US Patent Reform Slouches Towards Bethlehem

But at last it seems to be happening:

This case involved a guy who was trying to patent the concept of "mandatory arbitration involving legal documents." The USPTO denied the patent. After a failed appeal, the guy went to court, and CAFC is also saying that his concept does not deserve patent protection, with this being the key quote: "The routine addition of modern electronics to an otherwise unpatentable invention typically creates a prima facie case of obviousness." In other words, simply taking a common process and automating it on a computer should be considered obvious -- and thus, not patentable. This doesn't rule out business model or software patents by any means -- but it at least suggests that the courts are beginning to recognize that the patent system has gone out of control. The court also specifically addresses its own earlier State Street decision, suggesting that people had been misinterpreting it to mean any business model was patentable -- when the USPTO and the courts should still be applying the same tests to see if the business models are patentable. It then notes that a business model on its own shouldn't be patentable unless it's tied to some sort of product, and then states: "It is thus clear that the present statute does not allow patents to be issued on particular business systems -- such as a particular type of arbitration -- that depend entirely on the use of mental processes."

What Muhammad ibn Musa al-Khwarizmi Knew

Nice to see algorithms getting some respect:

Algorithms, as closely guarded as state secrets, buy and sell stocks and mortgage-backed securities, sometimes with a dispassionate zeal that crashes markets. Algorithms promise to find the news that fits you, and even your perfect mate. You can’t visit Amazon.com without being confronted with a list of books and other products that the Great Algoritmi recommends.

Its intuitions, of course, are just calculations — given enough time they could be carried out with stones. But when so much data is processed so rapidly, the effect is oracular and almost opaque. Even with a peek at the cybernetic trade secrets, you probably couldn’t unwind the computations.

Maybe; but the point is, they are just calculations. Which is why the idea of patenting any of them - as raw algorithms, business methods, or software - is, er, patently ridiculous.

Glimmer of Hope for US Patent System

A small step towards patent sanity has been taken:

The House approved the most sweeping changes to United States patent law in more than half a century on Friday in a victory for computer companies like Microsoft and finance companies like Goldman Sachs.

Specifically:

The measure passed by the House would change the rules at the Patent and Trademark Office so patents would go to the first person to file an application, not necessarily the first inventor. That would limit years-long disputes over who was the first to invent new technology and would bring the United States in line with other countries’ patent rules. It would also allow third parties to introduce evidence against applications and would create a system, called post-grant opposition, to challenge new patents.

In litigation, it would limit where patent suits could be filed so that cases are not concentrated in court districts deemed favorable to plaintiffs, create a new way to calculate damages to reflect the contribution of the invention to the overall product and allow immediate appeals of court rulings on the interpretation of patent terms while cases are proceeding.

Patent Joke of the Month

It is, of course, hard to choose from the rather crowded field of contenders, but this one certainly takes the biscuit:

An Information and Application Distribution System (IADS) is disclosed. The IADS operates, in one embodiment, to distribute, initiate and allow interaction and communication within like-minded communities. Application distribution occurs through the transmission and receipt of an "invitation application" which contains both a message component and an executable component to enable multiple users to connect within a specific community. The application object includes functionality which allows the user's local computer to automatically set up a user interface to connect with a central controller which facilitates interaction and introduction between and among users.

A system to create an online community - including, of course, that brilliant stroke of utterly unique genius, the "invitation application": why couldn't I have thought of that? (Via TechCrunch.)

Do Not Feed the Patent Trolls

Good point here about a big problem with the apparently welcome Peer to Patent project:

Helping patent trolls with their QA is like going through bandits' ammunition and throwing out the dud rounds for them before they try to rob you.

And sensible advice, too:

If you have Prior Art, print it out and put it in your safe deposit box. Make sure that the source is verifiable, but don't tell anyone what the source is. Don't say it's from "the June 1997 login;" or "comp.sources.unix in May 1986". If you want, borrow a tactic from Tim O'Reilly and tell people that you have prior art for a certain patent, but don't give attackers any more information than you have to.

More generally, perhaps the free software community should set up a shadow scheme that tracks all of these patent applications, and works to find prior art, which it then stores safely against a rainy day.

The Bad Boy of Genomics Strikes Again

When I was writing Digital Code of Life, I sought to be scrupulously fair to Craig Venter, who was often demonised for his commercial approach to science. Ind fact, it seemed to me he had often gone out of his way to make the results of his work available.

So it's with some sadness that I note that the "Bad Boy of Genomics" epithet seems justified in this more recent case:

A research institute has applied for a pat­ent on what could be the first largely ar­ti­fi­cial or­gan­ism. And peo­ple should be al­armed, claims an ad­vo­ca­cy group that is try­ing to shoot down the bid.

...

The ar­ti­fi­cial or­gan­ism, a mere mi­crobe, is the brain­child of re­search­ers at the Rock­ville, Md.-based J. Craig Ven­ter In­sti­tute. The or­gan­iz­a­tion is named for its found­er and CEO, the ge­net­icist who led the pri­vate sec­tor race to map the hu­man ge­nome in the late 1990s.

The re­search­ers filed their pat­ent claim on the ar­ti­fi­cial or­gan­ism and on its ge­nome. Ge­net­i­cally mo­di­fied life forms have been pa­tented be­fore; but this is the first pa­tent claim for a crea­ture whose genome might be created chem­i­cally from scratch, Mooney said.

This is problematic on a number of levels. For a start, it shouldn't be possible to patent DNA, since it is not an invention. Simply combining existing sequences is not an invention either. There is also the worry that what is being created here is the first genomic operating system: locking others out with patents maans repeating all the mistakes that have been made in some jurisdictions by allowing the patenting of conventional software.

US Patents: Is the Tide Turning?

Maybe I'm an incorrigible optimist, but these look hopeful signs:

The U.S. Supreme Court made it easier to challenge patents for failing to introduce genuine innovations, siding with Intel Corp. and Cisco Systems Inc. and dealing a setback to the drug and biotechnology industries.

...

The decision extends a Supreme Court trend that has put new limits on patent rights. In today's case, the justices heeded arguments from large computer companies and automakers that the lower court test, which centered on the requirement that an invention be "non-obvious," had given too much power to developers of trivial technological improvements.

"Granting patent protection to advances that would occur in the ordinary course without real innovation retards progress," Justice Anthony Kennedy wrote for the court.

(Via Slashdot.)

Two in a Single Stroke

This is cool: further proof not only that patents in general, and software patents in particular are stupid, but that at least some VCs are amoral parasites:

Earlier this year, we wrote about one such fund, Altitude Capital Partners who had quietly invested in Visto, but made it clear that it really only cared about companies who had patents that could be used in lawsuits. John points us to a Forbes article that talks about Altitude and other private equity funds that have raised hundreds of millions of dollars solely to support patent lawsuits. In other words, if you thought things were bad in the past, they're about to get much, much worse. Once again, almost all of these lawsuits aren't about protecting the interests of an inventor at all.

O rose, thou art sick.

Microsoft Patents "Lack Significant Innovation"

Now here's an interesting little cross-cultural spat:

The European Commission has warned Microsoft that it could impose further penalties in its ongoing antitrust case against the software giant.

The EC claimed on Thursday that Microsoft wants to charge too much for interoperability protocol licences that enable third-party software vendors to develop software compatible with Windows servers. In a damning statemement, the EC claimed that the protocols "lack significant innovation", even though Microsoft has been awarded patents on much of the technology in question.

And what's Microsoft's response? Why, to go running to Mummy:

Microsoft general counsel Brad Smith responded that other government agencies had found "considerable innovation" in Microsoft's protocol technology.

"US and European patent offices have awarded Microsoft more than 36 patents for the technology in these protocols, which took millions of dollars to develop, and another 37 patents are pending, so it's hard to see how the Commission can argue that even patented innovation must be made available for free," said Smith in a statement.

Hey Brad, you don't think this might be because the patent system in many countries is horribly broken, and regularly awards patents for obvious, trivial and otherwise inappropriate ideas? Just a thought.

Patent Abuse

Oh, look. Here's yet another reason to get rid of patents:

Guess what? Radio frequency identification tags are insecure. But don't demonstrate the technology's problems at a security conference. If you do, HID Global, a manufacturer of access-control devices, might sue you for patent infringement.

...

The use of patent law to prevent vulnerability discovery and discussion is bitter irony, because a fundamental purpose of patent law is disclosure: In exchange for the right to exclude others from using, making or selling a novel invention, an inventor agrees to make public all the details. Once issued, patents are a searchable public record, and expire after 20 years.

Patently Foolish

Oh, here's a good idea: let patent experts help decide whether or not to grant lots more patents. And if you need proof this is not going to be good for us, take a quick at gander who's endorsing the move:

The Business Software Alliance, whose members include Apple, Microsoft, Intel and IBM, was quick to hail the bill's approval.

Microsoft's Eternal Cheek

This is rich:

In this culture of instant information, some Microsoft Corp. researchers are pursuing a radical notion -- the concept of saving messages for delivery in decades, centuries or more.

The project, dubbed "immortal computing," would let people store digital information in physical artifacts and other forms to be preserved and revealed to future generations, and maybe even to future civilizations.

So, the company that more than anyone has tried to lock people into opaque, closed formats that will be unreadable in a few decades, let alone a few millennia, and which even now is trying to foist more of the same on people, suddenly discovers the virtue of unconstrained accessibility.

But to add insult to injury, it then tries to patent the idea. Earth to Microsoft: this is called openness, it's what you've been fighting for the last thirty years. There's a fair amount of prior art for the basic technique, actually.

Digital Library of the Commons

OnTheCommons has an interesting post about a new book called Understanding Knowledge as a Commons. This sounds great - see Peter Suber's comment below for details on open access to its contents. but sadly - and ironically - seems not to be open access (though I'd bet that Peter Suber's contribution to the collection is doubtless available somewhere),. However, tThis article did mention something I'd not come across before: the Digital Libary of the Commons.

This turns out to be a wonderful resource:

a gateway to the international literature on the commons. This site contains an author-submission portal; an archive of full-text articles, papers, and dissertations; the Comprehensive Bibliography of the Commons; a Keyword Thesaurus, and links to relevant reference sources on the study of the commons.

Among the list of commons areas, there is Information and Knowledge Commons:

anticommons, copyright, indigenous, local, scientific knowledge issues, intellectual property rights, the Internet, libraries, patents, virtual commons, etc.

Strange that free software is not included. But good, nonetheless.

EU: Not at All Patent to Me...

Well, I still don't really know what's going on as far as patents in Europe are concerned. But the Foundation for a Free Information Infrastructure (FFII), an organisation whose judgement I generally trust in these matters, seems happy enough with the latest vote in the European Parliament:

"We're 80% happy with the result" comments Jonas Maebe, FFII board member. "The main unfortunate artefact left in the adopted resolution is the fact that it promotes accession of the EU to the European Patent Convention, which would delegate most patent-related responsibilities to the civil servants of the Commission and member states. Overall our main concerns are however addressed and we would like to congratulate the many MEPs and assistants who very worked very hard for this result."

Bad, Bad Patent

First WIPO, and now this:

Businesses based on the licensing of patented technologies could be able to sue the owners of the patents while still using those patents if a biotech firm wins its landmark US case. The case could upend the basis of much US patent law.

...

While some argue that nobody is better positioned to assess whether or not a patent is valid than a licensee, there are worries that a precedent set by a MedImmune victory would create havoc. During the oral hearings one of the judges, Anthony Kennedy, said that a result in favour of MedImmune could "flood the courts" with cases.

I can't help feeling that this (possible) development is partly as a result of the constant hammering away by intellectual monopoly bores like me: even major companies like IBM and Microsoft are making noises about "bad" patents, and this case looks like it might be the tipping point for this discussion.

Microsoft Patents Free Software

OK, so it's the idea of free computing that Microsoft is trying to patent, rather than just free software. It's still doubly stupid. Stupid, because this is obvious, and hence non-patentable. Stupid, because it's not even novel (and hence non-patentable): it was tried during Dotcom 1.0, where it failed miserably. Make that triply stupid.

Who Ya Gonna Call? Patentbusters!

This blog has lamented often and loudly about the idiotic patents being granted, principally in the US (but with the EU trying very hard to follow down the same pitiful path). The question is, what can we do about it? Or, to put it another way, who are we going to call? - Patentbusters, of course, in the form of the EFF's Patent Busting project, which seeks to find prior art to invalidate bogus patent claims.

Mind you, with some of the top 10 most wanted, you have to ask why even this is necessary, so blindingly obvious are they. Take ClearChannel, for example, which somehow has a patent for

A system and method for recording live performances (e.g. music concerts), editing them into tracks during the performance, and recording them to media (e.g. CDs) within minutes of the performance ending.

Well, that must have been really hard to invent.

A Tiny Victory?

This is all deeply abstruse stuff, but the bottom line is that there seems to be a glimmer sanity in the European Commission's attitude to the European patent system, in the clash between the competencies of the European Union and the European Patent Office (which have little to do with each other), and even with regard to the patentability of software. Or are we being too optimistic?

I've Seen the Future - and It's Patented

The name Nathan Myhrvold probably doesn't strike fear into your heart; it may not even be known to you. But one day, rest assured, he will make Bill Gates look benign. Gates simply wants to own the software industry, and, as has been amply shown over the last quarter century, is prepared to do anything - including creating the odd illegal monopoly - to achieve that. But at least Gates has the virtue of believing passionately in the value of the software his people make; and at least they do actually make something.

Myhrvold's company, Intellectual Ventures, does not make anything. It will never make anything. For its domain is patents, and all it aspires to do is to create the world's biggest and most lucrative heap of patents to get the people who do actually make stuff to pay licences - whether justified or not - by threatening to sue them if they don't. Industrial-scale patent troll-dom, in other words.

Myhrvold once worked for Microsoft, and became very rich doing so. His new venture is based on an astute reading of the broken patent system in the US, and on how to play it in all its glorious brokenness. If you want the full details, read the excellent article in IP Law & Business, probably the best introduction to just how Myhrvold intends to do it.

He may well pull it off. His logic is impeccable, as you would expect from someone who is anything but a fool. But it is based on the past - a deeply-flawed past that threatens to bring innovation to a grinding halt in the US, and anywhere else stupid enough to acquiesce in the latter's demands that its own patent regime be imposed as part of trade agreements.

For all his cleverness, Myhrvold cannot see - will not see - that the future belongs to a different model for "intellectual property", a commons-based approach made famous by free software, though not invented by it (it's actually as old as the idea of the commons, which goes back to the Romans and beyond into the mists of time).

In fact, Myhrvold's likely success in bringing entire sectors to their corporate knees through the use of broad patent portfolios may have the ironic consequence of hastening the ultimate repeal of all the accumulated stupidities in the fields of patents, trademarks and copyright. For this reason, I wish him every success. Almost.

Paying the Price for Free Software

Sometimes, I think they do this kind of thing on purpose, just to annoy me.

Somebody writes some free software; somebody later "patents" the same software/idea (as if you could patent either software or an idea) - and then accuses the free software author of infringing on their patent and causing them financial loss by daring to give away an open source program, and tries to recover a mere $203,000 as compensation.

So let's count them, shall we? Patenting an idea, patenting something blindingly obvious, patenting something that has prior art - and then having the temerity to harass someone who actually came up with the idea first by demanding money they don't have from not selling the program because they generously give it away: only in America. (Via Right to Create).

I hope.