Friday, December 30, 2011

Preliminary ITC ruling on Microsoft's complaint against Motorola has obvious shortcomings

On December 20, 2011, Administrative Law Judge Theodore Essex announced his initial determination on Microsoft's ITC complaint against Motorola Mobility, finding the defendant's Android devices in infringement of four claims of a Microsoft patent. Until the ALJ published the redacted version of his detailed ruling (233 pages long) late yesterday, it was unknown how he arrived at this conclusion and why he didn't find a violation in connection with six other patents asserted by Microsoft at that stage.

I already got the feeling that Motorola wasn't comfortable with the defensibility of large parts of the decision when I saw that its lawyers opposed an extension of time for Microsoft to file a petition for review. Denying an extension during the Holiday Season, when pretty much everyone (even bitter rivals like Apple and Samsung, and, respectively, Apple and HTC) can agree on such extensions, is very unusual. Lawyers typically treat each other as colleagues, and try to be reasonably helpful and cooperative. I updated the related post today after I saw that the ITC granted Microsoft an extension by two business days (short of the five days that had been requested).

Having read the ALJ's initial determination, I'm convinced that this one will get a rough ride at the review stage or in the event of a subsequent appeal to the Federal Circuit. I obviously haven't been able to analyze all of the technical detail discussed in the initial determination, but that's not even necessary to see that the document has obvious shortcomings and even contradictions.

Parts of it look like an overworked ALJ mostly wanted to penalize Microsoft for having declined his invitation to narrow the case to a greater extent than it did when it dropped two of the original nine patents. It's anything but well-reasoned. Key parts of the decision are just based on Microsoft not having gone into enough detail on claims that Motorola didn't even contest in the first place. The ALJ disliked the style of one or more of Microsoft's witnesses, but even in a context in which a Motorola fact witness apparently confirmed what Microsoft claimed, he didn't want to attach much weight to that fact. The thing is not even well-written, and there are more typos than I've ever seen in a document of this kind. It's absolutely nothing for the ITC as an institution to be proud of in any way, and it lacks all of the characteristics of documents that hold out well when they are reviewed or appealed.

This is just my personal opinion. I haven't had the chance to discuss this document with anyone else, but I analyze so many documents of this kind that I'm confident that others will agree.

I'll summarize the key findings and then I'll give some examples of the aforementioned striking deficiencies.

The ALJ's conclusions of law

These are the ALJ's conclusions of law (quoted from section VIII of the initial determination):

  1. The Commission has personal jurisdiction over the parties, and subject-matter jurisdiction over the accused products.

  2. The importation or sale requirement of section 337 is satisfied.

  3. The accused products do not literally infringe the asserted claims of the '054 Patent, the '352 Patent, the '133 patent, the '910 Patent, the '376 Patent and the '762 Patent.

  4. The accused products do literally infringe the asserted claims of the '566 Patent.

  5. Respondents do not induce infringement of any of the asserted claims of the asserted patents.

  6. The asserted claims of the '054 Patent, the '566 Patent, the '352 Patent, the '133 Patent, the '910 Patent, the '762 Patent, and the[space missing, one of countless typos in the original text]'376 Patent are not invalid under 35 U.S.C. § 102 for anticipation.

  7. The asserted claims of the '054 Patent and the '566 Patent are not invalid under 35 U.S.C. § 103 for obviousness.

  8. The asserted claims of the '352 Patent and the '910 Patent are invalid under 35 U.S.C. § 103 for obviousness.

  9. The '352 Patent is not invalid for failing to meet the best mode requirement.

  10. The domestic industry requirement for the '566 Patent, the '133 Patent, and the '910 Patent has been satisfied.

  11. The domestic industry requirement for the '054 Patent, the '352 Patent, the '762 patent and the '376 patent has not been satisfied [period missing, another one of countless typos in the original document]

  12. It has been established that a violation exists of section 337 for the '566 Patent.

  13. It has not been established that a violation exists of section 337 for the '054 Patent, the '352 Patent, the '133 patent, the '910 Patent, the '376 Patent and the '762 Patent.

I said there are typos and the above 13 conclusions of law contain two of them. In addition, that quoted passage has an inconsistent punctuation. There are multiple lists of three or more patent numbers. In conclusions 6 and 10, there's a comma before the word "and" that precedes the last item of each list -- but not in conclusions 3, 11, and 13. In the remainder of the document, there are other inconsistencies. Obviously, a determination can be legally and factually accurate regardless of inconsistencies at this level, but courts and quasi-judicial agencies usually try to present their findings in a more polished fashion. Blog posts are a different thing -- they can still be fixed on the fly :-)

I've produced a table that shows the different findings for each patent. In order for a violation of the law governing the ITC to be found, a patent must not only be valid and infringed but there must also be a domestic industry for it. Quite frequently (also in a couple of instances in this investigation), the infringement question and the technical prong of the domestic industry requirement are equally affected by the construction of key terms in the patent claims.

































It's important to consider that a number of the above findings are based on the ALJ's disappointment over Microsoft having allegedly failed to elaborate in its briefs on questions that Motorola didn't even contest. In other words, some of those "findings" are due to whatever may or may not have gone wrong at the pleading stage, regardless of the actual merits underlying the related issues. I mentioned that before and will discuss it again further below, but I wanted to highlight it here as well since the above table does not at all serve as an indication of what a court of law might conclude. Microsoft asserted the same patents in a district court, and unless the dispute between Microsoft and Motorola is settled before, there will be a federal court ruling on these claims -- and the outcome could be a rather different one.

More on the obvious shortcomings of the initial determination

I'll go into more detail on some of the underlying issues as this process continues. It's pretty certain that there will be a Commission review, and there will be opportunities to go into more detail, possibly also on some of the technical issues. At this stage I just want to give examples of the shortcomings I identified, and which are so striking that I believe this preliminary decision will probably be modified or overturned.


I highlighted two typos in the 13 conclusions of law further above. The number of typos in that initial determination is really unusual. But not only their number, also their nature in some cases. For example, I found at least two instances in which "then" appeared instead of "than" (footnotes 26 and 39). I also saw various instances of singular-plural mismatch. Furthermore, there are missing or redundant words.


Further above I gave an example of inconsistent punctuation. Another example of an inconsistency that is highly unusual in a legal document is that the names "Motorola" and "MMI" (Motorola Mobility's stock ticker symbol) are used in confusing ways, such as this one: "MMI does not address the '352 technical prong in its brief. [...] As with its infringement for the '352 Patent, regardless of Motorola's lack of response, Microsoft must still show [...]" It would make sense to distinguish between Motorola (the name of the company when the investigation started), Motorola Solutions (an entity that was dismissed from the investigation), and Motorola Mobility. But in the sentence I just quoted as well as in numerous other instances, Motorola and MMI are used as synonyms. In a magazine article or blog post, one might want to use two names interchangeably for the sake of variation. But in a proper legal document, the primary objective is to have clear and unambiguous references.

Weighing of evidence

I believe that those reviewing the initial determination, or evaluating it at the stage of an appeal, will have doubts about whether the evidence, including testimony, was weighed the way it should be.

Here's a symptomatic passage:

"The ALJ finds that Dr. Stevenson's cryptic testimony and his evasive manner at hearing, which the ALJ found to be less then [sic] credible, leads the ALJ to give no weight to Dr. Stevenson's testimony on this point. As such, the ALJ finds this argument, which is based on his testimony, to be meritless."

That passage speaks for itself, but in case you didn't notice, the combination of "[t]he ALJ finds that" with "leads the ALJ to" doesn't make much sense.

In the infringement analysis concerning the '054 patent, the judge dismisses Microsoft's argument as "incoherent" and its key witness as "unreliable", but what's really amazing is that he doesn't even give much weight, if any at all, to the fact that a Motorola witness apparently confirmed something that Microsoft claimed. The ALJ argues that the relevant witness was just a fact witness, not an expert witness. In other words, Microsoft's witness is unreliable, and Motorola's witness is incompetent as soon as he agrees with Microsoft. This simply doesn't reflect favorably on the initial determination as a whole.

The need to discuss undisputed claims

In connection with the '566 patent, the ALJ says that "MMI has stated that it will not dispute infringement for the '566 patent" and, without citing any other facts, then finds that there is a preponderance of the evidence for infringement. But in connection with the next patent discussed, the '352 patent, the fact that "Motorola does not address infringement in its briefs" is not sufficient and "Microsoft still must show infringement by a preponderance of the evidence." In various contexts including this one, the ALJ accuses Microsoft of having omitted detailed discussions of certain issues in order to focus on those that were actually controversial:

"This is, quite simply, nothing more than an improper attempt to circumvent the page limitations set by the ALJ for post-hearing briefs."

In the same context, the ALJ stresses that "simply making conclusory statements and citing evidence with no explanation fails to constitute 'a discussion' of the issue in the post-hearing brief as required by the Ground Rules and is insufficient to carry a party's burden of proof". In connection with the '133 patent, he says the same thing in different words:

"While the ALJ may have been able to glean the meaning of the term 'manipulate' by culling through the record to find out exactly what Microsoft meant, this is no different from incorporating arguments by reference or simply citing testimony with no explanation. The ALJ declines to engage in such an exercise since it would eviscerate the need for any post-hearing briefing if the ALJ were required to 'figure out' the parties' argument from the record."

There are strict page limits in ITC proceedings. In an investigation in which a large number of patents are asserted, this limits the space available to the parties to discuss each patent. The ALJ clearly would have preferred Microsoft to narrow the case to a greater extent than dropping only two patents. I gathered that from the way he explained a six-week delay of this initial determination. It appears that this ALJ wants to show to Microsoft and everyone else that asserting many patents in parallel will backfire. The problem is that he bases significant parts of his decision on this instead of making substantive decisions. It remains to be seen whether the top level of the ITC is comfortable with this approach. It may not be.

The ALJ also complained that Microsoft didn't provide enough evidence in connection with the domestic industry requirement:

"The ALJ finds that Microsoft has failed to offer sufficient proof to meet the technical prong of the domestic industry requirement. [...] The ALJ notes that this is not a case of a fundamental unfairness in the law. Microsoft chose to rely on code for which it has no evidence [missing word] was ever implemented to meet its burden to prove the technical prong. This is not to stay that Microsoft was required to actually examine the source code of every device, but there must be some evidence presented as to how the device actually works."

Repetitious citation of Federal Circuit decision

In several instances, the initial determination says that dependent patent claims are not infringed if the independent claims they depend on aren't. In other words, a dependent claim has its own limitations (criteria) but also incorporates all of the limitations of the independent claim it depends on, and infringement requires that all limitations -- including all incorporated ones -- are met.

Each and every time that the ALJ makes that point, he cites a whole paragraph from a Federal Circuit decision. It would have been sufficient, and the more commonly adopted approach, to cite that decision only once -- where it comes up for the first time -- and to reference it briefly in all other instances. That's what other judges normally do. It's also what this same ALJ did in some other contexts (where he does make supra references).

If he suspects parties of circumventing the page limit, this repetitious and redundant quoting of a whole, relatively long paragraph is not a textbook example to follow either.

Some of this criticism may appear superficial. As I explained before, those are just indications -- some of them fairly strong ones -- of the initial determination being flawed in a variety of ways, and there will be some more discussion of substantive issues going forward, including technical stuff. This investigation curently has a target date of April 20, 2012.

As far as substantive issues are concerned, I've saved the best for last:

Android the "epitome of hardware dependent software"?!

Here's a quote that will surprise many people:

"Instead, the Android architecture, with its [REDACTED], is almost the epitome of hardware dependent software."

So he calls an operating system that runs on devices from several dozen manufacturers, and on a diversity of form factors, "almost the epitome of hardware dependent software".

In this particular ITC investigation, the conclusion is in Google's legal interest, but in any other context, Google itself would disagree strongly with any claim that Android is "almost the epitome of hardware dependent software"...

We'll see how this matter evolves. Expect a thorough review.

If you'd like to be updated on the smartphone patent disputes and other intellectual property matters I cover, please subscribe to my RSS feed (in the right-hand column) and/or follow me on Twitter @FOSSpatents and Google+.

Share with other professionals via LinkedIn: