Linux group asks Supreme Court to nix all software patents
Last week an advocacy group that champions the use of free and open source software, including the Linux operating system, asked the U.S. Supreme Court to invalidate all software patents.
The request was made by the Software Freedom Law Center in an amicus brief submitted in the case known as Microsoft v. AT&T, which is scheduled for argument Feb. 21. It is unlikely that the Court would take the advocacy group up on its invitation, if for no other reason than the Court has only asked the parties to brief narrower questions presented by the case, which focuses on the degree to which a U.S. software patent can be enforced abroad. Still, the center, which aims to provide legal peace of mind to developers and users of open-source software, is putting the lightening-rod issue on the Court's radar screen and hoping the Court will craft its ruling in a way that tees up the broader issue for future resolution.
One of the greatest threats to the acceptance of open source software by major corporate customers is the so-called FUD factor--the "fear, uncertainty, and doubt" that arises (justifiably or not) from the concern that patented proprietary software code might have found its way into the open source software at some stage, exposing the customer to liability for patent infringement. Because open source software is created through a decentralized process, with hundreds of individual developers making contributions, it is hard to guarantee against such an occurrence. If the SFLC could ever succeed in having the Supreme Court invalidate all software patents in one fell swoop, its mission would largely be complete.
The goal is not as pie-in-the-sky as it might sound. For many years the consensus assumption in the legal community was, indeed, that software, because it is just a set of instructions readable by a compatible machine, could not be patented without impermissibly carving out of the public domain fundamental laws of nature, abstract ideas, or mathematical algorithms, which the Supreme Court has previously declared to be off limits to patenting. That view was gradually reversed by a series of rulings in the 1990s by the U.S. Court of Appeals for the Federal Circuit, a special appellate court set up in the 1980s to handle patent cases, among others. But Daniel Ravicher, the legal director of the Software Freedom Law Center, argues in the group's brief that these Federal Circuit rulings are inconsistent with the earlier U.S. Supreme Court precedents in the area. He asks that they at long last be examined by the the High Court--and overruled.
The SFLC seeks to ride a legal wave of sorts. The Supreme Court now appears to be cutting back on years of expansive rulings by the Federal Circuit, which many perceive as having inappropriately maximized the rights of patent-holders vis-a-vis accused infringers. Last term, in the eBay v. MercExchange case, the Court struck down one such pro-patent-holder rule that the Federal Circuit had devised, and this term, in the recently argued KSR International v. Teleflex case, the Court is widely expected to pare back another. (For an earlier Legalpad entry on the KSR case, click here.)
The Microsoft v. AT&T dispute concerns a law that was passed in 1984 to plug a loophole that originally had nothing to do with software. The law focused on banning the practice whereby some U.S. manufacturers were effectively doing end runs around other people's U.S. patents by manufacturing the components of infringing inventions here in the U.S., but then shipping the components overseas for final assembly. The statute deemed such conduct to be infringing, so long as any "component" of the invention was being "supplied" from the U.S.
In this case, Microsoft (MSFT) has admitted that certain software in its Windows operating system violates an AT&T (T) patent relating to the digital coding and decoding of human speech. Though Microsoft concedes it must pay royalties for copies of Windows sold in this country, the question is whether it must also pay AT&T for copies of Windows that are installed on computers that are manufactured and sold abroad. In those situations, Microsoft loads a copy of Windows onto a "golden master disk," and sends that abroad. There, licensed "replicators" make a first-generation copy of Windows from the golden master disk. Then that first-generation copy is "installed" onto foreign manufactured computers--i.e., it is copied again. So, argues Microsoft, it's not infringing AT&T's U.S. patent for two reasons: (1) no molecule of anything that winds up in overseas computers is being "supplied" from the United States; and (2) software is not a "component" of a patented invention within the meaning of the 1984 law. Ravicher hopes to import his contentions onto the second prong of Microsoft's argument. He would like the Court to say: Right, software can't be a "component" of a patented invention because it's not patent-eligible to begin with.
The amicus briefs from the various groups who support either prong of Microsoft's position were filed on December 15. The most influential brief among them is clearly that of the United States (which represents the views of both the U.S. Patent and Trademark Office and the Justice Department), but the Business Software Alliance, the Software and Information Industry Association, Intel (INTC), Amazon (AMZN), Yahoo! (YHOO), and Autodesk (ADSK) have all weighed in on Microsoft's side as well. Amicus briefs on AT&T's side have not yet reached the deadline for filing, so we do not yet know who they will include. The best in-depth discussion (with links) I have seen of the Microsoft v. AT&T case is at the Patently-O blog, where the specific discussion is here.
The issue of Open Source software is a little more ambiguous than you state. Certainly one of the contentions of SCO was that its Unix source code (patented or not) found its way into an IBM Linux distribution. But a patent is an invention along with the steps needed to make that invention operable. In software, the source code can be completely different, yet infringe on a patent because the result is the same (as in Amazon's One-Click). The problem is the the Open Source community is not a monolith; many companies and individuals can leverage a single code base. If the implementation is challenged by a patent holder, it is likely that no one member of that community has the resources to defend it. An added issue is that after two decades, the Patent Office still does not know how to evaluate an application for a software patent.
I suspect that the software industry, including the Open Source communities, can live with any set of legal rules surrounding patents, as long as they are well-understood and consistently applied. That is not the case today.
Lets hope the court has some sense and go with the SFLC.
Software should not be patentable. The market would really take off without the fear of patenting math.
With regards to Peter's comments, the statement that "...a patent is an invention along with the steps needed to make that invention operable" is an incorrect statement. Please see 37 CFR in the entirety.
Personally, I thought the article was on point and well written. The interesting thing is that a seemingly radical brief is really not so radical afterall when viewed from a certain perspective. I believe the CAFC is widely regarded as exiting a 'settling' state after recently being formed.
The only thing that I wished would have been a mention of the patent pooling trend. I belive that Novell and IBM are part of such a trend. However, such a mention was probably beyond the scope of the article.
A well written article and a good read before I start my day.
Share and share alike? Not. Seems there is a bit of business discrimination going on here. Companies can protect logos, book titles, songs, movies, etc but software is somehow immune. Not sure why software is seen as different from any other artistic protection we provide. I understand the utopia open software and its ilk provide, but don't understand why people treat it any different then a song.
Henry, software is unique in that it's protected under copyright law *and* patent law. Under copyright law it's treated much the same as a song would be, i.e. people aren't allowed to copy source code from one program into another without the copyright holder's consent. SFLC is not arguing that this be changed.
The patent coverage is different, in that a software patent covers the set of steps required to achieve a certain result. Similarly, business methods can be patented in this way (an outcome I doubt many expected). This is what the SFLC (and many others) is arguing against.
This is a common confusion in the IP debate, but it's an important distinction to understand. In fact I thought Mr. Parloff was confused on this point as well when I read his statement, "concern that patented proprietary software code might have found its way into the open source software at some stage". But on further reading it seems he understands the difference and this was just a poorly worded portion of the article.
Henry, companies can already protect software just as they can protect books and movies - through copyright. The question is whether software is patentable, which is an entirely different form of protection.
Theoretically, software patents could reward and encourage innovation but in reality they are used more often as a legal tactic. Whether defensive or offensive, either way it wastes time, energy and money that could better be applied toward innovation. In that sense patents actually impair the innovation they are supposed to encourage. The lack any consistent method for evaluating patents exacerbates the situation by making rulings unpredictable.
I agree with Peter that in this case consistency and predictability of the law is paramount. A bad law enforced consistently is often better than a good law enforced inconsistently.
It's amazing to me how many people don't want others to get paid for their work. But they should get paid for theirs - of course. Music should be free. Software should be free. Anything you can steal should be free. What a world.
Henry in Seattle "don't understand why people treat it [software] any different then" "logos, book titles, songs, movies, etc".
But, Henry, songs and other artistic expressions are not patented, they are copyrighted. Most who oppose software patents do support software copyrights. Indeed, the open-source GPL license is based on copyright!
Everyone in IT knows x86 software boils down to a sequence of 1's and zero's stored and at some point used for a sequence of events. IF the courts are allowed to define "events" and software patents are allowed to govern royalites for those events, then the courts and patent office need to agree and define software patents "products" as either parts or a whole and that patents cannot be issued for past events.
For example: As I type n-o-w,(which is a event) chars appear on my screen, hence this is a past product that cannot be patent just because I created a new PC OS and called it "Be OS"
Additionally as the courts got wrong in the Microsoft Anti-trust case, a party cannot argue that a part is a really whole (they put IE on a Mac, but could not possiblity pull it out of Win95...!) when they have no patent on the whole(all events of PC OS)..
((hence the reason MS re-wrote the Win95 code during the course of the trail, so they could later demostrate it as a "functionality that cannot be removed" in Win95OSR2))
The Bottom line is sofware patents need to work more like trade marks not copyrights. Once it's there it's there and best of luck redesigning and re-patenting it(part or whole) before the next guy.
To say it as simple as possible:
A painting or song is art. Software is maths. Law of nature can`t be owned or sold... Users surely would be willing to pay for a usefull software and for service. To ease freeware use by giving more reliable legal bases would not be revolution. It would be evolution.
I have come across numerous software patents awarded for trivial inventions. For instance, displaying a digital image of a real estate property through a modem connection was patented in late 80's while displaying any other image via Compuserve or AOL was not. Currently a similar patent is about to expire that covers digital map based searching of real estate for sale plotted on a map via the internet while mapquest (map info, esri etc.) has been doing the same for years.
Copyright covers the intellectual rights of developed code already. It is my opinion that a software patent is rarely necessary or deserved. Current system allows those who exploit the system to collect royalties for trivial inventions in many cases from people who were not even aware they were infinging a patent.
If the constitution requires that things under patent protection be released to the public domian for certain purposes at a future date, wouldn't it be unconstitutional to protect a copyright to the point where no one could use it for any purpose when it was finally released?
For example, when Windows 3.1 is finally released to the public domain, will there be a computer on earth that can run it? If not, doesn't that mean that our current copyright laws are unconstitutional and need to be changed?
CNNMoney.com Comment Policy: CNNMoney.com encourages you to add a comment to this discussion. You may not post any unlawful, threatening, libelous, defamatory, obscene, pornographic or other material that would violate the law. Please note that CNNMoney.com makes reasonable efforts to review all comments prior to posting and CNNMoney.com may edit comments for clarity or to keep out questionable or off-topic material. All comments should be relevant to the post and remain respectful of other authors and commenters. By submitting your comment, you hereby give CNNMoney.com the right, but not the obligation, to post, air, edit, exhibit, telecast, cablecast, webcast, re-use, publish, reproduce, use, license, print, distribute or otherwise use your comment(s) and accompanying personal identifying information via all forms of media now known or hereafter devised, worldwide, in perpetuity. CNNMoney.com Privacy Statement.