The average number of references cited per patent has continued to grow over the past decade and now almost reaches 50 per patent with no sign of slowing-down. When the reference-count was much lower, many examiners apparently assumed that applicant had already ensured that the cited references did not disclose the patented invention. However, today neither patent applicants nor their patent attorneys have time to read through and fully consider such a large number of references. Obviously, examiners also do not have time to read 50 references. In my view, technology could offer a nice solution that would easily allow examiners to conduct an internal textual search that is limited particularly to references that have been submitted in this case and perhaps those previously cited in similar cases. So far, that search capability does not appear to be available for the examiners – especially for non-patent prior art.
Now, I should mention here that the ‘averages’ are a problematic figure for references cited because the distribution is quite skewed. Namely, there are a fairly small number of cases that cite a very large number of references, and those cases have a major impact on the results. The chart below shows a more nuanced version of the distribution.
I posted this in response to Paul Morgan’s post, but I want to also post generally:
Have they removed the penalties for not disclosing relevant art? I haven’t been involved in filing in awhile. I do not know if any of the new rule packages have eliminated those penalties.
While some parties’ submissions were always ridiculous, the draconian penalties for not disclosing was a strong incentive to submit everything of which you were aware that even tangentially related to the invention.
Lionel, check the en banc Therasense case.
No inequitable conduct unless one knows of materiality and intends, by withholding it, to deceive. Materiality is defined as in “but for, the case would not have been patented.”
Has no one suggested that someone claiming to be an inventor should be an expert in the art? If there’s too many documents related to your invention to read, maybe you didn’t need to invent something, you could have just used an obvious solution. Fact is, if there is so much information available, there is no need for the patent system. Patent protection is offered in exchange for disclosure, but clearly disclosure has no value if no one’s got time to read any of it.
ed, both you and your inventor have duty not to file claims you know are unpatentable over the art.
” … technology could offer a nice solution that would easily allow examiners to conduct an internal textual search that is limited particularly to references that have been submitted in this case and perhaps those previously cited in similar cases.”
Install Watson, but only let it read cited documents. …. and expect the results in the form of a question as in:
What is paragraph 24 and fig. 3, reference numeral 312
I would settle for the examiner actually reading the references that are applied.
The MPEP is clear that the examiner is not required to give the references submitted in an IDS any more consideration than the references that the examiner reviews as part of his/her own search. So if the examiner does a series of search queries with some root expanders and Boolean operators and gets X hits and picks some Y subset of those hits to review more closely, and then picks Z references to apply against the claims, how is that any different than dumping the 50, 100, 200, or 1,000 references an applicant submits into a product like Amberscope and then running the same search queries and picking out the references the examiner thinks are best? Who cares how many patent documents an applicant submits in an IDS? They get the same treatment from the examiner as patent documents that are not submitted.
To the extent that there’s no product that allows the examiner to do that for NPL, that could be a problem, but as patent literature far exceeds NPL in IDS submissions, I don’t see it as that much of a problem.
It usually isn’t. It’s really only the jumbo’s where you get 100+ that are problems worth talking about, and the prosecutions that people drag out so long that new art is cited and then sometimes prosecution has to be reopened etc. etc. Of course, as the avg goes up we’ll be seeing more and more problem cases.
That said, having to manually input even 50 patents can take a bit if you’re not a trained secretary.
Patents end 2 end is supposed to help address these things though.
Hey 6 if you are actually an examiner please contact STIC to learn about the basic functions in eDan because every examiner I know is using either the reference manager in eDan or the reference extractor button that is part of the OCR functionality in eDan to extract the patent numbers from the IDSes and put them in EAST.
NO ONE enters patent numbers manually. They’re automatically extracted and copied into EAST where they can then be keyword searched.
Yeah we use reference manager (but only for originally filed IDS. If you say a “reference extractor” button exists then I’ll have to look into that, thanks for the heads up.
One thing I’d recommend to any US examiner is that they look at the references cited in a foreign action. I’d say 90% of the time, the references cited in the foreign action are better than the references cited by the US examiner. Yet only a few times have I seen the US examiner use a reference cited by a foreign patent office. They seem to just initial the foreign references without review.
Yet only a few times have I seen the US examiner use a reference cited by a foreign patent office.
Examiner’s routinely rely on such references in chem/bio and have done so for years.
In addition, a substantial fraction of the art cited by chem/bio applicants is art that was cited by foreign patent offices. In addition, it’s advisable to include the arguments made by the foreign examiners and the applicant’s responses to those arguments. The docketing system of any law firm worth its salt will include reminders to get this information in front of the Examiner as quickly as possible.
Are you kidding? Foreign patent office searches are often the easiest way to make a rejection! Any Examiner not using them is wasting his own time. Maybe they consider it cheating? Seems like if someone has already searched, that’s a pretty good place to start.
On another note, all those people saying that Examiners need to do a better job might consider that Examiners should be paid better. Or based on some other metric. Most good examiners have much better career options outside the PTO, so there needs to be a way to selectively retain the good ones.
How much do you think somebody who’s done nothing but examine brake or clutch applications for 10, 20, or 30 years should make? A GS-14 step 3 primary (which somebody can reach in about 6-7 years) is making $120k+. You don’t think that’s enough? Here’s an idea: go out into the job market and see how much somebody is willing to pay you for that “expertise” somewhere else.
Do report back and let us know how it goes.
I get massive Information Disclosure Statement dumps in my art. As far as developing a more efficient way to comb through patent relations within the applicant cited prior art a company called Ambercite developed a product called Amberscope and it gives a very visual way to organize prior art based on citation analysis. Amberscope’s Citation analysis includes applicant citations and examiner citations, so it identifies the most pertinent pieces of art. At the office, we have the ability to turn those dumps into a search field and then search through that dump textually. We can also flip through that dump to see if the art is relevant. There is no easy way to search NPL cited by applicant, as the author pointed out, we just have to read them.
Glad to see a discussion of the legal fiction that large numbers of cited references are all fully read and considered by anyone in the application examination process.. Especially considering that the total volume of prior art doubles every ten years. Note the below admissions that inventors and clients cannot afford the time to keep up with it. Yet expecting examiners to do it all for them, in spite of the strong objection of patent attorneys and clients to increasing the very small U.S. application searching and examination fees that would allow them the time to do better prior art. Compare that with what it costs to get an EPO prior art search for example. Yet these are the same folks that strongly object to IPRs in which the much better prior art missed by the very limited PTO prior art search can finally get properly considered in the very few patents that matter.
What a bunch of twisted nonsense that was Paul.
1) The PTO can raise the fees and they can say it is for policy reasons. They can do this on their own.
2) IPRs for better prior art searches? Seriously? $1 million in spending for better prior art searches.
If Google Shadow Director Lee would only concentrate on her job, then she would see this as an imperative to fix.
Yes, the PTO can now raise some of its fees without Congressional approval, but that of course is twisting what I said. The point was that you and many others would strongly object to paying in U.S. applications what the EPO charges for their much better prior art searches. You claim to be entitled to govenment IPR or EPO quality prior art searches by U.S. application examiners, but you only want to pay a few hundred dollars for it. Have you priced thorough commercial [non-PTO] prior art searches for all the claims of an application lately?
As to the costs of IPRs, they are paid for by high fees by those being sued on their subject patents. They are rarely used by anyone not alreadly being sued or threatened with suits. IPRs are being extensively used because they are much cheaper than District Court litigation re patent validity issues based soley on prior art patents and publications that were never considered in the application. They have already greatly reduced the opportunities for patent trolls to shake down companies for large amounts of money on invalid patent claims, when before they had no alternative other than spending millions on full blown litigation.
In contrast, it simply does it make fiscal sense to spend large amounts of money on extensive prior art searchs on the 98% or so of patent applications that result in patents that are never asserted against anyone.
Note also that no one is denying that there are PTO examiners who could indeed be doing better prior art searches even within their limited alloted time. But engaging in fiscal unreality about the time and effort, and therefor cost, needed for a really thorough prior art search for modern technologies is not going to help.
>But engaging in fiscal unreality about the time and effort, and therefor >cost, needed for a really thorough prior art search for modern >technologies is not going to help.
Nonsense. Why do people write nonsense like this? Your words are like saying: the mediocre job we are doing now is about the best we can expect.
The EPO does a better job. The PTO can and must do a better job at prior art searches.
Paul, given the massive unbelievable atrocities from the SCOTUS, Obama, death squads, Google Federal Circuit judges, and Google shadow director Lee, I think the PTO could safely double the fee for prior art searches. Moreover, I think the PTO could do this as a policy reason to improve patent quality.
Let’s fact it. The examiner and the examination they do is where the dirt meets the road. The rest of this is third world adaptations to try and fix a broken system.
I meant rubber.
“Glad to see a discussion of the legal fiction that large numbers of cited references are all fully read and considered by anyone in the application examination process.”
I don’t think such a legal fiction exists. Or ever existed. Nobody that I’m aware of has ever operated under the assumption that every reference submitted in an IDS or cited on a PTO-892 was “fully read.”
Have they removed the penalties for not disclosing relevant art? I haven’t been involved in filing in awhile.
While some party’s submissions were ridiculous, the draconian penalties for not disclosing was a strong incentive to submit everything of which you were aware that even tangentially related to the invention.
This is another anti-patent ploy where the federal workers join in as it makes their jobs easier. But, fact is, the responsibility for EXAMINATION is on the Office. And an examiner that doesn’t learn their area should be fired.
That is the core of the PTO. You examiners becoming experts in your area. The core of your jobs is that.
The PTO often shifts examiners from one area of technology to another technology since backlogs amongst art units are not evenly distributed. For example, i was previously moved from computer networks to image processing software. I am currently being move from AU on image processing software to speeching signal processing because my last office action was on an application that was filed on May of 2014. Not bad for bunch of “federal workers are so lazy”, ehh?
The fact is, this happens often and we are required to learn whatever field we are being assigned to. Accordingly, the PTO does not require us to be experts, only knowledge enough to be one ordinarily skilled in the art.
Your standard appears to be, unless we become one of extraordinary skill in the art we are examining, we are “lazy federal workers”.
Yet, your standard for applicant differs. “By its very nature, the PTO should be the experts in the field and know the art—not the applicant.” at 1.1. Under this standard, applicant-inventors should not be experts in the field of their invention and they should not know the art in which they are trying to obtain a patent. To otherwise do so would be “The clients can’t afford it. We don’t have the time.” at 1.1.1.1.
Your standard is quite subjective and biased, eh?
Richard, the PTO should require you to be experts in your tech area. That is the core of examination and is the foundation of our patent system.
And the examiners I work with that have become experts in their area are usually much more efficient and happier at their jobs. The fact that the office has not told you that it your job means there is massive failure to be blamed on Google Shadow Director Lee.
the PTO should require you to be experts in your tech area.
What “tech area” is that? Banking? Hedge funds? Contract law? Real estate? Video rentals? Advertising?
I feel happier if i could stay in my current art as i generally know what the state of art of my current field is (not an expert). Alas, i am but a civil servant who must go where my superior directs.
The PTO does promote a certain number of examiners to be expert GS-15 examiners. Those examiners have PhD and spent their entire career in one field of technology.
However, to require all examiners to obtain PhD and promoting all examiners to GS-15 makes no economic sense. As i previously mentioned, re-assigning examiners to different technologies in accordance to the need of the PTO occurs often and it is simply efficient management to deal with differences in backlog. To require an examiner to obtain a PhD in one area and then re-assigning the examiner to another area to deal with backlog makes little sense.
Additionally, GS-15 expert examiners outputs a lot more office actions than a primary examiner. If a field of technology / AU has too many experts, then the backlog in that area would disappear a lot quicker and thus leading the office to reassign those experts to areas where their PhD matter little.
In view of the reality on the ground, if you become the “Highlight” (a term of art opposite to “shadow” in my field) Director of PTO to require all examiners to obtain PhD in one area, then this shifting of expert examiners to fields that they are not familiar with negates the advantages you so advocate.
“Additionally, GS-15 expert examiners outputs a lot more office actions than a primary examiner.”
Nonsense. The position factor for a GS-14 primary is 1.35 and the position factor for a GS-15 is 1.5.
.15 translates into quite a few bds.
Bullsh!t. At 20 hrs/BD for GS-12 expectancy (a pretty average number as far as expectancies go) a GS-15 gets 13.33 hrs/BD and a GS-14 gets 14.8. So in an 80 hour examining bi-week (which is a fairly rare occurrence) a GS-15 has to do 12 counts and a GS-14 has to do 11 counts (10.8) actually. One more count per bi-week. Or 26 more counts over the course of an entire FY. Assuming 80 hours of examining time per bi-week (again, a completely unrealistic assumption).
For 1,600 examining hours, a GS-15 examiner with a GS-12 expectancy of 20 hrs/BD would have to do 240 counts. A GS-14 examiner with the same expectancy would have to do 216 counts. 24 extra counts. Over an entire FY. Hardly “a lot more.”
You’ve been called on your nonsense.
Take my AU, as a GS-14, based on my hours, i need 59.85 BD to hit 100%. As a GS-15, i need 65.21 BD to hit 100% for the fiscal year.
Would one consider 5.36 BD a lot more? For GS-15 expert examiners, yes. The reason being: they are assigned to special dockets within their technology to deal with more complicated cases.
Thats the whole point of being an expert, to deal with cases no one likes to deal with because it is complicated. In my AU, while we don’t have such expert examiner, we do have special dockets that are harder than others (i.e., halftoning and color) so much so that our admin SPE asked for volunteers to do the docket. Factor in the complexity of the cases and the higher BD, on balance, it is a lot more than you would expect.
“Take my AU, as a GS-14, based on my hours, i need 59.85 BD to hit 100%. As a GS-15, i need 65.21 BD to hit 100% for the fiscal year.”
Is that for 1,600 examining hours or somewhere thereabouts?
“Would one consider 5.36 BD a lot more?”
No. Are you seriously trying to tell me that over course of an entire FY, having to do 11 more counts is “a lot more”?
Please.
“…we do have special dockets that are harder than others (i.e., halftoning and color)…”
According to one poster on this site, halftoning is so old and well known that it’s not even patent eligible subject matter.
“Factor in the complexity of the cases and the higher BD, on balance, it is a lot more than you would expect.”
It’s exactly what I would suspect, and which you yourself calculated, 11 (actually 10.72) counts more. Over the course of an entire FY.
“Or 26 more counts over the course of an entire FY. ”
Um, that’s quite a bit o BDs.
I’m not sure how you think you’re “calling me on my bs”.
Lulz, I forgot you were a lifer.
“Is that for 1,600 examining hours or somewhere thereabouts?”
1537 hours.
“No. Are you seriously trying to tell me that over course of an entire FY, having to do 11 more counts is “a lot more”? ”
Say person A, an inexperienced hand, was asked to assemble 10 widgets comprising components C and D in 1 hour. Say person B, though an experienced hand, was asked to assemble 12 widgets comprising components C, D, E, and F in 1 hour.
Would one consider the two additional widgets assembled by Person B “significantly more” than Person A?
Richard,
Your question highlights one MAJOR difficulty that the Office has LONG ignored: the widgets are not the same.
There is no way to answer your question because the system is set up NOT to recognize any difference from one widget to the next, thus there is no way to recognize (and account for) the proposition that you hypothesize.
I have long advocated to re-engineer this part of the Office business method.
“Say person A, an inexperienced hand, was asked to assemble 10 widgets comprising components C and D in 1 hour. Say person B, though an experienced hand, was asked to assemble 12 widgets comprising components C, D, E, and F in 1 hour.”
That’s a heckuva straw man you’ve built yourself there.
The difference between a position factor of 1.5 and a position factor of 1.35 is 11%. Not the 20% extra you incorrectly imply with your 12 widgets vs. 10 widgets strawman. And there is no difference in the “widgets” that a GS-15 examiner is asked to churn out from the “widgets” that a GS-14 examiner is asked to churn out. Read the application, conduct the search, and do a thorough examination and write up the OA. There are no “extra components E and F” required of a GS-15’s OA from a GS-14’s OA.
So a GS-15 examiner has to do 11% more work than a GS-14 examiner. Is that “a lot more”? To me, no it’s not. Especially considering that the GS-15 salary is 17.6% higher than the GS-14 salary at every step level. 17.6% more pay for 11% more work sounds like a pretty good deal to me.
“Are you seriously trying to tell me that over course of an entire FY, having to do 11 more counts is ‘a lot more’? ”
BTW, please do feel free to answer my question rather than wasting everybody’s time with silly straw man arguments. It’s a pretty simple and straight forward question: do you seriously believe that having to do 11 more counts over the course of 1537 examining hours is a “lot more”? That’s an extra .007 counts/hr.
Richard,
In an ideal world, what sort of tool(s) would you need to be able to quickly and accurately uncover prior art? What’s missing in the current applications?
In an ideal world, it would be nice for the applicant to claim what they invented so there is no need to find prior arts to read on broad claims.
I heard from my law professors about this legendary team that spent $10 million to find a piece of prior art on a library shelve in Norway…that would be nice.
In all seriousness, i do not think it matters because 99% of the patents do not get litigated. Those 1%, with a dedicated team of lawyers spending millions $ to find ways to invalidate it, chances are the patent would be invalidated no matter how good the examiner who issued the patent is.
However, i do think Inter Parte Review and the new ex parte reexamination makes up for a significant part of the deficiency of patent examination in that IP owners could increase the potential value of their patents because if their patent survives challenge, they have some ideas as to how far they could push their patent rights without spending millions litigating DJ actions or a normal trial.
And, as experts, they should be paid 1/2 or 1/4 what the attorneys, who you acknowledge are NOT experts, are paid? Yes, that seems likely to happen. Or, the actual experts will leave and enjoy a better compensated career outside the Office.
“Or, the actual experts will leave and enjoy a better compensated career outside the Office.”
Plenty do that. For those of you who decide to remain at the PTO, though, and whine that you don’t get paid as much as attorneys, the door is always open. If you want to make the same amount of money as an attorney, then go get a job as an attorney. If you do I’m sure you’ll quickly discover that there are many, many reasons why attorneys make more than examiners.
But I have a feeling that you’re gonna stay right where you are and console yourself with that ridiculous notion that “examiners do the same job as the attorneys, they just don’t paid as much.”
Keep telling yourself that. Maybe some day you’ll actually believe it.
Despite what you think, examiners do not think like that. We are not entitled ego-maniacs.
“Despite what you think, examiners do not think like that.”
I’ve heard enough examiners express that notion, in person and online, to know otherwise.
You miss my point. I’m not saying Examiners should be paid more. I’m saying that, for what Examiners are paid, you can’t expect the PTO to be able to hire and retain experts. You get what you pay for.
As to your other jab, I’ve worked both private firms and the PTO, and have found that an efficient Examiner gets paid very roughly about the same as a patent prosecutor.
An efficient Primary (6 years at PTO) can make 120k+ in 40hrs/week with 20 vacation, 13 sick, and 10 holidays off.
A patent prosecution associate may work 60 hours/week and get paid 180k+ with whatever vacation they can squeeze in.
And that’s if you can get a good prosecution job. A lot of patent prosecutors make far less than 180k for far more work than an effective Examiner. Of course, some in patent prosecution make much more than 180k.
And yes, the jobs are very different, though in theory they both revolve around applying the same set of laws to the application at hand.
“An efficient Primary (6 years at PTO) can make 120k+ in 40hrs/week ”
How does that happen? According to the pay scale I’m looking at it looks like you’d need to be step 9 for that to happen.
GS14, step 3 is $121,357.
Examiners are not on “standard” GS scale. See link to careers.uspto.gov
Why does virtually no other Patent Office in the world require the applicant to provide references to the Patent Office?
Answer: Because in other Patent Offices, the Examiners are supposed to be experts in their respective fields.
I do not know whether foreign Patent Offices has a rule about information disclosure. I do know that most Japanese specifications mentioned arts they are aware of and most relevant to what they are trying to obtain a patent in. See US 2006 / 0007465 A1, paragraph 5, for example.
I find this collaborative approach between attorneys and patent examiners more productive than treating each other as mortal enemies in a perpetual zero-sum game. As much as it is the duty of the examiner to become knowledgeable about the field they are examining, the applicant-inventor’s attorney also has a duty to be at least some what knowledgeable about the field they are trying to obtain a patent on.
I believe Rule 1.1 and Rule 1.3 require at least this much:
“A lawyer shall provide competent representation to a client. Competent representation requires the legal knowledge, skill, thoroughness and preparation reasonably necessary for the representation.”
“A lawyer shall act with reasonable diligence and promptness in representing a client.”
Even if you expect an examiner to be an expert in their field, the question remains why some applicants are submitting so much information. Even experts would require substantial amounts of time to review copious submissions of prior art.
Canada, India, Australia have their own versions of IDS submission requirements.
None of them have exploded into the US-style “cite everything anyone has brought to your attention else lose your patent” filings, though.
Canada, India, Australia do have their own versions of IDSes and similar patent office submission rules, but they don’t have the case law that the US does which forces patent practitioners to data dump everything when cross citing for related cases in order to avoid charges of inequitable conduct.
Wrong answer. References are provided in an application when they are available to avoid, in assertions of the resultant patent, being accused of inequitable conduct and risking unenforcability of the patent. That is the creation of the U.S. judicial system, not something originated by either the PTO or Congress.
Dennis,
Most patent search engines have cited references (both forward and backwards) available for review. With the move to the Common Citation Document (CCD), this data is available for US applications as well as issued patents. You can easily choose to review X, Y, or A refs from European cases and/or examiner vs. applicant citations from US cases.
This means that the patent documents provided by the applicant can easily be text searched separately from the rest of the 80 million patent documents. (This does not apply to NPL citations which are still a mess.) This can be done for an entire family of patents with only a couple of mouse clicks.
Unfortunately, the USPTO has not integrated the Common Citation Data into their internal (EAST) search engine, but it is available to the rest of the world through most of the high end commercial search tools as well as to European examiners using their internal search engine. The new Patents-End-to-End (PE2E) system which should roll out soon at the USPTO will incorporate this data. It should have been rolled out already but was a victim of the budget cuts a few years ago.
“However, today neither patent applicants nor their patent attorneys have time to read through and fully consider such a large number of references.”
O but the office has infinite time. Don’t ya know?
Of course examiners don’t have time to read all these references (I think that was one of Dennis’ points), but there’s rarely a penalty for over-reporting references, and a potentially large penalty for not reporting, so why not report marginally relevant references? And on the more snarky side, there are plenty of Examiners who are can’t or won’t read, understand and properly apply even one reference, so it matters little if they have to wade through a list of 5 or 50 references in an IDS.
“but there’s rarely a penalty for over-reporting references, and a potentially large penalty for not reporting, so why not report marginally relevant references? ”
Right sooooo, IDS rules then?
“And on the more snarky side, there are plenty of Examiners who are can’t or won’t read, understand and properly apply even one reference, so it matters little if they have to wade through a list of 5 or 50 references in an IDS.”
That’s true but by and large in the useful arts those examiners are the minority.
6, that is your job. Examine the patent applications. You are supposed to be the expert in that area. Man, you federal workers are so lazy.
I really am.
Look brosef, there was nothing the brochure for this job, or even in the job description that states that I should be spending a day or so going through 700 cited references from the applicant. And while that is rare and I’ve played along, that sht still needs to be nipped.
LOL. Where can I get my hands on one of those brochures, 6?
Not really 6. Whatever they submit is stuff that you should be responsible for anyway assuming it part of your art unit.
No, it is not. The “stuff” that they cite is just general background sht not even all that related to what they “invented” or “claimed”. I have yet to see a jumbo citation where the all the references were close to “relevant”.
Then 30 seconds per.
Note, that would be about 5 hours and 50 minutes spent reviewing each of 700 cites (assuming 6’s (high?) number and NWPA’s (low?) time estimate). And that’s assuming you can tell all 700 are the easy, background, garbage sort from the first 30 second scan. How much time do you all think would be reasonable for an examiner to spend reviewing cited art (keeping in mind they also need to do their own search)? Obviously it will vary by Art, but do people (on all sides) have a ballpark set of expectations?
Exactly. I was going to post to that effect but for some reason it was filtered. I thought NWPA was trying to show how it didn’t take all day, not exactly how it does take all day.
Do you even understand that the entire structure of our patent system is based on you doing your job. Knowing the art in your class.
Get to know your job. Spending time getting an MS in your field. Read the art in your area. That is what you are supposed to do.
A patent attorney cannot be an expert in all the different areas we write applications for and inventors can’t either as they are inventing and building.
Let’s be respectful, NWPA. Examiners can know how to do their jobs and still not have an entire week to peruse an IDS containing 50+ references.
I agree RH. That is why the examiner has to be the king/queen of the prior art for an art unit.