It is interesting that this poll does not mention the problem that I most care about: that the bar for "obviousness" seems to be way too low for software patents. It is so low that I actually worry in my every day work that I will accidentally violate patents simply by solving simple every day problems in obvious ways.
I would support some kind of software patents for truly novel inventions. However so many software patents that I see are almost insultingly obvious, even when you place them in context of their time.
So, yes to software patents if:
1. Shorter time period - 20 years was based on an industrial era time frame for physical manufacture and distribution
2. Much higher threshold for obviousness. Companies should have to declare problems they are applying for patents for publicly and fund a bounty for members of the public or any challenging organization to win by submitting solutions. Only when third parties fail to suggest the methods in the claims should something become patentable.
I would support some kind of software patents for truly novel inventions. However so many software patents that I see are almost insultingly obvious, even when you place them in context of their time.
So, yes to software patents if:
1. Shorter time period - 20 years was based on an industrial era time frame for physical manufacture and distribution
2. Much higher threshold for obviousness. Companies should have to declare problems they are applying for patents for publicly and fund a bounty for members of the public or any challenging organization to win by submitting solutions. Only when third parties fail to suggest the methods in the claims should something become patentable.