More on Patent Reform

March 16, 2011

It is spring break here so I had a little more time to poke around on the issues surrounding patent reform, particularly the First-to-File/Weakened-Grace-Period (FTF/WGP) issue that I wrote about yesterday  I was surprised that both the Kauffman Foundation and AUTM are in favor of the reform so am wondering what I am missing.  I still believe that it is a mistake and agree with this letter from Senator Feinstein.

Let me describe the situation that concerns me under the new law.

First, lets set up a scenario where a startup or university invents something.  Remember that invention in this case is often not a eureka! moment but rather an observation followed by improvement followed by more improvements and applications.  All of those might be patentable.  My key point is that invention of a new polymer, a new drug, a new type of solar cell, etc. is not like invention of a new paperclip.  Instead, the invention value improves as the scientist works on it.  For purposes of this example, consider a scientist that discovers a new polymer but is still working on how to better manufacture it, optimal formulation, and new applications that were never possible before.

Under the new laws, as I understand them, you would either have to be FTF or, to get a grace period to file, you can get a grace period of a year if (AND ONLY IF) you publicly disclose the invention.  So, the startup or university is faced with a choice – (1) file good provisional patents for every possible invention right away, (2) disclose the invention to the public to retain the grace period or (3) keep it secret, file a patent application later, hope that nobody beats you to the patent office.  Lets walk through each one:

(1) File good provisional patents on everything. Universities cannot possibly do this (if you disagree with me on this, you have never worked with university technology transfer to see how busy they are already).  A startup might be able to do it but it would have a cost.  I think that best practices for a VC funded startup might be to file provisional patents on EVERYTHING once a month, probably requiring 0.5 to 1.0 FTE plus increased legal expenses.

The Problem with (1) is that you can still be scooped. You invented the polymer, talked to half a dozen venture capital funds about it before getting funded.  VC funds don’t generally sign NDAs but you are going to have to tell them something.  Those funds sent your general, non-confidential description out to a number of really smart polymer scientists, some at large companies, to see what they think.  What are the odds that some of those scientists either connect the dots to figure out what you are keeping secret or, now that they know that your miracle polymer is possible, start to think about how to create one? Or what if they just start filing patent applications on the applications.  This can result is a much more limited business model (for instance, you own the idea on the polymer but someone else has patented applications).  As a VC, I can probably live with this but it does raise the risk that a large company positions its patent applications to limit the startup.

(2) Publicly disclose all inventions. First, though it is not explicit, I assume that while this preserves rights in the USA, the public disclosure is a bar to foreign patent filings.  That is a deal killer for VC funded deals.  You need to have rights into the rest of the world, not just the US.  Also, it has the same problem of letting the other guys know what you are doing, but worse – now you are showing the competition how to make your new, perhaps unoptimized polymer and letting them invent improvements before you have had a chance to do so.

A university may have other issues with this one – scientific journals usually require no-pre publication of research AND research funding is very competitive so, I doubt that the researchers will want to disclose exciting new work publicly before putting in more grant proposals in the area AND good researchers want to confirm and re-confirm before publishing, that can take months after the initial “invention”.

(3) Keep it Secret. That might be doable for a startup but I’ve always had my companies keep good lab notebooks and records because first-to-invent allows you to keep it secret without the risk of someone figuring out what you are doing and patenting it first.  In that case you can at least argue that you invented it first.  Why would a VC put money into a company when any researcher, anywhere in the world, could come behind you and take all the intellectual property rights? This is a particularly large risk with a transformative technology – the risk of leakage of information (at least hints that the other guys can follow) is high.  Universities have an even more difficult time with this one.  What happens when your graduate student goes for a job interview? They have to present their work.  What happens when you file for an NSF/NIH grant that 20 or so colleagues will read?

So, in my opinion, we are left with only one viable option for startups- continually file provisional applications (unless those get reformed away).  I can see this working (however with some expense to the company) for startups that have VC funding but for those startups that have yet to get funding, with limited budgets, there is a catch-22.  If they don’t spend the money, that they may not have, VC and other investors will be less likely to fund the startup because risk is increased.

I don’t see how universities win in this situation given the difficulty in continual provisional filing which is why AUTM’s support is puzzling to me.

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: