US Senate to Revive Software Patents with PERA Bill Vote on Thursday
slashdot.org
slashdot.org
And at the same time the patent lawyer and troll lobby was feeling panicked and started spreading money around. Now we have the lame duck congressional period and we're seeing the results of their money.
- we should convene actual groups of "persons having ordinary skill in the art", who are not shown the patent application, who are presented with problems, and asked to describe a range of viable solutions.
- we should also use groups of such persons to judge whether two solutions to a problem (at least one of which comes from a patent application) are not meaningfully different.
Given these inputs, I think it would become much more expensive to determine that something is genuinely inventive (and applicants should bear that cost), but it should be possible to weed out a lot of obvious or overly-broad applications. If someone in the field who thinks about this for k weeks and arrives at "the same" answer, it wasn't that novel. I think it should also be possible for a group of "persons having ordinary skill in the art" to judge that an application is seeking something overly broad and that a PHOSITA should know that something is obvious, and incur some extra cost to the applicant.
However, if we could convince ourselves that stuff being protected was genuinely inventive, though we might be in a very different world, it doesn't seem crazy to me that e.g. a bunch of the work in ML research in the past decade should have been patentable because it wasn't obvious. It was repeatedly surprising! Of course, the current frenzy of activity might not be possible if it weren't for the spirit of openness that the field is adapted to. If the transformer was patent protected, who knows where we'd be. But it's really difficult to argue that concepts like the transformer architecture, or diffusion models or even something older like CNNs were "obvious" when experts, people in active research were surprised when they first arrived on the scene with the performance they had. And the don't seem obviously less worthy of protection than any of a range of physical inventions.
So suppose to apply, one needed to pay for market-rate hiring of a jury of PHOSITAs, and all applications became more expensive -- then you'd only apply for the things that are strategically important and very likely to be approved. Overall the number of applications should go down and the quality of granted patents should go up. I think the main equity issue here is that someone who figures out something genuinely inventive and valuable is less likely to be able to pay for the process. And perhaps, how do you vet that someone is a PHOSITA acting in good faith?
Similarly, I have no idea on what basis you think that 300,000 applications would have to be bogus. We are talking about all possible inventions. I find a lot of the rhetoric on HN does not respect the problem-space patents exist in, i.e. coming up with comprehensible textual delineations in order to properly categorize literally everything that has not been invented yet. If you do not respect the difficulty of that problem, you are going to have trouble coming up with realistic solutions that actually achieve the goals of the patent system, which is something that the US constitution requires congress to implement.
Personally, I don’t think the patent system at all serves the purposes stated in the Constitution and I strongly advocate for it to be amended, to limit it to very specific scopes (perhaps medical devices and drugs).
The purposes stated imply that the publishing of a patent is a good alternative to keeping things a trade secret, and that while you couldn’t copy and sell a patented item you could use a patent to learn from and that it thus advances the state of the art. I think in reality though, all software patents are nonsense legalese and used in massive d*ck-waving contests between large companies. Nobody successful studied existing software patents as a way to advance their knowledge.
Any approach that doesn’t acknowledge the absolute perversion of patent law to serve completely different ends than intended, especially in the software market, is not in my opinion a good faith argument.
I'm sorry but this kind of hyperbole is ridiculous. Do you invent? Are you involved in infringement contentions? What is your basis for this view? This isn't an appeal to authority, I'm just trying to understand from where you have gotten this perspective.
Having a granted patent is of limited utility when it's so generic as to be easily voided as one you described. I don't think these kinds of threats hold significant water to anyone seriously involved in patenting, and I don't think it stifles the ability for anyone to invent something much more useful. That latter contention is a bit ridiculous. Prior patents are no constraints on future inventions. It's only the practicing of a patent that leads to infringement.
>The purposes stated imply that the publishing of a patent is a good alternative to keeping things a trade secret, and that while you couldn’t copy and sell a patented item you could use a patent to learn from and that it thus advances the state of the art. I think in reality though, all software patents are nonsense legalese and used in massive dck-waving contests between large companies. Nobody successful studied existing software patents as a way to advance their knowledge.
This is the kind of thing I see here a lot. I'm not sure how connected it is to reality. My impression is that this sentiment is born of tech-heavy media, which likes to catastrophize the patent system. I saw someone upthread say that the "Cloudfront" decision was monumental, it wasn't. It was a regular day validity argument, the sorts of which routinely sink bad patents. It did not make the legal news.
There is a truth to the idea that there should be less bad patents. That is obviously the goal. But I noticed you didn't really engage with what I described as the problem space patent has to deal with. Instead, you arbitrarily drew the lines at software patents, because of how you* feel about them, which is something likely resultant of the media you read and not your professional experience, from what I'm gathering.
>Nobody successful studied existing software patents as a way to advance their knowledge.
This completely misses the point. They are in the public domain, they are in the body of knowledge. Saying that people don't read every published patent, so it's a stupid system, really only betrays the fact that you aren't thinking about it very deeply. The alternative is what you described, a world of trade secrets where none of this is known. Acting like patents are dumb because you instead read things in textbooks isn't the slam dunk you think it is when the alternative was not reading about it at all, ever.
>Any approach that doesn’t acknowledge the absolute perversion of patent law to serve completely different ends than intended, especially in the software market, is not in my opinion a good faith argument.
Absolute perversion? Give me a break. There's not much for us to discuss if you're going to just make huge emotional equivocations and then call anyone who disagrees "bad faith." I'm not sure what you think that contributed to this conversation. There's no shortage of people here making extreme statements about the patent system based upon their general ignorance of how it functions, what it grants, and how they are applied. I see tons of false assumptions conveyed with these kinds of statements, and I don't think they help.
I still don’t see them being used to add any value to the world. Your claim that they’re in the ‘public domain’ doesn’t disprove that unless (1) they’re really being used for a learning purpose and (2) those learning outcomes outweigh the waste and damage we know they they cause. I can’t prove that the authors who create the materials people do use to learn are not reading patents as a major source of information, but you haven’t proven that they are.
Patents, including but not limited to software ones, are actually used very disproportionately by the biggest players in an industry, for amassing a war chest of patents that, if you do a good enough job bamboozling a judge, many people could be said to infringe. The reason for that is of course to do battle with other big corporations, so that when they come for you, you can end it in a draw and cross-license a bunch of them. Meanwhile if someone small tries to compete with you, unleashing patent hell can kill them outright because they have no patents with which to countersue you. Really cool tool “for the inventor,” patents.
I do question what, exactly, we gain from having this. How is this better than companies keeping some things as secret. (Let’s also just appreciate the absurdity of “secret methods” when it comes to software. We all know how software does things anyway. Actually implementing it well is hard. And we have copyright to protect code.)
You don't come across as reasonable, so I don't take it as much of a slight.
>I can’t prove that the authors who create the materials people do use to learn are not reading patents as a major source of information, but you haven’t proven that they are.
It's in the public domain. There's no disputing that it is in the body of human knowledge. That you argue something like this just shows you are not being reasonable.
>Patents, including but not limited to software ones, are actually used very disproportionately by the biggest players in an industry, for amassing a war chest of patents that, if you do a good enough job bamboozling a judge, many people could be said to infringe.
Of course, because no judge could ever reasonably be convinced that a patent infringes. It's only by bamboozling! All the changes to the patent system that have made it pro-big-tech come from the media you consume and the changes their funders put in place. PTAB is a great example.
> Meanwhile if someone small tries to compete with you, unleashing patent hell can kill them outright because they have no patents with which to countersue you.
LOL, get off tech media if this is something you think is actually an issue.
> Really cool tool “for the inventor,” patents.
You're just advocating that the patent system should be more pro-inventor, not that it should disappear. I wouldn't disagree with that at all!
Ostensibly the patent system exists to incentivize open publishing of new technologies, but it isn't worth squat if full of vague language written by lawyers.
That said... it would at least guarantee that you put in the minimum effort to actually do it, rather than just sitting on the idea and waiting for somebody to run afoul of your patent. And maybe we'll get a few patent examiners saying, "Explain again how a five-line chunk of code is patent-worthy?"
https://www.congress.gov/bill/118th-congress/senate-bill/214...
Looks pretty dead. Only 1 cosponsor and no action taken in a year.
Perhaps this one: https://www.congress.gov/bill/118th-congress/house-bill/9474
> Perhaps this one: https://www.congress.gov/bill/118th-congress/house-bill/9474
The OP/EFF says it's a Senate committee that's about to vote on "The Patent Eligibility Restoration Act (PERA), S. 2140."
I don't know tons about the government, but you linked a House bill that looks like it's still in a House committee, and I don't think a Senate committee would be talking about something like that.
Edit: I found this: https://www.judiciary.senate.gov/committee-activity/hearings...
It's an agenda for a Judiciary Committee meeting on Thursday that lists "S. 2140, Patent Eligibility Restoration Act (Tillis, Coons)."
I don't know where they got the information, as it is not mentioned in the webpage:
https://www.congress.gov/bill/118th-congress/senate-bill/214...
More recent: https://www.congress.gov/bill/118th-congress/house-bill/9474
These are committee votes, not the whole Senate.
Maybe big pharma and biotech should realize that large scale patent abuse is making people question the whole thing and start lobbying against software patents.
> I was just thinking the same thing. I used to just sort of accept that patents were a reasonable and necessary thing. Then after being in the actual business of software for a while as a startup entrepreneur and seeing the complete mess spurious patent claims cause, how patents are so often deployed by mega-corps to defend against startups, and the huge number of grifters the patent "industry" attracts, it's made me more open to the idea that just maybe the whole idea of patents is now more trouble than it's worth.
IMHO, that's taking it too far. IP is a solution that solved certain problems; but it's been so effective that those original problems are so alien that people overreact to the imperfections of the solution in extreme ways (like, "let's abolish it all").
So far as I can tell, the problem IP laws originally solved was that wealthy people wanted another way to create monopolies and extract rents. For that purpose, they certainly are effective.
It takes ~$1B to elect a president and millions for many of senators and congress(wo)men.
US congress's median age is 58.8 and median wealth is about a million dollars. Many are career politicians who are far removed from the lives of median Americans.
Remains to be seen what the Trump administration passes that helps the middle and lower class.
I'm submitting a patent for showing colors on the internet
The super-intelligent shade of blue, or more properly, Hooloovoo, is exactly what it says on the tin. Whether the creature is intelligent in general or just compared to other, lesser shades of blue is left ambiguous.
It can:
Refract in a prism (for special occasions). Turn purple when disgruntled about something. Meaningfully observe and interact with its surroundings, or it wouldn't be part of greater galactic society. So we're looking at some kind of energy creature that behaves like light, and has to be blue for some reason. Though there isn't much literal anatomy to speak of, I would like to know if you could possibly build a creature out of material that can pass through glass, and have it be both intelligent and equipped with senses.
Can light interact with light?
https://techrights.org/o/2017/08/31/iancu-nomination-and-sof...
Let's hope that with Musk it's gonna be different, but I have some doubts.
Edit: Got thrown off by the comments, that makes more sense.
The strategy might be to introduce it now, and use the fact that it failed the Democratic-lead lame duck session to sell it to the people in the next Congress.
Frankly, most legislation that favors the interests of wealthy donors is bipartisan.
For instance, the legislation that rolled back the banking reforms put into place after the financial crisis during the first Trump term.
> the bill, which was years in the making, was a rare bipartisan accomplishment at a time when Congress is gridlocked on almost all major issues.
Sixteen moderate Senate Democrats helped Republicans pass the bill. It was an unusual moment of political unity that sparked a public feud in which the Democratic Party’s progressive wing went to war with its more business-friendly centrists.
https://www.politico.com/story/2018/03/14/senate-passes-bill...
The moderate Democrats and the establishment Republicans have the same donors.