Why a PGR?

A picture is worth 1,000 words

I have lots of pictures. The one above sums it up. Here's my take: "We created this mess so you need to pay us to fix it."

The PGR was built for exactly this

A Post Grant Review is the mechanism Congress created to let the USPTO clean up its own mess. It lives in Chapter 32 of the Patent Act (35 U.S.C. §§ 321–329). The short version:

• You can challenge a freshly issued patent on essentially any ground of invalidity — §§ 101, 102, 103, and 112 — because § 321(b) lets a petitioner raise anything that could be raised under 35 U.S.C. § 282(b).
• But you have to move fast: the petition must be filed within nine months of the grant (§ 321(c)).
• The Board only institutes if it's “more likely than not” that at least one challenged claim is unpatentable (§ 324(a)). The procedural rules live in 37 C.F.R. Part 42.

In other words, the PGR was built precisely for this scenario: a patent that should never have issued, caught early, and corrected. Great. Except that it's the most expensive fee the USPTO has — more expensive than an IPR — even though its explicit purpose is to fix the Office's own errors. Errors that could have been avoided in the first place.

The grounds, not the process

I'll spend a lot of this blog complaining about examination mistakes and misleading statements, typos, etc. That explains how this system is flawed, but those are not the grounds for a PGR. You can have a perfectly good application become a patent with a horribly inept examination. Likewise, you could have a really good patent denied by an unreasonable examination. The process is wildly subjective, which is a problem in itself.

Sometimes you get a confluence of a terrible examination and a terrible patent, and the sum of those parts is what this story is about.

The proper grounds are §§ 102, 103, and especially 112 — with § 101 as background commentary, not a ground I'm raising. That proves the patent should never have been issued, but none of that addresses the inequity of examination or how that can go unchecked. The system is deeply flawed and needs fixing.

And one more thing worth flagging right now: the examiner didn't raise a single § 112 objection — even though the application is chock-full of § 112 issues. Written description, definiteness, enablement (or lack thereof): take your pick. We'll come back to that.

An analogy — Seatbelts

❮ ❯

I like analogies. You'll figure this out. As a society we've all decided wearing seatbelts is a good policy. The seatbelt itself is burdensome and uncomfortable. It slows us down and inconveniences us. The collective problem is that when we get in an accident, the accident externalizes that cost. The ambulance… the hospital… Forget the property damage. Forget what happened to the person not wearing the seatbelt. It's the medical bills and utilization of resources that are the cost to society. We made laws to require seatbelts. That reduces the worst of the negative outcome on the rest of society. It doesn't eliminate it — you can still drive like a maniac. But, if you crash, it's less likely you'll go to the hospital.

The examination quotas and push for allowance are the reckless driving (i.e. rushed examination). Bad/low-quality patents are the accident. The PGR is the ambulance ride and the hospital. There are no seatbelts! We all pay for this, but me most of all. And I wasn't even on the road.

A voice

I could file for ex parte reexamination, but then, like the third-party submission (stay tuned, I'll get to that), I'd be silenced: a third-party requester drops out after filing the request and has no further voice in the back-and-forth. I'd have to sit quietly by and trust the system. I did that once and it failed miserably.

Plus there are a bunch of § 112 issues that need addressing, and the PGR is the only venue for that.

But everyone would do PGRs

Yes — everyone realized how sloppy the USPTO can be and how often this safety valve would be used (and probably abused). Instead of fixing the underlying problem of examination and allowance quality, Congress figured there needed to be a self-regulating, self-filtering mechanism: an astronomical fee designed to keep frivolous requests away. My case is not frivolous, but I am shut out in a financial sense.

Would you pay $59,375^‡ to be right? Not to resolve anything. Not to recover damages. Not to stop infringement. Just to correct an error someone else made — to restore the exclusive right the patent was supposed to secure for me (by the Constitution via Congress), a right taken away by the same Office that then gets to cash the nearly-$60k check for a chance for that same Office to review their own sloppy work.

I didn't cause this error. In fact, I did everything right (within the mechanisms afforded by law) to help the USPTO not screw this up… but it wasn't enough.

Is it really the most expensive fee?

Yeah, probably. It certainly is in the USPTO fee schedule. I went looking for other fees and couldn't find an analogous one. The fee for drilling an oil well? $12,850. But that's for doing something productive (whether you like it or not, that's a fee for doing something). Likewise, there are fees for certifying drugs, airplanes, etc. All of which are for doing something productive. There are fees for screwing up, like if you spill a bunch of oil while drilling, but guess who pays for that?

As far as I can tell, the PGR fee is one of the few (if not the only) fees that the innocent party is on the hook for when someone else failed (i.e. the USPTO in its basic duty of examination). And it's prohibitive. By my math, it's nearly 150x what it costs to get a patent. That's at micro-entity rates (which carry a 4-application cap). Or, 75x the cost of a patent for a small entity with no cap. That means I could file 75 utter garbage slop applications for the cost of the PGR (just filing fees). What are the chances a few of those garbage applications sneak through? I wonder if I could patent the process of patenting garbage patents. I bet someone's already done that.

I've been quoting the fees alone

The true cost is closer to $300,000 when factoring in attorneys. The issues I'm raising here shouldn't require attorneys. Resolving them should simply require a competent and thorough examination. I hope to illustrate how this patent was neither competently examined, nor thoroughly examined. That shouldn't cost $60,000 - $300,000 to fix.

So how does a PGR benefit me?
It doesn't.

Here's the part that surprises people. The simple, honest answer is that a PGR does not benefit me in any tangible way.

I'm on the hook for a substantial fee and a mountain of work to draft the petition — all because the examiner didn't do what the MPEP plainly requires: conduct a thorough search of the prior art (MPEP § 904) and then properly analyze and apply it (MPEP §§ 2103, 2141). The art was right there. Applying it was the job. It's clear the examiner did not grasp the physical sensing characteristics at issue, or show even the most basic understanding of the field of the invention.

Assuming the best-case outcome — I figure out how to pay this punitive fee, I win, and the bad patent is canceled — what's my prize? I will have corrected a grievous error. That's it. If I lose, I'm in exactly the same boat I'm in today. Heads, I succeed on principle, and lose ~$60k; tails, I fail and I'm still out the ~$60k fee and the time.

It does not resolve the infringement mess

This is the part that should bother you as much as it bothers me: none of this resolves any infringement question.

Consider the absurdity. Their claim is so broad that — if you pretended it were valid — it would reach things that were already public before they started, my own disclosed work among them. A claim that sweeps in the prior art like that is invalid on its face: you can't claim what was already out there — that's the whole point of the PGR. But valid or not, the sheer breadth is its own problem: the very work I disclosed and watched them copy could be waved around as if it were grounds to come after me. That's a huge problem.

All because of sloppy work. Sloppy work by those charged with keeping that gate.

And if I succeed in this PGR and their patent is canceled? We still don't know whether they're infringing mine. The PGR doesn't answer that question. It just removes the cloud over whether I actually have a theoretical right to exclude (remember that exclusive-right bargain the Constitution lets Congress strike). More importantly, there is still confusion in the market. Who actually owns this technology? That damage is already done (and in my opinion, was the entire point of this approach). I should be able to say it's me, I'm the inventor, but now they get to say they have a patent too. That's not fair to the market. The PGR can't unring that bell.

What I'm actually accusing them of

First, what I am not accusing them of: infringement. I have never made that accusation — you can make up your own mind about whether anyone is infringing on anything here. I think it's all pretty obvious what's happening, but no accusations.

What I have accused them of is plagiarism.^† That one is clear, and it's easy to prove — you line up the text and look. Heck, I can show you with a simple Google text search. No claim construction, no dueling experts, no Markman hearing. Just two documents and your own eyes. That's coming in a few posts.

And if they're willing to copy text this blatantly, why wouldn't they copy the idea too? That's the fundamental problem here. The USPTO is so loose with issuing patents — and there's literally no penalty for trying — that copying has become the obvious play at this stage of the US Intellectual Property Game. Just copy. If you get caught, be more careful next time.

That's what this blog is going to show you. Bring your eyes.

Related dispatches

• Dispatch No. 4 — The first time this idea was good enough to copy — the earlier time this same idea was good enough to copy.
• Dispatch No. 6 — The Third-Party Preissuance Submission — the earlier, cheaper attempt to head all this off.
• Dispatch No. 24 — § 112 — They Never Said How — the § 112 ground the examiner never raised, picked up in full.

‡ That $59,375 isn't one line item — it's two stacked on top of each other on the USPTO fee schedule, at the large-entity rate for up to 20 claims: a $25,000 petition request fee just to ask, plus a $34,375 post-institution fee once the Board agrees to look. Add 20 claims and it climbs — each claim over 20 tacks on another $595 to file and $1,315 after institution. Fees change; this reflects the schedule at the time of writing. ↩

† “Plagiarism” isn't a legal term of art — there's no cause of action called “plagiarism” that you file in court. But you all know exactly what I'm getting at: copying someone's text and passing it off as your own. I'm using the word as a plain-English description of a documented fact, not as a legal conclusion. None of this is legal advice or argument. Everything on this site is purely my opinion. That's one constitutional right I should still be entitled to. ↩