A modest proposal towards patent reform

[unixronin]

I had a thought about patents while reading this C|Net article, which discusses the recent acquisition of Linux-related patents from Microsoft by the Open Invention Network, the fact that OIN was excluded by Microsoft from bidding directly for them, and speculates that Microsoft may have been attempting to "accidentally" arrange for the patents to fall into the hands of a litigious atent troll who might then go and file a court case against Linux developers and users based on the patents, much after the example of SCO.  It occurred to me that shenanigans of this type could be completely prevented in future by making an essentially-simple change to copyright law.

The change I would propose is this:

Patents cannot be sold.  They can only be licensed or released.  The original owner of a patent may, for an appropriate consideration, license a patent to any licensee exactly as now, and exactly as now that license may be exclusive or not, and may be for a limited term or for the life of the patent, but ownership of the patent always remains with the original patent owner.  Only the original patent owner may sell licenses to that patent; a licensee, even one with an exclusive lifetime license, cannot re-license the patent; but an exclusive lifetime license, once granted, cannot be revoked by the patent owner, exactly as though the patent had been sold ... as long as the licensee actually uses the licensed technology.  Patent licenses are valid only so long as the licensed technology is in use by the licensee; if you license a patent but then don't actually use the licensed technology within, say, one year, the license reverts to the patent owner.  The licensee has full rights to use the licensed patent, but the only right of transfer the licensee has is to release their license, in which case it reverts to the original owner.

Only the original patent owner has the right to litigate for infringement of the patent, though the patent owner has the right to delegate this power to a licensee of the patent to litigate against a specific infringer on the patent owner's behalf.  Thus, a licensee of a patent may not litigate against an alleged infringer of that patent without the knowledge and approval of the patent owner.

The owner of a patent may, at his or her discretion, at any time when no outstanding licenses to the patent exist, opt to release the patent, in which case the technology covered by the patent becomes public domain and no part of the technology covered by the patent may subsequently be re-patented.

If you are the original owner of a patent, this affects you in two ways:

1. You can no longer lose control of your patent.  Anyone you would have sold it to, you can now instead sell an exclusive lifetime license for the same price.

2. You can no longer transfer your patent to a troll who will litigate for you by proxy, and have plausible deniability that you intended it, because the troll cannot litigate under your patent without your permission.

If you are a would-be buyer of a patent, this affects you in three ways:

1. You have to get authority from the patent owner to litigate for enfringement, and you do so on the patent owner's behalf.

2. You cannot resell the patent or relicense it to others, because you don't own it in the first place.

3. You can't acquire the rights to a patent (under exclusive license) and then just suppress it because it would compete with a technology you control.  It's use it or lose it.

If you're a patent troll ...

...Well, then you're pretty much right out of luck.

• You can't buy up patents for the purposes of litigation, or to resell them to the highest bidder;

• You can't acquire patents merely to sit on them and prevent them from being used;

• You can't litigate on the basis of patents that you don't own, without the patent owner's approval.

The upshot:

Patents would still protect invention and innovation.  Companies who wish to gain exclusive control of an invention can still do so with the inventor's consent ... as long as they actively use it.  Individual inventors retain more control over their patents.  But patent trolls and patent speculators with no interest in actually using, developing or furthering the patented technology will be dead in the water.

Thoughts...?

Comments

[janetmiles.livejournal.com]

I'm not arguing with you, because on the face of it, this seems reasonable. But I'd like to poke at some possible edge cases, if you don't mind.

What happens if an entity which has purchased an exclusive lifetime license ceases to exist? Does that fall under the "use it or lose it" clause?

What if the entity is acquired by some other entity? Does the license transfer, or does it have to be renegotiated? Or would it depend on how the original license was sold?

What if the patent owner cannot do development itself, and also cannot find someone else to license the patent? That is, say an individual comes up with a legitimately new technology and gets a patent on it, but no company wants to develop it because the cost is too high.

[unixronin.livejournal.com]

What happens if an entity which has purchased an exclusive lifetime license ceases to exist? Does that fall under the "use it or lose it" clause?

If the licensee ceases to exist, any licenses purchased by that licensee expire with the licensee. They cannot be transferred as assets of the licensee as patents can now, because the licensee doesn't own the patents.

What if the entity is acquired by some other entity? Does the license transfer, or does it have to be renegotiated? Or would it depend on how the original license was sold?

I would think that would be an "it depends" situation, but my inclination would be to say that licenses default to being non-transferrable unless explicitly stipulated in the license terms to be transferrable. I see no reason transfer rights couldn't be negotiated with the patent owner in that situation, though.

What if the patent owner cannot do development itself, and also cannot find someone else to license the patent? That is, say an individual comes up with a legitimately new technology and gets a patent on it, but no company wants to develop it because the cost is too high.

How would that be different from the same situation now?

I have a couple of probably-patentable ideas floating around, but I can't afford to develop them myself, nor can I afford to patent them "on spec" in the hope that I can then sell or license the patent to a company interested in developing them.

[janetmiles.livejournal.com]

How would that be different from the same situation now?

I guess what I was wondering is if, under your proposal, the patent owner would lose the rights in a year of non-development (since you'd set it up that a patent license would revert to the owner if there was no use of the patent for a year). The more I look at it, though, the more I think that's not what you meant.

[unixronin.livejournal.com]

Ah. No, that possibility didn't even occur to me. If that were the case, all a company would need to do is wait the patent owner out.

[wyrdling.livejournal.com]

maybe that'd be better applied to licensees rather than owners.

[unixronin.livejournal.com]

I did. See above.

[lwj2.livejournal.com]

Sounds reasonable to me.

Might look at (hypothetically) a corporation which purchases exclusive use rights and is then sold.

To me, the usage would transfer with ownership of the corporation, as the patent use could be viewed as a corporate asset.

Also, for lifetime use by a corporation (which could, theoretically "live" forever) one might consider some sort of additional remumeration to the holder after a set period of time.

[unixronin.livejournal.com]

Might look at (hypothetically) a corporation which purchases exclusive use rights and is then sold.

To me, the usage would transfer with ownership of the corporation, as the patent use could be viewed as a corporate asset.

See the comment above which addresses basically the same question.

Also, for lifetime use by a corporation (which could, theoretically "live" forever) one might consider some sort of additional remuneration to the holder after a set period of time.

Well, patents don't last forever themselves. (What is it — 17 years, if memory serves?) That would be an issue the the patent owner and the licensee to agree on. There's no reason on earth why license terms couldn't stipulate "$X up front plus $Y per year", or "$X up front plus $Z% of gross sale price of each product sold which implements this patent."

[lwj2.livejournal.com]

You're correct, I'd managed to confuse "patent" with "trademark" in my feeble brain as far as lifetimes go.

[unixronin.livejournal.com]

And thanks to Disney and their Senator-for-hire.... :p

[ilcylic.livejournal.com]

Normally I would object, on the grounds that "If you can't sell something, you don't own it", but I already hate patents, so, if we can't just eliminate them entirely, we might as well try to reel in some of their worst effects.

[unixronin.livejournal.com]

Which is better: to sell something, or to lease out the rights to use it (possibly several times over) and still own it when the lease(s) expire(s)?

[ilcylic.livejournal.com]

Orthogonal, really. Government interference in property rights is bad. However, I don't accept that "Intellectual property" meets the definition of "property" to trigger my defence of it as such.

[unixronin.livejournal.com]

Then I don't really understand your hypothetical objection in the first place.

[ilcylic.livejournal.com]

That would be because I don't have one.

[unixronin.livejournal.com]

I think we're going in circles. :)

[ilcylic.livejournal.com]

"Normally I would object to (government intervention), but since I hate $THING_YOU_ARE_DISCUSSING, we should use your plan to reel in some of the worst effects of $THING_YOU_ARE_DISCUSSING."

Clearer?

[unixronin.livejournal.com]

Aaaah. Now I follow. :)

[freetrav.livejournal.com]

How does your proposal affect - if at all - the situation where the patentable work was done as (or as part of) a work for hire, where the normal case - now - is that the inventor is required by his contract with the employer to (for all intents and purposes) give (sell for $0) the patent to the employer?

[unixronin.livejournal.com]

That's a separate, and much more complex, issue. I don't see it as unreasonable to say that if an employer hires or contracts with an individual to perform a certain body of work, and during the course of that body of contracted work, done with the employer's resources on the employer's time at the employer's direction, a patentable innovation arises, there is a strong legal argument that the employer is the proper owner of the resulting patent(s).

On the other hand, if an individual who works for that employer, working outside of work on their own time using their own resources and not at the direction of the employer, comes up with a patentable innovation, then even if that innovation is directly useful to the employer and applicable to the employer's business, the individual, not the employer, is the rightful owner of the patent regardless of the fact that the individual happened to be employed by $EMPLOYER at the time.

[mrmeval.livejournal.com]

I'm not sure how to word this. The patent has to be in active use and making some money. If not you lose it in a year.

If you make X dollars from a patent and more than 1 percent of those is from litigation you lose all rights to the patent. You can forgo your money and still force the court to stop an infringer. You can donate the award to charity or leave it with the state.

Not!

[unixronin.livejournal.com]

I disagree. Consider the following scenario:

I invent a way to make an internal-combustion engine radically more fuel-efficient. (Ignore the theoretical limits of Carnot efficiency for now.) I try to make money from it myself, but despite converting my car and motorcycle to use my invention, and trying to market my invention to major automakers, I am unable to attract any investors or licensees. General Motors, however, quite unashamedly steals my invention and says "So sue us", in the sure knowledge that their pockets are so much deeper than mine that they can stall me in the courts until I go bankrupt. Their market share goes up 23% in a single year. My attorney files an initial brief showing GM's own marketing materials which exactly describe my invention in their new top-selling line of hyper-fuel-efficient cars, side-by-side with my patent, and petitions the court for summary judgement against GM. By a wonder of the legal profession, the judge grants the petition for summary judgement, and I am subsequently awarded thirty million dollars in damages.


Under your rule, I now have a choice: Donate the money to charity, give it to the state, or walk away from my patent.

I don't think so.

As for "the patent has to be in active use and making some money, or you lose it in a year" - in the scenario above, all GM has to do is stall me for a year, and bam, my patent is gone forever and they can use my invention for free with impunity. Again, I don't think so.

What if it's a valuable patent, but obscure or expensive to develop, and it takes me over a year to find a licensee willing to take it on?


No. This won't fly. You might as well abolish patents entirely. No company is going to license a valuable patent when they can just stall for at most a year and get it for free. Patents filed by individual inventors woulds no longer be worth the paper they're printed on.

[lebo-superman.livejournal.com]

Hmmm.... I like it. But, I think mrmeval is right that there needs to be an income cap on profits due to litigation. This is supposed to be about production, not prevention.

Also, freetrav raised a good question about how work-for-hire contracts would treat new patents developed under the contracts. I think that the person who does the work should get the patent, not a corporation. IMHO.

[unixronin.livejournal.com]

So you agree that if you invent something, and $MEGACORP blatantly steals your invention and says "So sue us", then proceeds to make a billion bucks off of it, you should have a choice between kissing your patent goodbye or trying to sue them knowing in advance that even if you do win the case you'll never see a penny of it? Can you really afford to throw that much money down a hole that you know in advance has no bottom?


Nay, fuck that. You might as well pass a law saying that any corporation may freely and legally steal any patent from any individual at any time for any reason without penalty.

[lebo-superman.livejournal.com]

No, nononononono. The income from litigation limit would only apply to licensees, not patent owners under your proposal. Or, that was what I was thinking of, anyhow.

[unixronin.livejournal.com]

Oooooooh. OK, I totally misunderstood what you meant.

My intent was that licensees would basically only be able to litigate on behalf of and with the cooperation of the patent-holder, and be required to share any proceeds (more specifically, any awards of damages purely for patent infringement, not necessarily awards for lost business) equitably with the patent-holder.

[smandal.livejournal.com]

Whenever someone owns something non-rivalrous (http://en.wikipedia.org/wiki/Non-rivalrous) but excludable (http://en.wikipedia.org/wiki/Excludability), he has an incentive to keep it that way. In this respect patent trolls and cable TV companies are quite similar.

The question is, when is defending one's turf just, when is it productive for society as a whole, and when is it just a money-grab? One standard is deadweight loss (http://en.wikipedia.org/wiki/Deadweight_loss), but I have no idea how to begin to compute that for patents.

To the extent our intuition is reliable, I think you are on the right track. Removing transferability and derivative patentability cut into non-productive turf-protecting, while letting the inventor collect his due.