To the defendant who doesn't have $500,000 or $400,000 (maybe $1-10k is available), whether it's court or IPR seems like splitting hairs. It doesn't matter what kind of legal proceeding, because they can't afford to participate in any legal proceedings at this level anyway. They also can't afford to pay for more than superficial advice.
What I'm getting at is: $500k is more than most people's entire assets, and more than the assets in many small businesses and organisations. With software patents like this one, most infringements are likely done by the "little guy", who can end up threatened when they publish code, or have to operate in fear of that happening, if they know about it. They simply cannot challenge the patent through any legal method that has a fee like $500k.
If the rational closed-source company will pay Unity's hypothetical fee because it's less than the IPR fee, and the open-source defendant doesn't have any choice because they can't afford any of the options... who is left that is both motivated and capable to use the IPR procedure?
In the case of this particular patent by Unity, the techniques are things people thought of decades ago, but like most "inventions" of this kind, it's only when you have a specific project (typically paid) where they'd come in useful that you typically get around to using it. Nobody has time to document and explain every little "invention of the day" every time they think of a new technique just to feed the prior art corpus. I tried that once. When you've come to see how low the threshold is for patentability in software (and to be honest, in some digital hardware too), it's seems literally impossible to write down every patentable insight as fast as they come once you're on a roll; you'd need a support team just to do the documenting.
It's upsetting to think up a technique, flesh it out, not have a use for it for 10 years (you're busy, right?), then find one day when you want to use the fruits of your own figuring out 10 years prior on a project, it cannot be used any more. Even more so when it follows naturally from techniques dev communities have been discussing for years in the field.
(A number of us also have stories of publishing a sketch of a technique on some forum, only to find it patented later. It happened to me with a texture-mapping graphics technique in the 90s, where I found a later patent actually cited my Usenet post; in practice that meant I couldn't develop it further because the obvious next steps were blocked.)
jlokier, if you're assuming I'm in favor of software patents, you haven't been paying attention.
Developers have been writing screeds like this for almost 40 years, but nothing changes. That's because the people in power don't want them to change. Do you want to just keep complaining, or do you want to do something?
> if you're assuming I'm in favor of software patents, you haven't been paying attention
I wasn't assuming that, but I was assuming from the way you wrote it, that you think $500k is an ok price for people to continue practicing the software engineering they were already doing. It might just be a language interpretation thing from the way you wrote about that. And because of this:
>> That's not court; that's an IPR [1].
I consider almost all the interesting defendents to be people who will not be able to afford the IPR procedure, and the above seems a mere technicality from that perspective. For those people, it might as well be a court case in spirit. I.e. something they can't afford, that involves a panel judging something and corporate lawyers making arguments that determine someone's future.
So it seems only relevant to people who might be serious about doing the IPR. That's what gave me the impression you consider it feasible for typical defendants. But then:
>> As for defense strategy: you're right, but Unity might want a quick hit. They don't want a lengthy proceeding where they might lose.
You wrote "but", yet seem to agree with what I wrote on each point here, which amounts to nobody having the combination of motivation and capability to use the IPR procedure.
> Do you want to just keep complaining,
If you're assuming I'm complaining, you haven't been paying attention. I'm saying that when it appears to be suggested (possibly by way it's written) that it's an ordinary and reasonable $500k fee for a defendant to overturn a patent via IPR, that suggests to me (and probably others) that the author thinks of defendants as the kinds of companies or individuals with that level of funding capability.
I probably didn't state my case well, but the gist is two things: 1. I think the the vast majority of interesting defendents (infringers and potential infringers) don't have that kind of funding capability. 2. A genuine question (which hasn't been answered): Who would pay for the IPR, when neither the small-time defendant or the company who can pay the patent holder's fee has both motivation and capability to do the IPR?
> or do you want to do something?
You appear to be suggesting an alternative course of action than writing on HN, something to do with people in power, but I can't guess what "something" you have mind. Do you have something concrete in mind, or is that just a generic complaint on HN that there's too much complaining? :-)
You seem to think that explaining the system is the same as justifying it. For example, saying "it costs $500K" constitutes minimizing that sum or calling it ordinary and reasonable. Neither is true. I don't need to pile on the condemnations in every sentence.
As for what "something" is: I happen to be retired while you probably will be living with the patent system for a few years yet. So don't just say "someone should do something." Whatever that "something" is, you'll be more motivated if you think of it.
> The defendant's standard tactic is usually to file for IPR, or PTAB reexamination of the patent, which is like a trial without a courtroom.
For it to be the standard tactic, that implies a worldview in which defendants have that tactic as an option, and that's where the impression you think most defendants have access to that level of assets comes from.
But once you've explained the fees, it becomes clear that the typical defendant's standard tactic isn't to "file for IPR, or PTAB reexamination".
I'm not sure what the standard tactic is, but it isn't the one they can't afford.
When I've spoken with other patent lawyers, they've been skeptical that working engineers/devs regularly "reinvent" things covered in patents by accident. They have said they don't believe such people exist. Yet in my line of work, it's a near daily occurrence; it wouldn't even be possible to write them all down some days, and often it's things that were figured out years or decades prior. I know this with confidence because occasionally I look up techniques I've developed before that I want to use in some project that warrants it, and then I'm dismayed to find them recently patented, sometimes more recently than when I first developed the techniques.
So hopefully you can understand why I might have the impression patent lawyers (some at least) imagine that most "defendants" (people pursued for infringement or under threat of such) are the type of people they have as paying clients, and that other defendants (who I think are the majority) don't exist.
> So don't just say "someone should do something".
But I haven't said that. You came close, though, thus asking what you had in mind.
My argument here is not that someone should do something. It's that most defendants won't pursue IPR/PTAB reexamination because it's not an option for them, and companies with sufficient assets to pursue the reexamination are better off paying licence fees, so who is it that's both motivated and capable to pursue the IPR/PTAB reexamination? If there isn't anyone, it's not going to be overturned, is it, and most infringers (accidental or not) are just stuck with it even it would be overturned on reexamination, aren't they?
The answer to that is a missing piece in your explanation of the system, which without it seems to say that it's "standard" to end up reexamining these sorts of patents and overturning them if appropriate. I would be very surprised if that really is what happens, but also genuinely interested if that's the typical outcome.
OK, you're offended by the word "standard." I get that. If I said that criminal defendants' "standard" tactic is to stall and hope the witnesses forget about it (I don't know if that's true, actually), you would be offended because really poor ones can't get out on bail? I wouldn't say "too bad for them." The legal system is well-known to be stacked against poor people on almost every level, and even Charles Dickens was not the first to write about it.
This is Hacker News, not Patent Law 101. I'm not obligated to explain every aspect of the system in a comment.
I gather you think there's something new or unique in your comments about poor defendants. There is not. People have been saying it since the 90s. You are preaching to the already-converted.
Finally, if you read my paper on SSRN (which I've cited here many times), you'd see that "programmers' standard toolbox of techniques" is what I called out as something missing from the law.
That paper got cited in the Checkpoint amicus brief to SCOTUS in the CLS Bank case. That's an example of doing "something" but I will admit it's not much.
If you really wanted to do more, you'd start or join a political movement to "abolish software patents." There already are such organizations.
What I'm getting at is: $500k is more than most people's entire assets, and more than the assets in many small businesses and organisations. With software patents like this one, most infringements are likely done by the "little guy", who can end up threatened when they publish code, or have to operate in fear of that happening, if they know about it. They simply cannot challenge the patent through any legal method that has a fee like $500k.
If the rational closed-source company will pay Unity's hypothetical fee because it's less than the IPR fee, and the open-source defendant doesn't have any choice because they can't afford any of the options... who is left that is both motivated and capable to use the IPR procedure?
In the case of this particular patent by Unity, the techniques are things people thought of decades ago, but like most "inventions" of this kind, it's only when you have a specific project (typically paid) where they'd come in useful that you typically get around to using it. Nobody has time to document and explain every little "invention of the day" every time they think of a new technique just to feed the prior art corpus. I tried that once. When you've come to see how low the threshold is for patentability in software (and to be honest, in some digital hardware too), it's seems literally impossible to write down every patentable insight as fast as they come once you're on a roll; you'd need a support team just to do the documenting.
It's upsetting to think up a technique, flesh it out, not have a use for it for 10 years (you're busy, right?), then find one day when you want to use the fruits of your own figuring out 10 years prior on a project, it cannot be used any more. Even more so when it follows naturally from techniques dev communities have been discussing for years in the field.
(A number of us also have stories of publishing a sketch of a technique on some forum, only to find it patented later. It happened to me with a texture-mapping graphics technique in the 90s, where I found a later patent actually cited my Usenet post; in practice that meant I couldn't develop it further because the obvious next steps were blocked.)