Slashdot Mirror
US Software Patents Hit Record High
Aditi Tuteja writes "US Patent and Trademark Office made a new record for the number of software patents awarded in a single year. The agency has issued 893 new patents yesterday. Pushing the total to 30,232 in this year. If this is the trend, more than 40,000 software patents will be issued this year, according to the Public Patent Foundation. The previous record was set in 2004. Several major technology vendors have pledged not to enforce their patents against open source projects. IBM for instance essentially donated 500 patents to open source projects last year. Earlier this year, the US Supreme Court overthrew a prior judgement that required a judge to issue an automatic injunction if he found that a patent was being infringed."
I know postgresql at one point unintentionally used one of IBM's patents. They removed/rewrote the code though, since they use the BSD license. Using IBM's patent would prevent people from close-sourcing it without negotiating a license from IBM.
Do you even lift?
These aren't the 'roids you're looking for.
In the recent decades it is common practice for patent applicants to submit an "Information Disclosure Statement" where they list prior art they submit to satisfy the foregoing laws and rules. And it's not uncommon to find an examiner using art cited by applicant in formulating prior art rejections. The real question, though, is one of state of mind of the applicant, i. e., the old question if "What did you know and when did you know it?" In asessing applicant's duty to disclose relevant prior art, how much effort to find relevant prior art must applicant show? It is not totally out of the question, even in this age of extensive internet search tools and data sources, that some public, but obscure, piece of prior art might exist, but not readily available. If someone only unearths the reference after searching for hundreds of hours (and quite possibly going to many dead tree sources or making hundreds of phone calls), are you really going to fine/throw in prison the applcant even if they can show they did make a decent search? And a borderline situation often exists in areas that have rapid development, where ideas can be published and spread very quickly. At what point does this become so commonly known that the applicant must be presumed to have known of it prior to the filing of the application, and thus be liable to criminal sanctions?
There's even a more complex situation where the question of obviousness is involved, as it most frequently is. A patent (or application) claim is unpatentable if it describes "subject matter as a whole [that] would have been obvious at the time the invention was made to a person having ordinary skill in the art to which said subject matter pertains" (35 USC 103). Who makes the judgment of what this ficticious person of ordinary skill in the art would know, and how he would judge the claim? It's a question of opinion to which different people, even with some credibility in the field, can disagree. What if the applicant, has the same prior art to consider, and decides it's not relevant, and thus, fails to cite it, but later some "expert witness" disagrees and states that, at the time of the invention anyone in the art would have known this? And, remember, the issue here is not that the claim is properly invalidated by this prior art, but that the applicant knew of the prior art, but did not disclose it; was this a good or bad faith omission?
It's hard enough just making the patentable/unpatentable decision under the current law, and it's a good idea to cite prior art even one is almost certain it's not relevant to the patentability determination, but routinely criminalizing such an omission seems a bit over the top.