Posted on 05/25/2006 7:58:13 AM PDT by ShadowAce
But if software patents are to be granted, I do agree with your RSA statement. The bar for "not obvious to a person skilled in the art" needs to be set pretty high. Right now any average programmer who writes program of any decent length is probably violating several patents. And any patents that are issued need to have a very narrow scope on the exact implementation.
If you look over on the MS-JPEG thread I mentioned the VirtualDub incident. The author wrote his own WMV support by reverse engineering the format. There is no way that should be allowed to violate a patent. Saying "we patented WMV decoding" is wrong, saying "we patent this particular method of WMV decoding" is more acceptable.
This is not accurate. I think the system is fundamentally broken because it presumes something that is clearly invalid. There is a false dichotomy here of two sides that are both refusing to address the reality that is causing the disagreement in the first place. It is very much like arguing whether it is the Tooth Fairy or Santa Claus that makes rainbows.
There are two facts here that all parties need to acknowledge before a constructive resolution will be found:
- There is no theoretical difference between software and hardware in any case, nor practical difference in many cases. Every software algorithm can be implemented in a mechanical machine and vice versa, making the implementation substrate immaterial.
- Patents are not an intrinsic good, being a somewhat arbitrary form of regulation. It may be time to revisit the basic assumptions behind that type of regulation as it currently exists.
Most parties on both sides of this argument are arguing a position that assumes one of these two things is not true. Hence all the heat with so little light. Continuing to make policy based on defective assumptions will not solve any problems no matter what the outcome.
Right there you are falling on the side of patenting software.
Wrong. I am making it an issue of patents, period. If you are going to allow patents, then software patents (a concept necessarily lacking anything remotely resembling a strict definition) should be allowed as well for the sake of consistency and well-formed law. Patents are all or nothing within the bounds of that which can be well defined. Software patents are impossible to strictly define, hence why this problem emerged to begin with.
You cannot have a meaningful discussion of property law based on a categorical distinction that does not exist even in theory. I do not care what the outcome is. Ban all patents, issue patents for everything, or only allow patents for categories with strictly definable boundaries. Everything else is handwaving that does not recognize certain basic theoretical realities of our universe.
I do not care what happens, but what I really abhor is that the whole argument and system is based on concepts that have no connection to reality on any strict level and which has no consistent treatment of the subject matter it is supposed to address. Make it clean and consistent, and I will be happy with any outcome. The current mess requires arbitrary and capricious decision making to function at all, which is why that is what we get.
Disclaimer: Opinions posted on Free Republic are those of the individual posters and do not necessarily represent the opinion of Free Republic or its management. All materials posted herein are protected by copyright law and the exemption for fair use of copyrighted works.