First, a lot of people have probably heard that Microsoft lost an appeal in a $290M case against Canadian firm i4i.

A federal appeals court on Tuesday affirmed a $290 million patent infringement judgment against Microsoft Corp. (MSFT) and reinstated an injunction that bars the company from selling current versions of its flagship Word software.

The U.S. Court of Appeals for the Federal Circuit said the injunction will go into effect Jan. 11.

After the jury verdict, U.S. District Court Judge Leonard Davis issued a permanent injunction that barred Microsoft from selling Word 2003 and Word 2007, which use a technology called “Custom XML” that is used to classify documents for retrieval by computers. Davis also ordered Microsoft to pay more than $290 million in penalties.

Barred from selling one of their flagship products? Wow. I managed to glean some idea about custom XML is, but I still can’t quite figure out what it does in the sense of providing any kind of useful or significant functionality to end users. It seems like one of those things that gets argued about in standards meetings, mostly by people who are more confident of their ability to keep up with developments in such intellectually inbred communities than in the wider world of real-life computing, and nobody cares until it hits the patent system. An interesting sidelight is that the original judgement seems due in part the result of a conflict between one of Microsoft’s lawyers and the East Texas judge. That’s a fight where it’s really hard to root for either side, but the practical consequence is that it might give Microsoft grounds for further appeal. By the way, how does a company from Toronto get to sue one from Redmond in a Texas court? That’s not only the wrong state but the wrong country.

So the news about Microsoft and i4i might be a mixed bag, but the other piece of news is quite a bit better. The US Board of Patent Appeals and Interferences (the what?) issued a ruling about the patentability of machines using mathematical algorithms. The ruling is from August, but it just became “precedential” this past Monday. I have no idea what that means, except maybe that those who find XML standards too accessible should consider a career in patent law. Anyway, here’s the good part.

The BPAI’s test for a claimed machine (or article of manufacture) involving a mathematical algorithm asks two questions. If the a claim fails either part of the two-prong inquiry, then the claim is unpatentable as not directed to patent eligible subject matter.

(1) Is the claim limited to a tangible practical application, in which the mathematical algorithm is applied, that results in a real-world use (e.g., “not a mere field-of-use label having no significance”)?

(2) Is the claim limited so as to not encompass substantially all practical applications of the mathematical algorithm either “in all fields” of use of the algorithm or even in “only one field”

In a nutshell, this reaffirms the principle that only the application of an algorithm – not the algorithm itself – can be patentable. This has supposedly been true all along, but many patents on applications have effectively become patents on algorithms because the applications have been defined so vaguely or broadly that no other application can escape claims of infringement. This ruling says that failure to limit the scope of such a patent properly might not only cause a specific infringement case to be dismissed but might also cause the patent in its entirety to be invalidated. This could fundamentally change the way people pursue such patents. Until now the incentive has been to make claims as broad and vague as possible, because there was no risk in doing so. Now, though, there’s a risk that over-reaching might destroy the patent’s value entirely, so there’s an incentive to be more specific. Only time will tell whether it actually works out that way, but it’s a good sign.

Now I’m kind of curious about which of my own (five) patents would survive this test. I doubt that they’ll ever be tested, and I won’t provoke my former employers’ legal departments by speculating, but it’s an interesting question.