Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

In theory? Yes. In practice? No.

For a lot of industries 20 years is an awfully long time to the point of complete irrelevance - and that "detailed information" is often vague enough to be unusable. Combine that with an overly-broad range of patent-able things and an overworked USPTO granting clearly invalid patents, and in practice (at least in the tech/software world) it's doing more harm than good.

Patents are no longer about protecting R&D investments. They have become more about patenting the vaguest concepts possible, in the hope of making a profit when someone else does an actual invention which somehow stumbles into your patents' wording.



Patents are no longer about protecting R&D investments.

Patents are always the tool of the moneyed and the lawyered, which is not the same thing as protecting R&D investment. There are exception, such as the guy who spent 12 years suing car companies over the windshield wiper, but that actually proves the rule because of time spent. It even apparently cause his marriage to break down. [1]

1. https://en.wikipedia.org/wiki/Robert_Kearns


To be fair, this appears to be a case of greed:

> Kearns sought $395 million in damages. He turned down a $30 million settlement offer in 1990 and took it to the jury, which awarded him $5.2 million; Ford agreed to pay $10.2 million rather than face another round of litigation.


Presumably, the point of a patent is to 'protect' the little guy which I would argued actually doesn't.

Not that I don't agree with you, what makes Kearns' pursuit 'greedy'?


He rejected $30,000,000.

Jury awarded 1/6 after litigation.

Ford payed double to get rid of this nuisance.

$30 million in 1990 was fuck you money. It’s still enough to comfortably retire on for a high school aged person.


Don’t forget that the lawyers probably took 33-40%, after costs.


He acted as his own layer, or at least, it's what Wikipedia reports


Inflation begs to differ. That much money left liquid for 60 years at 10% inflation will be worth less than 100k of today’s money.

Edit: sorry about the non sequitur. Bitterly learning about inflation and fiat recently.


I don't know when exactly it happened but you can be 100% sure anyone complaining about fiat currency has no idea what fiat currency actually is, has something to sell you, and if you don't get out soon enough you'll be standing there listening to them misunderstand Nixon at you.


Seems like kind of a non sequitur. Who would keep the money purely liquid? GP was pretty obviously talking about if it were invested, which statistically would have beat inflation using even pretty conservative options/funds. Using the 4% rule, it would have yielded $1.2m annually in 1990 value, and adjusted up for inflation every year thereafter. That's more than $2.8m/year in today's dollar.

Not only that, but in the odd scenario where it was "kept liquid", or kept purely in cash, it'd be worth $30m in today's money, not $100k. It's not clear where the $100k figure is even coming from. Even $100k/year doesn't make sense in any context I can derive.


What? That's not how inflation works. Even left to rot in a bank account it would still be 30 mil today. Not worth as much as 30 mil in 1990 but still worth as much as 30 mil in 2024 because it is still 30 mil.


The only way I can think of it making sense is if @syklep is from a country with a currency that has inflated 10% every year with respect to the dollar for the past 60 years, and that they assumed you'd buy local currency with the $30m when you get it and now would try to sell it back for dollars.

I don't think that currency exists though. The closest current example I can think of is if you're Argentinian and you bought $30m worth of Argentinian pesos in 1992 which would be worth ~$35k today. But that is due to relatively recent hyperinflation, not yearly 10% inflation, 10 years ago it would still be worth $4.3m, 15 years ago $8.8m.

You'd have to be seriously asleep at the wheel (or being strong-armed by local law enforcement, unfortunately) to keep your liquid currency parked in a hyperinflationary currency.


10% inflation? And left liquid and earning literally nothing?

I guess if he lit the money all on fire, it would be worth nothing, too.


You should learn some more.


$30m is still fuck you money today. Invest and live off the dividends and you’re pulling down an absolutely insane salary


Definitely agree — software patents should be abolished, full stop. They serve no purpose but to impede innovation and progress.

I’d argue the same applies to all patents; the world would be better off without them in general.


Maybe software patents require releasing the source code and last for a shorter term related to the pace of technological change. Then the incentive to share the advance is preserved instead of everything being a trade secret.


There’s not much in software that other people can’t figure out on their own though. Not too much social benefit in sharing.


I think Linux is a strong counter argument. And software might be simple to recreate, but it’s a shame to have so many smart folks spend time reinventing the wheel instead of building on each others work.


But the value in Linux is not the innovation but the sheer effort to (re-)create it. Patents don't protect that at all, copyright does (not that I am arguing for copyright, it also causes more problems than it solves).


Aren't the details of algorithms usually secret? E.g. my understanding is that the primary value that Tik Tok brings to the table is their recommendation algorithm.


The primary value of any social network is the network effect



I mean look at slides 8 and 9.

    There is no such thing as a software patent.
    There is no such thing as a rubber patent.
    There is no such thing as a steel patent.
    There is no such thing as an electricity patent

    There is only ... a patent.
OK fine, but there are patents covering applications of rubber, steel, electricity, and software.

So either you think "well it's not a software patent, it's a patent that covers the application of some software" is a useful point to make, or you're being pedantic for pedantry's sake. To the former, I don't think that makes a difference to someone being sued by a patent troll.


Is there a video that goes with those slides?


> overworked USPTO granting clearly invalid patents

Well no system can work that way - if judges in court were overworked and making invalid judgements, then the legal system would fail no matter what kind of laws you have and no matter what police does


These are pretty strange arguments. Why should an overworked USPTO lead to more patents? That assumes that the default is to grant the patent. If the default is to reject the patent, then an overworked office would not lead to more patents.

An overworked Supreme Court does not lead to more Supreme Court decisions.


> Why should an overworked USPTO lead to more patents? That assumes that the default is to grant the patent. If the default is to reject the patent, then an overworked office would not lead to more patents.

Former USPTO patent examiner here. I'll answer why an overworked USPTO will lead to more patents at present, but I make no claims that it should be this way.

The effective default is to grant patents. Why? Because the examiner has a finite amount of time for each application and has to have some sort of justification for a rejection. Unless there are some formal problems with the patent application, "I couldn't find prior art" means that a patent will be granted. Examiners could try "official notice" to basically say that they don't think it's novel or non-obvious without providing a reference, but that's easily defeated by attorneys. Examiners must provide a clear justification for a rejection.

If the amount of time an examiner has is too low (and it's far too low), that increases the chance that no prior art will be found, and consequently increases the chance that invalid patents will be granted.

Contrast that with the Supreme Court: The Supreme Court can decline to see a case. You can't do that as an examiner. You can try to have an application transferred, but that will just give it to another overworked examiner!


> The effective default is to grant patents. Why? Because the examiner has a finite amount of time for each application and has to have some sort of justification for a rejection.

The path of least resistance makes appearance once again. If we don't understand that this guides the default final state, we would argue about nothing constructive.


> Former USPTO patent examiner here.

Interesting. Did you enjoy it? Why did you stop?


There were some enjoyable parts, but I personally thought the job was quite stressful because of the high quota. Psychologically, I found it difficult to reduce my quality of work enough to meet the quota. Many examiners at the USPTO can knowingly reduce their quality to an appropriate level, but I wasn't one of them. (Others don't care about quality or aren't able to discern good quality from bad.)

I stopped for the simple reason that the job was far too difficult, and poorly paid on top of that.


I think we are mixing up two different ideas

Suppose an institution is overworked, it has two options - long queue or rush the job.

Courts understand their role is important, so you have a long wait, but they d0 the job properly.

Patent office, perhaps, rushes the job. Now whether they issue too many or too few patents is maybe equally bad, in my view, it’s screwed up either way.


The problem is, it is in the interests of the wealthy and powerful that both those things be true—that the USPTO be so overworked they can't adequately review most patents, and that the default be to grant the patent.

And the wealthy and powerful use that wealth and power to influence how government functions.

Thus, the current situation.


I think different patent lengths for different industries would do much to improve the current insanity.


I think full audited accounting of research costs should be included in the patent filing and that the patent should end either after 20 years or until profits from the invention (including payments from other companies licencing the invention) match 10x the costs.

And the maximum payout on violating a patent should be the remaining profit to end the patent.

Differences between industries then look after themselves.


Oddly all patent creators realize that extended Cabo vacations are the best way to brainstorm, and private Jet flights help with collaboration. Obviously only affordable by wealthy companies.




Consider applying for YC's Fall 2026 batch! Applications are open till July 27.

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

Search: