Verifiability
Verifiability is property of a scientific concept that means, that In the terms of the already accepted concepts, some specific experiment with some specific result, that confirms the concept, can be described.
Ability to verify
For verifiability, it is not necessary to perform the verification; but the way of verification should exist and formulated at the scientific analysis of the concept.
Verification (called also confirmation) of some scientific concept, hypothesis implies refutation of some alternative hypothesis. For example, confirmation of the First law of Newton may mean refutation of concepts of the inertial propulsion and that of the force-less translation; the failures of gravitsapa and varipend are verifications of the Law of Newton.
In religions, the rejection of any alternative concept is not required. For example, the existence of the Universe and its beauty are considered as evidence of creation of Universe by God. In pseudo–scientific concepts, the claims for the verification may exist, and often, as in the case of religions, no alternative concept, hypothesis is formulated, that would be rejected, refuted by such a verification.
The most valuable is rejection of concepts of high priority in the sense mentioned in Axiom S6 of TORI (if two concepts satisfying S1-S5 have some common range of validity, then, in this range, the simplest of them has priority). Such a rejection appears as confirmation, verification of another concept, that previously could have lower priority.
In Wikiedia, the term "verifiability" is used in a slightly different way, meaning the ability to find confirmation of the statement in the available literature [1].
Verification of some hypothesis should not be confused with its proof. If, within certain concept, some non–trivial hypothesis is formulated and proved by the deduction, then the hypothesis becomes Theorem. It non–trivial hypothesis is verified (refuting some commonly–accepted concepts), then it becomes theory.
Verification is important scientific tool, that reduces the amount of things, allowed within a scientific concept. Ideally, every scientific research (deduction, simulation, measurement) should confirm some hypothesis and reject some hypothesis in such a way, that any result of the research is interesting, has scientific value and can be used in the future research.
Reproducibility
The special case of verification is reproducibility. The effect, presented by the concept, is supposed to be reproduced at the reproduction of all conditions of the previously reported experiments or calculations. In particular, the reproducibility is essential for computational mathematics; the results of evaluation of some quantity, results of simulations are expected to reproduce, if the calculus is performed with different facilities (another hardware, another software, another code). For the tracing of any possible disagreements and errors, it may have sense to specify the software used and to provide the initial codes [2]. In order to omprove the reproducibility, to simplify simplifying the refutation of concepts, in TORI, the figures are supplied with generators, id est, codes, used for the generation. Stability of figures is verified, running the generators at different operational systems (some macintosh and linux); so, the figures are expected to reproduce also at running with other operational systems.
Reproducibility allows both, refutation and verification. If the result of a new experiment of calculus reproduces the previously reported results, this is considered as verification. If not, this may be matter for the refutation. In such a way, verifiability is closely related to refutability; these two qualities can be combined to the requirement of reprodicibility. However, the practice of analysing of the scientific (and pseudo–scientific) claims indicate, that lack of refutability is most often defect of the concepts suggested. For this reason, it worth to separate refutability in a special TORI axiom, and keep reproducibility as a special case.
References
- ↑ http://en.wikipedia.org/wiki/Wikipedia:Verifiability
- ↑ http://www.ams.org/notices/201306/rnoti-p679.pdf D.H.Bailey, J.M.Borwein, Victoria Stodden. Set the Default to “Open”. Notices of the American Mathematical Society, 2013, v.60, No.6, p. 679-680.