Quote:
Originally Posted by HarryT
It may or may not be patentable, depending on the jurisdiction. "Pure" software patents are almost never granted by the UK Patent Office, for example. In order to be patentable in the UK, a computer program has to be part of a solution to a "technical problem". A data compression algorithm used as part of a more efficient system for transmitting audio from one place to another probably would be patentable, for example, but the algorithm in isolation wouldn't be.
|
I was thinking more along the lines of something *really* new and unique, not a refinement of existing techniques. Say something based on map theory or other advanced math. By now even fractal encoding would likely be deemed obvious by a strict analysis. The problem, as you say, is the inconsistent application of the rules.
The people doing the patent approval simply aren't as versed in the subjects as the submitters so they can be snowed.
Edit:
Another problem is that obviousness changes over time.
50 years ago, digitally encoding audio and video was not an obvious solution to the distribution problem; it wasn't even viable. By the 80's it was viable for storage but not transmission. Today, going digital is the default first approach to almost everything. (3D printers point the way.)
Tone result is that looking back at past patents we need to look not only at the prior art in that specific field but also at the general state of technology at the time the patent was granted. Things change in 17 years and what seems unworthy of patent today might have been non-obvious at the time. Often I see people griping that a given patent is trivial or obvious... from the point of view of the present. Especially with computers. But in many cases they are tslking about patents that go back to the last century. Things have changed a bit since then.