Re: [Axiom-developer] Re: AMS Notices: Open Source Mathematical Software

[Michel Lavaud]

Maybe it would be clearer to say "so they extend their tolerance" ? 
Sorry for my poor English :-(
> >   # [...] so they translate their tolerance to errors in experimental
> >   # results toward tolerance to possible errors in commercial software
> >   
> Ah, OK. You meant gratuitous interpretation, I suppose ? An 
> experimentalist has to be tolerant to errors because errors are 
> inherent to experiments. In particular, for him, a possible error in a 
> program is just one among _hundreds_ of other possible causes of 
> errors. For a mathematician, a possible error in a program used in an 
> article is one among _zero_ other possible errors (if his proof is 
> correct, of course, as also the proofs of theorems his article relies 
> on).