Understanding (and Avoiding) Software Patents?
An anonymous reader asks: "I'd like to write some Free backup software, but this area is mined with patents. I downloaded one and tried to understand it, but the 'claims' section (arguably the most important part) is made up of utterly incomprehensible patentese, and I can't afford to hire a patent attorney to help me understand it. Are there any free or cheap ways to learn enough about patents to understand them, so I can figure out exactly what is patented and therefore avoid it?"
"How different does my software have to be in order to be non-infringing? The patent I tried to understand is Dantz's 5,150,473. Many, including Slashdot readers, have said what this patent covers, but from reading the patent itself, I would never have guessed. Also, there are lots of other patents to understand and avoid, so I'm looking for general information on how I can unravel it all into language I can understand."
It isn't so much "ignorance of the law" that provides some protection, but the ability to demonstrate that you were ignorant of the patented invention, because if they can show that you read the patent, that would imply that you ripped off the idea from it.
But if you haven't ever looked at it, you can make your alleged infringement look accidental, which might (in theory) even strike down the patent, by showing that it was obvious to someone familiar with the state of the craft.
http://alternatives.rzero.com/
Of course the easiest way to avoid willfully infringing any patents is to not know of any in the first place. But this will fall apart the first time you get a cease and desist letter.
The best way to know you aren't going to get nicked for crazy damages if you are found to infringe is to get an opinion from a patent lawyer that you aren't infringing. This doesn't mean you won't be found to be doing so at trial, but it does mean are you pretty much insured against treble damages. Unfortunately, this costs a crapload of money in attorney fees.
There is an article about "The Anatomy of a Trivial Patent" written by RMS. It may be a nice introduction to the topic if you want to read more complex patent texts.
Right, you mention a computer, so that what are you patenting is the computer, not the pure algorith itself. And of course, the difference is pretty much nonexistenct for practical purposes, but since we were working with a contrived example of just publishing source code, I think that was one of the few times that difference mattered.