- Peter Finnie: UK Innovation Is Under Threat Because Of Archaic Patent Law
- Eibhlin Vardy: Is There A Role For The Tort Of Unlawful Means In Patent Law?
- Alden Abbott: Patent Reform Is Critical To Protecting American Inventors
- Gene Quinn: USPTO Admits To Stacking PTAB Panels To Achieve Desired Outcomes
- Bruce Berman: Blockchain Patent Publications Picked Up Speed In August
Get a Job doing Patent Law
MM, just in case you missed this: link to nytimes.com
Leftists eating themselves. More news at 11.
I did see that. A funded think tank is ethically challenged because it wants money! So shocking. Nobody could have predicted that would ever happen because it’s never ever happened before.
I’m not sure how many times I have to repeat this but — because I know some people are still confused — I’ll repeat it again: I have no love for Google the Megacorporation and if it were up to me there’d be no companies with that amount of money anywhere on the planet. Unfortunately, Google isn’t unique. Not even close.
Also, the patent bar’s focus on Google as some sort of ultimate arbiter of everything is very silly. I’ve said this before a million times, too: Google could disappear tomorrow and nothing at all would change. The same issues plaguing the patent system now will remain and same tiny tiny group of entitled patent worshippers will find another entity to whine and cry about it.
Greg Subject matter eligibility for software is a good policy choice, in my opinion.
Why should logic be patentable? What’s the policy argument for patenting logic? Why was the awesomeness of patenting logic seemingly missed by patent applicants until very recently?
I’ve yet to hear a remotely convincing policy argument. “Keeps attorneys busy” and “makes it easier for rich people to get richer without getting dirty” aren’t great rationales. “It’s just the way it is and we don’t want to upset these entitled loud-mouthed rich people” is even worse.
So what’s the rationale?
Why was the awesomeness of patenting logic seemingly missed by patent applicants until very recently?
Some technologies require other technologies to arrive first. It did not make sense to invent cars until there was a network of roads on which to drive them. So to, it did not make sense to invent software until there was an ample supply of processors on which to run the software.
What’s the policy argument for patenting logic?
The genius of the patent system is that it does not require a central planner to direct investment for innovation. You create a reward system for developing new technology, and then you stand back and watch what technology is invented. It only makes sense to exclude categorically a given technology from patenting if you can know in advance that there are no innovations in that field that will be had with patents but not without patents. Given that we cannot know that about software (or at least I do not know that about software), I am inclined to include software within the realm of eligible subject matter.
I take your point that this stuff is not easy to search, but that really cannot be an argument for categorical exclusion. Software is no harder to search than industrial designs, and yet design patents have not brought great hardship on our society.
If there is a problem with software patents, it seems to me that the problem lies with the functional nature of so many software claims. The right remedy for this problem (to my mind) is to tighten up on §112(a) & (b) requirements, not on §101. The most sound policy choice for §101 is to leave it as a “coarse filter.”
>>If there is a problem with software patents, it seems to me that the problem lies with the functional nature of so many software claims. The right remedy for this problem (to my mind) is to tighten up on §112(a) & (b) requirements, not on §101. The most sound policy choice for §101 is to leave it as a “coarse filter.
I agree. But I struggle to see how it could happen.
I struggle to see how it could happen.
Same here. Now that the SCotUS has turned this into a §101 issue, it will be very hard to get the law back on track. Honestly, given the Court’s historical prediliction to resolve all patentability disputes on extra-textual grounds (Hotchkiss, Kalo, A&P Tea, Cuno, Myriad, Mayo, Alice, etc), I have dim hopes that any amount of statutory revision can right the ship unless and until the SCotUS is relieved of its jurisdiction over patent cases.
It only makes sense to exclude categorically a given technology from patenting if you can know in advance that there are no innovations in that field that will be had with patents but not without patents. Given that we cannot know that about software (or at least I do not know that about software),
How many sets of instructions for carrying out logic tasks by machine were written before the patent system was involved? Anything of value? How many new sets of instructions are written now without patents? Any instructions of value? What’s the ratio of new unpatented instructions to patented instructions? Where do skilled instruction writers stand on the issue of patenting logic?
If there is a problem with software patents, it seems to me that the problem lies with the functional nature of so many software claims.
Well, we’re talking about patenting instructions written for computers that were designed to carry out instructions. What sort of structure do you contemplate being described in the claims and how are those structures going to be evaluated in view of prior art structures? “If X then Y1” is in the prior art. How do we determine whether “if X then Y2” is obvious in view of that art? There are no unexpected results here. So how does it work?
Fundamental questions. Nobody wants to face them, we’re several decades into this “experiment” and the system is rotting from the inside as a direct result of it. Money that could be invested in all kinds of useful progress is instead being invested in cr @p “apps” for determining whether there is a hotel room within five miles with your favorite color of drapes.
Meanwhile the worst and least educated patent attorneys that ever lived whine and cry about how “confusing” it all is and insist that their precious entitlements were given to them by “Nature”. What in heck is going on here?
Your “fundamental questions” are disingenuous canards.
It is NOT that “nobody wants to face them.”
Rather, it is that YOU do not want to face the dialogue in ANY sense of inte11ectual honesty.
it is that YOU do not want to face the dialogue in ANY sense of inte11ectual honesty.
Says the super serious deep thinking guy who won’t admit that software is instructions for carrying out logic.
Keep the laughs coming.
Software is a “ware” and patentably equivalent to other machine “wares.”
It is a manufacture by the hand of man and just because it is written, does not make it any less patentable.
As YOU yourself have volunteered an admission against your interests in the past as you have acknowledged that you know and understand the exceptions to the judicial doctrine of printed matter.
Your “so serious-keep the laughs coming” poker tell of dissembling f001s no one.
Thanks for proving my point.
You had NO point to prove, thus, your statement is its usual non-sense.
Maybe you want to discuss that admission of yours…? I could point out (yet again) that simple set theory explication on how the exceptions to the judicial doctrine of printed matter work – even if you have admitted to knowing how that bit of controlling law goes.
What sort of structure do you contemplate being described in the claims and how are those structures going to be evaluated in view of prior art structures? “If X then Y1” is in the prior art.
Correct, “if X then Y1” is in the art, so if you are to fashion your claim in a manner to distinguish over the art, there will need to be a more complex setting out of the instructions. So far the CAFC has considered the disclosure of an algorithm to be sufficient structure for a step-plus-function claim, and that seems to me to be the right level of structure to specify in a claim. You can write a claim sounding in function, but you cannot expect it to cover any more algorithms than you have disclosed in your spec. By the same token, you cannot claim software the employs algorithms that you have not disclosed, because you cannot be said to have proven possession of those algorithms.
In practice, I expect that if this were the rule 90% of the problems you find in software claims would evaporate on their own. Once X’s claims are no longer assertable against all (e.g.) software to run escrow accounts, but rather only software that uses X’s algorithm, X’s claims will not be of commercial value to PAEs. In that world, far fewer such claims will ever be patented, because the value of such claims will be correspondingly less. The only software that will be patented will be of the sort where the given algorithm really is valuable itself. It would no longer be profitable to file patent applications merely for the purpose of running a shake-down, because the putative litigation target can (1) more easily recognize what is and is not infringing software and (2) can usually design around the infringing software, unless there is something genuinely genius about the infringing software (in which case, that is precisely the sort of innovation that we want patents to remunerate).
MM,
On the flip side, those logic instructions unquestionably result in performance of a computer-implemented method. Are methods patentable? Do they suddenly stop being patentable if they are being performed by a computer? Do they suddenly stop being patentable if the only physical effects are in a computer memory, a computer drive, and an electronic display and mostly aren’t discernible by a human without special tools? What if the effects happen at two remote devices and the method allows for more efficient or secure communication of data between two computing devices?
I’m not saying that I have the answers, but I think there are some really difficult issues to grapple with. To be honest, it’s at times difficult to even clearly phrase the problem that requires application of an abstract idea judicial exception in a software context, as most discussions seem to invariably get side tracked by strong policy-based opinions (on either side) that have little to do with a candid discussion of the mess that is 101 law.
I do think it’s likely the case that a large number of the patents and patent applications that draw the most ire from courts and onlookers could just as well be disposed of with 102 or 103, or possibly even 112 (I’m not fully prepared to have the 112 discussion). Obviously, it’s a challenge to search these adequately, and I’m open to policy based arguments regarding a reasonable compromise in this area, but to be honest it feels like there are a decent number that get by that probably shouldn’t, even with the limited tools available.
In this regard, I think your comment about a general lack of unexpected results in a software context is on point. I made a similar comment in another thread recently, nothing that I am surprised that examiners do not more often utilize the rationale that a claim merely combines prior art elements according to known methods to yield predictable results.
(As an aside, although generally correct for the patents you are referencing, it’s worth noting that there are certainly areas of software development where this probably isn’t accurate. AI comes to mind foremost, but even more conventionally any system with multiple processors or computing devices can start producing results that you don’t necessarily expect.)
More generally though, and this is why I ask the questions above that I do, if processes are patentable, then why can’t you patent a novel and nonobvious process that a computer is programmed to perform? This isn’t a matter of evaluating prior art structures, it’s just a question of whether the steps of the process are novel and nonobvious.
To be honest, it feels like there is a good chance that the problems being highlighted by software may have been baked into our system for awhile now. Software may be bringing attention to them, and amplifying their effect, but it feels like the same issues have always existed with respect to process patents. Some process patents, such as those that are tied to a new machine, may feel better, while others, such as business method patents, may feel worse, but I am not sure that our current law has any coherent theory for why a method implemented by a computer processor should be treated any differently than a method implemented by a robotic arm.
There may be great policy reasons why these should be treated differently, but in that case it would probably be more productive if we could be clearer about what rules are being created and why. Right now, it often feels like some parties (on all sides) are making an honest attempt to navigate difficult issues, while others (on all sides) have a preferred framework in mind and are trying to justify it. I can respect arguments from both the legal and policy perspectives, but it is not helpful when the two are conflated together, and even less helpful when some (on all sides) fail to appreciate or admit that this is what is happening.
As a patent lawyer, however, I am mostly focused on the legal issues. The policy stuff is above my pay grade, and I am not even entirely sure how I personally feel about it.
Are methods patentable?
Some are, some aren’t. Methods of carrying out logic on data certainly should not be. They were never deemed eligible in the past. And then suddenly we have some nonsense about “the essence of electronic structure” and voila! reason is chucked out the window.
Do they suddenly stop being patentable if they are being performed by a computer?
You’ve got it exactly backwards. Logic is *ineligible* subject matter. But “suddenly” logic starts being protectable with patents because … computer! And a computer is just a tool for carrying out logic on data. It’s as if writing was suddenly eligible because … ball point pen!
Do they suddenly stop being patentable etc etc
Again, you’re starting from the assumption that logic is eligible for patenting. It’s not. You also can’t patent an algorithm, and you can’t patent a correlation and you can’t patent information. All these things are abstractions. Reciting a “context” in the claim doesn’t fix the problem. It just obscures the problem.
if processes are patentable
Some are, some aren’t.
This fundamental stuff.
There may be great policy reasons why these should be treated differently, but in that case it would probably be more productive if we could be clearer about what rules are being created and why.
There are great policy reasons and the policies were baked into our system and they are discussed in the recent major Supreme Court cases addressing subject matter eligibility. In a nutshell, we don’t want a patent system that ties up the fundamental building blocks upon which all technological progress rests, i.e., information, and logical steps applied to information. Because the law itself is an abstraction that is tied up with information and logic it becomes far too easy for attorney “innovators” to game the system when that kind of junk is introduced, which is exactly what we’ve seen happen.
We want to incentivize anyone who figures out a FFT which operates in O(n) time to disclose their invention to the public.
Obviously, that motivation doesn’t support the current system which incentivizes anyone shameless enough to claim every way of achieving a functionally defined result.
Great comment.
Blockchain Patent
Every one of these patents is pure g@ rbage.
The partners have already hinted that they are already testing how the system can facilitate the letter of credit process
LOL Methods for generating a “letter of credit”? Deep, serious stuff. Totally not abstract! Nope. It’s super dooper “confusing” to even begin to question the seriousness of this “technology.”
Down the t 0i let the system swirls …
As of June 2017, the complete exodus of all computer programmers from New Zealand is complete, exactly as predicted.
From Peter Finnie’s essay
From Alden Abbot’s essay
Both of these authors are making broadly the same argument (if jurisdiction X does not change its patent laws, innovation will slow in jurisdiction X), and the argument makes no sense in either essay. Both authors are making the argument in the service of a good cause (broader views of what should qualify as “eligible subject matter”), but that does not make it a good argument.
If a U.K. inventor, working in the U.K., invents some software that is not patentable in the U.K.—but might be patentable in China—that is not actually a reason for the inventor to relocate the work to China. It is a good reason to file a Chinese patent application, but that is all.
U.K. citizens can get Chinese patents, as can U.S. citizens—even if the work of inventing is done outside of China. Therefore, the eligibility standards applied by the SIPO are no reason one way or the other for a business entity to relocate its R&D facilities in China.
There are several reasons why a company should or should not locate its R&D facilities in a given jurisdiction: (1) do the smart people who think up brilliant ideas find the jurisdiction to be a comfortable and pleasant place to live?; (2) does the jurisdiction legally permit the work to proceed? (not all jurisdictions permit, for example, growing transgenic plants); (3) is the new R&D facility near to or far from existing facilities?
These are only exemplary considerations. There are other relevant considerations that should motivate the location of R&D efforts. None of them, however, concern patent laws.
Patent laws are (or at least should be, to a rational decision maker) totally irrelevant to the decision of where to locate an R&D facility.
Interesting. Would that view also suggest locating product manufacturing in countries with weak patent systems to thereby be less likely to be sued?
[But would not labor costs and skilled employee availability greatly outweigh patent law considerations for both that and R&D location?]
[One other problem is that too many small company or individual inventors still tend to file patent applications only in their own country, even though no one country has anywhere near a majority of world trade any more.]
P.S. One relevant patent law issue to consider for R&D location is whether or not the country has a high compensation “inventor compensation law.”
One relevant patent law issue to consider for R&D location is whether or not the country has a high compensation “inventor compensation law.”
Now there you have me. That is a “patent law” consideration that could (at the tiniest margins) influence where one locates R&D. I should have said that subject matter eligibility requirements are irrelevant to a rational decision maker considering where to locate the R&D facilities.
“the argument makes no sense in either essay”
Actually, it makes eminent sense (you may not like the sense it makes, but that’s a different issue).
If an inventor can get a patent in a jurisdiction regardless of whether the inventor did the R&D work in that jurisdiction, then why should the jurisdiction’s rule on subject matter eligibility affect the decision where to do the R&D work?
Patent law is – and shall remain – sovereign-centric.
Do not confuse treaties of comity or other such aspects for the basic nature of patent law.
Treaties are part of the “basic nature of patent law” by now. There is not some hierarchy of statutes in the patent system. The PCT has just as much legal effect as Bloomer v. McQuewan at this point, and both are now become “the basic nature of patent law.”
Besides, none of this speaks to the issue of why a given entity locates it R&D facilities in one place or another (an outcome which is in no wise compelled by patent laws here or elsewhere).
No.
Treaties (for example) are not self-enacting.
Please Greg, inform yourself a little on the topic of the sovereign nature of our laws.
Treaties (for example) are not self-enacting.
Some treaties are and some are not. The distinction is irrelevant, however, because both the pre- and post-AIA patent act codifies the PCT in our national law. Once again, my basic point stands: treaties are part of the “basic nature of patent law” by now.
Your own reference of an Act codifying any treaty speaks to MY point (in case you missed that)
Further, there is NO “one world order” nature of a world patent.
You for some reason want to go down that odd Big Corp “view”
As to “Some treaties are and some are not”
Patent treaties are one that expressly is outside the reach of the executive power to make law through treaties.
It’s almost as if you have not been paying attention to one of the cornerstones of the Separation of Powers concerns with dealing with the statutory law that is patent law….
That is some artful dodging there. Do you ever plan actually to engage with the argument on the table?
Your accusation falls flat.
Instead of such pettiness, why don’t you actually engage the points that I have presented?
Oh, because you have no actual replies to those points, perhaps?
Then just say so instead of the dissembling.
[Y]ou may not like the sense it makes…
Did you not read my post #1? As I said, “[b]oth authors are making the argument in the service of a good cause (broader views of what should qualify as ‘eligible subject matter’).” I agree with their desired conclusion. I like their intended outcome well enough.
Subject matter eligibility for software is a good policy choice, in my opinion. I applaud authors who argue for statutes that embody that policy choice.
Not all arguments, however, are convincing arguments. The argument advanced by Mssrs. Finnie & Abbott only makes if one does not bother to think it through.
Especially taking Paul F Morgan’s observations into account… if the UK gets sloppier re. Patent protections, that is precisely where you keep your R&D (less prone to suit), while filing patents in China and licensing manufacturing there (assuming stuff is made in China anyway). If the system in China is good enough there would be no need to catch goods coming back into the UK… or even the US for that matter… for many goods the enforcement can be made at earlier stage of “making” which would be in China.
This raises interesting cross licensing approaches… if your competitor relies upon sales in the US but depends upon Chinese manufacturing… their larger US portfolio may not be as important as perhaps your larger Chinese portfolio… IF the level of protection and enforcement effectiveness is heavily weighed toward the Chinese patent system….
[Y]ou keep your R&D [in the UK] (less prone to suit), while filing patents in China and licensing manufacturing there… [F]or many goods the enforcement can be made at earlier stage of “making” which would be in China
I am sorry, but I am not quite following the logic here. If I hold a Chinese patent, but no U.S., U.K., or (e.g.) Polish patents, and I manufacture in China, then I can import to the U.S. fairly easily. However, if my competitor decides to manufacture in Poland and import to the U.S., they can also do that fairly easily. How does my Chinese patent do all of the enforcement work for me?
“ The argument advanced by Mssrs. Finnie & Abbott only [sense] makes if one does not bother to think it through.”
Actually just the opposite – as my response to your post at 1 (so yes, I read your post – all of your post) indicates.
I took exception to the part of your post that you got wrong.
You are still wrong on that part.
[M]y response to your post at 1… indicates…
Nothing. It indicates nothing at all. It is a bald assertion without even a fig leave of reasoning to support it. How many posts have you flung at this thread by now without even pretending to argue your putative point? When do you plan to get around to saying something meaningful?
You make a bald (and wrong) assertion and I am merely telling you that your bald assertion is wrong.
I “get” you feel differently, but that’s just your feelings.
You make a bald (and wrong) assertion…
Do you know what the words “bald assertion” mean? They do not mean “supported by a logically invalid argument.” They mean “unsupported by any argument.”
If you go back and consult my #1, you will see a supporting argument. It might, for what it is worth, be a bad argument, in which case you could make a genuinely interesting response by identifying the flaws. It cannot, meaningfully, be said to be a “bald” assertion, however. At least not without simply resorting to an extreme philosophical nominalism, in which words convey nothing but your own private sentiments.
No Greg – my reply hits upon an element of yours that is NOT supported.
“… without even a fig leave of reasoning…”
You might imagine that if autocorrect is going to save me from my typos, it might at least insert a more plausible save. Clearly, this was meant to be “… without even a fig leaf of reasoning…”
No sweat – I do not get hung up on the minor errors that accompany this forum.
This is not – and is not meant to be – a forum where “perfect” English usage is necessary.
There are some sAmeones that like to use THAT dodge….
😉
I would guess the strongest disclosures are written in the inventor’s primary language. So you might as well have your inventors living in a country where they can get patents. That could cause a slight preference in the allocation of R&D.
Greg, so long as were talking about software as opposed products that include software such as cars or phones, software can be run anywhere in the world and users can be located anywhere else in the world. This suggests that in a global environment, the situation is going to be complicated even further for enforcement purposes because some of the steps are going to be conducted in one country and other steps in a second country. In this environment, is there really direct infringement?
Given that patents are expensive and worldwide coverage is out of reach for most developers of software, I am not so sure people will be investing heavily in software in the future for reasons other than eligibility.
Comments are closed.