AT&T’s position is that its speech recognition software is protected by US patent, on the grounds that it constitutes a “component of a patented invention.” Microsoft was apparently granted license to utilize that software in the US, but AT&T contends that the duplication of that software outside of US boundaries, with the intent to sell the duplicates overseas, is a violation of that license. That view was upheld by the Federal Court of Appeals, and Microsoft is appealing that decision… Olson built a case against AT&T’s position: For foreign replication to work, a golden disk is shipped abroad to the replication service, containing the master of the Vista operating system that includes AT&T’s drivers. It’s not software at that point, Olson says, because no one can execute it. When it’s installed onto a hard drive, then it becomes software, and it’s the end customer who does that. If you accept that argument, Olson’s case goes on, then Microsoft did not copy AT&T’s software. And since the physical components – the hard drive, processor, and optical drive – necessary for the copying process to happen may or may not have been supplied by manufacturers in the US, then US courts may not have jurisdiction anyway.
This is confused on a number of levels, but the most fundamental error—and, frankly, the one that causes people to be a lot more sympathetic to software patents than they ought to be—is that Microsoft is accused of copying AT&T’s software. AT&T has never claimed that Microsoft’s software is in any way derived from AT&T software. They certainly don’t claim that Microsoft bundled AT&T drivers with Windows. If that were the case, AT&T would have an open-and-shut case under copyright law.
The claim, rather, is that some of the functionality of Microsoft’s software matches the claims in AT&T’s patents. This does not mean that Microsoft based its software on AT&T’s designs. It’s quite common for one company to patent an idea like “one click shopping” or “wireless email” and then for another company to have the misfortune of inadvertently developing a product that falls within the scope of the patent.
Although journalists certainly should do their homework, I don’t think it’s entirely fair to blame Beta News for the oversight. You frequently see stories on patent cases in which it says that Company A is accused of “stealing company B’s technology,” as thought they broke into company A’s offices and stole their blueprints. But in fact, in most cases, the only “technology” they share is a vague description in a patent application, such as “one click shopping” or “wireless email.” In many cases, the infringing company is not even aware of the original company’s products or patents until they’ve already released it and get a nasty-gram from the company’s patent lawyers.
I think that if journalists did a better job of explaining that, there would be a lot less public sympathy for the current patent system, and for patent trolls like NTP. People are naturally sympathetic to the plucky inventor who invents something and then has a big company copy his design without paying something for it. But in reality, that seems to happen pretty rarely. Far more common are inventors like Donald Stout who invent something, patent it, fail in the marketplace, and then start suing anyone who happens to have a similar product, regardless of whether the technology was developed independently or not. I think the general public would be far less sympathetic to patent trolls if that were clearly understood, and journalists writing about patent cases should do a better job of conveying that to the reader.
If you’ll forgive the shameless self-promotion, I think my write-up for Ars did a better job of explaining the legal issues in the case.