I've been rather too busy to respond, since one must be _very_ careful about setting up the criteria, but since there is additional interest I'll respond now briefly but carefully:
On Tue, Feb 11, 2014 at 3:31 PM, Jed Rothwell <jedrothw...@gmail.com> wrote: > James Bowery <jabow...@gmail.com> wrote: > > What I'm asking for is something similar to what I asked of proponents of >> alternative fusion technologies when writing up the fusion prize >> legislation back in >> 1992<http://www.oocities.org/jim_bowery/BussardsLetter.html> >> : >> >> "If you were considering competing for a cold fusion prize to be awarded >> for a reliably reproducible experimental protocol, how would you like to >> see that prize's criteria stated?" >> > > Well, it does not seem complicated. I guess I would say it has to be > replicated by at least two other labs; it has to work in one out of ten > runs; and it has produce a high signal to noise ratio. Exactly how high > should be defined by someone who understands statistics better than I do. > > It would be nice if they could have intermediate prizes for incremental > progress. > > One of out ten may not seem like much, but it is enough to make the > experiment reasonably easy to replicate. In fundamental research, there is > never any call for high reproducibility, only replicability -- which is a > different thing. IPS cell reproducibility is something like 1%. I think > Obokata has improved it a great deal with her new technique, but it is > still low. Improved reproducibility has no bearing on the scientific > validity of the claim, but it does make the research easier, and it is > needed for eventual commercialization. > The central problem in setting up a cold fusion prize is the cost of judging since replication is the foundation of judging and the cost of replication can be considerable. Moreover, the selection of those who are considered "skilled in the art" in the sense meant by patent validation (ie: that a patent is valid only if its disclosure allows those "skilled in the art" to realize beneficial use), must be controlled by the judging authorities. This is problematic since it is hard to know, a priori, what "the art" will be if one is trying to unconstrain the innovators in achieving the goal. Therefore it seems necessary that two conditions be met: 1) That the entrant will pay the costs of replication. 2) That the entrant will pay the costs of the negotiation of which teams are considered "skilled in the art" for the purpose of replication.
