Cloudflare Pays First $7,500 Bounties In War Against Patent Troll (cloudflare.com)

Cloudflare declared war on a group of lawyers that files patent lawsuits against tech firms, by offering bounties for the discovery of patent-invalidating "prior art." Now an anonymous reader writes: On Thursday, Cloudflare announced it has paid out the first $7,500 to people who discovered documents that could help invalidate Blackbird's patents. The money is part of a $100,000 war chest the company announced this spring... The company said it is ready to launch individual challenges to specific Blackbird patents. The company believes it has enough examples of prior art on US Patent 7,797,448, "GPS-internet Linkage" and US Patent 6,453,335 (the one asserted against Cloudflare) to lodge a challenge.
"We have received more than 230 submissions so far," Cloudflare reports, "and have only just begun to scratch the surface."

Comments

Those deck chairs won't self-rearrange.

[drinkypoo]

As long as pathetically obvious patents like these are granted in the first place, it's always going to be whack-a-mole. And as long as the USPTO is funded by charging fees for granting patents (and not just applying for them) the situation will persist.

Re: Those deck chairs won't self-rearrange.

[Entrope]

Where on the USPTO's fee schedule (https://www.uspto.gov/learning-and-resources/fees-and-payment/uspto-fee-schedule) do you see a fee for issuing a patent? There are fees for applying, fees for very large applications, fees to maintain an issued patent, but I see none for issuing a patent.

Re:Those deck chairs won't self-rearrange.

[stabiesoft]

I think the problem is the application fee is pretty cheap compared to the cost the submitter pays to the lawyers to submit it. The lawyers outgun the patent examiners and the result is the roughly 50% grant rate. The examiners are just not up to the task and as I recall, the patent office actually gives some of the fee to the treasury making the matter worse. If all the money for applications were used to examine it would be a start. Perhaps one approach would be you must file a bond of say 50K with each application and if the examiner finds prior art, the examiner keeps the bond. Now that would incentivize the examiner, probably too much but might be interesting to see what happens with the pendulum swung toward too hard to get a patent. Of course if the examiner does approve the patent you get the 50K back.

Re:Those deck chairs won't self-rearrange.

[Zocalo]

I think the problem is more fundamental than that. By paying patent examiners to approve patents, you're also effectively encouraging them to fail to do their jobs properly and just rubber stamp any random patent that they review; whether or not the examiners are competent or not doesn't enter into it. The fix is to turn that on its head and pay the examiners a commission/bonus for each patent that they can find prior art on or find a justifiable reason to invalidate the claim. Better yet, start with a reasonable low fee and/or bond, but have the number of previous failed applications result in successively higher submission fees/bonds for subsequent patent applications. That way you don't discriminate against smaller inventors who may not be able to afford large application fees, or even larger companies that file a lot of valid patent applications, but you do have a clear financial deterrant against the patent trolls that churn out an endless stream of frivolous and/or over-reaching patents.

Re: Those deck chairs won't self-rearrange.

[arglebargle_xiv]

Patent examiners are remunerated based on how many patents they grant. Disallow a large number of patents because they're trivially obvious and you lose pay, or possibly get fired. Rubberstamp everything that crosses your desk and you get a performance bonus.

This is not a system that produces good/valid patents.

Re: Those deck chairs won't self-rearrange.

[drinkypoo]

There are fees for applying, fees for very large applications, fees to maintain an issued patent, but I see none for issuing a patent.

Fees to maintain an issued patent after only three years, which amount to as much as the total filing fees. And anyone who wants to challenge the patent has to pay fees to cover those costs, so what are those fees for? What are you "maintaining"? Answer, the USPTO's budget.

Re: Those deck chairs won't self-rearrange.

[tepples]

Where on the USPTO's fee schedule (https://www.uspto.gov/learning-and-resources/fees-and-payment/uspto-fee-schedule) do you see a fee for issuing a patent?

Under "Patent Post-Allowance Fees" are several line items containing the term "issue fee". In addition, "fees to maintain an issued patent" apply only to issued patents, not to applications that do not result in issuing a patent.

Re: Those deck chairs won't self-rearrange.

[Entrope]

The thing a patentee is "maintaining" is his or her exclusive rights to the patented invention. One must to pay every few years, and those fees escalate over the potential life of the patent, to defer people from maintaining or asserting patents that are not in live use. Those maintenance fees are a small fraction of the USPTO's budget, but even if they were the majority, it would not make correct your claim that "the USPTO is funded by charging fees for granting [emphasis in the original] patents".

Re:Those deck chairs won't self-rearrange.

[stabiesoft]

I guess you missed where I said the patent examiner keeps the bond if they find prior art, IE the reject the patent. Agreed though my 50K bonus per rejection is probably too high. 5K would be more reasonable.

Re: Those deck chairs won't self-rearrange.

[Anonymous Coward]

Why the fuck should they be paid differently for accepting or rejecting an application? This isn't commission sales.

They should be paid by hour or salary and apply appropriate resources for each application fairly.

Re:Those deck chairs won't self-rearrange.

[Zocalo]

No, I didn't miss it, and I think the bond is potentially a good idea, but the problem isn't just turning it around to provide an incentive for examiners to reject bad patents - it's also figuring out a way to penalise the outright patent trolls and/or submitters of overly broad patents without placing an unfair burden on either lone inventors or the larger corporations/universities that produce genuine patent candidates in bulk. That's not entirely clear cut either, because for every pure patent troll, you've also got companies that apply for a mix of worthy and frivolous patents, and for many of those the cost of a bond (even one of $50k) is potentially just the cost of doing business. A company could afford to write off a lot of $50k bonds on rejected frivolous patents if they can get just one of them through and leverage it into settlements totalling ten of millions, or more, and that's the more fundamental issue with the system; the entire thing is too easily gamed by the big patent mills and the trolls.

The way to address that, and improve the signal to noise ratio across the board, is to increase the fees and/or bond for every rejected patent. e.g. say $5k for the first patent application, but if that gets rejected after any appeals then subsequent applications cost $10k, and it keeps going up, and up, and up, with each failed application and either doesn't come back down for quite some time, or maybe ever. That's necessary to prevent XKCD's bobcat scenario; large companies need to be prevented from gaming the system by trying to slip an overly broad patent through every now and then, then bringing the fees back down through more legitimate applications. Buying a patent portfolio? Congrats, you also bought their patent filing reputation and any increased fees that result from it. I'm still missing a way to deter patent trolls from filing each application as a new inventor or shell company to help keep their fees down, then transferring any successful applications to their patent pool though... Some kind of corporate registration and tax filings that could be cross-checked with other government agencies, perhaps?

Re: Those deck chairs won't self-rearrange.

[Zocalo]

In an ideal world, yes. The examiners would do their supposedly pretty thankless jobs, and all the bad patents would be rejected while all the good ones would be approved - and since it is an ideal world everyone (except perhaps the submitters of rejected patents) would agree on which are "good" and "bad". Unfortunately, the current system in the US (and at various other PTOs) is to incentivise either approvals or throughput, both of which discourage fair evaluations, while a flat-rate salary system is potentially open to abuse because of the sums of money involved for patent trolls, or did you imagine a PTO staffed entirely by examiners who would never, ever, consider a bribe? There needs to be checks and balances - which is why I mentioned an appeals process which would help prevent bad assessments and weed out bad/corrupt examiners - but I think you've also got to provide some form of motivation for the examiners to do a *good* job rather than a quick one, and absolutely not just rubber stamp anything that's on the right form. A reasonable financial incentive is one fairly obvious way to go on that, perhaps offset against a lower base salary, but might not be the only one.

Alternatively, or additionally, you could also add an extra step where rather than simply have the PTO mark a patent as approved it goes into a "pending" state when the patent application paperwork is published and anyone can submit prior art to negate its final approval. Once that pending period is up, the examiner(s) can review any submissions and make a final decision on whether or not to approve the patent.

Re: Those deck chairs won't self-rearrange.

[Theaetetus]

Patent examiners are not the USPTO. They are employees of the USPTO, and the USPTO is -- contrary to drinkypoo's assertion -- not "funded by charging fees for granting patents". There are extensive fees for applying for a patent, but no fees are due upon grant of a patent.

That's not true. The issue fees are not very high compared to the overall cost of prosecution, but they do exist. The issue fee is currently $960 for a large entity.

Re: Those deck chairs won't self-rearrange.

[Theaetetus]

Patent examiners are remunerated based on how many patents they grant. Disallow a large number of patents because they're trivially obvious and you lose pay, or possibly get fired. Rubberstamp everything that crosses your desk and you get a performance bonus.

This is not a system that produces good/valid patents.

Actually, patent examiners are remunerated based on how many patents they bring to a final disposition. That includes granting, but it also includes rejecting. Rubberstamp everything "rejected" and you'll get the same performance bonus.

Re: Those deck chairs won't self-rearrange.

[Theaetetus]

Where on the USPTO's fee schedule (https://www.uspto.gov/learning-and-resources/fees-and-payment/uspto-fee-schedule) do you see a fee for issuing a patent? There are fees for applying, fees for very large applications, fees to maintain an issued patent, but I see none for issuing a patent.

Try here, where it says "issue fee".

Re: Those deck chairs won't self-rearrange.

[Entrope]

So drinkypoo's comment is still essentially wrong, but somehow gets modded to +4 because it gives people something to feel outraged over? This is a large part of what is wrong with /. these days.

Re: Those deck chairs won't self-rearrange.

[parkinglot777]

Patent examiners are not the USPTO. They are employees of the USPTO, and the USPTO is -- contrary to drinkypoo's assertion -- not "funded by charging fees for granting patents". There are extensive fees for applying for a patent, but no fees are due upon grant of a patent.

To me, the word "granting" means "patenting process" regardless an application is granted. The overall cost to apply and get a patent is not cheap. And yes, USPTO is funded by itself from their own revenue. They have their own budget which comes from their own revenue (fee collections).

Re: Those deck chairs won't self-rearrange.

[Entrope]

You can insist that words don't mean what they do, but your mis-definitions would obliterate the distinction that drinkypoo was trying to make. He argued that the patent office should be funded by fees for things besides issuing patents, which is already essentially true. It's not clear that the "issue fee" is disproportionately high compared to the work it entails.

Re:Those deck chairs won't self-rearrange.

[currently_awake]

Switch to a patent registration system, where the USPTO just registers your patent claim and then the holder must prove the patent valid when they attempt to use it (with no assumption of validity). It would be just like today (rubber stamp patents) but easier to challenge.

Re: Those deck chairs won't self-rearrange.

[parkinglot777]

You can insist that words don't mean what they do, but your mis-definitions would obliterate the distinction that drinkypoo was trying to make. He argued that the patent office should be funded by fees for things besides issuing patents, which is already essentially true. It's not clear that the "issue fee" is disproportionately high compared to the work it entails.

Nope, you are the one who misunderstand the post, and then attempt to misinform others. "Granting" in patent field means you attempt to get a patent whether the application is approved.

And as long as the USPTO is funded by charging fees for granting patents (and not just applying for them) the situation will persist.

What does that tell you? It meant the opposite from what you said. If USPTO keeps themselves funded by the fees, then the problem persists. There is no such thing that you claim about USPTO "should be funded by fees" at all because currently USPTO is funded by fees.

Re: Those deck chairs won't self-rearrange.

[Entrope]

"Granting" in patent field means you attempt to get a patent whether the application is approved.

That is not what grant means in the patent process. The patent grant is when the government (or other body) officially recognizes the inventor(s) as holding exclusive rights to the invention. That happens after the applicant files a patent application, after the government reviews the claimed invention and existing art, and after the government approves the issuance of the patent.

If what you say were true, then there would be no sense in complaining that "the USPTO is funded by charging fees for granting patents (and not just applying for them)", because "granting patents" would include applying for them.

My point was that the USPTO is funded mostly by things besides the fees that drinkypoo complained about. According to the USPTO's most recent annual report, issue fees provide less than 10% of their income from fees (see the figure on page 40). Given that there are substantial expenses related specifically to the publication of a patent -- for example, costs to archive the patent essentially for the life of the USA -- it seems particularly frivolous to complain about the USPTO charging fees to issue a patent.

I feel so conflicted.

[Anonymous Coward]

On the one hand i hate patent trolls. But i also hate cloudflare since the daily stormer debacle.
I guess in the end i hope cloudflare gets a pyrrhic victory.

An entire $7500?!

[Gravis Zero]

Oh please, there are many millions of dollars on the table here. What they are paying out in bounties isn't even scraps, it's crumbs at best.

Re:An entire $7500?!

[Gaygirlie]

Who cares? Even crumbs is better than nothing when it avoids going into patent-trolls' pockets!

Re:Burn in Hell, Cloudflare

[that this is not und]

I have NoScript blocking Cloudflare. They seem like an outfit that think they are too big for their britches.