Re: what does a patent restrict?

On Fri, 22 Oct 1999, Andrew C. Greenberg <werdna[_at_]gate.net> wrote:
>
> On Thu, 21 Oct 1999, Mike Oliver <mikeoliver[_at_]home.com> wrote:
> >
> > [Re: software distributors] I just think that the proper analysis is
> > under indirect infringement principles -- which all require someone
> > to have committed direct infringement as a predicate.
>
> Well, I suppose what I am awaiting is a reason why Mike thinks this
> should be the case.

What the heck <g>, I'll try it. It comes down to this for me (and this makes sense only in the open source context) - if an article of manufacture claim is recognized valid for software solely as it exists in a storage medium, a distributor of open source code is liable for direct infringement, and has no defence that there are substantial non-infringing uses of the article. If, on the other hand, distribution of the article could only be claimed to be a contributory infringement (or active inducement) in connection with the making, use or sale of the machine in which the software operates, I believe that the distributor would have a decent defence because *in the open source context* one substantial use of the code that would probably not infringe the other process claims would be the educational value of viewing how the code works.

Viewed from an economic perspective, distributors are not the actors in a chain of conduct that constitutes infringement, that have great economic incentive to continue distributing in the face of an infringement claim. Their incentive is vastly disproportionate to the risk they face on direct infringement. Therefore they just cave. And they are more likely to cave if they are subject to a direct infringement claim than if they at least have the defences under section 271(c). Thus, I conclude (perhaps without real justification) that a software patent owner can scare off the distributors of the open source a lot easier than it might be able to scare off the publisher (who has a greater economic incentive to fight).

All of this, however, is probably pointless. First, its unlikely that only the article of manufacture claim will be in issue (none of the cases you cited involved article of manufacture claims or infringement of them). Second, as a practical matter, a distributor is not going to continue distributing an accused device just because he or she might have a 271(c) defense. They are going to cave, and probably even if the publisher provides a beefy indemnity clause. They will just distribute some other code that does not infringe.

This I guess also explains why I think that copyright law is the more appropriate legal construct that should be applied in the software distribution context. At least in copyright cases some nexus between the infringer and the software has to be shown (access and substantial similarity), there is no discovery on novelty or non-obvious issues that so consume and conflate patent litigation, and when infringement is found, almost always, the distributor had some level of knowledge of the work that was infringed -- some true culpability

How many software developers bother wading through the tens of thousands of ridiculous software patents to discover some nugget of technology. My totally subjective belief is very few. Why, because its a lot faster to just come up with a solution to a practical problem by coding it, instead of researching how others did it and then designing around a poorly drafted (and in my view often incomprehensible) software claim.

I think the 1st amendment advocates have it right -- that there ought to be a privilege to distribute open source software code for educational purposes free of the fear that such distribution will infringe a patent claim.

Like you said, it ain't over and we should watch this Amazon case closely.

-mike oliver
bowie & jensen, llc
<mikeoliver[_at_]home.com> Received on Wed Oct 27 1999 - 23:13:25 GMT