Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[patentatt]

In the recent TRANSOCEAN decision, Judge Moore wrote (for the majority):

It is not required to enable the most optimized configuration, unless this is an explicit part of the claims.

That strikes me as odd, if not incorrect.

It seems to conflict with the rule that enablement must be commensurate with the full scope of the claims.  MPEP 2164.08.  It also seems in tension with the court's recent holding in Sitrick v. Dreamworks, LLC, 516 F.3d. 993 (Feb. 1, 2008).

In other words: it is not clear if Judge Moore is saying:

1. the most optimized embodiment is / is not part of the claim scope
2. whether the most optimized embodiment doesn't have to explained in detail, or doesn't have to be enabled at all
3. if it is part of the claim scope, and it doesn't have to be enabled - why the hell not?

[petethebody]

Here, the Court was discussing what was claimed.  The alleged infringer argued that the device was not enabled by pointing to the well documented experimentation required of the patentee to actually make the device.  The alleged infringer argued, and the district court agreed, that 112 enablement required "Transocean to enable the invention to allow a person of ordinary skill in the art to take advantage of the 'timesaving' aspect of the invention."   However, at no point in the patent was "timesaving" explicitly claimed. 

You are extremely correct that the court seemed convinced that "timesaving" was not a part of this drilling rig invention.  Practically, that is difficult to swallow - how can this dual-assembly platform be useful if it is not more efficient?  No preferred structural or safety requirements appear to have been disclosed.  In fact, from the factual description of the patent in suit, it appears that efficiency is all that is discussed (the Fed Cir even says "Transocean attempted to improve the efficiency of this time-consuming process with the system described in the patents-in-suit.").  But remember, utility under 101 is a low hurdle. 

Here's how I see the opinion, and I may be very wrong: the question of enablement was one of law; undue experimentation, one of fact; increased optimization, even if well known as the extrinsic goal, must be an explicit requirement/goal of the patent, in order to require enablement; even so, the d.c. must still resolve the enablement issue on the device generally (w/o an eye to efficiency) by making findings on the underlying Wand factors of undue experimentation.  In sum, before you get to the issue of enablement of a specific embodiment (here the efficient one used), you must find whether the device generally is enabled. 

To me, this seems like it will fail for best mode on remand, even if enabled. 

[JimIvey]

Apparently, my prior firm litigated and won on the issue that the best mode need not be enabled, merely described.  They didn't tell me the cite for the case, and I never looked it up.  In addition, it was more than 15 years ago, so it might not be good law anymore.

Regards.

[petethebody]

You are right - best mode need not be enabled.  Best mode is a question of fact, reviewable for clear error.  All a court should look to is active concealment.  The policy seems a little off: "just tell us one way of doing this and you're fine.  you don't need us how to do the thing well" But, I understand that if every way of making it had to be enabled, patents would be much longer.

DeGeorge v. Bernier, 768 F. 2d 1318 - Court of Appeals, Federal Circuit 1985
"Not complying with the best mode requirement amounts to concealing the preferred mode contemplated by the applicant at the time of filing. In re Gay, 309 F.2d 769, 772-73, 50 CCPA 725, 730-32, 135 USPQ 311, 315 (1962). "[T]here is no objective standard by which to judge the adequacy of a best mode disclosure." In re Sherwood, 613 F.2d 809, 816, 204 USPQ 537, 544 (CCPA 1980). Instead, "only evidence of concealment (accidental or intentional) is to be considered. That evidence, in order to result in affirmance of a best mode rejection, must tend to show that the quality of an applicant's best mode disclosure is so poor as to effectively result in concealment." Id. The purpose of the best mode requirement "is to restrain inventors from applying for patents while at the same time concealing from the public preferred embodiments of their inventions which they have in fact conceived." In re Gay, 309 F.2d at 772, 50 CCPA at 730, 135 USPQ at 315. Compliance with the best 1325*1325 mode requirement exists when an inventor discloses his preferred embodiment. Id."
http://scholar.google.com/scholar_case?case=12930715075600667165&q=best+mode+enablement&hl=en&as_sdt=400000003

[JimIvey]

I finally read the linked blog.  We discussed this case at length recently in here.  You might search for "written description".

The Federal Circuit went further than the "written description" line of cases with that one.  In my opinion, they went too far.

No claim ever enables its full scope, well, virtually no claim.  Every claim that uses the transition, "comprising", covers every possible combination of the claimed elements with an untold number of other extraneous things.  Image a claim for a vehicle comprising 3 or more wheels.  How can a car with 300,000 wheels be enabled?  Or, 300,234, or any other ridiculously large number over 2?

As I decried in the discussion of the written description requirement, a much better tool for avoiding patent liability when a claim reaches where it really shouldn't is the reverse doctrine of equivalents.  The result is a finding of non-infringement, not invalidity.

In the Sitrick case, the patentee had an otherwise legitimate patent covering inserting one's image and audio into a video game.  If Dreamworks was found to not infringe by the reverse doctrine of equivalents, the claim could still be asserted against video game infringers.  Instead, the claim is gone.

In addition, I didn't see whether inserting an image and audio into a movie was substantially different than doing so into a video game.  With CGI, movies and video games are made in very similar ways these days.  Perhaps the claim was enabled (even without mentioning movies since it's so similar) and Dreamworks should have been on the hook.

But, I can't reconcile the ruling by the Federal Circuit with the long-standing, well-established precedent that says you don't have to enable each and every embodiment of the claimed invention.  Does anyone else understand how "full scope" is not synonymous with "each and every embodiment"?  More importantly, does anyone have any idea how one identifies a particular embodiment that must be enabled from a myriad of embodiments covered in the claims?

Regards.

[patentatt]

James,

I think you understand the issue exactly.  The problem is not so much a problem with Judge Moore's opinion, but a fundamental problem in enablement jurisprudence:

1. Patent claims typically recite comprising, meaning that they can read on things that include much, much more than what is "explicitly recited" in the claim.
2. Patent law requires you to enable the full scope of the claim, although earlier decisions indicated that you only needed to provide detailed descriptions of one embodiment.
2a. The CAFC has recently back off from the "one embodiment" rule and started to require more detailed description for multiple embodiments, in order to enable the claim.
3. But, because comprising claims have huge (potentially infinite scope) it is practically impossible to enable the full scope of the claim.

Judge Moore hints at an answer to this riddle when she writes you don't have enable features that aren't explicitly recited.  Of course, features that aren't recited are still part of the full scope of comprising claims.  So it seems that you don't have to enable the full scope of the claim (in practice).

There's another possible distinction to think about here too: whether the non-recited feature is functionally integrated with the recited features.  Consider a claim that recites "comprising A and B."  Consider an item that includes A, B, and C.  There are two possibilities.  First, C can be entirely unrelated to A and B - they are merely connected to each other.  In that case, Judge Moore's point makes more sense: it is probably good policy to not require the patentee to enable C in order to enable the claim.

But there is another possibility: C can be functionally integrated with A and B.  In fact, C can mean: performing A and B through the functionality of C.  In that case, Judge Moore's point makes more sense.  Suppose that an accused infringer makes the item of A, B, and C, but the patentee has not enabled doing A and B through C.  It seems to me that the patentee did not enable the full scope of the claim - and it becomes questionable whether it is fair to stop others from doing A and B through C.  This seems to violate the bargain of patent law, that you get what you made, because the patentee didn't really make doing A and B through C.

As another alternative, if the use of C to do A and B is new and nonobvious, you can invoke the doctrine of blocking patents: the original patentee gets a patent, but the accused infringer can also get a patent on A, B, and C.  If they want to do A, B, and C, they would need to cross license.

I do not know of much scholarship on this issue, although there was a great lecture on the subject at my old law firm by a professor/litigator from (I believe) the University of Minnesota.

But I bet you can find some good scholarship on the issue by searching SSRN for "enablement":

Heightened Enablement in the Unpredictable Arts
UCLA Law Review, Vol. 56, p. 127, 2008
Sean B. Seymore
Vanderbilt Law School
Date Posted: March 10, 2008
Last Revised: November 6, 2008
Accepted Paper Series
157 downloads

3    Incl. Electronic Paper    The Enablement Pendulum Swings Back
Northwestern Journal of Technology and Intellectual Property, Vol. 6, p. 283, 2008, Washington & Lee Legal Studies Paper No. 2008-40
Sean B. Seymore
Vanderbilt Law School
Date Posted: May 12, 2008
Last Revised: November 6, 2008
Accepted Paper Series
96 downloads

4    Incl. Electronic Paper    Rethinking Enablement in the Predictable Arts: Fully Scoping the New Rule
Stanford Technology Law Review, Vol. 3, 2009
Bernard H. Chao
University of Denver Sturm College of Law
Date Posted: March 30, 2009
Last Revised: April 1, 2009
Accepted Paper Series
46 downloads

5    Incl. Electronic Paper    The Formal Structure of Patent Law and the Limits of Enablement
Berkley Technology Law Journal, Forthcoming
Jeffrey A. Lefstin
University of California - Hastings College of the Law
Date Posted: April 9, 2008
Last Revised: November 24, 2008
Working Paper Series
90 downloads

Relatedness.

[JimIvey]

Judge Moore hints at an answer to this riddle when she writes you don't have enable features that aren't explicitly recited. 

But I don't think the claim at issue explicitly recited movies or video games.  Here's the claim from the court opinion:

    54. A method of integrating a user voice image into a presentation output, the method comprising the steps of:

    sampling a user's voice;

    analyzing the sampled user's voice to provide user voice parameter data representative of the user voice image;

    storing the user voice parameter data; synthesizing and interjecting the user's voice into the presentation

    output responsive to the user voice parameter data comprising the step of associating a particular predefined character image within the presentation with the user's voice so that when the particular predefined character is speaking, the user voice parameter data is input as a model to a voice synthesizer that effects the integration of the user's voice into the presentation output as associated with the predefined character image.

Well, I didn't find the part where the opinion says that only features recited in the claims must be enabled.  But, I think I have a better sense of the reasoning.

The technology seems to require access to pre-composition data such as "a particular predefined character image within the presentation."  In a video game, the image is composed in real time and there's programmatic access to the constituent objects in the video playback.  In a movie, all you have are the pre-composed images that make up the frames of the video.

So, why not just rule that the claim does not cover movies?  There used to be a guiding principal that claims are to be construed to be valid, even if it results in non-infringement.  That doctrine appears to be entirely gone.

Of course, it didn't help that the Spec referred in several places to audiovisual presentations such as movies and video games.  It also didn't help that the patent owner asserted that the claim(s) did cover movies.  Still, the overriding principle that claims are to be construed to be valid, if possible, is entirely missing in action.

My intuitive guess is that two things are at work here.  The first is that the patent holder can waive the doctrine that claims are to be construed as valid if possible, by arguments made in litigation.  The second is that part of the problem is that other parts of the claim make no sense in the movie context, particularly, the "particular predefined character image within the presentation."

sigh...  Then, how did Dreamworks infringe?  Did Dreamworks sell movies with other voices dubbed in? 

I give up...  If only I could get a client to ask me about this and pay for a comprehensive study of this case....  Maybe I could offer to write an article for my old law journal....

Regards.

[patentatt]

You're talking about Sitrick, which was an earlier decision that heightened the requirement for enablement.  Sitrick is the third in a series of enablement cases that expanded the enablement requirement (Liebel-Flarsheim Co. v. Medrad, Inc., Automotive Technologies International, Inc. v. BMW of North America, Inc., and Sitrick v. Dreamworks, LLC).

In contrast, I was originally talking about the most recent Transocean decision.  That is where you find Judge Moore's language about "explicit part of the claim scope" ("It is not required to enable the most optimized configuration, unless this is an explicit part of the claims.").

There used to be a guiding principal that claims are to be construed to be valid, even if it results in non-infringement.

Yes, Phillips says the doctrine of construing claims in favor of their validity is a "doctrine of limited utility in any case" (paraphrasing).  But see the discussion of recent CAFC decisions here, where at least one case seems to have invoked the rule more liberally:

http://www.717madisonplace.com/?p=2678
http://www.717madisonplace.com/?p=2723

BTW I recommend the law review articles that I mentioned above, esp "THE FORMAL STRUCTURE OF PATENT LAW AND THE LIMITS OF
ENABLEMENT."  They help explain these issues.