The Economist on Patent Reform

ar1550 writes "The Economist recently posted an opinion piece on the state of patent systems, describing not just the mess that is the USPTO but flaws present in Europe and Asia. From the article, "In 1998 America introduced so-called 'business-method' patents, granting for the first time patent monopolies simply for new ways of doing business, many of which were not so new. This was a mistake." The article also describes the difficulty of obtaining legitimate patents. "

Re:One-sided article

[Jumbo Jimbo]

The motivation for allowing business-method patents, described in this paper from Harvard Law School [harvard.edu] is that innovations were too easy to copy - essentially the same motivation behind the original idea of patents.

It seems that a good idea in principle may have resulted in legislation that is not working in practice because of a flawed framework / companies taking advantage (your choice). Not that I agree with the idea of business-method patents in the first place, but this may make the idea behind them clearer.

Re:One-sided article

[Rude Turnip]

The grandparent post was referring to the reason for business method patents, not patents for real things.

Re:One-sided article

[3StrangeAllies]

The big turning point in Business Method Patents, as someone stated, was State Street Bank & Trust Co. v. Signature Financial Group, Inc. - 1998 [umd.edu]. The key point in this change of attitude was, according to Judge Rich, that "business methods have been, and should have been, subject to the same legal requirements for patentability as applied to any other process or method" (refering to In Re Schrader, 2 F.3d 290 (Fed.Cir.1994).
Somehow, it makes sense -- the general set of criteria for patentability should apply to most legal subject-matter (not sure that it would be wise to grant a patent to a new process to produce Cocain ;). Hence it is not the business method patentability in itself that seems flawed but the patent prosecution that let stuff go through without the proper checking. Especially regarding the facts in State Street, the mathematical process and business method really showed an innovative process, and it would have been counter-productive to bar this process from patentability...

Just my two €urocents...

Re:Superficial article

[Halo1]

1. It address the European patent policy in one line with the European Parliament. Clearly, the author is unaware of the complete separation between the European Patent Office and the European Union. Therefore, the European Union as such (and even less the European Parliament) can issue any decision with respect to the European Patent Convention.

Since all European patents have to be enforced in national courts under national laws which have to be in line with European directives, it does not make sense for the European Patent Office to grant patents which will obviously not be enforceable. In fact, a representative of the EPO said last Wednesday at a conference [eifonline.org] organised by the European Internet Foundation that whatever the outcome, the EPO will respect the directive by the EU.

Re:Who wrote it?

[yiangouk]

The Economist [wikipedia.org]

"Articles, which are often heavily opinionated, almost never carry a byline. This means that no specific person or persons can be named as the author. Not even the name of the editor (currently Bill Emmott) is printed in the issue. The author of a piece is named in certain circumstances: when notable persons are invited to contribute opinion pieces; when Economist writers compile surveys; and to highlight a potential conflict of interest over a book review. The names of Economist editors and correspondents can be located, however, via the staff pages of the website."

Comment removed

[account_deleted]

Comment removed based on user account deletion

Re:One-sided article

[Vicegrip]

Everything is reproducable once you've seen it. Thats the whole idea behind patents.

Apparently Jefferson disagreed with you:

From the Economist:
PATENTS, said Thomas Jefferson, should draw "a line between the things which are worth to the public the embarrassment of an exclusive patent, and those which are not."

And this [earlyamerica.com]:
Jefferson, a strong proponent of equality among all people, was not sure if it was fair or even constitutional to grant what was essentially a monopoly to an inventor, who would then be able to grant the use of his idea only to those who could afford it. His feeling that all should have total access to new technology was one of the reasons he never took out a patent on his own inventions. This is consistent in his belief in the natural right of all mankind to share useful improvements without restraints. He felt that inventions can not, in nature, be a subject of property and that the promiscuous granting of patents was not only against the theory of popular government, but would be pernicious in its consequences. (Curtis, 1901) In fact he referred to patents as "embarrassments to the public" (McLaughlin, 1989).

The Gauntlet

[rumblin'rabbit]

A serious problem with patents is how difficult it is to determine whether something is covered by a patent. Here are some of the obstacles:

• You have to know how to do a patent search. This is fairly mechanical, but many don't know how to do it.
• You have to know what keywords to search on. There can be many different names for the same things, particularly between different disciplines (eg, singular-value decomposition, principal component, Karhunen-Loeve, reduced rank, and eigen-whatever can all mean the same thing). Even experts in the field might not be able to think of all possible terms.
• You have to know how to read patents in general. This is no small thing.
• You have to be able to understand the particular patent you are reading. Good luck. Patents can be remarkably opaque. I've read patents in areas that I'm expert in, and been left with only a vague idea of what they were about.
• After you understand the technical aspects of the patent, you have to analyze exactly what is and is not covered by the patent. In many cases this can take considerable legal expertise. It may even require the services of a patent lawyer.
• Even when you have determined that an invention is covered by a patent, is the patent valid? For example, a three-year-old patent might describe a method that's been well known for a decade - in other words, it's prior art. A common occurrence is that only part of the patent might be invalid. Do you take the risk of ignoring the patent?
• Has the owner of the patent been paying the patent maintenance fees? Has there been a judgement overturning or limiting the patent?

I'm sure others can add to this list.

Re:would this fix the bulk of the problem?

[PMuse]

Basically, keep things as is, but limit the patent term to,say, 5 years. After that patent owner can extend it to the full 17 year term but make the extension EXPENSIVE (say, 40K per patent)...

The system you suggest already exists and has existed for decades, albeit at about 1/10th the costs you propose. It is called Maintenance Fees [uspto.gov]. See 37 CFR 1.362 et seq. [cornell.edu] These fees are due at 4, 8, and 12 years after issuance. Big companies are charged higher amounts than small ones.

Re:the problem isnt so much the system

[Aceticon]

A well designed system does not rely on the the assumption that no participant will abuse it. When abusing a system is beneficial to the abuser, any system that has no explicit definition of abuse and/or no negative-feedback (read punishment) on abuse will be BE ABUSED - abusing such a system will only provide rewards and no penalties.

This is as valid for a set of rules encoded as a construct of laws as it is for any set of rules determining usage/access of/to a system.

I sugest some form of punishement should be introduced for patent applications deemed to be frivolous, duplicate or over encompassing.

Re:patents for the rich/poor

[Halo1]

You'd get large companies spawning fake small companies and later buying them up (together with their patents), or simply taking licenses on those patents. And your proposal would not solve the litigation problem in any way. Have a look at the (later slides in) presentations given by Brian Kahin [ffii.org] and Jim Bessen [ffii.org] at the FFII conference [ffii.org] from last week.

Re:The Gauntlet

[back_pages]

What you say is true, however let me try to give a helpful response.

• You have to know what keywords to search on.
  It's far easier to search if you use the classification. Rather than hoping you get the right terms, it acts like an index. This alone should narrow down your search to 100-5000 patents, and don't forget they are all cross referenced before being issued and related patents are listed on the front page.
• You have to know how to read patents in general. This is no small thing. You have to be able to understand the particular patent you are reading.
  While this is a problem, the law states that if the patent does not clearly convey to a person of ordinary skill in the art how to make and/or use the invention, it is legally deficient. I'm not making an excuse for all patents, but while 99.9% of the patents may be cryptic, you might be surprised how readable they are when you are "an person of ordinary skill in the art". Again, I'm not excusing all patents.
• Even when you have determined that an invention is covered by a patent, is the patent valid?
  This one is easy. If the patent has not expired or been abandoned, it is valid until a judge says otherwise. All live patents are presumed valid until proven otherwise in court.

Systems

[Antony-Kyre]

First of all, I like the idea of http://fairtax.org/ [fairtax.org], but here's a very simplified idea that follows the same logic.

Whether we keep or get rid of income taxes, it doesn't matter for this idea...

Fix it so 30% of the total federal tax revenue is redistributed. If this means raising taxes, so bet it. If this means cutting wasteful spending, so be it. But with spending over a half trillion per year on "defense", I'm sure we could cut a big part of that, although some would disagree.

Semi-free college education at the least. Do something like this...

In exchange for a free college education, the person would pledge 5% of their income for 20 years to help pay for this. So not only would someone be contributing to the regular tax system by having a better job, but they'd be putting 5% of their income into this system. 5% of $0 is $0, so if you don't earn a dime in a given year, no big deal. And it's over after 20 years, hopefully by age 42 or so.

Personally, I think we should get rid of the IRS and get rid of property taxes, and simply go on a consumption/sales tax system, provided that the rebates are kept.

The 30% thing I mentioned earlier, let's play around with it. If current federal income taxes results in $2000 billion, 30% of that would be $600 billion. We could cut a large chunk out of the DoD, and modify some tax brackets too.
-
Now we got $600 billion among nearly 300 million citizens. That's $2000 per person, but we don't want people having babies just to get the money. Solution: Limiting it to just adults would result in about $2666 per person, if there are 225 million adults. Or maybe just limiting it to anyone who is age 5 and up, which is kind of like anyone who is school age.
-
If a college tuition is $4k per year, this could very well help. This could help seniors too. And those who are homeless on the street who get $0 now, this would surely help.

Inventor's view on the subject on AlwaysOn

[jernst]

I just recently wrote a piece for AlwaysOn with a similar view, from the perspective of the inventor.

http://www.alwayson-network.com/comments.php?id=57 63_0_5_0_C [alwayson-network.com]