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?

·