12:01 Tuesday

Wait One Day, Lose Two Days! Not Free!

Today brings the best teaching example I've seen on the importance of promptly responding to Office Actions.

Every patent practitioner knows that there are statutory periods to respond to USPTO Office Actions. Typically, they are six months. Miss those deadlines and you could be up a creek, needing to petition to revive your application, if you can.

However, the PTO generally reduces these to a "shortened statutory period" of three months (two, for certain actions). So if you're late, you simply pay an increasing fee for each month you are delayed, up to the statutory limit. And unless you: a) work only with clients/inventors who respond promptly to your emails (and pay their bills on time); b) don't ever work on litigation matters; c) never take a vacation; and/or d) have been lucky enough to never be jammed with a crisis of someone else's (or even your own) making; then you have almost certainly availed yourself of this "convenient" option provided by the PTO. (In fact, certain litigious patentees in pre-GATT days were known to always use the maximum allowed extensions, to keep an application pending as long as possible.) That first extension month is yours for the low, low price of $150 – and if you are a small entity, it's $75. What a bargain! See 37 CFR 1.17(a).

But there is a bigger price, of course: patent term. Most of us (?) appreciate that these delays can negate extra patent term that would otherwise have been granted under Patent Term Adjustment for PTO delays. Fewer, however, seem to take into consideration the more obvious problem that a delayed response also delays the issuance of the patent. Every day you wait to file that response is, in theory, a day the Examiner is not acting on it, and one more day until that Notice of Allowance arrives.

Combining these costs can yield disastrous results. Enter today's patent, stage right:

U.S. Patent No. 8,110,187 Purification and characterization of cytotoxic lymphocyte maturation factor and monoclonal antibodies thereto Assignee: Hoffman-LaRoche, Inc. Issued: February 7, 2012 Filed: October 8, 2002 Continuation-in-part of ... 07/455,708, filed Dec. 22, 1989. USPTO Delay: 1,057 days Applicant Delay: 392 days Total PTA Adjustments: 665 days.

Commentary: First things first: this patent is DOA. Like the Army patents discussed here a while back, Roche's patent expired a few months ago. Twenty years from the priority date (12/22/1989) plus 665 days PTA = October 18, 2011. Oops. The difference in this case, however, is that the Applicant's delay was solely responsible for the stillbirth. Here's how:

Pretend that Roche had filed all its responses without extension, but that the PTO still had its delays (and that the Examiner's docket was uniformly distributed during that time). Then the patent probably would have issued about 392 days sooner, or January 11, 2011. That is still more than twenty years from the priority date, but there would have been 1,057 days of PTA extension, resulting in an expiration date of November 13, 2012 (normal expiration of 12/22/2009 + 1,057 days). Total patent term would have thus been 672 days (11/13/2012 minus 1/11/2011). Coulda, woulda, shoulda.

So, instead of having two years of an enforceable patent, Hoffman-LaRoche has nothing but the receipts it spent on prosecution. And those receipts probably aren't small – there were three appeal notices, two RCEs and one 453-page submission in an unsuccessful attempt to provoke an interference. Add to this that three different law firms were involved (it was originally filed by the erstwhile Pennie & Edmonds firm, passing briefly through Jones Day, to its current location), along with the clarifications in PTA calculations following the Wyeth v. Kappos case, and perhaps there is some understanding as to how the expiration date issue might have gone undetected.

But all the Applicant delay here occurred by April 2009. At that point, with the terminal date in sight, Roche pressed forward with an Office Action response and full appeal briefing to the BPAI. By the time the reply brief was filed in October 2010, it was a foregone conclusion that the patent could not issue with any term. (Prior to the appeal, Roche could have received a maximum of 226 days extension, to August 5, 2010. Roche gained an additional 439 days for its appeal pendency, and only then because the appeal eventually succeeded in reversing the rejections.)

Bottom line: 392 days of delay results in the loss of 672 days of term for Roche. Valium, anyone?

There are certainly times when extensions are necessary, and even useful. But we need to be mindful of their consequences with respect to term, especially in fields where a single day of a patent's life could have significant value.

BPAI to HP: Equal Protection Clause doesn't save claims from 101

For a move even Stephen Colbert would admire for its ballsiness, we extend kudos to Hewlett-Packard, who argued the following in its appeal brief of App. No. 09/976,630:

 

No mention of In re Nuijten, which was decided several months earlier. Or maybe HP thinks that it is unconstitutional to apply KSR, too, since there are oodles of obvious claims out there in issued patents...

Today, the Board of Patent Appeals and Interferences responded appropriately in its Ex Parte Haines decision:

Section 101 rejection of claims 21-27   Appellant alleges that the USPTO has issued patents to others that contain “propagated signal claims.” Appellant contends that the § 101 rejection of the instant claims is a violation of Appellant’s rights under the equal protection clause of the Fourteenth Amendment to the U.S. Constitution. App. Br. 14.   However, Appellant has not cited any authority in support of the novel legal argument. The U.S. Patent and Trademark Office is not part of a state (or local) government, but an agency of the U.S. Department of Commerce. Cf. U.S. Const. amend. XIV, § 1 (“No State shall . . . deny to any person within its jurisdiction the equal protection of the laws.”). We are not persuaded that the USPTO has violated Appellant’s rights to equal protection of the laws under the Constitution.[FN1] [FN1]: Absent showing that the agency acted pursuant to some impermissible or arbitrary standard, an argument based on the equal protection component of the Due Process Clause of the Fifth Amendment would also fail. See In re Boulevard Entertainment, Inc., 334 F.3d 1336, 1343 (Fed. Cir. 2003) (TTAB decision that affirmed the refusal to register marks containing the term “jack-off” did not violate equal protection principles, even though the USPTO had previously registered similar marks for other applicants).

Patenting the Virtual World: Webkinz

There's few situations more embarrassing than when your six-year old son figures out that you were playing his video game the night before.  Okay, maybe when your wife stumbles upon you after midnight playing your son's video game...  Alas, I admit to suffering both of these shames last summer when I became fascinated with my children's Webkinz.

For those of you with kids between 5ish and 10ish(?), I don't have to explain.  For the rest of you, Webkinz are Beanie Babies meets Tamagotchi meets facebook (kind of).  In the "real world", you spend a few bucks and get a cute cat/hippo/gorilla/kangaroo/pterodactyl/etc. animal to cuddle.  Inside the toy's package is a code that is then entered into a website to unlock a "virtual" version of your animal.  In the virtual world, you feed and take care of your pet, and earn "kinzcash" by playing games.  The cash can be spent on things like room decorations, food, or gifts that you can "send" to your networked friends.  It's pretty wholesome stuff, not expensive, and the commercialization factor, while present, isn't off the charts (though there is temptation to buy "real" clothing accessories, which come with their own codes, to then see those garments magically appear on the virtual version).  And there is one small, though real 'hook' to your wallet: your registration on the site is good for a year, extendible by... buying and registering another Webkinz.  But for under $10, it's well worth it.

Of course, what I found most intriguing was that there are several patent numbers that appear on the Webkinz website.  Last week, Howard Ganz, creator of Webkinz and grandson of the family-owned company's founder, was granted his sixth Webkinz patent.

U.S. Patent No. 7,604,525 System and method for toy adoption and marketing Assignee: Ganz (Woodbridge, Ontario, Canada) Filed: January 22, 2009 Issued: October 20, 2009 9. A method that provides a virtual world presentation comprising: registering a toy; using a computer system for providing a website associated with the toy, said website having a user account associated with said registering of the toy; in response to and subsequent to said registering, using the computer system for displaying a virtual replica of the toy representing said toy on said website, wherein said virtual replica of said toy is manipulated by a user accessing the website via the user account; in response to said and subsequent to said registering, using the computer system for granting access to the user by way of the user account to at least a subset of features on the website in connection with the virtual replica of the toy; and in response to said and subsequent to said registering, using the computer system to facilitate caring for the virtual replica of the toy presented in the virtual world presentation, wherein said caring for the virtual replica of the toy comprises, and maintaining both of a health and happiness of the replica of the toy via a plurality of maintenance activities, which include all of feeding the toy, shopping for the toy, and playing with the toy, and taking the virtual replica of the toy to a virtual health and welfare facility, wherein playing with the toy comprises participating in an activity on the website in combination with another virtual replica of another toy appearing on the website that is controlled by another user via another user computer over a communication network, wherein an appearance of the virtual replica of the toy is animated using video components for the animation, said animation being used during participation in the activity on the website in combination with the another virtual replica of another toy.

Commentary:  There was an Examiner's Amendment that deleted a clause in the last element, leaving a difficult parsing job (the [, and] before "maintaining" probably should have gone).  We can ignore that for now, in light of all the other good things to discuss here.

• Accelerated Examination.  Ganz's first patent was from an application filed in 2004.  That same application has served as the parent for all of his patents.  Of course, the AE program wasn't around until 2006, and Ganz has now taken full advantage in this continuation patent.  Notably, this patent has five earlier-filed sibling applications that are still pending.  And Ganz unsuccessfuly tried AE on a couple of these, getting petitions denied.  This time they got it right.
  

• Bilski Guidelines. In June 2009, the examiner hit Ganz with a 101 rejection: "As explained in the Interim Guidelines in view of In re Bilski, the first step in determining whether a claim recites patent eligible subject matter is to determine whether the claim falls within one of the four statutory categories..."  That sounds fine, and it is certainly in line with the current Guidelines published in August 2009. (In fact, this same Examiner just rejected a related Ganz application under 101--in October 2009--with exactly the same language.)  But to my knowledge, in June there were no post-Bilski 101 guidelines other than John Love's January 7, 2009 memo, which was silent as to particular steps, and only referred examiners back to MPEP 2106 for specifics.  Was there another, unpublicized set of Interim Guidelines adopted before August?  My guess is that the August guidelines were available and in use for a couple months before going public.

• Machine? Transformation? Both?  To overcome the 101 rejection, Ganz amended the claims, partly as marked above, with several references to computer systems and stressing the linkage between real and virtual worlds (e.g., there was an element reciting a "virtual adoption process" that was subsequently removed by the examiner).  Ganz argued this satisfied not only Bilski's machine prong, but its transformation prong, as well, following Bilski's nod to Abele: "A transformation of raw data into a particular visual depiction of a physical object on a display is sufficient" to satisfy the transformation test, argued Ganz.  

That's not so clear here however. The visual depiction in Webkinz is really a generic animation of "a" Webkinz toy, and not necessary your Webkinz toy.  That is, the same graphic used on the website to display your Gecko is also used for the thousand other kids' Geckos out there.  That's quite different from the x-ray data of Abele, which really was derived from an actual physical article.  So does this virtual image really "particularly" depict a physical object?  It's unclear.  And we don't know if this argument succeeded for Ganz: the examiner allowed the claims without any 101 discussion.

• Scott C. Harris? Yes, the same Scott Harris of Fish/Google fame and who argued the In re Ferguson paradigm non-shifting case appears to have been involved here.  You won't find his signature on any of the papers, but the Examiner notes an August interview with Harris in the Notice of Allowance, where Harris authorized an Examiner's Amendment.  Sure enough, PAIR reveals that Harris is now associated as an attorney of record on this application, working with the Cleveland-based Pearne & Gordon firm.

• Claim strategies.  It's worth looking at the claims in the other Ganz patents.  For example, the parent, U.S. Patent No. 7,534,157, claims the physical toy as an apparatus, but all the elements are directed to the computer ("A toy for utilizing a computer system for providing a virtual world to a user of said toy, the computer system comprising:...").  Sister Patent No. 7,425,169 includes method claims from the perspective of the end user's actions ("A method comprising: first using a first registration code obtained from purchasing a first toy, to access a website;...").  Other patents in the portfolio cover variations on the method claims. Having different claiming approaches like these allows for a wide scope of overlapping claim coverage, and for potential infringement under both direct and indirect theories.

[UPDATE 10/26/09: See comment below on the "transformation" question in light of last week's Ex parte Gardiner decision.]

Be All You Can't Be: Army Patent Ambushed by Friendly Fire

You may recall the brouhaha back in the Reagan era when it was discovered that the Navy had purchased some very expensive ($600) toilet seats.

I remember comedian Emo Phillips around that time joking about his letter to the IRS, noting that the value of the paper clip he’d used to attach his check should cover any deficiency in his return, based on Pentagon spending figures.

Today, it appears the Army may have bought some much more pricey toilet paper to match the Navy’s purchase.  How?  By obtaining a patent that expired five years before its issuance.

Here's the relevant excerpts from the patent’s face:

U.S. Patent No. 7,604,811 Oral-intestinal vaccines against diseases caused by enteropathic organisms using antigens encapsulated within biodegradable-biocompatible microspheres Assignee: The United States of America as represented by the Secretary of the Army Issued: October 20, 2009  Filed: December 10, 1996 Continuation of application No. 08/396,986…which is a continuation in-part of application No. 06/590,308, filed on Mar. 16, 1984, now abandoned…

Because this application was filed after June 8, 1995, its term is 20 years from the filing date of the earliest reference specifically referred to for a priority claim: March 16, 1984.  In other words, this patent that issued today actually expired March 16, 2004 – five and half years ago.  35 U.S.C. Section 154(a)(2).

    "But maybe there is Patent Term Adjustment available…" Nope. The application was filed prior to May 29, 2000.  Nothing there.

    "But maybe there is Patent Term Extension available under 37 CFR 1.701…" Nope. There was no interference, secrecy order, or appellate review.  And even if it were available, it could only extend up to five years – to March 2009.

    "But maybe there is an extension pending FDA approval of the vaccine under Section 156…" Nope.  That would also be a maximum extension of five years under Section 156.

This patent is DOA.  If anyone has a good explanation as to why prosecution of this patent was pursued, please share.  (And note that if your answer is "to get the app published," there were other, less burdensome routes available, such as voluntary publication under 1.221.)

So let's see where your tax dollars went over the last five years in pursuing this stillborn:

• Four responses to non-final office actions
• Three responses to final rejections, including two RCEs
• Two Examiner interviews
• One terminal disclaimer filed
• One notice of appeal
• One appeal brief
• One petition to correct priority
• One issue fee payment

That's over $5,700 just in PTO fees.  In this case, those fees are effectively a shifting of government funds from one entity to another.  But it’s hardly an efficient use of resources.  And I wouldn't be at all surprised if the Army spent three times this amount for the attorney's corresponding time.

And of course, there are several interesting and crazy twists:

• First, you did read that list correctly.  The applicant filed a TERMINAL DISCLAIMER in May 2006 – a year after the patent had already expired –to obviate a double patenting rejection.  That's like a baseball team staging a press conference to announce it's not exercising the option to keep the player who retired the year before.

• Second, the Army filed a petition in October 2007 to correct the priority claim to get around an obviousness rejection.  The cited art was a PCT publication from overlapping inventors, and the petition corrected the priority claim to include the underlying 1991 U.S. application of that PCT app.  That is a different app from the fatal 1984 application. But the petition clearly includes the chain going all the way back to the 1984 application, which appeared as a priority document in today’s patent app as originally filed.  In other words, if it wasn’t apparent up to that point (hard to imagine), the Army filed a petition two years ago which on its face evidenced that any resulting patent would not be entitled to any term.

• Third, and perhaps most interesting, the priority claim to the 1984 reference may not have been even necessary.  In the last of the final rejections, the Examiner noted explicitly that the 1984 application "do[es] not teach the specific CFA/I peptides employed in the vaccine and method of the instant claims.  Accordingly, the priority date of the instant application is the filing date of its immediate parent, U.S. Patent Application 08/396,986, filed 3/01/95."

This raises a question: if an application is a continuation-in-part and none of its claims are supported by the ancestor application, should its term be determined by the date of that ancestor app? (Set aside that it makes no sense to file such a C-I-P…)  I don’t think so.  The language of the statute is clear that it is the “specific reference” to the earlier app that determines a patent’s term.  Indeed, the Army could have disclaimed priority to the 1984 application and removed that specific reference.  See MPEP 201.11 III(G) (which amazingly anticipated this type of post-GATT situation).  But it looks like now they’d need to get it reissued. And the Army would probably be estopped from disclaiming priority earlier than April 1991, the date of the underlying PCT application.  Would it take less than a year and a half to push a reissue through? And it doesn't feel right that the Army should serendipitously benefit from the examiner's unsolicited priority statement, either.

• Lastly, and true to form with common perception of government efficiency, care to guess what the Army did prior this patent's issuance?  That's right: it filed a continuation application.  

Sure, it's not Charmin.  But really, how abrasive do you think that official U.S. Patent is?  I'll trade you a couple paper clips for it...

Contest: Longest claim ever?

Last week the PatentlyO blog ran a guest post by Stephen Schott arguing for a repeal of the "single sentence" rule.  Schott alluded to the existence of at least one patent claim consisting of "a 250 word sentence [that] would drive even Proust to distraction."

IBM today was issued U.S. Patent No. 7,593,860.  The patent contains four independent claims, the shortest of which contains 1,019 words.  Here is the longest, weighing in at 1,053.  (And we can leave the Bilski discussion alone for now.)

Can you find any patents with longer claims?  Submit your entry by 12:01 a.m., Friday, September 25.  Appropriate prize to be determined (initial thoughts: a copy of any Faulkner novel; an audio of a Joe Biden speech; a recording of Wagner's Ring Cycle... you get the idea.)

U.S. Patent No. 7,593,860 Career analysis method and system Assignee: IBM 31. A process for integrating computing infrastructure, comprising integrating computer-readable code into a computer system, wherein the computer system comprises a computer readable medium, wherein said computer usable medium comprises job related data, wherein said job related data comprises a first specified job title and a first list comprising a first plurality of required skills for said first specified job title, wherein said job related data further comprising a plurality of additional job titles, and wherein the code in combination with the computer system is capable of performing a method comprising:  receiving by said computing system, job candidate data, said job candidate data comprising a job candidate and a second list comprising a second plurality of required skills related to a current job held by said job candidate, wherein said current job held by said candidate differs from a job associated with said specified job title;  receiving, by said computing system, first job market data, wherein said first job market data comprises first internal market data and first external market data, wherein said first internal market data comprises first capacity planning data associated with an entity and first workforce management data associated with said entity, and wherein said entity is associated with said first specified job title;  storing within said computer readable medium, said job candidate data and said first job market data;  comparing, said first plurality of required skills with said second plurality of required skills to determine a first set of common skills between said first plurality of required skills and said second plurality of skills;  applying, by said career analysis tool, a first set of weighting factors to said first set of common skills, wherein each weighting factor of said first set of weighting factors is applied to an associated common skill of said first set of common skills based on a predetermined relative importance for each said associated common skill with respect to said first specified job title;  calculating, by said career analysis tool, a first score for said candidate by dividing the number of said first set of common skills by the number of said first plurality of required skills;  comparing by said career analysis tool, said first job market data with said first plurality of required skills to determine a first set of market valued skills from said first plurality of required skills, wherein said first set of market valued skills indicate a first group of critical market place skills required to perform a specified job associated with said first specified job title, wherein said first group of critical market place skills are linked with first business unit skills associated with achieving first business priorities comprising growth in revenue and innovation, and wherein said first group of critical market place skills are aligned with a first specified market direction and a first specified business strategy for said entity;  reviewing, by said career analysis tool after a specified time period, said first set of market valued skills;  determining, by said career analysis tool based on results of said reviewing said first set of market valued skills, that said first set of market valued skills are outdated;  receiving, by said computing system in response to said determining that said first set of market valued skills are outdated, second updated job market data, wherein said second, job market data comprises second internal market data and second external market data, and wherein said second internal market data comprises second capacity planning data associated with said entity and second workforce management data associated with said entity;  storing within said computer readable medium, said second updated job market data;  comparing by said career analysis tool, said second updated job market data with said first plurality of required skills to determine a second set of market valued skills from said first plurality of required skills, wherein said second set of market valued skills differ from said first set of market valued skills, wherein said second set of market valued skills indicate a second group of critical market place skills required to perform said specified job associated with said first specified job title, wherein said second group of critical market place skills are linked with second business unit skills associated with achieving second business priorities comprising growth in revenue and innovation, and wherein said second group of critical market place skills are aligned with a second specified market direction and a second specified business strategy for said entity;  applying, by said career analysis tool, a second set of weighting factors to said second set of market valued skills, wherein each weighting factor of said second set of weighting factors is applied to an associated market valued skill of said second set of market valued skills based on a an amount of time that each said associated market valued skill has been determined to be a market valued skill;  calculating by said career analysis tool, a second score for said job candidate by dividing the number of said second set of market valued skills by the number of said first plurality of required skills;  comparing by said career analysis tool, said first score to a first numerical value;  comparing by said career analysis tool, said second score to a second numerical value;  comparing by said career analysis tool, said first plurality of required skills with said second plurality of required skills to determine a set of required skills of said second plurality of required skills that are not comprised by said first plurality of required skills;  generating by said career analysis tool, a training program for said job candidate to obtain training associated with said set of required skills;  determining by said career analysis tool, results of said comparing said first score to said first numerical value and said comparing said second score to said second numerical value;  determining by said career analysis tool based on said results, that a first analysis score for said job candidate in relation to said first specified job title will be calculated;  calculating by said career analysis tool, said first analysis score, wherein said calculating said first analysis score comprises adding said first score to said second score;  comparing said first analysis score to a first predetermined value; and  performing an action for said job candidate based on results of said comparing said first analysis score to said first predetermined value.

Happy Birthday, Beauregard

Twenty years ago tomorrow, U.S. Patent No. 4,864,492 issued to IBM for a “System and method for network configuration.”  On its face, nothing appears out of the ordinary with this patent.  However, lurking within its claims is this:

10. A computer program product having a computer readable medium having computer program logic recorded thereon for establishing a network configuration in accordance with a plurality of protocols of a network architecture, said computer program product comprising: means for incorporating the protocols of the network architecture into a table of rules; means for determining the validity of user input data with said table of rules; and means for feeding back to the user at least one of said rules in said table of rules if said user input is determined to be in valid.

 

And thus, the era of the “computer readable medium” claim was born (albeit with 112/6 "means for" language seldom seen in today's CRM claims).

For those unfamiliar with the “software arts,” here’s the nutshell on CRM claims: rather than simply claiming the method (i.e., the steps performed by a computer program) or a computer system (e.g., the hardware configuration that can be used for performing a computer program), the CRM claim essentially tries to protect the floppy disk (remember those?) or CD, DVD, flash drive, etc. on which the program is stored.  For practical purposes, these claims allow software companies, for example, to go after manufacturers of competing programs on theories of direct infringement—they made and sold infringing disks.  When software claims are limited to systems and methods performed via the end user’s computer, often only indirect infringement theories are available.  More importantly, it may be easier to demonstrate infringement by showing a sale of a disk than by finding an actual instance where method steps were performed.  And of course, that sale would still infringe the CRM claim even if the disk were taken to Mexico and the encoded method was actually performed there -- not so with a straight method claim.

The attorney who prosecuted the ’492 patent, the former Marilyn Smith (now Dawkins), also prosecuted a more famous patent for IBM, U.S. Patent No. 5,710,578 to Beauregard et al.  This patent’s prosecution, of course, went up to the CAFC on the question of the patentability of such CRM claims under Section 101.  In re Beauregard, 53 F.3d 1583 (Fed. Cir. 1995).   As a result, CRM claims are often referred to as “Beauregard claims”. (Note, however, that that the Beauregard case was vacated, and its merits were thus never ruled upon by the CAFC.)  Interestingly, Dawkins apparently left IBM and now works for (John) Deere & Co. in Moline, Illinois.  But she co-authored a 1995 IDEA journal article on CRM claims that is loaded with examples and explanations.   Dawkins notes that there was an earlier patent, 4,853,962, that claimed the CRM as a multiple-dependent claim of a method:

10. An information storage device tangibly embodying a program comprising instructions adaptable for execution on a programmed machine, wherein the method of any one of claims 1-9 is performable through the execution of said instructions by the machine.

CRM claims are commonplace today.  How common, you ask?  Consider the following charts showing the history of CRM claim issuances, in absolute numbers as and as a percentage of total patents issued:

The upturn of the hockey stick in the later half of the 90's corresponds directly to the Beauregard case working its way up through the courts, casting more light on the claim format.  But it is almost certain that these charts undercount the actual number: there are many variants of language used in drafting CRM claims, e.g., “machine readable medium”, “computer readable storage medium”, etc.  And now, in light of In re Nuijten, you are seeing more tweaks to the standard language to make clear that the medium is limited to a physical storage device.  It is impossible to capture all these permutations with reasonable effort (using the public PTO database, anyway).

Growing even faster is the rate at which CRM claims are being filed, (note the gap in data sources beginning in 2001, when the PTO started publishing pending applications)::

The last data point shows that 5.5% of all applications filed in 2007 (that subsequently published) contained CRM claims.  That's a hefty percentage, but still far short of what we saw for Jepson claims in the early 1980's.

We can only guess how CRM claims will fare after the Supreme Court’s forthcoming decision on Bilski.  Already we have seen inconsistent results from the BPAI.  In Ex Parte Li, for example, the Board gave tremendous weight to Beauregard claims and found that they were outside Bilski’s purview.  Other panels, however, such as in Ex Parte Cornea-Hasegan, have found that CRM claims may just be exaltations of form over substance, and thus are subject to the Machine or Transformation test.

But until things are sorted out, a CRM claim will likely remain as standard ammunition in the software practitioner’s arsenal.

Five for Netflix

It’s hard to believe that Netflix is already eleven years old.  The company that Reed Hastings started after paying a $40 late fee to Blockbuster now stocks over 55 million DVDs that it rents to 10 million subscribers.  I’m sure that if my wife and I ever find some of that time that seems to have evaporated once we had kids, we’ll join up.  (Not that we're complaining...we do love our kids.  Really.)

Surprisingly, until today Netflix only had four patents to its name.  One of them, U.S. Patent No. 7,024,381, was famously asserted against Blockbuster the day it issued, April 4, 2006.  That case settled a year later while Blockbuster’s summary judgment motion was pending.

But today Mr. Hastings and Netflix received their fifth patent.  And no, they don’t need to send back a patent to the USPTO in the supplied pre-paid envelope before receiving number six.

U.S. Patent No. 7.546,252 Approach for managing rental items across a plurality of distribution locations Assignee: Netflix, Inc. 1. A method for managing digital versatile discs (DVDs) carrying movies in distribution locations, the method comprising the machine-implemented steps of: identifying a first set of one or more DVDs located at a first distribution location, wherein the first set of one or more DVDs carry a first set of movies that has not been requested for rent by rental customers associated with the first distribution location; causing the first set of one or more DVDs to be sent from the first distribution location to a designated distribution location; identifying a second set of one or more DVDs located at a second distribution location, wherein the second set of DVDs carry a second set of movies that has not been requested for rent by rental customers associated with the second distribution location; and causing the second set of one or more DVDs to be sent from the second distribution location to the designated distribution location, wherein the first, second and designated distribution locations are distinct distribution locations..

Commentary: First, the obvious question.  Is there a Bilski problem?  (And even though the Supreme Court has granted cert, it’s still the law, at least for now…)  For now, I'll posit that this claim sails clear:  The DVDs here are apparatuses that are probably “particular” enough to satisfy the machine or transformation test.  It helps that the claimed method makes active use of the movies contained on the DVDs, too.  In that way, the DVDs aren’t just any ol’ object to be distributed – it matters that the DVDs carry unrequested movies.  This “ties” the DVDs to the method through the identifying step.  So perhaps this claim is a good example of how Bilski might not have killed all business methods.  (Of course, obviousness may be a different story, right?)

Or perhaps not…

Compare claim 4:

4. A method for managing rental items in distribution locations, the method comprising the machine-implemented steps of: a computer identifying a first set of one or more rental items at a first distribution location that is not needed by customers associated with the first distribution location; the computer causing the first set of one or more rental items to be sent from the first distribution location to a designated distribution location; the computer identifying a second set of one or more rental items at a second distribution location that is not needed by customers associated with the second distribution location; and the computer causing the second set of one or more rental items to be sent from the second distribution location to the designated distribution location, wherein the first, second and designated distribution locations are distinct distribution locations.

The claims are nearly identical, but with two key differences.  Think subject-object.  Claim 1 is a “general” method, where the subject is anything or anyone, but the object is “particular” apparatus (DVDs containing movies).  (Yes, the preamble says "machine-implemented steps"...but is that really limiting?)

Claim 4 is reversed: the subject is a computer, which performs the method on a “general” object (rental items).  It seems clear from recent cases that the CAFC intended to kill the Claim 4-type scenario where the subject “computer” isn’t particularized.  And it isn't even close here.  (The “computer” terms were added in an Examiner’s Amendment, by the way, with no explanation given, in April 2009.  In re Bilski had been fully digested by this time.)

But what about Claim 1?  Should it fall by Bilski's sword, as well? I’ll leave the discussion open.

Some other interesting prosecution notes:

• Priority disclaimed.  The app was originally submitted as a C-I-P claiming priority back to Netflix’s first patent, 6,966,484, which was filed in September 2002.  During prosecution, Netflix amended their specification by deleting the priority claim language.  PAIR does not indicate any parent application.  So by doing this, Netflix may have gained over a year of patent term (this app was filed in December 2003), but at the risk of potentially exposing itself to some intervening art or earlier on-sale activity.  I suspect this disclaimer, which occurred in September 2007, was provoked by the then-looming rule change for continuation applications.  Those rules, to have taken effect Nov. 1, 2007 but for the Tafas injunction, would have limited continuation practice and the number of claims.

• Term Extended? You'd think that because of the disclaimed priority, the claims would be hit with a 103 rejection over the '484 patent, or at least an obviousness-type double patenting rejection leading to a terminal disclaimer.  Not so.  Instead, Netflix actually had term extended by over 800 days for PTE.  Go figure.
  

• Weaker presumption?  What do we make of the Examiner’s statement in the Notice of Allowability:
  

“Applicant’s voluminous IDS received August 28, 2008 has been given a cursory review.” 

Should an accused infringer need to present clear & convincing evidence to show invalidity in this situation where the Examiner essentially admits he did not read the references?

• Undue Multiplicity.  The only claim rejections were under 112 for indefiniteness under the “undue multiplicity” rule.  See 37 CFR 1.75(b) and MPEP 2173.05(n).  The examiner found that the 126 claims originally presented “would tend to obfuscate, confuse, and becloud the claimed invention.” 

Undue multiplicity rejections are not so common anymore, in part because they are almost always reversed by the BPAI when challenged.  The conventional wisdom is that they were historically used by examiners to trim workload, requiring applicants to pick some subset of representative claims for examination. (Indeed, the doctrine was cited in Tafas for the proposition that it is possible to place at least some limits on number of claims in an application.)

It has been a recent curiosity of mine whether an alleged infringer could challenge a patent’s validity in a civil litigation under this Rule.  After all, if “the net result” of the repetitious claims tends “to confuse rather than to clarify” the subject matter of the invention, then it should not matter if this challenge is made by an examiner during prosecution, or if it is made by a lawsuit defendant after issuance -- either the claims are indefinite or they are not, and the statute applies equally regardless of timing.

In a similar style to the successful argument reviving prosecution laches with Lemelson, I suspect there is some case out there with a patent having a ton of claims where it could be argued, “If undue multiplicity doesn’t apply here, then it doesn’t apply anywhere.”  It's a longshot, but the right district court judge just might bite...

Any takers?

·

Endangered Species: the Jepson Claim

In one of my posts last week, I noted that a featured patent included Jepson claims.  Jepson claims are not dead (either "merely" or "really most sincerely"), but they may be on their way to extinction.

Jepson claims have been around since at least 1917.  Such claims typically begin with a preamble of already-known elements ("In a system for making widgets with elements A, B and C"), then a transitional phrase ("..the improvement comprising..."), followed by the inventive elements presently claimed ("...elements D and E...")  It has been said that in a Jepson claim, "the claim preamble defines not only the context of the claimed invention, but also its scope." Rowe v. Dror, 112 F.3d 473 (Fed. Cir. 1997).

Jepson claims bear some similarlity to Eurpoean claim forms that use "characterized by..." language.  And like other claim forms, Jepson claims have their pros and cons.  Wikipedia (admit it--you look there, too) actually gives a good summary, noting that Jepson claims can be useful in allowing the Examiner to hone in quickly on points of novelty, particularly in crowded arts.  This could result in fewer Festo-inducing arguments and amendments.  But the big downside, of course, is that there is an admission that the preamble's limitations are known in the art, and presumably (rebuttable) known by others for 103 purposes.  See MPEP 2129.  In a post-KSR world, where the apparently best obviousness attack is to show that an element is missing in the cited references, that doesn't leave much wiggle room.

It is not news that use of Jepson claims is atypical.  Landis writes that "the occasion is rare when it is the recommended format" and that "It is seldom used in practice."  Judge Newman echoed this thought over twenty years ago when she noted that the Jepson form of claim "occasionally continues to be used", and was thus an exception to "the modern style, [where] patent claims no longer merely claim the salient features, the 'heart' of the invention." Pennwalt Corp. v. Durand-Wayland Inc., 833 F.2d 931 (Fed. Cir. 1987) (en banc) (Newman, J., dissenting). 

Less known, however, is that the use of Jepson claims has been dropping at a breathtaking rate.  When Judge Newman made her comments, nearly 8,000 patents with Jepson claims were issuing annually.  Today that number is closer to 1,000.  The remarkable graph below represents the number of patents which include the word "improvement" in their claims, shown annually since 1976.

The law regarding Jepson claims has been pretty stable over this period, so the root cause of this drop-off is probably found elsewhere.  Perhaps it was simply a natural process as new attorneys/agents entered the practice and opted to try claiming more broadly.  Or perhaps there was a growing perception that courts were looking with increasing favor at broadly drafted claims.  I'm not sure.  But if you have a good theory--or a compelling argument why you think Jepson claims will/should have a big comeback--please share.

Anybody care to start a "Save the Jepson claim!" campaign on Facebook?

[UPDATE 6/4/09 1:56PM: The graph below shows the issuance of Jepson patents as a percentage of total issuances.  Peak is over 14% (that's 1 in 7 patents, friends) in 1980, dropping steadily to about 0.7% in 2008]