Senate Passes Landmark Patent Reform Bill

inkscapee writes "The US Senate is congratulating itself for passing a 'landmark' piece of patent reform legislation. Some key elements are 'first to file' instead of first to invent, and ending fee diversion, which means fees paid to the Patent Office will actually fund the Patent Office. Curiously, this practice has resulted in a backlog of 700,000 patent applications. The House is reportedly working on a similar bill, and soon harmony and rationality will triumph."

Comments

wait

[CSFFlame]

Isn't first to file REALLY bad? It helps patent trolls doesn't it?

Re:wait

[Gumshoe]

Isn't first to file REALLY bad? It helps patent trolls doesn't it?

Yes. It's a disaster. In effect if you invent something you now have to patent it or possibly suffer the consequences from a patent troll. It's a money spinner for the patent office if nothing else.

Re:wait

[IP_Troll]

First to file encourages people to file for patents sooner rather than waiting for someone else to file and cutting them off with a first to invent claim.

First to file doesn't favor patent trolls, they can't patent the invention if someone else was using it publicly and didn't bother to patent it.

If anything it cuts off patent trolls because they can't keep inventions a secret waiting for someone else to file a patent, and then usurp that patent from the original filer.

Re:wait

[skids]

They used to file, then before they got approved, refile with amendments and just keep doing that until someone with something close enough to their patent came along, then refile with one last final amendment to make their patent close to what the unfortunate victim had started to market.

That was called a "submarine patent" and they fixed that a while back. This is probably aimed at killing the practice you just mentioned, which was what trolls turned to after that.

Re:wait

[Jane Q. Public]

First-to-file does favor companies and corporations indirectly, if not necessarily patent trolls.

Backyard and basement inventors (who make up a surprisingly large percentage of true innovation) in many cases do not have the money to file right away. Although the patent search situation is much better than it was some years ago, it can still take money, research, and time to file, and the services of a patent attorney, while not technically necessary, can be an advantage.

These are things that many companies and corporations have readily at hand, or can easily afford to do. Not so the independent inventor.

There are a few protections in place, but there is still no doubt that first-to-file favors the big players over the little. That is a very bad situation, and I argue that it is worse for inventors in America than the problem it is supposed to solve.

Re:wait

[h4rr4r]

No, you just have to publish to so you can provide prior art. The whole point of the patent system is to provide a limited monopoly in exchange for showing the public how to make your widget. So either patent or publish or STFU.

Re:wait

[h4rr4r]

So then they should publish their works. The patent system was created to move knowledge into society. If you want to keep your widget a secret you do not deserve protection. It was not made to protect anyone or let anyone profit from anything.

Re:wait

[93 Escort Wagon]

Flying car
Flux capacitor
Food in pill form
Warp drive
Pills in food form

  My offspring's gonna be rich...

Sorry, I just printed out your post and faxed it to the Patent Office - I win!

Re:wait

[anyGould]

From TFA:

The Senate bill would transition the Patent Office to a so-called first-to-file system, bringing the U.S. system in line with the patent regimes of much of the rest of the world. The shift would confer patent rights on the first inventor or company to file an application, rather than the current first-to-invent system.

So, since the rest of the world already does it this way, I'd say you'll have to look elsewhere for your troll problem.

Re:wait

[ObsessiveMathsFreak]

The patent system moved knowlege into society by allowing the inventor to continue to profit from his invention.

And the whole system is based on two key assumptions

1. The profit motive is the best way to encourage innovation , and more crucially
2. The benefits of the system would outweigh any other costs it inflicted on society.

You can argue about the first point. Personally, I think that innovation would still occur by without the patent system, but the point can be argued.

The second point is really where all the modern problems with the patent system lie. The huge costs of this system, direct and indirect, are have a deleterious effect on society, knowledge and innovation in general. It's become a matter of question whether the amount of innovation the patent system encourages is really worth the additional costs and indeed damage it imposes.

Re:wait

[srobert]

You win because you got there first. But ... (flux capacitor). Still not too late.

Re:wait

[Jane Q. Public]

"Yes, but it achieves the goal of moving that knowledge into the culture."

Not if there isn't any knowledge to move into the culture. Countries that did not have a patent system for citizens have been excellent examples of just that.

"That is a nice reasoning, but it exists to promote the useful arts, not to make you rich."

It isn't "reasoning", it's historical fact. The patent system was put in place, yes to promote the useful arts and sciences... BY allowing artists and inventors to profit from their works. If you haven't gotten that by now, you are pretty slow. And it doesn't take reasoning to prove it. There are reams of historical documents that say as much.

Re:wait

[zzatz]

Very wrong. Patents exist to provide an incentive to the inventor for making knowledge about his patent available to the public. It's not about inventors; it's about growing the body of public knowledge. A patent is the reward for publication, not a reward for inventing.

Inventors can profit when they keep their inventions secret, for certain types of inventions. Keeping a better mousetrap secret wouldn't work, because anyone could buy one and figure out how to make it. But if the invention is a better way to make a mousetrap, secrecy could be more profitable than a patent. First to market is another powerful reward that does not depend on patents.

The government should not be in the business of picking winners and losers; the market, with certain notable failure modes, does a much better job. On the other hand, providing incentives to people to act in ways that benefit the public at large is the principle role of the government.

Re:wait

[jvillain]

This would make sense if every one didn't avoid reading patents for fear of getting dinged with treble damages.

Does not Affect Prior Art Doctrine

[Sonny Yatsen]

I'm sure someone's going to start asking whether a First-To-File system affects the prior art doctrine and whether it means big companies can steal ideas from open source projects and patenting them. Let's dispense with some misconceptions.

Misconception 1: This destroys the prior art system.
* This isn't true. A prior art will still cause an application to be denied under 35 USC 102. This means that if any sort of prior art is published (i.e. available to the public) that would anticipate or render an application invalid, it would still operate to render the application unpatentable. Remember, the law requires all patents to be "novel" and "nonobvious".

Misconception 2: This would mean big companies can steal ideas from open source projects and file applications on them.
* This isn't true either. The open source project would function as prior art against the later application. Even though there is a first to file system, it doesn't mean that the first person to file can steal ideas that were out there and use it as their own.

The first to file system only really works in a very specific context - where you have two inventors who filed an application on almost identical types of inventions within a short period of time. Under the current system, there has to be a very fact-intensive and time consuming process of determining who was the first inventor - which means going through years of lab notebooks and correspondences to pin out the priority between two applicants. This is very expensive, very time consuming, and it's also taxing on the court system and the USPTO. A First to File system makes it much easier - priority can be determined within seconds of looking at the filing date.

Re:Does not Affect Prior Art Doctrine

[Anonymous Coward]

I don't know much about the new legislation, but considering that the provision in current law, that "... all patents to be "novel" and "nonobvious"." has been frequently ignored, I'm not sure the specific context you mention would be the limiting factor in practice.

Re:Does not Affect Prior Art Doctrine

[Drakkenmensch]

Remember, the law requires all patents to be "novel" and "nonobvious".

What will become of the dozens of geniuses who patent laser pointers as "cat exercise systems"?

Re:Does not Affect Prior Art Doctrine

[Cyberax]

This isn't true. A prior art will still cause an application to be denied under 35 USC 102. This means that if any sort of prior art is published (i.e. available to the public) that would anticipate or render an application invalid, it would still operate to render the application unpatentable.

Yeah, sure. Like USPTO concerns itself with prior art searches. Or like you're going to waste hundreds thousands of dollars in multi-year lawsuits if a patent troll approaches you with 'an offer you can't refuse'.

Re:Does not Affect Prior Art Doctrine

[MozeeToby]

So... how is that different from now? Patent trolls trample over prior art with the system as it is today, the change to first to file doesn't change that in the slightest. All it does is reduce the costs associated with a review if two similar applications are received at nearly the same time. Incidentally, this is exactly how most of the Western world's patent systems work, and they have no more, and often quite less, trouble with patent trolls than the US does.

Re:Does not Affect Prior Art Doctrine

[vldragon]

If there are multiple inventers coming up with the same thing at aproximantly the same time you would think the invention wouldn't be patantable as it should be considered obvious by the fact that more than 1 person came up with it.

Re:Does not Affect Prior Art Doctrine

[BenFenner]

The first to file system only really works in a very specific context - where you have two inventors who filed an application on almost identical types of inventions within a short period of time.

Why is that even a specific case to consider that needed clarifying? Wouldn't two inventors coming up with almost identical types of inventions within a short period of time necessarily evoke the "obvious/trivial advancement" disqualification?

Re:Does not Affect Prior Art Doctrine

[Anonymous Coward]

Old system:
I invent something, somebody catches wind of it and puts in a patent application, then I send in a patent application showing I've been working on it for longer. I get a patent.

New system
I invent something, somebody catches wind of it and puts in a patent application, then I send in a patent application showing I've been working on it for longer. They get a patent because they filed first and I can't afford to take them to court to get it reversed. I lose. Or I do have enough money, they lose the patent and now NOBODY has a patent. I needed patent protection so I could license out my product. I still lose.

Re:Does not Affect Prior Art Doctrine

[knight24k]

I'm patenting my laser pointer as a cat bowling device.

http://www.youtube.com/watch?v=aFztjgfDWDA

Re:Does not Affect Prior Art Doctrine

[TheoMurpse]

on examination by a tech person instead of a lawyer, many would be thrown out before they got very far

You realize PTO agents are necessarily "tech people," right? You don't need a law degree to be one, but you absolutely must have a technical degree (well, there are a few exceptions if you just took a bunch of chem/physics in college but didn't major in it, but for example, I have a BS in abstract math and a law degree, and I cannot become a PTO agent because I lack the requisite tech background).

Not to mention the fact that, to become a patent prosecuting attorney, you must sit for the Patent Bar. To be eligible to sit for the patent bar, you basically have to be an un-degreed science genius, have taken a bunch of science classes in college that basically measure up to a BA in the subject, or have a bachelors, minimum, in a science/engineering discipline.

So the suggestion that the patent process is not dominated by tech people is specious unless you lose your tech credentials by virtue of getting a JD.

Just as an example, 100% of the friends from law school who work in the patent field have an engineering BS if not a MS/PhD. The least-credentialed I know has a BS in ME from MIT.

Re:Does not Affect Prior Art Doctrine

[Sun]

I needed patent protection so I could license out my product. I still lose.

Too bad for you. You should have done one of two things:
1. Made your invention public, either through a patent (a provisional patent costs very little money) or through publishing.
or
2. Made sure no one gets wind of it.

The patent system was established to give inventors to publish their inventions so that society at large can benefit from it. You attempted to to circumvent that, and, indeed, lost. Why is that not a win for the system, again?

Shachar

Ha

[cbs4385]

For a patent, it should be the first to walk into the office with a working prototype, and that implementation is what the patent should cover.

Reform?

[Anonymous Coward]

There's nothing reformatory about this bill. The patent trolls and lawyers are well pleased.

Re:Reform?

[countertrolling]

There's nothing reformatory about this bill.

You can't "reform" slavery

Re:First to file versus first to invent?

[Sonny Yatsen]

Then the patent is invalid on the basis of 35 USC 102(f):

"A person shall be entitled to a patent unless -
(f) he did not himself invent the subject matter sought to be patented."
http://www.uspto.gov/web/offices/pac/mpep/documents/2100_2137.htm#sect2137

first to file standardizes things world-wide

[rritterson]

Other comments have taken to clearing up some of the misconceptions regarding first-to-file versus first-to-invent so I won't duplicate them. However, one thing not yet pointed out is that the vast (VAST) majority of the rest of the world uses a first-to-file system.

By switching our system, it reduces the burden on an inventor (and thus the legal cost) of obtaining a world wide patent as the systems become closer to the same. And note that Europe has not considered switching to first-to-invent as a way to combat patent trolls, which says something about how much the USA switching will help/hurt trolls.

That's not real reform

[Waffle Iron]

Backlog, schmaklog. The real reform we need would be to reduce the number of patents issued by orders of magnitude. The bar for patentability should be raised from "not blatantly obvious to below-average freshman engineer" to "that's freakin' genius". That would simplify things for everybody, eliminate most of the huge burdens on society involved with accounting for tens of millions of extant patent claims, while still ensuring that people with genuine Big Ideas get rewarded.

IMO, it would be an improvement to make the patent system a reality show like The Apprentice or American Idol. Allocate something like 100 possible patents each year to each field of industry, then have juries (not bureaucrats) review all of the applicants in rounds, make the would-be idea monopolists defend their claims in public. Keep voting applications off the island until the few truly worthy patent candidates still stand.

Re:First to file versus first to invent?

[necro81]

If memory serves, the person who actually invented it may still file for a patent for up to one year after the first public disclosure. Generally, businesses will still try to file a patent before the first public disclosure, but that is largely a strategic decision, not a legal one.

More info on the bill

[billstewart]

The Senate bill is S.23, aka "America Invents", sponsor Patrick Leahy, who's been trying to get patent reform done for years.
Bill status query at thomas.loc.gov (not sure if these are persistent), Computerworld article, National Journal with some brief comments from pro/neutral/con parties, SF Chron article.

Silicon Valley businesses large and small were mostly against it, IBM was for it. Dianne Feinstein attempted an amendment to remove the First-to-File part, but voted for it anyway after that failed. Barbara Boxer voted against.

The US patent system has been first-to-invent for a long time, while Europe has been first-to-file. There's lots of other detail, largely intended to reduce the amount of patent litigation, improve the coordination with non-US patents, potentially improve the problems with patents on things with prior art and obviousness, and affect some tax issues."

Re:More info on the bill

[Drishmung]

First-to-invent is more moral... OK, you can stop laughing now. Yes, I suppose you're right, this is the patent system as it actually exists rather than as we'd like it to exist...

In practice, first-to-invent does not result in a more moral outcome. It encourages deceit ("I invented it last year, before this similar device entered the market. Honest, I did.") and litigation. It's more complex, so it's slower and more expensive. Most (?) other countries have first-to-file. As far as I'm aware none have ever contemplated switching to first-to-invent.

It's a first step. Nuke business process and software patents, stop litigation tourism and mandate loser pays court costs and things might even start to improve.

First-to-File prevents publish-then-patent

[billstewart]

One big difference between the First-to-File system and First-to-Invent is that with First-to-Invent, you can publish your invention and then file the patent within a year of publishing it, while it's very difficult to do that with first-to-file. I'm not sure how important that is in practice; one major impact it had in the US was the RSA patent and other patents that were affected by another US quirk, which is that the military can declare your patent application to be classified and prevent publication (nearly forever), and Publish-then-Patent made it possible for R, S, and A to get their work out.

Re:First to file versus first to invent?

[TooManyNames]

Nope. An inventor has up to a year after public use/offers for sale to patent. If, during that time, someone else files for patent before the inventor (say one of the purchasers), they'd get priority with first-to-file, not the actual inventor. Now, 102(f) is there to safeguard against fraudulent claims of inventorship, but if it can't be proven one way or the other, the inventor is SOL. Under the old system, the inventor could have used notebooks and other materials to prove his claim to the invention (being first to invent). Of course, it wouldn't be easy for the inventor to prove he was first to invent, and it would cost him to do so, but he would still have had the capability to prove he invented first even if 102(f) was unprovable. Now, the only hope for him is to prove beyond a doubt that 102(f) holds.

Novelty and publication

[tepples]

In effect if you invent something you now have to patent it or possibly suffer the consequences from a patent troll.

You appear to claim that the law has been changed such that publication of an invention outside of a patent no longer makes the invention not "novel". Can you cite a particular piece of language in the bill supporting your claim?

Re:Novelty and publication

[Anthony Mouse]

I think it will have the opposite effect you suggest, bringing MORE secrecy rather than less. You dare not let anyone see even an early demonstrator or model, or even talk about something you have in the works for fear they could run out and file a patent EVEN WITHOUT inventing anything. If you don't have to prove you invented it first, you can shut down your competitors with mere paperwork, while you labor in the back room to actual make something.

I haven't read the bill, but generally speaking, I don't think first to file really does that. The only way someone can do that under first to file is by breaking the law by claiming they invented it when they really got the invention from your publication. And if the troll is willing to break the law then the problem you're expressing still exists under a first to invent system, because the troll can just claim they independently invented the invention before you did and produce a bunch of backdated documents "proving" as much.

First to file even improves on the situation because it allows the actual inventor to simply file a patent application before they publish the invention, thereby precluding anyone else from lying and claiming they invented it first, because no one (other than a bona fide independent inventor) can file before the inventor if the inventor files before publishing the invention.

Of course, what would be even better is "first to publish" (and filing a patent application counts as publication). That way everybody has an incentive to publish ASAP because publication is the only way to reserve a priority date, and publication is much easier for smaller inventors than filing a patent application, but publication is much harder to fake than a bunch of backdated notes because you can verify it with e.g. archive.org.

Why do you ask if you already knew the answer?

[Lead Butthead]

From TFA -

Chuck Grassley (Iowa), the Ranking Republican on the Judiciary Committee, and Orrin Hatch (R-Utah) were original co-sponsors of the bill.

The moment I saw THAT name, I already know this bill is bought and paid for by special interests. This is the same MAFIAA loop that tried to rape us with the INDUCE act. I think we'll all be so very f_cked if this f_cker passes. Epic Fail does not even begin to describe the this...

It kills submarine patents

[rsilvergun]

but it creates it's only share of problems described above. First to File would be fine if our patent office would stop rubber stamping stuff like 1 click shopping...

First-to-file is good, not bad

[Bruce Perens]

Currently, a patent applicant can pre-date their application by 364 days and claim that they made the invention before the Open Source project, standards committee, or whatever they are eavesdropping. With first-to-file, there is no more pre-dating of inventions, so this loophole is closed and the Open Source project, standards work, etc. is more defensibly prior art.

There has been no prosecution for lying on a patent application under perjury laws since 1974, when the patent office closed its enforcement department. Lawyers sometimes get penalties if they are caught in inequitable conduct, but not the lying applicants. Thus, there was no penalty for pre-dating your invention.

Re:First-to-file is good, not bad

[Bruce Perens]

They can do everything you are afraid of today, but they can pre-date their invention by 364 days, so that it is even harder for you to prove that you sent an email disclosing this particular invention before they "invented" it.

So, I think it's better for the system to make them file their invention with a date concurrent with or after your email, rather than a year before you. You have a better chance of killing that patent.

It would be nice if public disclosure was enough to prevent someone from filing a patent, but that ignores the fact of patent thickets, in which someone patents many different (and possibly trivial) variations of what you invented. If you want to stop that, sit down and think of the variations yourself, and make a protective disclosure of them.

IMO, the system is still set up so that most of the incentives are for the bad guys, and there are little penalties for being a bad guy. This isn't the last reform bill needed.

Re:First-to-file is good, not bad

[Bruce Perens]

I don't think you are considering the cost and difficulty of arguing precedence over a pre-dated application in a first-to-invent regime. Most Open Source projects and many standards committees simply don't have the funds to bring that to court. First-to-file removes the opportunity for the applicant to perjure in this way, and makes it so much less likely that the question would have to be litigated.

Re:First-to-file is good, not bad

[Bruce Perens]

Yes, but that's perjuring themselves to the court rather than the patent office, and judges are very much more likely to bring perjury charges and the peril to the perjurer is substantial. Also, it puts the burden of proof on the patent holder. And the patent holder would have to show that the material was not published, but was made available to that particular party in some way.

Bill promotes publish-then-patent

[Anonymous Coward]

Actually, if you read the bill, this isn't the case. The provisions give you a one-year window for your own public disclosure still, but that disclosure will bar anyone else immediately. So in fact, publish-then-patent will be even more advantageous: you can disclose and then have a year during which nobody else can beat you to filing unless they themselves disclosed before you. In the current system, you still have the one-year window to put pressure on you to file, but there is no advantage to disclosing because competitors need merely be diligent about reduction to practice or filing to get around your disclosure.

conflict of interest

[shadowofwind]

fees paid to the Patent Office will actually fund the Patent Office

Doesn't this mean that an even larger number of ridiculous patent applications will be approved, since everyone's salary will be dependent on the available budget? Now there's an incentive to massively expand the number of patent awards, with no comparable incentive for quality.

Re:How does the current "Prelminary Patent" fit?

[shutdown -p now]

Does anyone know how this is affected by the present legislation?

No, the ability to publish and then file within one year is still there. In fact, changing to first-to-file means the defense is even stronger, as now someone can't claim that they have "invented before you published" - all that matters is whether they published or filed before you did.

What the reforms should have been

[labnet]

- patent owners must be commercially utilizing their patent to keep it. (within 4 years of patent application)
- software and business process should not be patentable (ie 1 click)
- accidental infringement be not initially punishable
- the bar for novelty to receive a patent be way higher than it is now.
- patent application / renewal fees be based on number of patents held
eg x = v * n(n+1)/2 where v = the base patent cost.
Lets say v = $100
1 Patent = $100/annum to maintain
2 Patents = $300/annum to maintain
10 Patents = $5.5k/annum to maintain
100 Patents = $500k/annum to maintain
1000 Patents = $50M/annum to maintain

This would force very large companies / trolls to only keep their best patents and toss out the dross polluting the patent system.
You would need rules to stop companies spawning sub related companies to get around the intent.

One idea for patent reform

[istartedi]

We should just go to compulsory licensing. If you think that your product might infringe a patent, you place 10% of the sales price in escrow. You keep a record of each sale (something most businesses do anyway), and at the end of the year you have an escrow account that's 10% of gross sales. If you think you are in the clear, you escrow nothing.

OK, are you with me? Those who claim that your product violates a patent have 1 year after your fiscal year ends to make a claim against the money you escrowed. If they don't make any claims, you pocket the money with interest.

If any patent holders make claims during that year, at the end of that year (as long as two years from the first violating sale), two things may happen.

1. You may disburse the escrow funds to claimants, based on the number of patents violated. e.g, You violate 2 Farbco patents and 1 Spimco patent, Farbco gets 2/3 of the money and Spimco gets 1/3. This gets a bit more complicated if you have multiple product lines, but you get the idea.

2. You may contest the claims, but if you do, you are subject to the usual civil claims process. If you didn't escrow anything, this is your only option.

Most people would probably opt for (1), unless the claims are really frivolous.

A system like this would have a number of consequences. First, software given freely would have an escrow of zero--immunity from suits. Per-seat service contracts (ala Red Hat) might be a different story. Yes, commercial free software would have to maintain the escrow. Really, I'd like to see software patents just go "poof!", but one thing at a time. Secondly, "suppressive" patents would not exist. If there realy is a car that runs on water, all I have to do is build it and escrow the money from sales. This might even silence the conspiracy theorists. Ditto for things like restrictions on large battery packs for electric vehicles. Nobody can stop me from building it--they can just tax me 10% for the duration of their patent. If my car is 10% better than theirs, it's worth building. If they aren't building the car, they still get paid; but they can't just sit on the patent at the behest of Big Oil.

Each application should be a competition

[morkk]

The patent office posts on it's website the summary or abstract of the application and invites knowlegable practioners of the art to invent a comparable system within some given period (possibly determined by the complexity of the invention).

If an alternative invention is submitted before the end of that period then that application must have been obvious.

If the application is repudiated within 1 hour the applicant is forced to eat their application documentation!