About the Authors

  • The Authors and Contributors of "Patent Docs" are patent attorneys and agents, many of whom hold doctorates in a diverse array of disciplines.
2018 Juristant Badge - MBHB_165
Juristat #4 Overall Rank

E-mail Newsletter

  • Enter your e-mail address below to receive the "Patent Docs" e-mail newsletter.

Contact the Docs


  • "Patent Docs" does not contain any legal advice whatsoever. This weblog is for informational purposes only, and its publication does not create an attorney-client relationship. In addition, nothing on "Patent Docs" constitutes a solicitation for business. This weblog is intended primarily for other attorneys. Moreover, "Patent Docs" is the personal weblog of the Authors; it is not edited by the Authors' employers or clients and, as such, no part of this weblog may be so attributed. All posts on "Patent Docs" should be double-checked for their accuracy and current applicability.
Juristat #8 Overall Rank


« Manufacturing Alliance Sends Letter on Patent Reform to President Obama | Main | Patent Office Hosts Roundtable on Deferred Examination: The Opposition »

February 11, 2009


While Mr. Weiss' suggestion that the courts should play a role in setting standards for the USPTO is laudable on its face, as seen in the entire KSR debacle, merely because a court sets a standard does not mean it has introduced clarity. Kevin hit it right on the nose with the quote from Graham.

Kevin: "Generally, the patents that are mentioned in this regard are frivolous (methods for playing with cats or swinging on a swing), which are the patent equivalent of a vanity publication."

This bit of rhetoric on the part of patent bulls has become even more of a red herring than the swing patent itself.

Regardless of whether the swing patent had any impact on the US economy, the point of the swing patent is that its issuance pointed to a deep systematic flaw in our patent system. To some extent, that flaw has been addressed by the Supreme Court in KSR (although anyone who doubts that the swing patent was also anticipated needs to have their head examined). Now what remains is for the PTO and the courts to apply KSR diligently.

"Other complaints about "bad" patents tend to be either philosophical (for opponents of human gene patents, for example) or political (by industries that would rather expropriate than license technology developed by others)."

In fact, Kevin, anyone who keeps up with public commentary on bad patents (see, e.g., the PatentlyO blog or various threads on this blog) knows that the complaints are neither philosophical or political.

A Texas court found Hyundai liable for infringement of USPN 5,367,627 and awarded $34 million in damages. Claim 1 of that patent was obvious and unpatentable when the PTO issued the claim. It's obvious and unpatentable now.

I can point out your brushing aside of these basic facts because I keep a close eye on what has been issued and what is being issued. But it makes me wonder how much other misleading information is in the post. I simply stopped reading after the mischaracterization.

Dear Art:

Your comment raises some important issues that I’m happy to address.

When I say the issue is philosophy or politics, I mean this: there are on the order of 2 million patents in force at the moment. And there are (as you rightly note) howls of protest on the blogosphere and elsewhere that there are a spate of “bad” patents amongst those 2 million. My question is simple - if that were true, then where are the statistics that back up the assertions? Where are the thousands of patents being invalidated, by re-examination or in litigation? Where are the scholarly articles assessing why these patents are “bad”? If there are so many, it should be easy to survey extant patents and illustrate why and how the Office has done such a bad job. But whenever anyone talks about a “bad” patent, all we get are the cats and the PB&J sandwich patents.

The portion of the commentators on Patently-O making these assertions are no support for the argument about “bad” patents - there tends to be conclusory statements that this or that patent is invalid, but little analysis. And the same is true for commentors on our blog - lots of heat and little light.

Now, I agree that there are patents that have been improvidently granted - I cannot see how it could be otherwise, being a product of human activity. Even if the Office hadn’t hired 6,000 inexperienced new examiners over the past 5 years, and even if attrition didn’t rob the Office of examiners just when they had become experienced enough to avoid the kinds of errors that might lead to improvidently-granted patents, it would be shocking if some percentage of issued patents did not have at least one claim where reasonable people could differ on its validity. Paradoxically, these “new” examiners are more likely to refuse to grant a claim that should be allowed (since the PTO doesn’t track this statistic) than they are to grant a claim that they should not (which the Office actively works to avoid). And according to Mark Lemley, the more experienced an examiner becomes, the more likely it is that s/he will grant a claim that is invalid (although I have my doubts about that claim).

The existence of patent claims that should not have been granted is even more likely since the arbiters of what the claims mean - ultimately, the CAFC judges - are no part of the process of having the claims granted, and whether a claim is valid or not depends on the scope of the claim as it is construed. But even under these circumstances, the echo when the question is posed regarding all these “bad” patents is deafening.

To clarify, when I say philosophical, I should have added the following consideration. Is a patent a “bad” patent if it has at least one claim that is overbroad or otherwise invalid, or if it does not have at least one claim that is valid? The number of “bad” patents that exist will vary widely depending on how you answer that question.

In the interest of having everyone on the same page, here is claim 1 of USP 5,367,627:

1. A computerized method of selling parts for particular equipment specified by a customer, comprising the steps of:

a) receiving information identifying a customer's parts requirements for the equipment, comprising the step of receiving equipment application information, comprising an identification of the equipment with which one or more parts are to be used;

b) electronically specifying information identifying a plurality of parts and specifications for the parts;

c) gathering parts-related information for one or more parts within the plurality of parts which meets the customer's requirements, comprising the step of electronically associating at least one of the parts within the plurality of parts with the received equipment application information; and

d) receiving the gathered parts-related information and compiling the parts-related information into a proposal meeting the customer's requirements.

On its face, the patent cites 10 U.S patents (are they all invalid, too?) and other references. The $34 million represented a 2% royalty; the damages award did not include enhanced damages for willfulness nor did the non-practicing plaintiff obtain an injunction. I could not find an appeal to the CAFC, and I note from my research that Hyundai also lost a $25M verdict to Texas Instruments. Was that patent invalid for obviousness too?

(On the substantive validity issue, you may be right - I don’t have the technical background to judge. But since this case was decided in 2008, after KSR, and Hyundai didn’t bother to appeal, maybe they don’t share your certainty.)

So I’m sorry you stopped reading. I will not blindly defend the patent system when (if) it goes seriously off the rails, but I won’t excoriate it based on anecdotal, unsupported, politically and economically motivated attacks that lack evidentiary support.

Thanks for the comment.

"Is a patent a “bad” patent if it has at least one claim that is overbroad or otherwise invalid"


You also ask: "My question is simple - if that were true, then where are the statistics that back up the assertions?"

That statistics lie in the fact that it is frightfully easy to find patents that are invalid. Every week. I recall some time ago a biotech patent that issued that was "front paged" here, for reasons that are mysterious. Apparently, it was a "big deal." And the first claim was very poorly drafted and ridiculously broad. But that fact went un-noted (except in the comments, where it was reluctantly agreed that it might have issues).

"whenever anyone talks about a “bad” patent, all we get are the cats and the PB&J sandwich patents"

But Kevin: that's false. I just cited an awful patent that had nothing to do with cats or PB&J patents. In fact, you wrote: "On the substantive validity issue, you may be right - I don’t have the technical background to judge."

Can I ask what "technical" background you need to assess this patent? What part of the claim do you not understand? Maybe you come from a different class background than I do. I once worked at a job where I sold parts to people. And I was using computers to organize things long before this patent was filed.

Dear Art:

We differ fundamentally on the premises of the argument. I think is very unlikely that any patent will issue that does not have at least one claim that turns out to be invalid. This is because the drafter and the examiner will never be certain to know all the prior art. Different examiners will be more or less likely to grant broad claims or narrow claims, and in any event whether a claim turns out to be valid or not will depend on how its terms are construed. So if your position is that one invalid claim makes a bad patent, then you will find many more bad patents than I do.

Even so, I am curious at your estimation of the frequency of bad patents. 1%? 5%? 10%? If you believe it is a greater figure, then please back up that claim with more than your anecdotal belief that it is so easy to find "bad" patents, even using your definition. But frankly, except for perhaps air traffic controllers, what area of human endeavor has 1-5% error rates? And then take into account the complexities of new technology and the semantic limitations of describing it?

The claim you reference from our pages was this one:

1. A composition comprising:
(a) a first nucleic acid which
(i) comprises between 5 and 100 nucleotides, of SEQ ID NO: 1 or a complement thereof, and
(ii) decreases the level of a POSH mRNA and/or a POSH polypeptide when introduced into a cell, and
(b) a pharmaceutically acceptable excipient.

Now, as we discussed at the time, not the greatest claim ever written. But there were other claims in that patent, such as these:

3. The composition of claim 1, wherein the first nucleic acid comprises between 15 and 30 consecutive nucleotides of SEQ ID NO: 1 or a complement thereof.

4. The composition of claim 3, further comprising a second nucleic acid that hybridizes to the first nucleic acid under physiological conditions to form a double stranded nucleic acid.

5. The composition of claim 4, wherein the first and second nucleic acids are ribonucleic acids.

These claims, particularly claim 5, should be valid. So we return to our difference between what constitutes "bad" patents.

Getting back to the patent you cited, did the patentee assert this claim? Was Hyundai adjudged to infringe it? If there were narrower claims that were asserted and that were the basis of the infringement, then what is the point (besides the philosophical one) of citing it? And if claim 1 was the basis of Hyundai's liability, why wasn't that judgment appealed? Post KSR, wouldn't it make sense to try to invalidate it at the CAFC (if it were so obvious)?

Finally, without contradicting your recollection, I have had the experience of having clients make similar statements regarding such claims. But I have never had the experience of having any of my clients be able to find the art they thought would invalidate the claim. If your recollection is better, perhaps you would care to share it with the rest of us?

Thanks for the comment.

The comments to this entry are closed.

May 2024

Sun Mon Tue Wed Thu Fri Sat
      1 2 3 4
5 6 7 8 9 10 11
12 13 14 15 16 17 18
19 20 21 22 23 24 25
26 27 28 29 30 31