Blog post

Microsoft’s Word Patent Woes May Be Broadly Shared

By Brian Prentice | August 12, 2009 | 10 Comments

Few things must be more emotionally paradoxical for the anti-patentistas of the world than to see Microsoft get clobbered on a patent infringement action. Do you rejoice because one of the great advocates, and enforcers, of software patents has gotten a strong dose of their own medicine? Or do you fret at yet another example of a patent system gone mad?

Oh coward conscience, how dost thou afflict me!

And boy, has Microsoft gotten clobbered! In case you haven’t seen the news, the infamous U.S. District Court of East Texas has issued an injunction against Microsoft selling or importing any copy of Microsoft Word that can open .XML, .DOCX, or DOCM files containing custom XML. That is on top of damages, determined in May, of $240,000,000 ($40,000,000 of which is for willful infringement) plus interest and per diem penalties, all for infringing on i4i’s patent relating to a system for manipulating the architecture and content of a document separately.

My normal reaction to these rulings is ho-hum. Given the mountain of poor quality software patents, and the well-earned reputation of the East Texas district court as the venue of choice for patent cases, it’s usually a matter of time before rulings like this are overturned on appeal. Indeed, the first thing that sprung into my mind when I started to look over the patent was formatting tags in WordPerfect circa 1982.

But the more I read through the patent claim the less confident I was with my initial reaction. In fact, I think this one might actually have some legs. Keep in mind is that this claim was filed back in 1994. The claim considers the existing state of the art at that time – formats like TROFF, RFT and SGML – before asserting that:

…in sharp contrast to the prior art the present invention is based on the practice of separating encoding conventions from the content of a document. The invention does not use embedded metacoding to differentiate the content of the document, but rather, the metacodes of the document are separated from the content and held in distinct storage in a structure called a metacode map, whereas document content is held in a mapped content area. Raw content is an extreme example of mapped content wherein the latter is totally unstructured and has no embedded metacodes in the data stream.

One thing seems clear to me – this is not a typical rubbish software patent that earns its filer a 20 year monopoly on the dead obvious. Fifteen years ago this would seem to me to have been an innovative idea. That is, of course, assuming that yet more detailed forensic software engineering examination finds a problem with the novelty of this claim.

But, if the validity of the patent is upheld then the immediate question is whether this will also impact ODF. If so, then this turns out to be a significantly more important issue and one which will crystallize the fury of the anti-patentistas. No longer will this be the source of some Schadenfreude at Microsoft’s expense. This will be seen as yet another attack on open standards and open software.

There is another interpretation that I fear will be lost in the noise. That is some introspective consideration of whether there is actually a rampant disregard in the software industry for other’s property rights. If it is not just .docx but also ODF that infringes then that could be seen as some pretty significant oversight, potentially even arrogance, on the part of Microsoft, Sun Microsystems and OASIS. And given that Microsoft was aware of i4i’s patents, one wonders why they didn’t just buy them (at a significantly reduced price then what they might end up paying now) and then target ODF for license agreements like they’re doing with their patent infringement claims against Linux.

As with all these patent issues there’s a lot more water to pass under the bridge before we can make a final judgement. But one thing seems pretty clear to me. Whether you agree with software patents or not the fact is that they exist. And that means we all need to understand the associated opportunities and risks and manage them accordingly.

The Gartner Blog Network provides an opportunity for Gartner analysts to test ideas and move research forward. Because the content posted by Gartner analysts on this site does not undergo our standard editorial review, all comments or opinions expressed hereunder are those of the individual contributors and do not represent the views of Gartner, Inc. or its management.

Comments are closed


  • Marco says:

    The suit and verdict are ridiculous. I just read the patent and cannot believe the verdict was as it was. My more detailed analysis:

  • FredericBaud says:

    Just another piece of evidence that the patent system is a massive destruction of wealth. For reallocating a rent between 2 agents, the system is destroying value for the entire community in the form of long hours of pointless haggling.

    Whatever is described in this patent, how can we imagine that we had to reward someone thinking of connecting XML and Doc documents, in whatever details, and if we hadn’t done so, we would have continued to live in a medieval world.

    And concerning disregards in others intellectual property rights, this is the only thing I think will come as a positive evidence that these rights do not serve any purpose.

  • F says:

    “but rather, the metacodes of the document are separated from the content and held in distinct storage in a structure called a metacode map”

    TeX has had “styles” since its inception, and these are nothing more than “metacode maps”. I have a TeX manual that is probably older than the author of this piece.

    Fifteen years ago, this would only have been an “innovative idea” only if you ignored one of the most popular text formatters at that point in time.

  • Tim Acheson says:

    I am still digesting the facts. I will continue to update this web page as I progress. When finished, this page will provide a simple analysis of the facts, and offer an informed prediction about the likely outcome of the appeal:-

    I can offer technical expertise and experience of the type of computer operation in question. I am consulting with contacts who have specialist knowledge in the other key areas, patent law and patent interpretation — patent attorneys and editors on publications about scientific patents.

    At this stage the story has been widely-reported in the mass media and online, and commented upon by the general public. There is a lack of informed analysis.

  • FredericBaud says:


    Interesting (unfortunately with apparently a bit of tragedy) to see at the same time one of your colleague pondering on the effects of a court decision back in Dec 2008 about infringement by Emptoris of 2 patents owned by Ariba. I don’t know the details, and don’t know if Emptoris had what it would take to develop into a contributing company, but extracting $7M of Flesh to feed a company like Ariba did certainly not help the cause for innovation.

  • Michael Wasserman says:

    I read the patent. It seems to boil down to this: markup is inline, that is, it interrupts the data stream it describes; inline markup can be inconvenient; to take the markup out of the data stream one can create a separate map indexed to points in the data stream being described.

    That seems to me to be about the most obvious and trivial solution in the world. Using indices to reference portions of a data stream was described in the literature at least as early as 1973. Communications of the ACM, 16:622 (

    Also, I cannot see a threat to the Open Document Format itself. ODF is XML. The patent is for a technique that is basically not XML. (Although I suppose the ODF method of recording editing history might be conceived of as infringing.) It may well be the case though that OpenOffice–and many, many more programs–use an index map technique that is covered by the patent.

  • Tom Anderson says:

    In the claim, they walked through how a knowledgeable person would manually read a file that is encoded with SGML. Then they claimed that their patent covers every possible automatic approach to doing that? It’s so ridiculous. Patents don’t protect mental acts. Patents that claim to cover automatically through a computer that people already do should not be defended as “state of the art”. Patents are supposed to cover new and inventive things, not automatization of things that are done already.

    I might as well patent every process that people today do mentally that is not yet done by a computer.

  • makomk says:

    Nope, zero threat to ODF from this particular patent. ODF itself does its formatting in a very similar fashion to HTML and the other SGML-based document formats, and that’s pretty clearly not covered by the patent.

    In fact, core OOXML isn’t at threat either (though it’s more at risk from other patents than ODF, since it does its formatting in an oddball fashion that isn’t HTML/SGML-like). The infringing feature is something called “Custom XML”, where you can stick an arbitrary XML file in the Word file and integrate text from it into the document proper.

    The key elements of the patent are that the document is split into two parts – one containing just text, the other containing formatting – and that the formatting part references the bits of text to which it applies via their locations. Custom XML has these key elements (though the location is actually implemented as a full XML path).

    ODF doesn’t. While they are adding their own answer to Custom XML in the next version, it doesn’t work the same way. The key idea of their RDF-based solution seems to be that you insert the text inline into the document XML itself like normal text, and mark it up with metadata that allows it to be extracted later. It may additionally allow the incorporation of things from seperate RDF files into the document (hard to tell from the docs), but it uses unique xml:id attributes added to the XML elements to glue everything together.

    Plus, even if ODF did infringe (which it doesn’t), it wouldn’t necessarily be at risk of being sued. i4i isn’t just a patent troll – they’re angry with Microsoft because this new Microsoft Word feature was killing off the market for their own core product based on this patent (and Microsoft was entirely aware that it would, as demonstrated by internal memos). Microsoft has a long history of deliberately pulling stunts like this – it’s why they have a bad reputation in the IT industry.

  • Al Macintyre says:

    The i4i patent is an idea that instead of storing documents as one large file that combine both content (e.g., text) and formatting intermixed, they instead separate the content from the formatting markup into separate files, allowing content to be changed independently of style. However, the exact structure of the separation is not detailed in the patent, just the concept of separation.

    This idea has been embodied in IBM code, that some people might call legacy code, for many decades, on IBM systems known as S/38, AS/400, iSeries, using a concept IBM calls “external definitions” of files and other computer objects. IBM’s prior art could invalidate the i4i patent and Microsoft patents.

    With “external definitions”, the layout of a file or table or other IBM object type, is stored separately from the actual content of the file. The layout says how big the thing is, a fixed size, or if it has variable size, what can go in each of the field columns such as text, ISO compliant date, floating point, integers with how many decimal places, urls, images, whatever.

    Prior to IBM inventing “external definitions”, IBM software used “internal definitions” where the layout was defined by the various programs that accessed the files, and these definitions could be contradictory, not complying with relational data base rules. IBM systems still support this, so legacy code can be ported to the newer systems.

    With “internal definitions”, the content was separate from the layout, and that concept goes back to the dawn of IBM computers, 50 years or more ago, before even when Microsoft was founded.

  • Michael Kay says:

    This case could be a critical nail in the coffin of software patents. It illustrates everything that’s bad about them: inhibiting innovation, inhibiting adoption of open standards, rewarding people who had an idea but didn’t know how to make a success of it in the market, putting the user community in the cross-fire, trying to work out whether two pretty abstract software designs are in some sense technically equivalent when there is no theoretical basis for determining the equivalence of two designs. More than anything else, there’s a good chance it will be a wake-up call for Microsoft, making them realise that software patents aren’t just a weapon they can use against little guys, but a weapon that little guys can use against them. If that makes Microsoft see the light and realise that software patents are 100% evil, it will be a victory for us all.