[virtuowl]

I always feel that patents or at least the way patents work is really bad for humanity as a whole and hindering our progress. On the other hand there has to be some way for inventors to benefit, are there any good alternatives?

[rayiner]

Have you ever worked in a field where patents are an important part of the market structure? Telecom, pharma, etc.?

I spent my engineering career in telecom, and patents are a key part of the model. Those industries don't have the equivalent of a Google or an Apple that can bankroll expensive R&D on the back of massive consumer revenue. They're no unicorns or network effects where you can take a 20% cut on every app that gets sold to your captive user base. Instead, companies pay teams of very expensive PhDs to develop things like error correction systems for modems. They patent those things, use the patents to either extract licensing revenue or, more importantly, keep out free riders who didn't contribute to developing the technology. Then it's all obsolete in five years and they do it all over again.

Markets are good, and patents, as imperfect as they are, provide an infrastructure for integrating expensive R&D efforts (which don't result in tangible property), into markets, in a way that avoids free-riding. And it also results in market structures that are sane. I can just go buy a Broadcom Wi-Fi chip. Broadcom can recoup its investment by just selling me a chip, and count on patents to keep out free-riders. Their business model doesn't require them to make a whole consumer end-product like a phone, or tie their hardware to advertising, to be viable.

In the absence of patents, you get trade secrets and vertical integration. Apple doesn't care about patents because it doesn't sell you chip designs or even a chip, it sells you an iPhone. Google doesn't care about patents because it doesn't sell VP9 or Blink--those are just a means for furthering its advertising business.

Take a step back and think about the business models you want to see more of. Do we need more Apples and Samsungs, which want to sell you a complete end-user product tied to their ecosystem? Or do we want more ARMs, which are happy to just sell you a chip or even an IP core? Wouldn't it be nice to bring back Netscape? Where you could just exchange money for a web browser, instead of going to google.com and it bugging you to download Google's web browser for "free?"

[monocasa]

> Those industries don't have the equivalent of a Google or an Apple that can bankroll expensive R&D on the back of massive consumer revenue

What are you talking about? The biggest players in the telecom market report some of the largest profit in the country, both in real terms and as a percentage of revenue. They're just so entrenched that they view any R&D as a cost center if they can't use it to keep others or of the market.

[rayiner]

If you look at a pure telecom company like Verizon, they're at $13.6 billion in profits (for 2017) on $125 billion in revenue. Apple is at $50 billion in profits on $229 billion in revenue (for 2017). So Apple's margin is almost double.

Also, "consumer revenue" is the key word. The biggest R&D players in telecom don't sell to consumers. Intel and Qualcomm sell chips (and ARM just sells you IP). They don't make the chip in house and sell you a complete phone. At the same time, a company like Verizon or Comcast that sells to consumers isn't vertically integrated and doesn't do its own R&D. They buy all their equipment from other companies that do the R&D. You don't have vertically-integrated company that uses consumer revenue to bankroll the R&D. You have specialization and market transactions mediated by, among other things, patents.

Contrast web technology, where there’s no patents and no markets. Nobody sells browser engines. There is no specialization. It’s all driven by Apple, Microsoft, and Google, who invest in the R&D to further their consumer-facing platforms.

[monocasa]

Verizon (and most telecoms for that matter) plays games by being hundreds if not thousands of tiny partnerships, which obfuscates their profit.

[rayiner]

Verizon is a holding company. It's structured as a holding company with numerous subsidiaries because it's got physical property all over the country, and is subject to literally thousands of different regulatory regimes: federal, 50 states, thousands of city and county utility boards. But its profit figures are reported on a consolidated basis. (And if you have hard evidence to the contrary, man do you have a valuable lawsuit on your hands.)

[monocasa]

It's setup in slightly different ways for different niches in the market, but on the mobile side it's a holdover from original FCC regulations on how ownership of spectrum worked. The FCC wanted to encourage competition, so they heavily emphasized partnerships with little mom and pops for each local region of spectrum. After the FCC removed those regulation requirements, those partnerships were still useful as sort of bellows of profit so that the C level can manage their growth curve.

This is all not really private information, is how pretty much all of the telecoms are structured, and isn't illegal.

[natch]

>What are you talking about?

I suspect what the poster is talking about is telecom equipment manufacturers, not service providers. Yes service providers reap obscene profits. Equipment manufacturers are typically much more challenged by competition and commoditization.

[monocasa]

It used to be that the service providers did massive amounts of R&D. Hell, Bell labs was one of the most impressive R&D laboratories of all time. The service providers just decided that it was better to buy themselves protective laws, and to squeeze out any R&D than to try to actually compete.

This isn't a sign of how important patents are, but a sign of how the extreme rent seeking behaviors of the telecoms greatly manipulated many aspects of the market that they're in.

[rayiner]

AT&T (of which Bell Labs as a subsidiary) was a vertically-integrated nationwide monopoly that bankrolled its R&D using its massive consumer revenue. Most people would consider the current market structure, where service providers are separate from equipment manufacturers, to be an improvement.

[monocasa]

And yet, it was the fiscal year of 2017 that has had the historical highest profit margins, not their years of (highly regulated) national monopoly.

https://www.thestreet.com/story/14243906/1/how-at-amp-t-mana...

[dctoedt]

> patents, as imperfect as they are, provide an infrastructure for integrating expensive R&D efforts (which don't result in tangible property), into markets, in a way that avoids free-riding

Preventing free-riding is just one means to the constitutional end of promoting the useful arts; it's one that imposes tremendous transaction costs on society. Another means to that end is for the law to say to innovators, what you've done here is nice, but your bringing it forward simply doesn't provide enough net societal benefit to warrant legally prohibiting your competitors from copying your work — so if you want to keep whatever competitive advantage your innovativeness gives you, then you'd better keep innovating, because your competitors might be gaining on you. (Pace Satchel Paige [0] and Thomas Jefferson [1].)

When a patent is issued, most of the time it's a single, junior civil servant who literally makes legally-binding industrial policy for the entire nation. At a minimum, patent applicants seeking such a private industrial policy should have to do, and document, a thorough prior art search, in the same way that Ph.D. candidates must do and document a literature search for their dissertations. The existing patent examination process is like requiring a Ph.D. candidate's adviser to do the literature search, and requiring the degree to be conferred if the adviser doesn't turn up anything — and of course in the patent world there's always the factor that "the antlike persistency of solicitors has overcome, and I suppose will continue to overcome, the patience of examiners, and there is apparently always but one outcome." [2] A false-positive issuance of a patent has somewhat-greater ramifications for society than a false-positive issuance of a doctorate.

[0] "Don't look back. Something might be gaining on you." http://www.satchelpaige.com/quote2.html

[1] In a famous 1966 opinion, the Supreme Court, citing Thomas Jefferson, said: "The patent monopoly was not designed to secure to the inventor his natural right in his discoveries. Rather, it was a reward, an inducement, to bring forth new knowledge. The grant of an exclusive right to an invention was the creation of society—at odds with the inherent free nature of disclosed ideas—and was not to be freely given. Only inventions and discoveries which furthered human knowledge, and were new and useful, justified the special inducement of a limited private monopoly. Jefferson did not believe in granting patents for small details, obvious improvements, or frivolous devices. His writings evidence his insistence upon a high level of patentability." Graham v. John Deere & Co., 383 U.S. 1, 9 (1966), https://scholar.google.com/scholar_case?case=910565259149730...

[2] Lyon v. Boh, 1 F. 2d 48, 50 (S.D.N.Y. 1924) (Hand, L., J.), https://scholar.google.com/scholar_case?case=969659756696519..., rev'd on other grounds, 10 F.2d 30 (2d Cir. 1926), https://scholar.google.com/scholar_case?case=121904095703982...

[ScottBurson]

> what you've done here is nice, but your bringing it forward simply doesn't provide enough net societal benefit to warrant legally prohibiting your competitors from copying your work

The interesting thing to me about this is that it is fundamentally not a legal judgment, but an economic one. I think patent examiners should be required to have some background in economics along with whatever legal training they get (and, of course, their subject matter expertise). Similarly for judges deciding patent cases.

Anyway, as I've posted here on HN occasionally, I think the problem of figuring out at patent application time whether the invention is obvious or not is too hard, particularly for a junior civil servant. Instead I propose that the first step in an infringement proceeding should be for the patentee to supply objective evidence, along the lines of the Graham factors, of nonobviousness. The burden of proof should be on the patentee, and the court should evaluate this evidence on its own, without deferring to the examiner. What do you think of this idea?

[dctoedt]

> The interesting thing to me about this is that it is fundamentally not a legal judgment, but an economic one.

It's a mix of both. Unfortunately, too many patent lawyers seem to subscribe to the natural-rights theory of protecting inventions: If you invented something, why then just naturally you should be able to prevent others from making, using, or selling it without your consent — as though conceiving and writing up the idea was the most important part of bringing the benefit of the idea to the public. Too often, the other facts involved in effective innovation seem to take a back seat.

(In the 1980s and 1990s, at meetings of the ABA Section of IP Law, I don't know how many times I'd hear Don Banner [0], a former chair of the section and former commissioner of Patents and Trademarks, hold forth on the importance of protecting "the little man from Little Rock," i.e., individual inventors.)

> The burden of proof should be on the patentee, and the court should evaluate this evidence on its own, without deferring to the examiner. What do you think of this idea

I think that would be the sounder public policy — keeping in mind of course that the patent examiner is a neutral expert.

[0] https://en.wikipedia.org/wiki/Donald_W._Banner

[ScottBurson]

> I think that would be the sounder public policy

Excellent! I think you're the first person here who has agreed with me :-) It seems like a subtle tweak, perhaps, to many people, but I think it would make a huge difference. Rayiner thinks it's unnecessary after Alice, but I'm not convinced; subject matter is a different aspect of the problem. (Also, though I welcome the effects that Alice is having, I'm not sure I think it's correct. The problem is that word "abstract". As software engineers, we know that abstraction is a hierarchy, and the abstract/concrete distinction is not absolute but is relative to the level within that hierarchy that one is speaking at. Any idea, in short, can be argued to be abstract.)

> keeping in mind of course that the patent examiner is a neutral expert

A neutral expert, but also one whose incentive structure encourages them to close cases. They don't get bonuses for fighting off obstinate applicants with bad patents.

[_rpd]

Require mandatory licensing of all patents on 'fair and reasonable' terms (as decided by some sort of market). Require that inventors get a share of the licensing fees (non-transferable until valued by the market).

[BraveNewCurency]

The point of patents is to entice innovators to PUBLISH their inventions.

A patent should be about some process that is so complex that nobody else has figured how to do it before (i.e. some secret trick). When it's published, humanity is better off by publishing it now (even if it's locked up for a little while) so people can investigate related solutions (some of which might not infringe).

But the patent office is allowing patents on obvious things, so the public gets no value in their publication. (Actually negative value, since people have to waste time reading them and companies are afraid to develop products).

[philipov]

Don't allow corporations to own patents. Limit their ownership to individual humans.

[NeonVice]

How exactly would that work when I work for a corporation who is paying me to invent? If I am successful in my invention, then I just take my invention and run. This would stifle research investment.

[monocasa]

This is how it works in Germany. Patents aren't transferable in Germany, even from employee to employer, but you can licence them to your employer.

[walterbell]

Shared ownership? Both parties take risks.