by D.Crouch
The Senate IP Subcommittee heard testimony today on the topic of patent quality:
Promoting the Useful Arts: How can Congress prevent the issuance of poor quality patents?
Testimony from PTO Commissioner Drew Hirshfeld; Professors Polk Wagner, Melissa Freeney Wasserman, Colleen Chien; and Former PTO Acting Driector Teresa Stanek Rea. Sen. Tillis began with his remarks with a statement that about the “madness” of patent eligibility, including the Chamberlain case that found a garage door opener to lack patent eligibility. “No one can blame the USPTO” for this situated by the courts. That said, Sen. Tellis is looking for a consensus option.
In his testimony, Commissioner Hirshfeld began with an interesting statement on new examiner time allotment program:
Starting this month, all examiners began receiving additional examination time tailored to specific attributes of an application, including the overall number of claims, the length of the specification, and the number of pages in any filed information disclosure statements. Also starting this month, examiners with the least amount of examination time in our production system also began receiving additional time to align their time allotments with the requirements for current patent examination.
Beginning next fiscal year, we will utilize an updated process for assigning patent applications to patent examiners. This process will automatically match each application to the examiner best suited to examine the application, taking into account the complete technological profile of the applications, the work experience of each patent examiner, and the workload balancing needs of the agency.
Prof. Wagner:
One lesson here, I think, is to tread quite carefully when seeking to improve patent quality. The small firms and individuals who desperately need the patent system to bridge the gap between great idea and successful technology can be (and often are) disproportionately impacted by reforms that are of little consequence for large companies. . . . .
Many patentees have incentives to make their patent documents as unclear as possible.
Prof. Wasserman:
Indeed, invalid patents are unnecessarily reducing consumer welfare, stunting productive research, and discouraging innovation. . . . Patent Office’s fee schedule should be restructured to minimize the risk that the Patent Office’s revenues will be insufficient to cover its operational costs and to diminish the Agency’s financial incentive to grant more patents when revenues fall short.
Ms. Stanek Rea:
But we do not want to rely exclusively on reviewing the quality of a patent only after it has issued. We also need to focus on front end modifications during examination.
Prof Chien:
[Congress should] encourage and expect the USPTO to engage in bold, persistent, and rigorous policy piloting and evaluation.
Re: “Sen. Tillis began with his remarks with a statement that about the “madness” of patent eligibility, including the Chamberlain case that found a garage door opener to lack patent eligibility.”
How anyone could pick the actual claims* [and spec.] of THAT particular patent to defend, especially after its Patently-O blog discussion, is quite remarkable. Surely there must be better examples of patent claims that deserved a full trial instead of a 101 rejections than that one?
*A signal from any old wireless system that any old door or window is open, like various old burglar alarm systems.
“especially after its Patently-O blog discussion”
That Patently-O blog discussion did NOT go in the direction you indicate Paul.
we all know where that thread went. nowhere. its going nowhere until information is placed securely in law and/or prosecution & litigation procedures.
the Chamberlain case is another easy one in my scheme:
is it a method? Yes
is the result of the method only an item of useful information ? Yes (the state of the door).
does the utility arise by human consumption of the information? No.
The door-opener consumes the information. Eligible.
Obvious as the sunrise, utterly non-inventive, and we need a procedure to establish that for somewhat less than one-million dollars if s tupid patents are not to be weapons of mass economic destruction.
the Chamberlain case is another easy one in my scheme:
is it a method? Yes
Er, no. Claim 1 of the patent at stake in Chamberlain:
There may be a thousand defects with this claim, but it is quite clearly a claim to a machine, not to a method.
Re; “Obvious as the sunrise, utterly non-inventive, and we need a procedure to establish that for somewhat less than one-million dollars if s tupid patents are not to be weapons of mass economic destruction.”
That was the rationale for enacting PGRs, IPRs and CBMs. There is no need for that kind of expense for claims like Chamberlains. But D.C. 101 motions, if promptly considered and successful, can be even faster and cheaper, if filed right after suit. An IPR can still be filed within one year of the complaint if the 101 motion is not granted.
The Ends do not justify the Means, Paul.
Marty – you are so in love with your own paradigm that you are not even bothering to look before you leap.
The only perfect patent is the one that . . .
…does not exist — nor should such “perfection” be an actual aim.
Ben notes below that: “So the PTO thinks that it’ll improve quality for applications to receive additional examination time. But this adjustment is only being applied to the most simple inventions (think simple mechanical devices such as fishhooks and screws)” [and is for disposal rates rather than quality improvement].
That is echoed by other sources. I.e., that high tech internet or computer related applications will NOT get increased examiner examination hours. Even though the “bad patents” tag seems most often applied to widespread litigation assertions of patents with very broad functional claims in those very technologies!
Corporations outsourcing their patent application filing and prosecution to patent attorneys with little or no knowledge of the corporations technologies, and filing greatly over-broad claims missing what is validly patentable, is another, increasing, quality issue.
“That is echoed by other sources. I.e., that high tech internet or computer related applications will NOT get increased examiner examination hours.”
To be clear: like ALL applications, computer related applications will get additional time based on application specific circumstances. For example, if an application has 4+ independent, it’ll get an extra hour above the basline. If an application’s specification is 150+ pages, it’ll get an extra hour above the basline. So if an application has 30 independent claims and an 1100 page specification it gets two extra hours over the baseline.
In contrast, simple mechanical inventions have received a baseline increase on every application.
Thanks for the clarification, Ben.
Thanks Ben, for that caveat re all examiners getting an extra hour for more than 4 independent claims or a 150+ page spec.
A long overdue start on a good idea, but of course I cannot help wondering if a Hyatt-style application with a couple of hundred independent claims would still get only one [or maybe two] extra hours of examination unless assigned to some specially set up examination group or put on some back burner?
“or put on some back burner?”
Paul, you are such the Office apologist.
Did you not understand that phrase, or not understand sarcasm?
Maybe hold up a bigger “S” sign (your post is not distinguishable from your typical “Office Supporting” posts).
Pardon potential repost…
Maybe hold up a bigger “S” sign (your post is not distinguishable from your typical “Office Supporting” posts).
SPEs already have some discretion in giving additional time for exceptional cases. I suspect a 300 claim case (not properly restrictable) would induce a call to a TC director, and a substantial amount of additional time would be allocated to the examiner.
And I’ll say it again. The way to improve quality is move the entire PTO to a place like Pittsburgh.
Bring back TSM so that objective reviews can be performed.
Better allocation of applications so that examiners focus on small areas to become experts.
Probably some form of evaluation of the examiners work and/or searches performed by contractors.
Again, a lot of these “professors” are making a lot of (some millions) from burning down the patent system. There is no ethical constraints on their conduct. Few of them have any real experience with patents, innovation, science, etc.
The whole thing is a joke. It is like asking the criminals to come and explain why there should be no laws against stealing.
The key to improving patent quality is searches and finding reasons to combine. This is something where good automation tools can help. The other key is to get examiners understanding their AUs better. The more experienced an examiner typically the better quality patents.
The whole thing is ridiculous.
A parade of “people” coming up to Congress saying, “I am paid very well to destroy the patent system by large corporations and here is my testimony they prepared and/or approved for me to give. I have no ethics and regularly lie, but that is OK as my university doesn’t care as long as I pull in a good reputation and money from SV. If you play along, Congresspeople, you will be rewarded too.”
Reading her remarks, it doesn’t appear that Wasserman has ever been involved in the process of developing a new product and bringing it to market, let alone the patenting aspect of that process. And she mentions false positives alot – i.e., patents (or, if she’d had an inkling of how patents are constructed, patent claims) that the PTO shouldn’t have granted – but, save for one sentence, she doesn’t discuss false negatives, viz. patent claims the PTO should grant but doesn’t. That one sentence is limited to stating that granting more examination time doesn’t result in a higher rejection rate for valid claims, and cites to another publication of hers. Nowhere does she discuss the cost to society of the problem of failure to allow allowable claims. Reading her remarks, I feel as if she’s still living in 2009.
Terry Rea, who has had experience working with real businesses, and who also worked in the PTO, mentions the need to improve the up-front quality of examination (rather than the knocking out bad patents after they’ve been issued), but her five double-spaced pages are too few to get into details. A wasted opportunity.
Who picks these “witnesses”?
The incredible thing is not only who picks them but that they can testify and be part of this without disclosing where they are getting their money and where their universities are getting money.
Was Colleen Chien promoted because of her ideology and the connections of San Jose State with SV companies? Can you file an ethics complaint against Chien that will be taken seriously by her university? No. Has Chien disclosed her sources of money? No.
So some bought off clown pulls in huge some of money to pro statute herself to the SV firms to burn down the patent system.
Although I can quibble with Chien’d lead in (it is NOT the Office job to determine what is truly an advance – executing what Congress has set forth is what the Office job is and it is the Congress that determines what an advance is — as well as “promote” is NOT limited to the notion of “advance” as Chien would use), her input is fairly directed – and substantively so – to what the Office DOES control: its examination tactics (and measurements thereof).
At first glance, her input here is more than decent.
It is given by a person that does not deserve a seat at the table.
Meh, let’s focus on content.
I really like the testimony of Prof. Wagner.
I agree that the quality of patents would be improved if the background discussed the best prior identified during prosecution in a meaningful way, if the summary of the invention discussed the practical utility of the invention and the evidence supporting it, if the detailed description was limited to embodiments or other material related to the granted claims, and the claims were separating the non obvious improvements over the prior art that has been uncovered from the elements known in the analogous art.
I suggest that the issue fees could heavily penalize specifications that are not amended to be clear and concise so that even engineers can rapidly ascertain the invention covered by the granted claims.
Wagner: “ Unfortunately, the phrase “poor quality patents,” is too often deployed as a shorthand for “patents I don’t like” or “patents that might be used against me” rather than as a useful description of a specific systemic problem.”
Does this remind you of Malcolm, Marty, the anti-software patent sycophants, and anyone else?
Wagner: “ Unfortunately, over time, a combination of the structure of the patent system and the caselaw has yielded a set of incentives that push patent applicants in exactly the wrong direction from a patent quality perspective.”
…leading into the reflection of the reality of court induced patent profanity.
There is a reason why the Jepson claim format is a dinosaur. I have not seen any recent usage studies of that “optional” claim format, but if I recall correctly, we were heading well south of 1%.
The late Ned Heller (among others) proclaimed that format as the “answer,” without regard to reality.
In that regard though, Prof. Wagner also strays into the lack of touch with reality with providing an answer of Congress forcing changes without correcting the profanity (and merely calling these forced changes “[moves that] creating the conditions in which it is clearly in a patentee’s interest to invest in patent quality.”
Likewise, Prof. Wagner shows a penchant for ir-reality with the “provide a glossary” approach — ignoring the “turtles all the way down” problem with using words to define words (which, being words, would then need another turtle layer of words to define those words) — all the while ignoring the reality that words — as they are at the top level — are simply a limited tool .
So while I see some good intent, I also see flaws.
The problem is that Obama appointed activist judges that are trying to burn down the patent system so any opportunity to invalidate claims is used. The CAFC has misused the background section.
While I get your point about Obama and the extreme influence of the SV established companies during that time (how many undocumented meetings were there with the USPTO and Big Money SV…?), the notion of patent profanity is most definitely not limited to the Obama era appointees.
Having a “legislative forcing” in a vacuum without regard to both past, present and likely future judicially induced patent profanity is a non-starter.
The corrections for patent examination quality needs to distinguish from the “bad patents” mantra fed by anti-patentists and a return to the realization that strong patent rights (both carrot and stick) should be the aim.
As Prof. Wagner notes: going for “perfect” is a sure fire way to NOT meet the goals of having patents in the first place.
I agree that the quality of patents would be improved if the background discussed the best prior identified during prosecution in a meaningful way…
True, but what would be an even more effective means for improving patent quality would be for the Office to require Jepson claim format. The PTO has the authority to make such a requirement (just as they have the authority to require that one present one’s claims in English).
In other words, directly addressing the known prior art in the specification text is good. Directly addressing the known prior art in the claim text is even better. Do not let the applicant hide the ball. Make the applicant state up front what the applicant perceives to be the advance over the art. If the applicant does not present the claim in Jepson format, send a formalities notice and refuse to examine it (just as one would with a claim set presented in Spanish) until the claims are put into Jepson format.
This would scarcely cost either the office or the applicant anything. It would, however, improve quality of examination almost immediately, by making it easy for the examiner to focus the search on the most relevant points.
Wouldn’t this be a pain to enforce? What’s going to stop the gameplayers from playing games?
Through the notice and comment process, one promulgates a regulation requiring the applicant to place the prior art elements in preamble and the putatively novel elements in the claim body. Through a second notice and comment process, one promulgates a regulation making it a violation of the duty of candor knowingly to place an old element in the claim body.
If someone wants to play games with their formulation of the Jepson claim, it is their own funeral when it later comes out in litigation that they knew that this or that element in body was old in the art. As for those who simply refuse to formulate the claims in Jepson format, treat those claims just as one would a set of claims presented in Chinese (i.e., keep sending notices of defective application papers and refuse to docket them to an examiner until the formalities are observed).
I truly doubt that the patent office has the authority to turn an statutorily optional claim format into a hard legal requirement.
Not only is such not their role, this is exactly the type of “over-reach” that was sounded in the testimony to Congress.
(it would be interesting to see an updated trend of usage of Jepson claim format)
“What’s going to stop the gameplayers from playing games?”
Did you mean the gameplayers within the patent office, Ben?
You kind of dismissed out of hand a similar point on a previous thread when I pointed out a flaw in your desires. It seems like you are doubling down and ONLY looking at “games” that the other side MAY be playing while clenching tight your eyes to the games that your own side plays in.
Altogether not surprising, mind you, given that you also seem to see nothing wrong (at all) with the likes of SAWS.
Oy you struck a nerve. Every patent attorney/agent in the US should be force-fed Strunk & White, daily. Verbosity, boilerplate, redundancy, and unneeded disclosure clogs the system with noise and drives up costs for everyone.
Well, some of these have been created in reaction to the way courts construe the claims. As Prof. Wagner points out, claim interpretation by the courts may need to be first straightened.
[Q]uality of patents would be improved if… the claims were separating the non obvious improvements over the prior art that has been uncovered from the elements known in the analogous art…
Is this a call for mandatory Jepson claims? If so, I agree 100%.
Expunge from the system patents that protect logic and information. Patent quality will become 1000 times better overnight because the PTO hasn’t figured out how to examine those kinds of patents and they never will.
Your feelings are noted.
As is the total lack of any actual tie to innovation protection law.
If Malcolm can’t see or understand the human-configured billions of atomic parts of a machine in action then it must be ethereal magic!
Pretty much.
If you can’t describe the novel structure of a single working embodiment of your “invention”, then you aren’t in possession of it. The fact that you can’t describe that embodiment is your problem, not the problem of the patent system. The rules exist for good reasons.
Now, if you believe that a very special new exception should be made because you and your favwit kind of “invention” are sooooo important, then Congress should recognize the exception as such and develop a coherent set of rules for examination of these super special “inventions”, along with formal paradigm for claiming theses special “inventions”, and a specialized database of prior art. What should never have happened is what did happen: building a nonsensical house of legal cards on the mythical foundation that data processing logic “is the essence of” actual structure and allowing that “essence” to be claimed instead of actual structure.
It’s always amazing to watch allegedly intelligent attorneys struggle with the basics (or pretend to).
You seem to be saying that (1) only structure can be claimed, and (2) it can only be claimed at a fine level of granularity. (1) is wrong. (2) can be inferred because it is trivial to describe many structures, as in “a compressor”, or “a decoder”, or “a multiplexor”, or “a clustering algorithm”, or “a routing table”, or “a translation lookaside buffer”, or “a ray tracer”, “an FFT” etc etc etc. These are no different than describing, say “a compound with element A and element B”, or “a solvent”, or “a transcription factor”, or “an aluminum alloy”, or “a viron”, or “polylactic acid”, etc etc etc. These things don’t need to be described at a lower level (molecular or atomic) because enough people know enough structural features of these things. Our job as attorneys isn’t to provide a verbal blueprint or MRI of the material world. Our job as attorneys is to communicate clearly and efficiently to the relevant artisan, of whom we can assume a baseline knowledge and nomenclature.
He’s beyond wrong — and yet again pulls his Trump act and pretends that his view is the “basics” and all others are somehow out to “grift” the system.
PTO Examination needs a major shakeup if things are going to improve. Poking around the edges is a plan to preserve and justify the status quo.
I think the count/docketing systems should be rebuilt so that every subsequent action is done by a random, technology appropriate examiner. Examiners would act as quality reviews of other examiners. If a subsequent examiner can’t go final, the system can flag the original examiner’s work as likely defective and needing review. Instead of a tiny fraction of actions being reviewed, every rejection would receive a review.
“every subsequent action is done by a random, technology appropriate examiner”
An absolutely horrible idea.
“Examiners would act as quality reviews of other examiners.”
We’ve already gone over this — still a bad idea.
I do not wish to advance a strong thesis here, but I notice that according to the limited data that we have, it seems that EP claims are invalidated—when challenged—at about the same rate as U.S. claims—when challenged. This despite the fact that the examination process is fairly different (and much more resource intensive) than the U.S. process. That suggests that what we are doing now is maybe as good as can be done within the realm of the possible.
Of course, here I am drawing a conclusion from an examination of only two jurisdictions. I have no idea what the rates are for CN, JP, & KR. If any of those jurisdictions see a lower rate of invalidations for granted claims, that would suggest that their models might constitute a case-study for others to follow.
“That suggests that what we are doing now is maybe as good as can be done within the realm of the possible.”
This make perfect sense.
Enough already with this false “poor quality patents” narrative.
Enough.
“That suggests that what we are doing now is maybe as good as can be done within the realm of the possible.”
I suspect the current PTO management would strongly agree with you, given their program of elaborate but superficial changes. It looks designed to impress a legislative spectator without changing anything meaningful.
I think changing examiners throughout prosecution would have benefits beyond improving patent resistence to challenges. Based solely on every examiner having their own standards, we’d expect repeated random assignments to cause the office’s aggregate standard to converge (you only have to convince the easiest of x examiners over y rounds that the case is allowable.). Beyond that, exposure to other examiners work product should cause individual standards converge (examiners largely work in silos, with most examiner-on-examiner review being primary-junior review rather than peer review. Exposure/review of peer work should develop consistency accross examiners in a way that primary-junior review cannot.). I’m not sure why this seems like such an unworthy idea. It could be done with limited additional resources.
I think that yours is an interesting idea. I am an empiricist, so my thought is that the PTO ought to test it out. Over a set of five years, it should randomly assign ~50% of incoming applications to be examined just as they traditionally have been examined, and the other ~50% to be examined as you propose. Then ten years down the line, one can check the subsequent invalidation rate from the two groups and see if one group or the other has a meaningfully better quality.
Suspect you are building castles on sand here, Greg. At the EPO, 4% of issued patents get opposed. That means 96% do not. What do we know about the “quality” of the 4%? Anything? And even if we do, what does it tell us about the quality of the other 96%? Anything?
Not much because we know nothing of the reasons why an Opponent files an opposition at the EPO within 9 months of issue. So, the broad outcome of oppositions at the EPO (1/3 of opposed patents revoked, 1/3 maintained as issued, 1/3 maintained in narrower form) is consistent with Opponents assessing their chances of success before filing their oppositions. With an opposition that fails, you are worse off than you would have been if you had not opposed at all.
I have a better idea. For litigated patents, let us scrutinise the prior art relied upon in the PTO and then that relied upon by the Courts. If the patent is revoked by the court on prior art that the PTO should have found when examining the application, that tells us something about PTO search quality. If would-be Opponents cannot find any art better than that already on the PTO file (by the way a common situation here in Europe) that also tells us something About PTO search “quality”.
I think that you misconstrue the reason for the comparison MaxDrei.
Let’s first establish whether or not the premise advanced by Greg holds true, before dismissing that premise as “castles built on sand.”
Also, I do not see merit in your proposed comparison of prior art search.
Are you really not expecting to merely see a confirmation that established well-monied entities can far more extensively search for prior art on the select items enforced against them? One needs no “study” to be able grasp that obvious outcome.
You would be better off exploring that point from Greg that you wanted to dismiss — if indeed the “quality” by revocation metrics are similar (but your prior art search metric is different), what then would be EPO lower quality driver?
You have to have one given the premise of “equal quality as measured by revocation.” You cannot have “same overall quality” and one facet be improved without some other facet being degraded.
I know that you like your hobby horses, but the overall picture here shows that riding those horses alone is just not enough.
At the EPO, 4% of issued patents get opposed. That means 96% do not. What do we know about the “quality” of the 4%? Anything? And even if we do, what does it tell us about the quality of the other 96%? Anything?
Sure. And the same is true in the U.S. (only a tiny fraction of U.S. granted patents get challenged in either IPR or court). I agree that these are imperfect data by which to assess “quality,” but still and all they are the best that we have.
The PTO director is obliged to steer the car by looking in the rearview mirror. This hard to do, and it should come as no surprise that the results have an unsatisfactory feel. Still and all, this side of the eschaton, one cannot get perfection. What we have might well be the best that is humanly possible in an actual patent system that can actually function.
“For litigated patents, let us scrutinise the prior art relied upon in the PTO and then that relied upon by the Courts. If the patent is revoked by the court on prior art that the PTO should have found when examining the application, that tells us something about PTO search quality. ”
Appreciate the suggestion but it will never work. EPO uses PSA, which prohibits hindsight. In the U.S. you have 275+ APJ’s who see an IPR petitioner cobble together 5, 6, 7+ references to establish the “obviousness” of a claim and the APJ’s say, “Wow, you’re right!!!”
Without a system that requires a challenger to identify D1 and at most 2 additional references that bear on the inventive step/obviousness analysis any comparison of the art used by examiners and art used by challengers would be meaningless.
Yes, I see that. But what’s to be done about it? Could the Congress amend the patents statute to require obviousness to be decided by the EPO’s PSA methodology?
Sure, MaxDrei, those are a pair of glasses that Congress could decide to put on.
Wagner: “ And indeed my research shows that we should be skeptical of claims that we can spend our way to a patent quality utopia; in a comparative study with the Japanese patent system, which has a much higher ratio of examiners to applications and a longer tenured examiner corps than the USPTO, there appeared to be little difference in either the real or perceived problems with patent quality.”
“examiners with the least amount of examination time in our production system also began receiving additional time to align their time allotments with the requirements for current patent examination.”
So the PTO thinks that it’ll improve quality for applications to receive additional examonation time. But this adjustment is only being applied to the most simple inventions (think simple mechanical devices such as fishhooks and screws). Why?
Is it because of an epidemic of poor quality, simple mechanical patents? Or is it because they can make token adjustments to these fields without perturbing “stakeholders”?
My understanding (from an inside source – a substantially high one at that) is that this has everything to do with work-product flow.
This actually is not directed to any specific art unit, nor is necessarily art unit driven (unless you want to consider “backlog” of an art unit to be art unit driven).
Senator Tillis: “Derp some dude on the Internet told me derp that garage door openers were ineligible DERP and I believed it derp because I am simple minded derpity derp what’s a claim anyway? derp just words!!!”
Gob help us from these f-kin know nothings.
Yay, Malcolm’s Derp Dance.
I didn’t watch the hearing . . . did anyone in attendance point to ANY such allegedly “poor quality” patents?
Or was everyone just continuing with the illogical, without- foundation, no examples, “but I know it when I see it!” intellectually-dishonest, “poor quality patents” dance yet again?
That said, Commissioner Hirshfeld and Prof. Wagner know of what they speak.
And here’s something Congress could (will?) do:
Restore patent Section 101 eligibility to pre-unconstitutional Alice / Mayo . . . so that the hard-working examiners will have more time to look for prior art.
Hirshfeld comments reflect my LONG expressed views as to obtaining better patent examination quality.
Great to see!
(ditto Rea’s comments as to inanity of having a post grant “quality” system untethered to where “quality” is actually actionable and controllable by the office: with examination itself)
[Congress should] encourage and expect the USPTO to engage in bold, persistent, and rigorous policy piloting and evaluation.
What kind of waffle is this? What does that even mean, “bold… piloting and evaluation”?
It certainly does NOT mean hiding another NON-patent law political rant style hyperlink in the comments.
About 20 years ago, the EPO held a series of conferences aimed at defining patent “quality”. How does the USPTO define patent quality. In particular, is it a breach of quality standards to issue a claim that is new, useful and definite, but where the non-obviousness of its subject matter is of very dubious “quality”?
Or putting it another way, how does one distinguish expeditiously and fairly, during ex parte examination of hundreds of thousands of claims, between obvious and non-obvious engineering (predictable) subject matter?
From my point of view as an in-house attorney as a patent-owning entity, a quality patent is one that affords commercial exclusivity (i.e., a patent that can both survive a validity challenge during assertion and cover the products that a competitor would like to bring to market). I do not think, however, that the PTO should use that standard to measure quality, because (1) the PTO cannot reasonably be expected to know the nature of the applicant’s competitor’s business and (2) should not care as between two U.S. businesses which wins and which loses.
For the PTO’s purposes, “quality” should mean “granted patents survive validity challenge,” and “supportable claims go to allowance.” Of course, no more than a tiny fraction of granted patents will ever face a validity challenge, so the “data” by which the PTO can assess “quality” will be never be better than very imperfect. Moreover, each successive directorship will be “steering” by looking in the rearview mirror (the patents being challenged now will mostly have been granted by in Dudas’ administration or earlier). Still and all, perfect quality (and no one should expect perfection) would mean that (1) 100% of patents survive litigation challenge; and (2) 100% of applicants’ valid claims receive a notice of allowance from an examiner.