Controversial Opinions on Software Patents - The Good, the Bad, and the Ugly

I think that copyright is already good enough.

Most things on our society were created on a time without patents. And no one died because of it.

Patents only allow big companies to crush the little guys that don't have enough money to fight them until death on the courts.

As it is now, every stupid idea, even it exists since the dawn of mankind is patentable. Not to speak for the people that just create patents for everything they can remember of, without creating anything new, and then suit however does a real implementation of their idea.

Where would be mankind today if patents were allowed in old Greece, or in the Renaissance days?

The only reason we have the piano, is because its author decided not to patent it.

The idea that software is patentable is relatively new. The legal basis of software patents is that processes can be patented. If someone were to invent a new algorithm for searching strings, this is something that I would consider patentable. But this is not what people are complaining about.

The issue is that patents are being issued for ideas that are not novel and/or are trivial. Fundamentally these are invalid patents under current patent law. Unfortunately, it costs money to defend against even a blatantly invalid patent. But this is a symptom, not a cause. The root cause is that these patents are issued in the first place. Some have proposed that the reason this is happening is that the patent office's funding comes from the patent application fees and not from the federal budget like it used to. This creates a conflict or interest that drives the office to focus on the number of filings and not on the quality of the patents.

I'm not in the position to assert that this is correct but I find it to be a compelling theory.

Martin Fowler is just wrong. The point of patents was to encourage people to share their secret inventions. By rewarding inventors, it was hoped that the general standards within an industry would improve as people learnt and adopted each new invention.

Here I am Mark the dyer. I invent an improved dying process. I keep it secret for obvious reasons and none of the other dyers gain any knowledge. Then patents come along. I publish my new technique. All the other dyers adopt it - paying me a fee to do so. Everyone wins.

If you read old patents from, say, 100 years ago comparing them to today's the difference is striking. Those old patents clearly disclose the invention and make very specific claims with respect to proven processes which are often already in use. The examples are clearly detailed.

Now read some modern patents. You are mired in legalese. Any actual process is written out in such general terms that attempt to broaden the patent in any and every possible direction. Examples are so vague that one is unlikely to reproduce any process by simply relying on the text. One must use one's expertise to guess and infer what's actually going on. Claims are multiple and general and are often made with respect to imagined applications - not just the actual application disclosed.

The point of a modern patent is not to provide competitors with knowledge but to provide lawyers with weapons to use in a court room.

"...I invented an improved dying process...". Sorry man, but we as software developers have been "inventing" solutions every day based on other solutions!

Today, protecting your invention is only one of the reason for filing a patent. There are three other and probably more compelling reasons:

- Measure of R&D progress
- Defensive measure
- Startup valuation

setandbma.wordpress.com/2011/01/13/abc-of-paten...

André: Sorry man ...

LOL - as they say - I was using that as an illustrative example to compare patents, as they originally were, with patents today. I never invented any 'dying process' - I'm a developer too [why did you think I'm here?]. Oh dear, irony, metaphor and analogy just don't work too well on the internet.

Udayan B. - The 3 other compelling reasons you give are not compelling at all. The compelling reason why patents were created was to share knowledge and discourage people from keeping secrets. In that manner, and that manner alone, patents are useful.

So-called 'intellectual property' is a one size fits all solution. A new drug takes decades and hundreds of $ millions to develop. There would be no new drugs without patents. In software, patents are perverse - not helping the development of new technology but, rather, are applied in an attempt to fleece successful companies or to strangle competitors.

Copyright allows publishers to PREVENT the publication of books (as well as publish new ones) but we would probably have fewer new books of worth without copyright. As the critics[1] of copyright have correctly pointed out the terms granted to copyrights are absurd. No software anywhere requires a term of more than five years. After a short while, the software we write is either obsolete or has been rewritten so many times that it's completely different. Given that the copyright applies to all new code and that successful software (in use) is in constant development - such useful software would never go out of copyright - even with a 5 year maximum term.

Economists are at odds with each other. One camp argue that strong IP laws are essential to the development of industry. Another camp point out that the 19th century boom in the USA and 20th century boom in China occurred in an environment of rampant IP 'piracy'. If anyone wants to convince me - give me references to empirical economics research. Everything else - even my writings - are just supposition.

There was a reasonable editorial in the FT this week making the point that governments use the proceeds from patent grants as a form of taxation. Patents should have a far more rigorous approval process. I would propose something akin to what happens in peer-reviewed journals before publication - the patent fees could be used to pay the peer reviewers. At least with such a system, if we must have software patents - please let them be approved by people who understand software development.

[1] E.g. see: Liberals: Lessig(2002, 2006), Vaidhyanathan(2003), Socialist: Bettig(1996) and various articles by 'libertarian' writers in Reason magazine.

Some categories of patents seem to largely do the opposite of what was originally intended. The original intent was to give society access to an idea which might otherwise have died. Software patents are probably seldom used to find useful ideas. They are used instead to prevent others from using obvious ideas. They are a barrier to human progress.

(Some ideas are not obvious and took genuine R&D. Protection for those seems reasonable.)

Recent changes in US legisaltion may make the situation worse.

If patent clerks were prosecuted for making mistakes we might get some control over this runaway train! It would also be a good idea to plug some of the holes in what the patent clerks know, and not to let the good ones be taken by patent lawyers.

Mark,

Society's reason for having a patent system is to foster progress by encouraging people to share knowledge in return for a brief monopoly.
An individual's (or firm's) motives for filing a patent application probably include all of those that Udayan mentions, and some others.

The others are really the problem- because broadly-written patents insufficiently scrutinised by patent offices allow the holder to deny others to make use of inventions that the patent-holder didn't make, and perhaps didn't even imagine. In some cases they catch inventions already made by someone else, but not noticed by the patent examiner. They reward people for employing patent attorneys, rather than for inventing.