reform which does not change this approach will have only limited effectiveness, as patents will simply increase their complexity to make them more difficult to examine, until junk makes it through again. The USPTO must have the ability to say no.

P.S.: I feel I should bring up non-obviousness - prior-art is not the only disqualifier. It should disqualify all of these "do specific but obvious thing, on a computer" "inventions", prior-art or not. Perhaps the filer could be required to explain why their invention is non-obvious, saving the examiner some time.

[btrettel]

> No! The USPTO is not a defending army trying to keep out barbarians. It grants patents. If it does not have time, or resources, then it does not grant patents. If the patent examiner feels they don't have enough time, the patent is denied - too bad. You can re-file, paying a fee for extended examination.

Rejections can not be arbitrary. If the examiner can't find prior art but wants to reject the application, what are they going to write in their office action, the response to the patent application? "I couldn't find anything, but I don't like your application, so I'm going to reject you." As a junior examiner, everything I did had to be approved by someone. There's no way that would get approved. I'd be told by my primary examiner to go back and find something if I want to reject, and by the way, I don't get any extra time.

I haven't heard of any examiner being fired for poor quality, but repeatedly writing rejections with no basis would probably get an examiner fired for poor quality eventually.

And the applicant can appeal. As far as I'm aware the examiner gets no time at all to respond to an appeal! (I didn't get any time for the one appeal that I had.) The examiner does get time to respond to an RCE or continuation. It doesn't take too many appeals to seriously impact an examiner's productivity metrics, and if the examiner falls below 95% of their target for too long, they're likely to be fired. [0]

A lot of attorneys complain that the USPTO is far too harsh. And I think they have some fair points. Only about half of patent applications are issued as patents within 3 years. The vast majority of applications (over 80% last I checked) are rejected in the first action. (I rejected every application I received in the first action.) The process needs to be fair. So there need to be mechanisms to prevent the USPTO from simply rejecting everything to be safe. If the USPTO rejects everything, why even bother having a patent system? (Note that I'm not defending the USPTO's current system, which I think is absurd to give the examiner no time for an appeal.)

> P.S.: I feel I should bring up non-obviousness - prior-art is not the only disqualifier.

I'm afraid that you don't know what you're talking about here. Obviousness is a prior art rejection, and it's the most common type of prior art rejection. An examiner could go on "official notice" and just declare something obvious or known without prior art, but as far as I'm aware that never works.

[0] The way these metrics work (roughly) is that certain actions get a time added to a counter. Then your "production" is calculated as the amount of time you earned (the counter) divided by the amount of time you spent in examination.

[car_analogy]

You are telling me how things currently work, I am talking about how they should work.

I never said to reject things arbitrarily. Only if there is not enough time for an examination. Think of it as putting things on a backlog. If faster turn-around is wanted, more funding can be provided. Lack of funding should only make the work slower, not broaden patentability.

> I'm afraid that you don't know what you're talking about here. Obviousness is a prior art rejection

I'm sorry but this is absurd, and if the patent office functions using this definition makes it no less absurd. Why would US patent law require both non-obviousness and lack of prior art, if the two are the same? The fact that claiming obviousness without prior art "never works" in the USPTO is just evidence of its deep dysfunction.

I sympathize with working in a bureaucracy that seems built to produce bad outcomes, but conflating arbitrary rejections with asking for more time/putting things on a back-log, obviousness, and prior art, really does not inspire confidence. It smells like reasoning deformed by a broken bureaucracy, rather than sensible application of principles or the spirit of the law.

[btrettel]

> Why would US patent law require both non-obviousness and lack of prior art, if the two are the same? The fact that claiming obviousness without prior art "never works" in the USPTO is just evidence of its deep dysfunction.

In US patent law, the word "non-obvious" refers to differences between the claimed invention and prior art. See 35 USC 103: https://www.law.cornell.edu/uscode/text/35/103

I agree that this conflicts with the colloquial use, which may be why European law says "lacks inventive step" instead of "non-obvious".

I get the impression that "prior art" to you means that the claimed invention was previously disclosed, which is a rejection under 35 USC 102. But that has to be exactly as disclosed. It's a high bar to meet.

> It smells like reasoning deformed by a broken bureaucracy, rather than sensible application of principles or the spirit of the law.

This speculation is neither necessary or helpful.

[car_analogy]

> In US patent law, the word "non-obvious" refers to differences between the claimed invention and prior art. See 35 USC 103: https://www.law.cornell.edu/uscode/text/35/103

> I agree that this conflicts with the colloquial use

The law you cite says, distilled:

A patent [..] may not be obtained [..] if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious [..] to a person having ordinary skill in the art to which the claimed invention pertains.

This is, if anything, a superset of the colloquial use of obvious. It covers not only everything obvious to practitioners in the field, but also things that are only obvious if you are familiar with every latest invention in that field.

To clarify: You seem to believe that looking at it in terms of "differences" narrows the scope of what is obvious. Why? "Prior art" is, literally, the sum of human knowledge and invention (up to some date). To say something is obvious in terms of differences from prior art means that, if there is an obvious way (to someone skilled in the art) to accomplish the claimed invention using any prior knowledge or invention, then that invention is itself obvious.

How could you read this any other way? How could something be obvious, but become not-obvious if you referred to prior art?

[lostdog]

> the root cause of most bad patents is obviously the lack of time patent examiners get.

It takes about 30 seconds to realize that the above patent has zero novelty.

I'm not going to disagree with the patent office being underfunded, but the fact is that whoever granted this patent had a complete lack of critical reasoning skills, and no amount of time was going to help that.

I see this all the time with the patent office. They'll approve any madlib-style patent without thinking about it at all. Any patent of the form "______ wirelessly" gets approved. "________ with a computer" is the other offender. Sure, maybe they reject it on the first pass, but it always makes it through the second pass.

Want to know the truth about how patent approval works? Talk to a patent attorney who files them. Here's some true things about getting your patent accepted:

1. Novelty doesn't matter at all.

2. The lawyer doesn't need to understand anything about the technology to write up the patent. They do need a few of your buzzwords, but that's it.

3. Patents always get rejected the first time, and then if your lawyer is any good they always get accepted the first time.

I've got a fistful of patents myself, and the acceptance of the patent application had no relationship with the inventiveness of the invention.

[zimpenfish]

> It takes about 30 seconds to realize that the above patent has zero novelty.

But can you find good corroborating evidence in 30 seconds? I'm pretty sure the applicant's lawyers would have a field day with "Patent Officer X rejected this due to their own critical thinking", no?

[Aachen]

> This is not a matter of caring about the quality of the work. I'm confident that few critics of the USPTO would do a better job than current examiners under the same time constraints.

I doubt anyone criticises the workers at the bottom, if/once they're aware of said time constraints. Rather, the caring about the quality of the work is something the person who decides on the time constraints clearly doesn't do.

It is a matter of caring, but yeah of course it is decided by a higher-up, not every person working there being individually lax. (I don't expect google/malboro/shell/... employees to be individually evil either, for example, even if the companies clearly could be better-respected if they cared about doing their work in society well.)

[mcherm]

Another possibility would be to create some system with an opportunity for an adversarial legal process. Allowing experts in the field to weigh in (providing prior art and/or professional opinions on patentability) might take some of the burden off the patent examiners themselves.

[btrettel]

One form of this already exists: https://www.uspto.gov/patents/initiatives/third-party-preiss...

Almost no one uses it, unfortunately.

[thayne]

> The most effective way to eliminate bad patents would be to increase the amount of time patent examiners get.

I disagree. I think the most effective way to eliminate bad patents would be to eliminate patents altogether. If there are no patents then there can be no bad patents:)

More realistically, it would be more efficient to stop bad patents before they make it to the patent office and waste patent officers' time. And the best way to do that is probably to reduce the incentives to file bad patents. I don't know what the most effective way to do that is though, maybe expire patents sooner, limit the number of patents an entity can file in a period of time, increase the fee for each additional patent filed, a penalty for patents that are rejected or invalidated, require you to actually produce (or license to someone else to produce) the thing you patent in order to keep it, etc.

[sfifs]

Why can't all patent applications be mandatorially opened for a public comments period for a period of say 3 months for anyone else to file any objections? The act of patent application provisionally would protect the idea till the patent is granted.

Then the examiner could look at both the application and the comments and decide whether to grant or reject the application.

[btrettel]

US patent applications are open for third-party prior art submissions. https://www.uspto.gov/patents/initiatives/third-party-preiss...

Almost no one sends in prior art.

[bonzini]

Where can patent examiners look for prior art?

[btrettel]

Patent examiners can look basically anywhere a member of the public can and some other places.

The internal search tools check a lot of patent databases and are quite good, but take time to learn.

The USPTO has a lot of subscription databases as well, including fancy AI/ML-based ones.

Many examiners will also search normal search engines like Google, though this can be tricky for legal reasons. If the application was not published yet then examiners are not allowed to get very specific in the search and other search engines as that could release confidential information to the search engine. The USPTO has agreements with the subscription databases to keep the searches confidential but no such agreement exists with Google.

In my view making new search tools like https://www.priorartarchive.org/ would not help the situation too much. It would be better to integrate more databases into the existing USPTO tools, as they are designed for serious power-searchers, and would make the new databases more visible. The internal search tool is much faster than the alternatives and operates by keyboard. Point-and-click search is much slower by its nature. Speed really is critical when time constrained and I think that is something not appreciated outside of patent organizations.

[PeterisP]

What I'm reading in your description is that the process is designed to find cases where someone has patented the same thing before, but not designed to find things that should be fundamentally unpatentable ('obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art') and many other people are doing the similar thing in their products but not trying to patent it (because it shouldn't be patentable) and not explicitly writing it up in public blog posts with the exact same keywords.

Presuming that "ways to wirelessly share photos online such as through social media" has not been patented before, the best place to look for prior art would be the actual existing social media products which have ways to wirelessly share photos online instead of historical patent data; an effective search would have to be for actual prior art (i.e. products and solutions), not descriptions of prior art (patents and webpages).

[deno]

Why is the secrecy required if the patent is going to end up public anyway? If you could pre-publish to the public (with a verifiable timestamp) then anyone could submit a challenge and the patent would be only be granted if it’s actually novel and remains unchallenged for some time. IDK I must be missing something because this seems too obvious a solution...

[btrettel]

Not all US patent applications will be published. The applicant can pay extra to keep it secret unless it is granted.

As for why the applications aren't published as soon as possible, I'm not sure. I suspect people might want some time before publication to test the market. If there is no market demand, I know that some law firms will abandon the application. But if it's worthless, then why not publish it?