From: mailbox@dmitry-kazakov.de   
      
   On 2022-11-29 18:37, Ben Bacarisse wrote:   
      
   > In my view, if you need to check the arguments, that's a separate   
   > function because what to do with bad data is so often   
   > application-specific.   
      
   Yes, validation is a part of the measurement process. Usually datatype   
   in which sensors report data and ones used in computations are   
   different. Validation happens upon conversion and then, ideally, invalid   
   values are excluded per design.   
      
   > Just normalising and carrying on is only one   
   > strategy, and one that can delay detecting bugs.   
      
   Sure. E.g. IEEE 754 float design is a perfect example of such bugs when   
   computations with non-numbers just continue until too late.   
      
   --   
   Regards,   
   Dmitry A. Kazakov   
   http://www.dmitry-kazakov.de   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)