Yet Another Ridiculous Software Patent

by on June 6, 2006 · 10 comments

I have a hard time writing about stories like this one without getting shrill. ZDNet is reporting that Net2Phone is suing Skype for violating its “point-to-point internet protocol” patent.

What does it cover?

A point-to-point Internet protocol exchanges Internet Protocol (IP) addresses between processing units to establish a point-to-point communication link between the processing units through the Internet. A first point-to-point Internet protocol includes the steps of (a) storing in a database a respective IP address of a set of processing units that have an on-line status with respect to the Internet; (b) transmitting a query from a first processing unit to a connection server to determine the on-line status of a second processing unit; and (c) retrieving the IP address of the second unit from the database using the connection server, in response to the determination of a positive on-line status of the second processing unit, for establishing a point-to-point communication link between the first and second processing units through the Internet. A second point-to-point Internet protocol includes the steps of (a) transmitting an E-mail signal, including a first IP address, from a first processing unit; (b) processing the E-mail signal through the Internet to deliver the E-mail signal to a second processing unit; and (c) transmitting a second IP address to the first processing unit for establishing a point-to-point communication link between the first and second processing units through the Internet.

Where to begin? This describes an absolutely pedestrian networking protocol. There’s nothing remotely novel or non-obvious about two computers communicating directly with each other without using a server. I don’t even understand what the “invention” is supposed to be. If you asked a random CS major how to implement a peer-to-peer network application, he’d probably come up with a description like this in about 10 minutes.

Some people seem to think that these kinds of bad software patents are anomalies–that there are good ones as well, and that we ought not to throw the baby out with the bath water. Yet I’ve seen news accounts of more than a dozen examples of bogus patent suits in recent months, but I have yet to see an example of a legitimate software patent. There’s an awful lot of bath water here, and I’m having trouble seeing the baby.

  • Ted

    There is no baby. We need to pull the plug.

  • matt

    i have been a computer programmer for nearly 20 years. i’ve written more code than i care to even try to recollect. yet in all that time there is nothing that i’ve written that i would consider patentable. to me software patents are as ridiculous as saying a certain arrangement of lego blocks should be patentable.

    the only people that profit from software patents are patent attorneys.

  • Brian Moore

    Well, there’s one way to find out. Some kind of review of all existing software patents is obviously in order. Ones like this obviously need to go — you can’t patent that… you just can’t. Let’s get rid of all the crazy, asinine patents like this and see if we have anything left. I’m willing to grant there might be some…

    Oh yeah, and revoke the bar licenses of lawyers who used any of those patents as grounds for a case.

    Or maybe just summarily execute them. You can’t go wrong with that.

  • Ted

    There is no baby. We need to pull the plug.

  • matt

    i have been a computer programmer for nearly 20 years. i’ve written more code than i care to even try to recollect. yet in all that time there is nothing that i’ve written that i would consider patentable. to me software patents are as ridiculous as saying a certain arrangement of lego blocks should be patentable.

    the only people that profit from software patents are patent attorneys.

  • Brian Moore

    Well, there’s one way to find out. Some kind of review of all existing software patents is obviously in order. Ones like this obviously need to go — you can’t patent that… you just can’t. Let’s get rid of all the crazy, asinine patents like this and see if we have anything left. I’m willing to grant there might be some…

    Oh yeah, and revoke the bar licenses of lawyers who used any of those patents as grounds for a case.

    Or maybe just summarily execute them. You can’t go wrong with that.

  • Ted

    In arguments over this issue in the past, software-patent proponents have raised examples such as the RSA alogorithm, which was a nice bit of non-obvious work which is very easy to copy once it is disclosed (I believe the shortest implementations of RSA fit within 20 or so lines of code).

    Of course, I remember my college mathematics professors mocking the RSA patent back in the early 90s — not because the algorithm is bad or obvious — but because it is, after all, just math. The RSA patent was essentially a patent on raw mathematics. Imagine if the first guy to figure out triangulation had been able to patent that (or at least its application to measuring stuff at a distance).

    All math becomes, at some point, useful in some application. If we allow patents on particular applications of mathematics (such as understanding the difficulty of factoring large primes and its applicability to assymetric encryption, i.e. RSA), then ALL of MATH becomes patentable.

    Yet we have traditionally considered math to be facts of nature, and therefore unpatentable.

    The contradictions continue. Software patents are an absurdity, unless you want to open up all ideas and facts as privately ownable.

  • http://www.techliberation.com/ Tim

    In addition, the RSA patent was granted more than 20 years ago. If that’s the most recent example they can come up with, good software patents must be truly few and far between.

  • Ted

    In arguments over this issue in the past, software-patent proponents have raised examples such as the RSA alogorithm, which was a nice bit of non-obvious work which is very easy to copy once it is disclosed (I believe the shortest implementations of RSA fit within 20 or so lines of code).

    Of course, I remember my college mathematics professors mocking the RSA patent back in the early 90s — not because the algorithm is bad or obvious — but because it is, after all, just math. The RSA patent was essentially a patent on raw mathematics. Imagine if the first guy to figure out triangulation had been able to patent that (or at least its application to measuring stuff at a distance).

    All math becomes, at some point, useful in some application. If we allow patents on particular applications of mathematics (such as understanding the difficulty of factoring large primes and its applicability to assymetric encryption, i.e. RSA), then ALL of MATH becomes patentable.

    Yet we have traditionally considered math to be facts of nature, and therefore unpatentable.

    The contradictions continue. Software patents are an absurdity, unless you want to open up all ideas and facts as privately ownable.

  • http://www.techliberation.com/ Tim

    In addition, the RSA patent was granted more than 20 years ago. If that’s the most recent example they can come up with, good software patents must be truly few and far between.

Previous post:

Next post: