jump to navigation

The swarm strategy, 2 years later May 25, 2011

Posted by Brian Schar in Patent prosecution, USPTO.

When the PTO proposed the continuation rules that never went anywhere, and after the insanity of McKesson that purports to require applicants to cite office actions from other applications in the family in a particular application, I thought about how one could deal with the increasing insanity and chaos of PTO procedure.  The biggest problem with McKesson was that, taken to the extreme, it would prevent any application in a family from ever issuing.  In a family tree with, say, 8 members, one of them would always have an outstanding office action, such that you would have to file an RCE in any related application that had received a notice of allowance, or such that you would have to withdraw any related application that was pending issue.  As a result, the continuation rules would have killed off application families, because it would have restricted the ability of applicants to file RCEs and continuations to allow the PTO to actually consider the references it was requiring under McKesson

Thankfully, the continuation rules died the ignominous death that they deserved.  Nevertheless, the McKesson problem continued.  Further, the 20-year-from-filing term made CIPs significantly less valuable then they had been in the past.  The fundamental problem with McKesson was the coupling of applications together via a priority chain.  The solution was to decouple applications, and thus was born the swarm strategy.

The swarm strategy is to file multiple small applications independently of one another, each directed to an individual, small piece of an overall invention.  Rather than filing an omnibus application and spawning divisionals and continuations to pursue details and narrower claims, the swarm lets one file 4-5 applications where one would have been filed previously.  Instead of an 80-page omnibus, five 15-page applications could be prepared in the same amount of time.  Because each is directed to a different aspect of an invention that is not disclosed in the other applications, they do not claim priority to one another, are not related under McKesson, and thus the McKesson problem is overcome.  Further, the odds of one of the 4-5 application sized “swarm” getting issued within a 3-4 year period are greater than the odds of a single omnibus application getting issued in the same time frame.

I’ve pursued the swarm strategy for a few years.  On the plus side, I file fewer supplemental IDSs, and have a much reduced need to file RCEs solely to have ridiculous, irrelevant art from related applications considered.  (That’s particularly useful now, with RCEs going right to the back of the line.)  Swarm applications, generally being 20 pages or less, are picked up for examination substantially faster than longer cases.  Examiners apparently have a 6 month or so window for picking what application they want to work on next, and smaller applications are always preferred – especially at the end of a quarter.  The claims are more focused and the Examiners tend to understand the applications better.

However, the downside of the swarm strategy is that the short applications tend to be directed to rookie examiners.  By and large, I like the rookies, most of whom seem to be earnest young men and women who honestly want to do a good job.  However, rookies aren’t always trained very well, and don’t always know quite what they’re doing.   As a result, the first time one of those cases gets actual adult supervision is often only after an appeal brief has been filed.  Another issue is poorly-thought-out obviousness rejections.  Where an application in a swarm is directed to an improvement to an invention that’s disclosed in a previously published application or a previously-issued patent, where that publication or patent is incorporated by reference, that incorporation by reference draws the obviousness foul.  The obviousness rejection is typically along the lines of “it would be obvious to improve published application 2009/XXXXXXXX as claimed in claims 1-10.”  These rejections can be overcome, but they are annoying.

I am considering changing the swarm strategy to lengthen my applications rather than incorporate material by reference.  An example: Where published application 2010/XXXXXXXX has been improved with a feature A that is claimed in a subsequent application, the previous short-application swarm strategy application would simply incorporate 2010/XXXXXXXX by reference, to ensure enablement, and would then only disclose the new feature A that was not in 2010/XXXXXXXX.  However, if the new application were written like a CIP to the earlier published application, but did not claim priority to that application, that application would have the identical content, and require the same amount of work, but would be longer, and would not need an express incorporation by reference statement.  (Of course, prudence would require that the earlier published application 2010/XXXXXXXX be identified in an IDS.)  The longer application would be less likely to be diverted to the most junior members of the examining corps, and would be less likely to draw the easy (and wrong) obviousness foul.

In two more years I’ll post how that’s going.


No comments yet — be the first.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: