Paul Allen sues just about everyone for patent violations

Interesting tech news. Seems like he’s claiming some pretty core stuff is patented:

The patent system is broken, and does more harm than help to encouraging invention

Seriously, how can you really patent stuff like this?

Specifically, the patents cover:

– “Browser for Use in Navigating a Body of Information, With Particular Application to Browsing Information Represented By Audiovisual Data.”

– “Attention Manager for Occupying the Peripheral Attention of a Person in the Vicinity of a Display Device.”

– “Alerting Users to Items of Current Interest.”

Those are concepts. The patentable part should be the code and not the concept. It would be like patenting the idea of a word processor.

I know nothing about patent law of course, but on the surface all this seems absurd.

I used to admire Paul Allen, because the idea of getting super-rich and then quitting and having super-rich fun (sports teams and science fiction museums! why not!) is great, and I admire someone who can get outside of the “must get EVEN RICHER” mindset.

But patent trolling, ugh. Why couldn’t he just get into dog-fighting if he had impulses toward being a slimy jerk?

I’m actually somewhat glad to see lawsuits like this, Oracle vs Google, etc and hope this trend continues because while these suits are stupid, having big names go after each other on patents instead of just doing backroom cross-licensing deals should eventually force the USPTO to get its shit together on software patents.

I am 95% certain Oracle/Google will be settled with a backroom cross-licensing deal.

This kind of stuff is more blatant, because the company suing has no products and can’t be intimidated by patent counter-suit… but I also don’t think it’ll “heighten the contradictions” enough to push anyone over the edge into sensibly abolishing software patents, alas.

The PTO has kind of done stuff to address software patents, but there’s very little that they can do about it.

You can’t patent a concept or idea, only an invention that has been “reduced to practice.” You have to, in the patent, teach someone else how to do it, with specific examples of the invention reduced to practice. So, for example (real world something I worked on) you can’t patent the idea of putting an anti-cancer drug inside of something that takes it directly to the tumor, to focus the cell killing to the specific tumor and thus reduce the toxicity of the drug to the rest of the body (great concept!) - you have to provide an actual working example of a specific anti-tumor agent contained in a protective vectoring agent, tell how to make it, and give data on how well it actually worked and how it killed the tumor with less side effects than a control you compared it with.

The purpose of the patent system was to have new ideas and technology revealed so that others could practice it, rather than kept as a secret. In exchange for teaching the world how to do it, you got protection for a certain amount of time (used to be 17 years, now 20.)

What’s the point of being innovative if there’s no incentive to improve X if some company owns the patent to the concept of X instead of just one method of achieving X?

That’s exactly the question the patent office should be asking themselves and the lawmakers, Eric. The current way is not helping innovation the least.

While I agree with the purpose of the patent system in general, “Reduced to practice” is not a higher barrier than or clearly distinguishable from “concept” for many software inventions, which is why patents are so problematic in that field.

Yeah, I have to admit that I am ignorant of the actual state of the art in software IP practices. My world has been primarily chemistry, Bio, engineering, etc.

But let’s say you wanted to patent, say, windows that automatically pop up ads when a user types specific words. I would assume you’d have to submit examples, as in all other fields, in which you demonstrated that you’ve actually reduced that to practice (and how) - for example, actual code of some type? What has always been stressed whenever I was in the patent office in Washington (we’d go personally make a case for some applications in which it was obvious the examiner just wasn’t “getting it”) is that you can only invent “things” and not “ideas.” The definition of what is a “thing” has broadened over the years, but the point was an idea was absolutely not patentable in and on its own.

Well, for your proposed patent the patent claims would contain sample code and a very detailed description of what the code does, as well as detailed descriptions of alternate implementations of the same idea. If you do that (and I’m sure such a patent exists) you’ll have an infringement case against anyone else who comes up with the same concept, because many aspects of the implementation are implied by the concept. Meanwhile, other people come up with the same concept-- in no way making use of the information supposedly made available by the patent, and they are suddenly susceptible to an infringement lawsuit.

It’s very difficult to release any software that does not infringe on some patent or other. Software patents don’t protect innovation–they provide a barrier to entry for new players (who don’t have a lot of money to pay lawyers, or a pool of patents that could be used for retaliation) or, when wielded by patent trolls that don’t themselves produce any product and as such are not susceptible to retaliation, a tax on innovation.

The standard is that you have to describe it in sufficient detail that someone “skilled in the art” could reproduce it. Thus, logical diagrams or pseudo-code are generally adequate.

Right. Which is why the trick is always to write the patent in a way that allows someone to reproduce it, without actually giving away “the best” part of the invention. <G>

I’ve been involved in patenting for about 30 years, but I admit I don’t really understand how it works in practice in areas such as software, and it gets fuzzy in my head when I see what people are trying to do with life forms.