Skip to content
 

Leahy Versus Albedoman and the Moneygoround, Part One

Edward Wyatt reports:

Now the Obama administration is cracking down on what many call patent trolls, shell companies that exist merely for the purpose of asserting that they should be paid . . . “The United States patent system is vital for our economic growth, job creation, and technological advance,” [Senator] Leahy said in a statement. “Unfortunately, misuse of low-quality patents through patent trolling has tarnished the system’s image.”

There is some opposition:

But some big software companies, including Microsoft, expressed dismay at some of the proposals, saying they could themselves stifle innovation.

Microsoft . . . patent trolls . . . hmmm, where have we heard this connection before?

There is also some support for the bill:

“These guys are terrorists,” said John Boswell, chief legal officer for SAS, a business software and services company, said at a panel discussion on Tuesday. SAS was cited in the White House report as an example of a company that has spent millions to defend itself against what it believes are frivolous lawsuits.

20 Comments

  1. Ian Fellows says:

    Best to just get rid of software patents.

    Any software is an instance of an algorithm, and algorithms are math. You can’t patent Math.

    • It’s not just software patents that are problematic, but also business method patents, like “one click ordering.”

      Even without patents, code is automatically copyrighted. Oracle was suing Google most recently for infringement of the copyright and patents on the Java API:

      http://en.wikipedia.org/wiki/Oracle_v._Google

      • Mark Palko says:

        Going back to the This American Life story:

        “A group called Innovatio has been going around suing coffee shops and grocery stores for violating its patent on providing Wi-Fi in a public space. They say they own that idea. And if your coffee shop is providing Wi-Fi, you owe them money. They say they’ve made a strategic and business judgment not go after people using Wi-Fi in their own homes, quote, “at this stage.” Another company called Boadin Technologies, LLC, has sued a bunch of media organizations, saying that one of its patents is infringed every time a word or phrase on your computer autocompletes.”

    • phayes says:

      You can’t patent math but you can patent the use(s) of math. It is economic quackery though. Seeing what was in that patent trolls link reminded me of the story of how the FFT was protected from the patent medicine:

      “In fact, Cooley says, the Cooley-Tukey algorithm could well have been known as the Sande-Tukey algorithm were it not for the “accident” that led to the publication of the now-famous 1965 paper. As he recounts it, the paper he co-authored with Tukey came to be written mainly because a mathematically inclined patent attorney happened to attend the seminar in which Cooley described the algorithm. The lawyer, Frank Thomas, “saw that it was a patentable idea,” Cooley explains, “and the policy of IBM and their lawyers was to be sure that nobody bottled up software and algorithms by getting patents on them.” A decision was quickly reached to put the fast Fourier transform in the public domain, and that meant, in part, publishing a paper.” http://web.archive.org/web/20060426151241/http://www.siam.org/siamnews/mtc/mtc593.htm

  2. Mark Palko says:

    Nathan Myhrvold’s company features prominently in this week’s TAL and if anything comes off worse than it did in the first installment.

    http://www.thisamericanlife.org/radio-archives/episode/441/when-patents-attack

    And a quick word of warning, if you’re thinking about adding a podcast, someone has a patent on that one.

    • Andrew says:

      It’s impressive how that guy managed to snow so many reporters for so many years. I guess part of it is that people want to be nice to a billionaire; after all, he might spill some money on to them. Also there’s the mystique of the lone genius.

    • Bill Jefferys says:

      According to the TAL episode, they aren’t going after nonprofit podcasts.

      Apparently the guy that they are going after is getting money for his podcasts, even if it’s not a lot.

      (Doesn’t mean they won’t change their minds of course).

      But the whole thing stinks to high heaven. I agree, S/W patents have to go. (In the old days the “podcast” idea and many of these other “patents” would never have gotten patents because there was no working model…or because they were obvious…apparently there were over 5000 people trying to implement podcasts (I think) at the same time as this guy in New Hampshire…if you have 5000 people doing the same thing, it must be pretty obvious.

      Same thing for the business method patents, as Bob Carpenter says.

      Where I live (Vermont) a law was recently passed that will help…if a patent troll comes after a Vermont company, the state will take the side of the Vermont company and, I think, defend them, apparently at state expense. But a national solution is required.

  3. Fernando says:

    What about academics that patent specific software techniques? Should academics be in the business of patenting estimators, measuring instruments, and procedures?

    At this point I have not made up my mind one way or the other but would like to hear readers’ opinions.

    PS I wish I’d patented OLS. At one cent per regression, and given the pervasiveness of specification searches in the social sciences, I would by now be able to hire Bill Gates as my butler…

    • Phil says:

      I’m wrestling with this issue a bit myself. I work at a government research lab, where I work on methods for analyzing “smart meter” data from large buildings to do things like automatically detect malfunctions in the cooling system, and quantify the effectiveness of energy savings retrofits. There’s actually a pretty big industry around this stuff, believe it or not, but most of the methods were developed back when the only energy data were from monthly bills. Having 15-minute or 1-hour data has the potential to change everything, and obviously will do so (and is doing so already).

      In December I started working on a new modeling approach, implemented using Stan (it’s great!), and within less than a month I had something that I think is a huge advance on current practice. Unfortunately I don’t have a project to support that work, but I continued working on my own time (yes, really) for a couple more months and then gave a talk about it. People from several companies watched the webcast of my talk, and contacted me later. Some of them agree that my new approach might be really useful (and thus economically valuable). But in three months of applying for grants I have not been able to get support to work on this more, although I can squeeze a little bit of work on it into a few upcoming projects. Even worse, I am having trouble keeping funded to work on anything at all — to put it bluntly, I could lose my job!

      Yesterday I spoke with someone from our Technology Transfer department to discuss my options (which include things like: quit and start a company of my own; quit and join one of the companies that is interested in this stuff; find a company (or group of companies) to pay me to work on it as part of my job, with the company ending up with all of the intellectual property rights; find a company or companies to pay me to work on it as part of my job, with the company getting a non-exclusive license to the technology (i.e. the software); drop the whole thing; etc.

      One thing I found is that I may have already shut off some options by giving that talk: I have “publicly disclosed” some of my work, which means I can’t patent it, if it was ever patentable anyway. I mostly feel that as a government employee, work that I do should be freely available, so I’m not really bothered by this. But I only _mostly_ feel that way, not completely.

      Similarly, it _mostly_ seems ridiculous that it’s possible to patent an algorithm. Should someone be able to patent OLS? Newton’s method? Come on. But…but…I have developed something valuable, isn’t there _any_ sense in which it’s “mine”? Sure, I can copyright the _code_, but it’s only a couple of hundred lines of Stan; given a good understanding of the algorithm, a skilled Stan user can write it in a week.

      I think we may have to accept that certain things, even very valuable things, simply can’t be monetized, or at least not for a fraction of what they’re worth. If you come up with a really funny joke, other people will repeat it, maybe millions of people, perhaps for years, and you have no way to get compensation for that. Meanwhile, other people who create things that provide much less value to many fewer people are able to get paid for it.

      I’d say all of this falls into the category of “well, whaddyagonnado?”

      • Fernando says:

        Phil:

        I think you should patent it, not as software but as an apparatus for measuring something.

        The reason is that if you don’t maybe some company will, and, before you know it, you are paying them to use your creation. Later you can license it using Creative Commons, or for $.

        If you’d like send me your best grant proposal and, if I think it has potential, I may be able to raise some money for you.

        I am in the process of setting up my own company, and I too am planning to file a patent. Might as well kill two flies with one swat.

      • Fernando says:

        BTW my understanding is you have one year from public disclosure to file a patent, so you might still be in time. Check with a lawyer.

        • Ian Fellows says:

          Phil, I sympathize completely. You can still build a company off of ideas that others can copy, and almost every company is built this way (Facebook, exxon, your local flower shop). If you can build a turnkey solution and build strong service relationships with your clients, then that is the important part (and the part that takes the most work). If a competitor can do it better, then that is the (cruel) free market working.

          Fernando, Creative Commons generally refers to copyright, though it looks like they do have a draft in the works for a Patent License http://wiki.creativecommons.org/CC_Public_Patent_License

  4. nottrampis says:

    okay I am starting up the Andrew Giles fan club down here.

    I might call you the Castle of Statistics!!

    please keep it UP!

    • Anonymous says:

      This guy has all the signs of being a robot link harvester, misspells your name, has nothing but vacuous “good job” type posts, and links to a blog that looks like a link hub whose only purpose is to bring in links from important sites (to raise page rank) and then maybe have outgoing links to sites of commercial interest to his clients to boost search engine traffic.

      • Bill Jefferys says:

        Yeah, delete the original comment and leave the comments on the comments…or at least delete the fake link.

  5. nottrampis says:

    I hope you blokes are not talking about me.

    I put up most but not all of Andrew’s articles as they are very good for people who are not stat people to read and understand.

    They are what I believe to be the highlight of the week.

    you mightn’t believe that but I do.

    It is something I do for people who do not have the time to read everythig on a lot of topics especially this one.

    Think of it as a poor man’s Mark Thoma!

    to my mind Andrew produces quality and quantity.

    I haven’t misspelt his name either!