Should Patents Be Harder to Get?

Here's a very interesting viewpoint I noticed on Groklaw. It was published anonymously:

I am not a software person, but I have written and studied a lot about software. I've been annoyed that Groklaw has (seemingly) presented the idea that it is only software patents which is a problem. The problem in today's society is ALL patents.

Patents (letters patent) were invented a long time ago. A time when there was much less information than now, much fewer people, and less cross-disciplinary knowledge.

The over-riding concern about patents originally, was that this knowledge (the new patent) was exceedingly hard to acquire, and their was a tremendous barrier to entry into the market. Various patent laws were enacted, and things like obviousness to skilled practitioners got lumped into prior art.

My specialization is Materials Science and Engineering. Which is one of the least populated parts of engineering. And of late is getting contributions from chemistry, physics, electrical engineering and engineering physics, and some other places.

As with copyright, the original idea was that this new thing/concept/data/realisation/... was unique, expensive to develop, expensive to reproduce, and hence there was an argument for allowing the inventor to possess a monopoly on this knowledge so as to recuperate the development costs.

With many areas of Copyright, there is little or no barrier to entry, which was a huge reason to enact Copyright. And people who work in Copyright can continue to fight that battle.

In patents, there is always a barrier to entry, as their is an expectation of some idea being reduced to practice. If I look through the news section of one of my MSE journals, it isn't unusual to see 2 (or more) news items about some new patented technology. And they all look to be either the same, or very similar. And if you decide to go looking at research in that small area of Materials Science and Engineering, you will probably find tens or even hundreds of university programs investigating that same topic. You will also find commercial organisations investigating this, that happened to have published something which indicates they are interested in the same topic. How many commercial organisations are investigating this topic, and haven't published anything?

When patents first came out, if this knowledge wasn't protected, it was lost to society forever. This was the justification for granting a monopoly. I'm in one of the least populated parts of engineering, and for any given patent, you can often find tens of similar patents, and if you look you will find their are hundreds or even thousands of research programs (university/government, or commercial) working in this area.

Having various jurisdictions (governments) move to "first to file" is an indication that non-obviousness to people skilled in the art is being disregarded. For the most part, there is no need (on the part of society) to register a patent for anything.

What this is showing, is that even if the person (group, company, ...) that got the patent, hadn't of even applied, that same knowledge (or something very similar, or possibly even better) would soon be in the marketplace (presumably under some different patent). What this shows, is that more often than not, the knowledge is not novel, and it is obvious to a skilled practitioner of the art (and "art" is getting fuzzied by the cross-disciplinary nature of just about everything). There is no reason for society to grant a monopoly for this new invention, as it would be invented by someone else soon, if this particular person didn't claim to invent it "now".

In my field, there still are "inventions" by a single person that would not be immediately forthcoming if that person wasn't there to do the work. And typically these are theoretical topics, for which one seldom gets a patent. Once this person publishes their work, it is often seen that zillions of ideas get "realized" and patents granted. There is no non-obviousness to any of the patents granted: once the original theoretical work was published, most (typically all) are just obvious consequences, and should never be granted "letters patent". There is no prior art, but there are so many people skilled in understanding that particular kind of knowledge, that there is no incentive to society to grant a monopoly, to avoid having the knowledge lost forever.

If this is so obvious in my narrow area of engineering, why are there patents in Electrical, Mechanical, Civil or any of the other larger branches of engineering?

The only thing that comes to my mind, is that there are very few MBA graduates that know how to produce a better product and better service, the only thing they know is how to exploit an monopoly.

I have long known that patents were completely antiproductive in software. The job of any good software writer is to come up with the right way to solve all the many problems posed in getting a system to work. And they never look at any existing patents. They look at existing knowledge - data structures, algorithms, and so on. But not patents. Only very rare software writers, who already know what they are looking for, use an algorithm knowing that it is, or might be, patented. An example might be highly mathematical algorithms such as JPEG, etc. But the key ideas in them are mathematical, and isn't mathematics supposed to be unpatentable? (Which open up the whole question of whether patents are an arbitrary privilege for some kinds of inventors, and which is denied to others.)

A key moment for me with patents was the discovery that James Watt held patents prior to 1800, which he used to block the development of high-pressure steam engines, which meant that they had to be big and bulky and could not be practical for moving engines such as railway locomotives. When Watt's patent expired, the railway age could at last begin. Have they ever been fair or beneficial?