Patents Are Eating the World and Hurting Innovation

http://blogs.hbr.org/2014/06/patents-are-eating-the-world-and-hurting-innovation/

by hype7 • 12 years ago

153 54 12 years ago

blogs.hbr.org

dougabug• 12 years ago

The expression "X is eating the world" is eating the world.

kaonashi• 12 years ago

I can't believe X ate the whole thing.

recursive• 12 years ago

No, it's still eating.

tomsthumb• 12 years ago

Agreed, it is to be considered harmful.

anotherevan• 12 years ago

Yes, if you use a headline like that, you're doing it wrong.

jgilliam• 12 years ago

I read the headline as "Parents are eating the world and hurting innovation."

jrells• 12 years ago

Me too! I was very interested, then very confused.

hartator• 12 years ago

haha same here! I was going to post about it!

Alupis3• 12 years ago

I couldn't agree more with the article. The US Patent system is broken and needs revisions to meet today's marketplace.

Patents are awarded to non-practicing entities and people who can't act on nor implement the very patent they were awarded.

Idea's should not be patentable. If you can think of something, so can I. Especially when it comes to software, since software is nothing but an idea written down.

Idea's that oneself cannot implement and/or act on, should not be patentable. To be awarded a patent, you should, at the very least, be required to provide a working POC. I'm tired of companies being awarded absurd "future thinking" patents that have vague wording along the lines of "something will happen when the user does something, etc".

Patent Trolls (aka Non-Practicing Entities) should not be able to hold patents if they are not using and/or implementing them. How can you collect money on an idea for years/decades but not have the capacity yourself to perform the service/make the product the companies you are suing are doing?

The US needs drastic patent reform. It is stiffling innovation.

josaka2• 12 years ago

The chart in the article, showing patent suits filed as a function of time, is a little misleading. It appears to hockey stick in 2011 (after a long upward trend). It's worth noting that in 2011, the rules on how many defendants you can group in a suit changed, causing what used to be a single suit to spawn multiple separate suits, even though the number of companies affected didn't necessarily change. I think a better chart would lists defendants as a function of time.

Natsu• 12 years ago

I remember a similar upwards trend when you look at the graphs of money spent on patent lawsuits over time. So I believe it's a growing trend, even if the change to the joinder rules may explain part of the spike.

I went looking for the chart, but all I can find are articles saying that Google & Apple spent more on patent lawsuits than R&D. I think that is illustrative of the problem, albeit in a different way.

ximeng• 12 years ago

What is the best way to get data on who is litigating, who is being sued, and over what patent? For only 5000 lawsuits it would be relatively easy to understand the scope of the problem.

WildUtah• 12 years ago

It would be nice to just read in the court filings, which are a public record, and produce data extracts to explain the world. Unfortunately the PACER system -- officially a public agency -- charges you by the page for those public documents.

Once you've got them, they're free for you to share, but sharing projects threaten the revenues of the system and the publishers (Westlaw and Lexis) that prosper from the lack of public access to public data. Those publishers and PACER have worked hard to block and shut down and even bring unrelated federal prosecutions against people who organize free systems.

One of the reasons Aaron Swartz was a major target of federal prosecutors even though his nominal accusers didn't want charges brought is that he was active in promoting free access to federal court documents. [0]

Companies like Lex Machina [1] work with the officially supported publishers to offer analysis of patent lawsuit filings and patent office documents (also officially but not really public) to answer questions like yours for very wealthy clients.

[0] http://www.theverge.com/2013/2/8/3968824/aaron-swartzs-battl...

[1] https://lexmachina.com/

josaka2• 12 years ago

For coarse data, the PricewaterhouseCoopers annual patent lit studies are pretty good: http://www.pwc.com/us/en/forensic-services/publications/2013....

shiven5• 12 years ago

Here is a hypothetical solution:

(0). Make patent issue fee ridiculously expensive (100k+ ?).

(1). Increase the salaries of every patent examiner to ridiculously high levels, supported by the extremely high fee.

(2). Make every patent examiner personally liable for approving a patent, to the extent of a lien on their personal assets. No bankruptcy allowed (as for student loans).

(3). If any patent is later proved to be carelessly approved, infringing on prior art or broader than it deserved to be, the patent examiner should be held liable and pay damages for every patent they approved that was struck down.

That might help cut down the indiscriminate issuance of patents that should never have been allowed in the first place.

kefka• 12 years ago

I'd gladly take the job. I'd even buy an ergonomic "DENIED" stamp.

It's the only stamp I'd use.

ww520• 12 years ago

Why not hold the patent filers liable and have them paid damages for their struck down patents? Why blame the examiners? Blame the source of the problem.

Codhisattva• 12 years ago

Liability is the antithesis of American business. (The act of incorporating is in part a shield against accountability and liability.)

shiven5• 12 years ago

It is the examiners who allow patents through. I would go so far as to argue that just because I ask for something, from the public, via the government, I shouldn't just get it. It is the responsibity of the examiners to deny a non-deserving ask.

neurobro2• 12 years ago

Somewhat less draconian:

1. Make patent applications free of cost, so anyone can play unless they start clogging up the system.

2. Applicants must wait 2^n days between applications, where n is the applicant's number of rejected applications within the past 1,000 years. Non-human persons must be at least 1 year old before applying.

3. Examiners are paid approximately the nation's median income, but also receive rights to 1% of license fees for patents they approve. If an approved patent is ever overturned in court, the examiner is fired and loses all future royalties for any other patents they approved. Those royalties then go to the patent office. Budget surpluses go toward scientific research or maybe education grants.

4. Patents assigned to biological human inventors expire after 20 years. Others expire after 5 years.

Alupis3• 12 years ago

How about -- No Patents?

Patents are ideas... if you can think of it... so can someone else.

Patents were intended to protect someone's investment into an idea. Today, they just serve as cash cows.

With no patents, a company would be forced to continuously innovate and stay ahead of the curve -- which would replace their patent warchest with a warchest of talented individuals and innovators instead.

Terr_• 12 years ago

Problem: Garage-Inventor Bob's patent will be thrown out, because if it gets challenged Bob will probably lose since he has no big pockets nor legal clout. (Not because the idea itself is good or bad.)

Later, SuperMegaCo will submit a marginally different formulation of the same basic thing, and it'll get approved because they're probably going to win any long expensive court-battles.

fpgeek• 12 years ago

I agree that changing the incentive structure would help, but I don't think targeting individual examiners is the best way to do it. I think the right target is the patent office itself.

The only thing you've listed above that would change the patent office's incentives are the increased issue fees - which would only encourage them to let more bad patents through. Yes, individual examiners would push back because of their personal liability, but they're really trapped between a rock (their boss) and a hard place (getting sued in the future).

On the other hand, if you made the patent office the target of all the penalties for issuing a bad patent (legal fees, damages, etc.) then they'd take care of examiners themselves (not measuring by patent-processing rate, allowing additional time for aggressive prior art searches and so on).

shiven5• 12 years ago

Sure, sure. (Guess I missed the forest for the trees and all that...)

But in either case, first the patent granting authority needs to be held grossly liable to prevent this rampant abuse, then and only then will subsequent measures be effective.

However, targeting the examiners focuses on the personal responsibilty for their actions. If you institutionalize that responsibility (patent office vs patent examiner) you create a situation where the perpetrator (examiner/boss) hides behind the collective back of the patent office and the taxpayers are again left holding the bag of liability.

A person or a discrete group of persons approved the granting of this right to the applicant (or the denial of freedom to non-applicants). Why should they be absolved of their individual responsibility, for their individual actions?

lyoNick• 12 years ago

How about a reform that tie patents with the R&D costs? placing the incentive where it's actually matter

i.e:

* for each patent filed, one should provide an 'expanse clause' of the R&D expanses

* if one file a patent then those expanses became taxable (the same way warehouse stock is?)

* when applying one can choose to hold the patent for a limited some of years (say 5) WITHOUT dependeing on ROI, OR hold the patent untill ROI exced 10 fold but no longer then 10 years. (number cand depend on general public priorities {medical , novelty , etc.} )

* the 'expanse clause' should be limited to what R&D that is strongly tied with the current patent, but can include failed R&D as long as they are strongly tied to the successed one.

* patent must reach a working implementation within the 1st year of filing, or they are revoked.

* maybe have two types of patents: - one like the above for those with working implementation, - the other for 'consept patents' that have a fix price for filing, and can be held for up to 12 months before one must file a 'regular patent' with a working implementation, and if faild to do so, it is revoked retroactivly so it can be used to sue.

[*] P.S. not a native english speaker, so commetns on grammer and spelling mistakes are actually welcome.

kvb• 12 years ago

it's actually matter -> it actually matters

expanse -> expense

pjc502• 12 years ago

(2) Make every patent examiner personally liable for approving a patent, to the extent of a lien on their personal assets. No bankruptcy allowed (as for student loans).

This is never going to happen and would be an abusive illegal employment practice if it did.

md2be• 12 years ago

Make every patent applicant (and patent agent/counsel) liable for filing patent claims that are found invalid whether issues or not.

keithpeter• 12 years ago

You could make the agents self-employed on a bounty basis rather like the old Lloyd's Names

http://moneyweek.com/how-to-become-a-lloyds-name-42723/

However, the inevitable Black Swann event caused changes even in that very high risk system...

"Historically, Names had unlimited liability to the risks they were underwriting. In the late 1980s/early 1990s, Lloyd’s entered the most traumatic time in its history. Unexpectedly large legal awards in US courts on asbestos, pollution and health hazard policies (APH) served up huge losses to many Names, a large number of whom lost their shirts. Names’ numbers dropped from over 30,000 to below 10,000."

I think just giving patent examiners enough time and resources to actually verify the claims would probably cut down a lot of the problem. That and a specialised court like

http://www.justice.gov.uk/courts/rcj-rolls-building/patents-...

instead of a general purpose district court sitting with a lay jury

mellisarob• 12 years ago

true but initiates need to be taken by the legislative if they wish to overcome these patents.

obsurveyor• 12 years ago

The only fix patents really need is a decrease in time. Make it 2 or 3 years instead of 20(retroactive, no grandfather clause) and dump software/DNA/genetic patents. Go back to patents being actual inventions instead of ideas and processes.

jrkelly• 12 years ago

If you drop DNA patents would you recommend it be covered by copyright? Tough if there is bothing preventing direct copying.

pjc502• 12 years ago

Since the thread is full of insane suggestions for reform, I'll make some sensible ones:

1) non-physical implementations or items already covered by copyright should not be patentable. Pure software and media carrying it should be treated as speech and creative works; it should not be possible to infringe a patent simply by writing or running software.

2) re-introduce the requirement to deposit an example of what's being patented. A set of photos plus manufacturing drawings in usual electronic form should be acceptable as a substitute to avoid a storage problem for the patent office. (Want to patent a chip? Send Verilog or GDSII along with your application)

3) following from that, if there is a software component in the claims then the source must be provided in the patent application.

niels_olson• 12 years ago

What has a better return on investment? A Comp Sci BS or a JD? I'm betting the Comp Sci has a higher mean, median, and max.

jonnathanson• 12 years ago

Should be pretty easy to find aggregate stats on both.

Law follows a bimodal distribution, though, so you'd need to factor that into account. Basically, there are two types of law degrees, and two types of lawyers: those from top-N schools, and everyone else. Those who land prestigious clerkships, and go on to practice things like securities or corporate M&A at top BigLaw firms, and those who don't.

When we speak of a "JD," we can't really speak of it as a homogeneous class. You could argue that this effect could be found with any degree, for instance, that a CS degree from Stanford leads to much better outcomes than a CS degree from a mid-tier university. But the effects of school, class rank, career track, and prestige in the legal profession are near-total, and the divide in outcomes is quite dramatic.

HeyLaughingBoy• 12 years ago

The combination is likely a far outlier. Considering that with an undergraduate degree in CS, the JD can become a patent lawyer and IP lawyers are among the best paid, that's probably the best bang for the buck.

jayvanguard• 12 years ago

Have there been any proposals aimed at reducing the transferability of patents? Perhaps the lifetime of the patent could halve each time it is transferred to a new legal entity along with some reduction in possible damages.

fancyfancy• 12 years ago

I agree that patents are making innovation difficulat. This company seems to be changing that, and may help solve the problem. I thought it was interesting, at least! Bideas: http://bit.ly/bideas_com

Patents Are Eating the World and Hurting Innovation

Related Stories