This Kat has (perhaps an oversized) sympathy with the difficulties that confront a court in ruling on a patent infringement matter. As such, this Kat is increasingly trying to place his instinctive response to be a loyal and honest critic (after all, weren't we all first taught in law school how to criticize a court decision) within the context of the challenges of patent litigation. The subject-matter of a patent infringement case is devilishly difficult. Patent claims are by their nature the product of wordsmiths, with all the attendant uncertainty that comes with using language to describe inventions. Often, the invention describes technology of varying complexity. Even if the court is added by a phalanx of experts to help it, the ability of any court to understand the technology in full is limited. Moreover, this Kat's decade-long involvement in rendering UDRP panel decisions has given him a new perspective on the adversarial process and the challenges of rendering a decision, even within the stripped-down context of the UDRP dispute resolution framework. Sometimes, there really can be more than one reasonable way to render a decision.
Pondering these issues, this Kat dosed off and was soon deep into another IP-inspired dream. I dreamt that I was a judge in some future time, charged with ruling on patent litigation matters. I was in my chambers, writing (well, not really writing, since paper and pencil had long been consigned to the British Museum and the Smithsonian Institute) a new opinion. The way it worked is that I would orally compose the judgment, paragraph by paragraph. At the end of each sentence, the computer would, based on an algorithm that had been developed by the Council of Computer Elders, evaluate my contents. The algorithm was proprietary and secret, and it was maintained under the most fastidious conditions of multi-level security. None of the Council of Computer Elders had access to the entire algorithm, which itself was compiled on the basis of another algorithm, which consolidated the collective work of each of the Council of Computer Elders.
With respect to my discussion of the factual aspects, the algorithm would match my contents with the courtroom record, including an objective weight that was given for the testimony of each witness. Relying on the algorithm, a red light would shine on those portions of the draft opinion that were not consistent with the record, inviting me to recompose until I got it right (in years past, a slight electric charge had been applied, but its use had been discontinued.) With respect to my legal arguments, the algorithm would compare my proposed contents with the entire corpus of judgments and legislation and produce from one to three possible conclusions. Here, as well, the red light would light up until I reached the "correct" conclusion.
There was no default—I had to compose and compose until I got the decision "right" (in fact, advancement up the judicial ladder was based largely on the average amount of time taken to reach the "right" result, weighted for differences in the scope and complexity of the cases.) The end result was that here, as always, I would ultimately reach the "right" result. I remember going through five iterations of the judgment before me, still struggling to avoid those annoying red lights ("only three paragraphs to go"). And then—Mrs Kat woke me up for my morning constitutional.
This Kat's dream offers a future (utopian or dystopian, as each Kat reader prefers) view of how we can achieve a "better" jurisprudence for patent litigation. What we should be doing in the here and now, however, remains very much an open question.