On 10 May 2014 17:37, "rjf" <fate...@gmail.com> wrote: > > > > It seems to me that the "reproducibility" should be with respect to the same conditions as the original publication. That is, someone who says "I'm telling the truth because yada yada Sage version x.y.z on machine q.p.... should provide not only the commands, but version x.y.z and machine q.p on which it was run. (Provide to whom? )
IF I am understanding you correctly, that would be extreamly unusual and inappropriate. I admit that I may have may have misunderstood you. I believe that you are saying to allow others to verify the results you need to supply both the source code and computer hardware to others. Why should research with Sage be l performed in a way different from how research in any research area I can think of? I have worked on medical physics, RF engineering, optics and never worked that way and only seen a limited subset of it. It would be economically impossible to do this. How much money would you need to put into a research grant application if one of the objectives was to provide computer hardware to one or more people that might want it? The nearest I have seen to this sort of thing is where a research group measures something (the device under test or DUT) then passes the DUT to other research groups who measure the same DUT. When all groups taking part of the study have measured the DUT using different equipment in different laboratories with different people, the results are compared. It is just not practical to distribute the same computer hardware, and in any case it might cause inaccurate conclusions to be reached. Remember the Pentium bug, where an Intel CPU gave an error computing certain floating point operations? Making the source code available is both practical and useful, especially if it can be compiled on multiple hardware platforms with different compilers. That would reduce the chances of erroneous results from CPU bugs, or similar. Dave -- You received this message because you are subscribed to the Google Groups "sage-devel" group. To unsubscribe from this group and stop receiving emails from it, send an email to sage-devel+unsubscr...@googlegroups.com. To post to this group, send email to sage-devel@googlegroups.com. Visit this group at http://groups.google.com/group/sage-devel. For more options, visit https://groups.google.com/d/optout.
