What should happen with OOXML/ODF after the i4i patent?

By Rick Jelliffe
March 23, 2010 | Comments: 5

Alex Brown has a recently blogged on Document Format Standards and Patents. Some points of interest: Alex expects the customXML feature should be taken out of the new OOXML Strict (the dialect of OOXML which represent what National Bodies actually want Microsoft and implementers be transitioning quickly to, from the older ECMA/Office 2007/Transitional dialect); Alex also suggests it would be a good idea to clarify ODF's exposure too.

The i4i patent seems, as I have blogged before, completely bogus. But Microsoft's legal team conducted their challenges on legalities rather than technical grounds. So we are stuck with the patent standing, until the USPTO grows a brain.

Also of interest to standards wonks is Alex's description of the ISO/IEC IPR regime.

The US patent system is a continual threat to innovation and co-operative, market-enhancing activities such as standards development: it is not an abstract thing but a real complication in the standards world. To my mind, the US system is a form of State Capitalism, discredited , where organs of the state pick winners in an incredibly arbitrary manner. Utterly counterproductive.


You might also be interested in:

5 Comments

Rick, you can't blame i4i on the USPTO. The patent was upheld in court. It was found to be valid. Remember, the USPTO doesn't make the law. Congress does. USPTO is merely the agency that carries out that part of the law. And they don't judge patent infringement either. The courts do. So in this case, I think the USPTO did an exemplary job, since their issuance of the i4i patent was subjected to extraordinary scrutiny in this case by a well-heeled litigant, and was ultimately found to be valid.

Also, for a form of US "State Capitalism" it is ironic that the patent holder is a Canadian. Note that something like 40% of all US patents are of non-US origin, largely from Japan, UK, Germany and France. That doesn't look like "discredited" to me. In fact it looks to be globally rather popular.

Rob: Actually, I don't think the courts did look at technical details in any detail: SAMBA's Jeremy Allison's recent talk on FOSS and IP issues mentions that attacks based on prior art are rarely successful (requiring more expertise than a jury reasonably possesses) so defences are more usually based on claims that the patent was not violated.

IIRC the actual grounds of the first appeal are pretty clear if you read them.

And as I have given multiple examples of what looks like prior art to me, as have many others surprised old-timers, your claims of "extraordinary scrutiny" are laughable. That was one of my frustrations.

That the patent holder is Canadian is irrelevant; a typical deflection. The point is that this is the USPTO that issued the patent. The US law is driving the problems. Yes, this may be because of the US government, but it is they who are in turn being driven by well-heeled patent-pusher corporations, such as your employer, the largest patenter.

The more that the US relies on patents, the more that it encourages its own economic stagnation.

Rick, this is no different than any other trial. If the topics are beyond the knowledge of a typical juror, then both sides will bring in their expert witnesses, and try to persuade the jury via those means. You see the same thing in medical malpractice, accounting fraud, etc. Even criminal trials will bring in experts to interpret physical evidence.

Studies have shown that patent litigation cases are decided in favor of the patentee 58% of time. The remaining 42% seems a bit large for you to call dismiss it "rarely successful".

In any case, you need to distinguish an attack on novelty versus an attack on obviousness -- two different criteria in the US. To attack novelty you need a _single_ prior art reference that teaches all features of the claim. And you need to do this for every independent claim.

If you start picking multiple prior art references and putting them together, then you are going after obviousness, not novelty. At this point the question is whether the combination of prior art elements is obvious from the perspective of a "personal having ordinary skill in the art" at the time the invention was made. Presumably you have more than ordinary skill in this field. So saying that this is obvious to you and unnamed "old timers" is not relevant. And then there is the common hindsight bias. Many things are obvious when looked at 1y years later. The question is what was obvious back then, when the invention was first made.

I'm not saying that there are not junk patents out there. But in this case, Microsoft took their best shot and lost. By all appearances they knowingly appropriated an invention from a small Canadian company and got caught. They were sued and lost. They appealed and lost. Microsoft does not skimp on legal representation. They were well-represented. I think this case is a poster child for how the patent system is supposed to work. If this was not involving a standard that Microsoft paid you to promote we wouldn't even be talking about it.

To suggest that a feature should be taken out of an approved international standard because a potential supplier of software to support the standard in unable to support a key feature is absolutely shameful.
To suggest that a feature should be taken out of an approved international standard because my paycheck comes from a Company that stole technology to implement the feature and was caught in the act is 'brown nosing'.
To suggest that the patent is bogus is to expose your ignorance of patents, the patent process and the complexities of patent enforcement.
To suggest that a patent is bogus because my paycheck comes from the Company that was caught infringing the patent is 'earning my keep'.

Dave: To suggest my paycheck comes from Microsoft is to show either you are an idiot who does not do any research or a troll. Grow a brain, or at least some manners.

In fact, SC34 (and its predecessor) has had a practice for at least the last 15 years (and perhaps its 30 year history) that it would not standardize encumbered and royalty-bearing technologies: we have only been interested in things that are clearly implementable by open source developers. The i4i patent throws a cloud over this.

My comments that the patent is bogus is not a statement about its legal status, but about its technical content. And I have documented several other occurrences by other parties, including Microsoft, of what I would consider mis-appropriation of previously standardized technology. Are you saying that ISO HyTime's clink is not relevant? Or have you absolutely no idea of the technical argument?

News Topics

Recommended for You

Got a Question?