Live data from Hacker News

Delaware judge discovers hidden entity recruiting people to be patent trolls

ipde.com

161–170 of 360 posts

Re: Delaware judge discovers hidden entity recruiting people to be patent trolls

#161
post #48

Earlier quoted context omitted.

But real inventions can occur in software too? Inventions that takes serious money, expertise and so on. Some examples: - Rust memory management - New models in machine learning that enables amazing results - Protocols

I'm skeptical rust memory management is such a big invention by them and would bet some money the ownership model already existed before in papers at the very least. And even if not just imagine having to pay Dahl or Alan Kay to implement OO in your language, which was by every possible metric more innovative than rust's memory management. Even if it would have stopped java from being a thing it's just not worth it.

Those things got invented anyway though, without being patented. The right question, in my opinion, is whether there are important software innovations that would have been significantly delayed, or kept secret, or never discovered at all, if software patents weren't a thing.

And if so, do they outweigh legal costs and chilling effects of patent trolling, and the inefficiencies of people having to engineer around patented ideas.

Intellectual property is a fairly artificial concept. It's quite "big government" when you think about it - government grants a monopoly on the application of an idea. It is only worth preserving if it makes the country richer.

Re: Delaware judge discovers hidden entity recruiting people to be patent trolls

#162
I'll take a very unlike myself opinion in this - I'll support patent trolls, but only if their exploits are increased by a thousand fold and punishes the entire patent system for being incredibly inept, antiquated, and against any and all forms of innovation. We need serious and far ranging complete patent reform (or revolution) and if deliberate brute-force use of the horrendous laws and regulations that have allowed those entities to proliferate over the last several decades finally forces changes, then bring on more of them.

Short term they are horrendous, long term we may need them to force positive changes that benefit and are on the side of small-time inventors and problem solvers, and not the mega-corporations (patent trolls with massive capital).

tl;dr: Patent trolls are bad, the patent system that enables them is bad, we need to force massive patent reform.

Re: Delaware judge discovers hidden entity recruiting people to be patent trolls

#163
post #136

Earlier quoted context omitted.

What? What you are proposing has nothing to do with the perfect vs the good, it is a massive fantasy completely ignorant of how things actually work. Fees are merely a rounding error in the overall cost of patents, prosecuting (obtaining) them, pursuing cases against "violators" or defending them. The fees are in the $hundreds to small $thousands of dollars [0]. It typically costs $20,000 to $50,000 in patent lawyer…

Help me understand how cost scaling doesn't fix the problem of "players with money can simply massively burden the minor players, often to bankruptcy." If it becomes more costly for players with more money to burden minor players, then would they not burden minor players less?

For the small players, you aren't reducing the costs

This includes but is not limited to 1) the costs of the courts. 2) the costs of the attorneys to defend, 3) the costs of the technology and executives to mount a proper defense, 4) the opportunity cost to the small defending company which COULD OTHERWISE be focusing it's resources on something productive like a new product or support (vs defending a bogus lawsuit).

Again, the courts do not even begin to scale - the courts are the opposite of scalable.

The new class of enabled patent trolls may lose one case, but they'll win others, possibly because they've bankrupted their victim, and can continue to plague the rest of society.

Moreover, the courts are massively inconsistent. Again anti-scaleable. The inconsistencies get worked out through appeals to higher courts. This literally takes decades and hundreds-of-thousands to millions of dollars per case. They do not get worked out when it is state cases, which is why we have already have venue shopping (you know about the East Texas patent scam courts, right?).

And your so-called "solution" to charge big players orders of magnitudes more cannot be gamed? Simple, make a small company, pay the small fee, buyout later. Or syndicate the fees, or get rounds of investors to handle the fees, since the return is virtually guaranteed. Ya, then you change the fee structure (years later) and the game begins again.

The fees are only a rounding error in the costs, and if you think exponentially higher fees cannot be gamed, I'd like to talk to you about a fantastic deal on an oceanfront property in Kansas, because you are so friggin gullible.

No, automatically granting patents with some weird scaled fee structure and letting the courts sort it out is one of the worst ideas ever. You would literally lay waste to entire sectors of innovation. It would be only a few years before people would soon be screaming for proper centralized regulation and a patent office capable of judging obviousness and prior art; congratulations, you've just re-invented what the founders invented nearly 250 years ago.

Seriously, you are making a nice demonstration of how ignorance of how a real system works creates the illusion of finding wonderous solutions.

For every highly complex system problem there are a huge variety of simple solutions, all of them wrong. Congrats, you just found one.

Re: Delaware judge discovers hidden entity recruiting people to be patent trolls

#164

Current USPTO patent examiner here. The most effective way to eliminate bad patents would be to give examiners more time, say double the time across the board to start. If an examiner can't find prior art in the little time they're given, and they have no other reasons to reject the application, they'll have to grant it. The amount of time was (basically) set in the 1970s based on data from the 1960s. There have been…

> And while search technology has improved, it hasn't become orders of magnitude better. Are you saying that search technologies of 2022 are not orders of magnitude better than the search technology in the 1970's? Search tools in the 1970's were basically manually panning through microfiche. This was the time before even relational databases were commercially available, let alone full text search across thousands (or…

To be clear, I do think that patent searching has improved. My point is that it hasn't kept up with the shear number of documents available.

You have a point about the switch to computerized searches. Searching by examiners used to be looking through paper files, not microfiche, though. But patent search technology hasn't improved appreciably since full text patent search was introduced in the 1990s. There have been a lot of new documents to search in the past 25 years alone, yet there have been few changes to examining time.

Also, full text search is probably not as useful as you think for patent examination. For many technologies it's the best way to search, but for many others (like a lot of what I examine), text search is much less useful than simply flipping through a ton of documents, looking at the drawings. It's common that important details exist only in the drawings or are most easily spotted in the drawings. So this limits the improvement one gets from switching to computerized searches.

And, there's a big advantage to paper documents that wasn't maintained during the switch. From what I'm told there were a lot of useful notes from previous examiners written on the patent documents. Those notes were entirely thrown out when the USPTO switched to computerize searches. They could have digitized them, but USPTO upper management isn't known for thinking ahead. That's a loss of a lot of institutional knowledge. At present there is no way for examiners to share margin notes, and there should be.

(Again, like my other comments here, this is just my opinion, not that of the USPTO or US government.)

Re: Delaware judge discovers hidden entity recruiting people to be patent trolls

#165

Earlier quoted context omitted.

But that's not how invention works : what about advances that seem trivial after the fact... yet nobody has thought of for centuries, despite all the "ingredients" being available ?

What about patents are non-transferable and only valid, if the issuer actually makes use of them? The moment you let them rot in a drawer, you invalidate the patent. That would disable companies whose sole purpose is to "troll" others with patents they aquired from whomever, and it would also disable companies from inventing something just to block off this path to competitors, while never making use of it itself to…

I give one example of 18th century misuse (or rather, too broad of a patent granted) in a parallel thread :

https://news.ycombinator.com/item?id=33505728

Some decades later, however the (much less broad ?) patents over steam engines able to produce circular motion, seem to have forced Watt to differentiate and improve his "2nd engine" so much that it could be differentiated enough from the other patents :

https://technicshistory.com/2021/10/10/the-steam-revolution/

Note also that Watt was bad at business, and only owned 1/3rd of "his" patents.

Our post-modern issues seem to me to instead stem first from :

- it being too easy to shelter liability (and tax fraud !) behind complex company trees

- a failure of antitrust and lack of goal/duration-limited companies allowing companies to grow waaaay too big to the point where they have enough power to heavily weigh on governments (note though that Watt had already managed to convince the British government to give a 25 year extension on "his" "1rst" steam engine)

Re: Delaware judge discovers hidden entity recruiting people to be patent trolls

#166

Earlier quoted context omitted.

IANAL, and was not proposing specific legislation. I used "shell company" in colloquial sense - a token / minimal / empty legal organization, used by lawyers & their ilk for what most honest & reasonable non-lawyers would see as duplicitous or malicious purposes.

The corporations in this case a) hold assets and b) provide liability limits. That is the basis of all corporations. So we’d need to get specific about which practice is duplicitous. Is it the treatment of IP as an asset? The transferability of patents ? Limited liability of patent defense? I am no expert in any of those things but I bet there is value in each and bad unintended consequences of each. If I described t…

>If I described to a lay person that there was a hacker website frequented by lots and lots of people with an active forum community, you’d likely get at least some of them to suggest banning it…

Different things. In this case you have a pathological outcome in the process. "Free real estate" as it were because you've got an organizational structuring that is essentially censure proof by the Court unless you pierce the corporate veil.

Remember, corporations are suffered to exist as a courtesy, and a means to distribute risk, but our Court system is very much tuned to "there is a person, get them in here now, and lets get this sorted."

If the Court has to choose between being ineffectual and exploitable, and piercing the veil in order to get to the bottom of duplicitous behavior, one can only hope the Judge will settle in and push the issue.

Re: Delaware judge discovers hidden entity recruiting people to be patent trolls

#167

Earlier quoted context omitted.

The main problem is searching for prior art. There are AI search tools but they usually aren't good in my experience. It's rare that an AI search tool will return prior art that can be used in a rejection. Usually the prior art these tools find is related but not good enough. But sometimes (perhaps through random chance), a solid reference can be found. I found a 102 reference (the strongest kind) a few weeks ago for…

Are synonyms not built in to the search tool itself? "car" also finding "automobile" etc?

I have no idea as these tools rarely explain how they work. There are a lot of AI patent search tools and I'm sure each of them works differently.

Also, don't underestimate how difficult finding synonyms is. Many examiners, myself included, keep lists of search queries with a lot of synonyms to use later. I've been doing this for nearly two years now and my saved search queries keep growing. I don't expect this to end anytime soon.

(Again, like my other comments here, this is just my opinion, not that of the USPTO or US government.)

Re: Delaware judge discovers hidden entity recruiting people to be patent trolls

#168
post #90
post #78

Earlier quoted context omitted.

Every patent should be automatically granted and the courts can sort out the winners if there is a dispute. That is the system we have moved to.

> the courts can sort out the winners if there is a dispute. This will skew the issue towards deeper pockets. whoever has more money for court battles will win the patent case. Is that what we want?

That's already the reality.

Re: Delaware judge discovers hidden entity recruiting people to be patent trolls

#169

Current USPTO patent examiner here. The most effective way to eliminate bad patents would be to give examiners more time, say double the time across the board to start. If an examiner can't find prior art in the little time they're given, and they have no other reasons to reject the application, they'll have to grant it. The amount of time was (basically) set in the 1970s based on data from the 1960s. There have been…

I think a much more efficient method would be to make the rule that if a citizen finds prior art more than 5 years older than the filing of the patent, then the patent holder must pay $1000 to the citizen and the patent will be invalidated.

If the patent holder does not pay within 6 months, then all the patent holders patents since then will be invalidated.

It has to function without involving the courts as they will bleed anybody who are no rich.

This is simple, gruesome, but effective. I just can't imagine the politicians wanting this efficiency.

Re: Delaware judge discovers hidden entity recruiting people to be patent trolls

#170

Current USPTO patent examiner here. The most effective way to eliminate bad patents would be to give examiners more time, say double the time across the board to start. If an examiner can't find prior art in the little time they're given, and they have no other reasons to reject the application, they'll have to grant it. The amount of time was (basically) set in the 1970s based on data from the 1960s. There have been…

I think a much more efficient method would be to make the rule that if a citizen finds prior art more than 5 years older than the filing of the patent, then the patent holder must pay $1000 to the citizen and the patent will be invalidated. If the patent holder does not pay within 6 months, then all the patent holders patents since then will be invalidated. It has to function without involving the courts as they will…

Outsourcing of justice starts to be really popular in US I see.
Post reply on HN