I've had several requests to do this, but have been rather lazy to do it.

I am a U.S. patent attorney and have done pretty much anything you can think of that relates to patents (procurement, due diligence, searching, invalidation, etc.). I mostly help startups build patent portfolios for valuation purposes in the U.S. and in foreign countries, and also help large companies (US and foreign) obtain patents for their research/product labs in the U.S. and abroad as well. I have also filed my own patents before as well.

Although I have interacted with pretty much any country that one would reasonably consider for obtaining patents, the countries I'm most familiar with are Japan, Taiwan, EPO and Germany, and to a lesser extent South Korea and Australia, although I am not a patent attorney for those countries.

Disclaimer: I am also not your patent attorney so nothing posted here should be construed as legal advice.

Subscribing as I'm super interested in the process

I'm no expert on this topic so perhaps I'm missing simple things that you can explain.

The whole patent system seems to be holding the US back in the form of much higher costs for products while benefiting a very small number of people. Also there seems to be very little evidence that patents actually increase innovation. - https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5664960/

It seems to me like a gigantic positive for humanity when India ignores pharmaceutical patents or when China ignores software/hardware/manufacturing patents and makes substantially cheaper products.

Rather than adjusting the US patent system it seems more likely that other countries will be forced into cooperating. Do you think this will be a positive thing? If not what changes could be made to the patent system that would improve it for use worldwide?

Can any of these microcap patent troll stocks actually hit it big or is it a dated business model? Next time we here XYZ awarded $500m in MSFT patent suit, should we immediately think short?

I'm not sure the measurement should be if patents increase innovation, but rather would innovation occur at all in a system with no patent protection, particularly in low trust societies like China or India. It is much easier to copy a competitors solution than it is to develop your own, and we saw a lot of that in China in the past when they didn't really have a functioning patent system, so it kills a lot of incentive to innovate when your competitors simply duplicate your solution in fairly short order.

Flipping the question around, when was the last time you saw something innovative from China, or from the Indian pharmaceutical industry?

One improvement that should be made is that the standard of examination in the U.S. needs to improve. Oftentimes, the examiners get hired are bottom of the barrel new grads that have no concept of the technology field or the state of the art. The policy of letting the courts worry about validity should be eliminated; at least we should match the standard of Japan or Europe.

Collaborative search systems may help. The U.S. is testing a pilot program for U.S. examiners to collaborate with Japanese and Korean examiners for applications that are cross field into those countries. U.S. examiners have a lot to learn from their foreign counterparts IMO; which is why you hear about these patent problems mostly from a U.S. perspective and very rarely from an European or Japanese perspective.

It's definitely dated. It is significantly harder to get to the situation to win multiple hundreds of millions, and I suspect you will very rarely ever see that happening going forward from a troll.

The troll business has always been about trying to settle for less than the cost of defense by filing nuisance suits on products based on patents that are barely relevant to the product. Part of the past success was because defending yourself was rather costly (sometimes even 5-10m per suit), and also because the Eastern District of Texas is very patent friendly and awards rather ridiculous damages for plaintiffs.

Two things that killed the troll business are inter party reviews (IPRs) which allow the defendant to try to kill the patent before allowing the trial to proceed, and the TC Heartlands case which took away the Eastern District of Texas as a viable venue. IPRs have a rather large success rate (80ish percent), and cost fractions of a million dollars, so that has driven the price of settlement way down.

The ones that you see survive that process and end up with some damages nowadays have at least some form of legitimacy, although they might not necessarily collect on everything. It does depend on a case by case basis though, particularly if the damages award comes from a judge and jury that had no concept of how patent damages should work (see Carnegie Mellon vs. Marvell).

I (most likely) invented a novelty item that would have exceptionally, exceptionally niche commercial appeal if any, but it would be hilarious to actually have a patent for it. What's the cheapest place to get one for the lulz and approximate cost (US citizen)? I don't trust George Foreman.

Generally speaking, if you invented it in the U.S. then you have to file it in the U.S. first. However, U.S. is also generally the cheapest place to obtain a patent for a U.S. citizen if you are a small entity or micro entity, as you can get 50-75% discount on USPTO fees.

Small entity is for solo inventors or companies with less than 500 employees. Micro entity is for the above + you make less than 3x the median wage level, so something like <$150k annually.

As for the cheapest price for getting a patent on the novelty item, it depends on what you are trying to achieve and the nature of the item. If you care mostly about protecting the look and feel of your novelty item, you can get a design patent fairly cheaply. That can possibly run you less than $3k or even $2k all told depending on its complexity, and you can obtain them similarly cheaply in other countries too after you obtain a foreign filing license from the USPTO permitting you to file your application in foreign countries. Most of those costs would be from hiring someone to prepare professional looking drawings to draw out perspective views of your item. If it is a complicated product, then it may run you more than $3k if you have many perspective views that you need to prepare to understand what the product looks like, or if you have variants that you need to protect (the charges will be per extra figure needed, and also per variant if the Patent Office forces you to divide your application into multiple applications).

Any law firm with patent attorneys not trying to rip you off would give you something roughly resembling the estimates above. I suppose you could save some money to do it yourself if you know a guy who could make patent drawings for you, as the text for such design patents is very basic (look up any design patent and you'll see they all follow a very similar text description structure). The only time you would really need an attorney is if the patent office rejects your patent application, or for assistance in guiding you to meet all the formal requirements for the application.

If you are trying to protect the functionality of the item, then you will have to obtain a utility patent, which will run you a lot more and would definitely need attorney assistance.

Does a patent basically verify that a device, if built to the description, will actually work as claimed?

Thanks.

No. As a practical matter, the examiner is going to assume the applicant is acting in good faith and the invention really works as claimed, unless it's just obviously ridiculous like a perpetual motion machine or whatever. The logic here is that there's little harm to the public if someone gets an exclusive right to an invention that doesn't work.

Also note that a patent doesn't mean that the technology works better than existing solutions. If you invent a new pain medication that's 90% less effective than aspirin, it could possibly be patentable. People like to advertise that their product has a "patented technology," but really that's not actually saying all that much.

Yep, Cranberry Tea is right.

In theory, patents are supposed to disclose enough for someone to make and use the invention as described without undue experimentation.

In practice, Examiners are mostly assuming that you are operating on good faith, as it is rather difficult for them to prove that your invention is not enabled by your description unless it's something very blatant.

Thanks.

Thanks. I'm looking at something loosely described as a quantum battery (except it's not a battery) and the science is way over my head. So I was wondering if perhaps since it was granted a patent that might give it some validity.

I would say in this case the patent shows that the inventor isn't an obvious crackpot, but it's no guarantee that the invention works as advertised. One would hope the examiner did a little digging to make sure the science isn't totally absurd, but there's no way he/she can thoroughly vet the technology.

subbed! Awesome thread

That's a resounding no. Getting a patent is a good way for frauds to seem more convincing and con investors.

If I invent something in Argentina, can I patent it in the US and protect it against some folks seeing my product in Argentina and trying to take over the US market?

Has knowledge of patent law informed your investing or trading? Made you money (apart from a handsome lawyer's salary ). Anything that the media or analysts get very wrong?

Yes. Easiest way would be to file for a PCT patent application with WIPO as the receiving office, designate all countries and then file in the U.S. within 30 months. Other way to do it would be to file in Argentina first and then file it in the U.S. and claim priority back to the Argentinian patent application within 12 months.

I am not sure if you can file directly into the U.S. first as I do not know if Argentina requires a foreign filing license or has export control issues regarding exported technology. That's something an Argentinian patent attorney would be able to tell you. But you should almost certainly be able to file in the U.S. after filing in Argentina at the 10 month mark afterwards or so.

There were three incidents that I remember.

One that I didn't play because I was a dumbass was for the patent troll ACTG. When the American Invents Act passed, ACTG's business model became completely dead. At the time, it still had a 2b market cap and was a no brainer short forever. But alas, 5 years ago I had a philosophy of never just going straight short, and ACTG had no options.

Now it is at 200m market cap and that sucks :P

Another that I remember, but I can't actually find the case now, was the SCOTUS ruling on some pharmaceutical patents for Mylan awhile ago. SCOTUS ruling basically narrowed the scope of some of the claims fairly significantly, but the analysts misinterpreted the ruling and the stock went up 10% in the morning. We had an internal memo about the case that indicated that the ruling was the expected one and it wasn't very good for Mylan since it rendered half of the patents to be useless, and the guy who wrote it added the throwaway line to the effect of "and yet, the stock is up quite a bit in the morning. Wonder why that is the case." So I bought some weekly puts for lulz.

IIRC it dropped to up 2% that day and I ended up with a ten bagger.

The third one was Carnegie Mellon v. Marvell, in which Carnegie Mellon was rewarded something like 1.2 b in damages because somehow the lawyers of Carnegie Mellon convinced the judge and jury to apply damages on Marvell's foreign (non U.S. sales), even though U.S. patents don't apply to foreign sales typically. If it were just U.S. sales, damages would be 300m to 600m so they were very off. I went long MRVL, and it eventually recovered from the initial shock of the ruling and sold it later at +40%. Marvell went to settle for 750m eventually, and I am still baffled as to why they paid so much.

Currently I hope DGLY turns into the next Carnegie Mellon v. Marvell since it seems to have all the same factors that made Carnegie Mellon vs. Marvell so ridiculous, only this time DGLY actually has a product.

I'm also a patent attorney (practicing 20 years), but won't steal your AMA thunder. In fact, I'll even throw a couple of questions your way:

1) How do you envision the IoT, AI and blockchain patent situations playing out? Will it be litigation hell, or do you think there might ultimately be pools and FRAND licenses in order to help pave the way for more immediate innovation and commercialization?

2) What, if anything, do you see happening to curb the newly in vogue "bottom feeding" troll practice? This is where a questionable patent might be purchased for a pittance and then the troll creates LLCs to sue companies in batches (not all at once, since that would encourage joint defense) with the goal of getting maybe $10000-$50000 from each company such that there is almost no way to turn down the offer due to even the smallest effort to fight likely resulting in higher outside counsel expenses?

3) Now that the dust has settled, what do you see as the best and worst changes to have emerged from the AIA?

Always cool to see another patent attorney, since it's a very niche field and there are not so many of us in the grand scheme of things.

1) I actually work pretty heavily in all three fields. I think IoT will go the SEP/pool route since there will be a need to have a standards body to figure out how communications work between IoT devices among other things. Oddly enough, in view of Delrahim's comments about how FRAND royalties should be handled and how the Trump administration views SEPs in general, I expect more SEP litigation to determine what FRAND ought to be. Already we are seeing IoT pools like Avanci being very aggressive about licensing their patents and stretching the definition of what constitutes IoT, so to answer the question, I do see standards bodies and pools taking over IoT, but I also see it possibly resulting in more litigation hell, as odd as that might seem.

My experience from AI/Blockchain leads me to believe that it's the wild west, especially since a lot of the underlying technologies are already public domain. It will result in "X on the internet" patents all over again, only with "use AI to do X" or "use blockchain to do X" instead, especially in fields in which Alice is not a concern. They will definitely be the next area that trolls/NPEs will target and suck into litigation hell, especially AI IMO. Have you seen otherwise?

2) Probably nothing . Maybe more Microsoft-like clauses offering protection from patent trolls for customers that use their enterprise solutions, which includes a license to all of Microsoft's patents. Whether Microsoft would actually defend such people or just pay the 10-50k is unknown though, although maybe they can justify it as defending their IP. At least it's a far cry from the 100k-500k settlements before the AIA.

3) The best change is the micro-entity fees, as that opens up filing in the U.S. from smaller foreign startups that meet the standard.

I have mixed feelings about the IPR system. On one hand, it is a great and cost effective way to kill troll patents. On the other hand, precisely because of that, you can also suffer a lot of nuisance IPR suits that have fairly low merit but that you still have to defend against. This happens quite often between foreign manufacturers choosing to be douchebags to each other in the U.S., even though the underlying licensing dispute or product line may be far less than the cost of the IPR, or larger companies trying to bully nano-cap or startup companies.

Say I went back in time to like 1998 and that plastic tennis-ball thrower that people use to play fetch with their dogs hadn't been invented yet. How much would it cost and what would it entail to capitalize on this idea that I'd stolen from the future?

Could I just submit a few drawings and explain what it does and that it's made of polyethylene or whatever? It seems like that should cost about $100 bucks, but what you say above about the minimum being $2k (and I don't even know if you'd classify this thing as needing a design patent or utility patent) makes me think there's much more to the process.

This is what I'm talking about, by the way:

I'd reject that on the grounds that it's clearly just a Jai alai thrower.

But yeah, getting a patent is always an expensive proposition; a few grand is standard even for simple inventions.

If you only care about the look of the product, then a design patent would still run you at least a few thousand throughout the process, since there are several fees that you will encounter (filing fees, fees related to possibly getting rejected by the examiner, paying for the patent to issue, etc.)

If you are trying to patent a novel aspect of it, then it will cost a lot more to write in the patent why it is novel and for discussing the novel aspects with the Examiner, including the filing/issue etc. fees.