Do you have any thoughts on SCOTUSblog's take on it, that it revolves around a question of whether the standard of review should depend on whether the USPTO considered a particular piece of evidence? --> http://www.scotusblog.com/2011/04/court-takes-up-standard-of...
Their take is that Microsoft is only asking for the standard to be lowered with regard to prior art that was not considered by the USPTO. Microsoft argues that the case for deference is greatly weakened if the USPTO never evaluated the specific prior art in question; but that deference should still be given w.r.t. anything that the USPTO considered.
IBM's amicus brief opposes that approach, because they argue it'll create a mess if the standard of review depends on details of a specific patent's administrative record, and because it will encourage patent filers to just dump huge reams of potential prior-art into the USPTO record, to make sure that anything that could possibly be found will count as having been already considered by the USPTO.
I think that a half-way approach would only create chaos and would satisfy no one. Its main appeal is that it would give the Court a middle ground by which it could say, consistent with general principles of administrative law, that the USPTO should be accorded great deference but not insofar as it did not in fact consider the evidence subsequently used to challenge its decision. That works logically but, as noted by IBM, would likely only complicate the issue beyond bounds as a practical matter.
Microsoft likely fashioned its petition along these lines owing to a dictum in KSR v. Teleflex that said as much and also to try to appeal to the incremental jurisprudential approach of the current Supreme Court (decide only what is absolutely necessary for a given case and leave it to future cases to do more). The irony of this is that Chief Justice Roberts, who seems to specialize in incremental jurisprudence, has recused himself from the case and will not participate in the decision.
I believe the half-way argument is weak and will not hold sway. If the Court goes for reform, there is no reason it should not go all the way on this issue. This is particularly true given that nothing in the statutory language limits the Court's power to set an appropriate standard. Unlike Bilski, where it faced a challenge to business method patents in a context where Congress had expressly said you could have them, there is nothing here that stands in the way of the Court's making a coherent ruling based entirely on what is appropriate for the case. I am hopeful, therefore, that this case will help bring some sanity to this troubled area of the law.
One of the amicus briefs urged that courts should give deference to the examiner's decision only as to prior art that had been discussed in writing by the patent examiner, thus providing a written record of reasoned decision-making that could be reviewed by a court. That brief relied on the Supreme Court's holding in the Zurko case [1] that the standard of review in the Administrative Procedure Act applies in appeals from decisions by the USPTO just as much as in other administrative appeals.
Footnote: The Zurko patent [2] was one of the early software patents, for a method for executing trusted commands in an untrusted environment. The lead inventor was Mary Ellen Zurko, aka "mez," who then worked for Digital Equipment and is now with IBM. (Disclosure: I drafted that patent.)
Their take is that Microsoft is only asking for the standard to be lowered with regard to prior art that was not considered by the USPTO. Microsoft argues that the case for deference is greatly weakened if the USPTO never evaluated the specific prior art in question; but that deference should still be given w.r.t. anything that the USPTO considered.
IBM's amicus brief opposes that approach, because they argue it'll create a mess if the standard of review depends on details of a specific patent's administrative record, and because it will encourage patent filers to just dump huge reams of potential prior-art into the USPTO record, to make sure that anything that could possibly be found will count as having been already considered by the USPTO.