>> We are asserting that there exists a context in which they are the
>> same. Question: In what context is an R-magnitude and a B-magnitude
>> the same thing? Answer: when you count photons without regard to
>> their wavelength.
> Photon counting and the logarithmic ratios of the magnitude scale are
> very different beasts. If you are going to pursue photon counting you
> are likely to end up with something resembling IRAF's QPOE interface.
> Similarly with expressing world coordinates - there is an irreducible
> core of complexity. The way to deal with this is to embrace the
> complexity, not attempt to implement false simplicity.
OK I am showing my ignorance of what magnitude exactly means. But it sounds like you are taking a hard line that the concept of "magnitude" is meaningless by itself, and has no place in the UCD list. Surely it has some rigorous definition in terms of photons or flux or energy or SOMETHING? And therefore a context in which they can be compared.
And what is wrong with world coordinates? We have UCDs for CRVAL and CRPIX and CTYPE and all those good WCS keywords. Are they also so poorly defined as to be useless? They are well-defined in the Calabretta papers, aren't they?
>> The problem with namespace is that it reduces interoperability --
>> there was a fear that everyone would simply use their own namespace.
>
> You aren't describing a problem - you describe an opportunity.
The namespaces for UCD was my suggestion. I was voted down. Received on 2005-06-29Z19:20:17