Many of the folks commenting on that last post know what they're talking about, but not everyone who opines on drug industry patents does. (My fear, as I said yesterday, is that Boldrin and Levine might be in that category). So I thought I'd lay out some of the ground rules that people outside the pharmaceutical world might not be familiar with.
First off: it is not difficult to figure out what is in someone else's pill. Modern instrumental techniques make can make this an about an afternoon's work - a couple of days for a tougher one. And as for the formulation - all the stuff in there other than the active ingredient itself - that's usually easy to reverse-engineer, too. For any practical purpose, without a patent there is no barrier here.
Next, while the process of producing a given drug is often the subject of separate patents, there are still, in most cases, multiple ways to skin the chemical cat. Generic drug companies like Teva and Ranbaxy, among others, are famous for being able to come up with processes that are equivalent (and in some cases superior) to the patented ones. It's rare that there's a barrier to innovation here.
Third, it should be remembered that patents must thoroughly disclose the subject of their claims. Patents on drug substances must provide specific, detailed instructions on how to make the compounds in question and on their physical properties. One "skilled in the art" should be able to step right up and whip up a batch - and in fact, those skilled in the art often do, to get a look at the competition's compounds or to try to improve on them. It can be hard to see which of the claimed compounds is the one that's being developed, true - but by the time it gets into the clinic it's often clear, and by the time it gets to the FDA, all must be revealed.
And finally, please, please take some time to understand the difference between an academic discovery and a finished drug that makes it to market. Talking about how "gosh, most medicines are found through government-funded research anyway, right?" will not convince anyone who's done drug development that you know what you're talking about.
There, that should help clear out some of the underbrush. Workable suggestions for something other than patents on drug substances, anyone?