Matthew John Charles wrote:
> Hi Jan,
>
> That's not a bad idea. For one or two kinds of calibration it would
> probably be overkill, but if we want to leave open the option to
> generalize this later that would be a good way to do it. Dunno...
>
>
> Cheers,
>
> Mat.
>
>
Hi Mat,
in software design you only deal with three quantities:
0, 1, many.
If you have to do it twice, there is a good chance that you'll have to
do it more often later.
With decisions like this I feel we should adopt Python's development model.
New questions are discussed on the dev list to accumulate suggestions
and get a general feeling for preferences and requirements and then a
final and ultimate decision is made by the BDFL (benevolent dictator for
life).
Cheers,
Jan
--
Jan F. Strube -- University of Oregon
Stanford Linear Accelerator Center
bldg. 48, rm. 244, MS 35
(650) 926-2913 phone
8522 fax
|