The "Aharonian Test"

I name this test in honor of Greg Aharonian of PATNEWS who has been raging about the crap coming out of the USPTO for longer than I've been blogging. The test is simple: if it fails at its most basic function, it's not suitable material for patent discussion.

In an email sent out today Greg has pointed out that the PTO's database of its own patents fails this test. The patent database is both accessible to the public and used by examiners. As I noted last time, one reason we might point to for bad patents being issued by the PTO is that examiners do not have adequate tools to do their jobs properly.

So, what is the most basic function one uses an existing database for? Searching. In fact, the PTO provides a public Web interface that allows you to search the database. In Greg's email he suggests using this interface for a simple query such as "ICN/PA" (that link should perform the query for you).

This query asks for patents where (at least one of) the inventors are in the country of Panama. Go ahead, give it a try. I'll wait.

Right, so if you're like Greg and me you'll notice that the PTO's database fails to perform even this simple query properly. Some of the answers appear to be because of substring matching (bad implementation of the search algorithm); others appear to be because of incorrect data in the patent records themselves (bad quality control). Regardless of the cause, it's clear that this tool fails here, which leads one to wonder how badly it fails on other queries. If patent examiners aren't finding relevant prior art maybe we ought to give them the right (tested) tools and (quality supervised) data to make that possible, eh?