On 10.01.2008, at 11:16, Oliver Jennrich wrote:

> I'm wondering if updating the constants without a possibility to  
> revert to an earlier version is a good idea.

Generally this is an important point for reproducibility.

However, personally, I do not care. Under normal circumstances I want  
to use the current "correct" values. If I want to reproduce some old  
calculation I use the same old GSL version to really get the same  
calculation results - after all the code also changes continuously.

> Would it be useful to have something like
>
> GSL_CONST_MKSA_SPEED_OF_LIGHT_2000
> GSL_CONST_MKSA_SPEED_OF_LIGHT_2006
>
> and
>
> GSL_CONST_MKSA_SPEED_OF_LIGHT
>
> to be either of the two, depending on a #define?

The default value should definitely be the latest, but one could  
envision to have
   GSL_CONST_CODATA_1998
define the values according to that standard. However, someone needs  
to implement it and it needs to be maintained as time goes and new  
standards are published.
It should also be made clear that everybody uses the latest values  
for scientific calculations...

Greetings,
Jochen
-- 
Einigkeit und Recht und Freiheit                http://www.Jochen- 
Kuepper.de
     Liberté, Égalité, Fraternité                GnuPG key: CC1B0B4D
         Sex, drugs and rock-n-roll