[Ifeffit] database (was: Cu oxides spectra)

[Matt Newville]

Carlo,  Thanks for the offer of help! 

Dave, thanks for the suggestions, I think those are all good.

In fact, I think none of this is that hard to do, but it will
take some time. For what it's worth, I think the data should be
stored in a 'proper' relational database, if for no other reason
than to make it easier and faster to search.  It will also make
many of the 'fancier' features (user comments) easier to
implement.

At this point, I think the most immediate need is (as Dave
mentioned yesterday) to identify a reasonable format for the
data at both upload time and at download time.  For upload, we'd
certainly want a web form where fields could be filled in or
parsed from a file following a 'Standard Data File Format'.

There have been a couple attempts at a 'Standard' data format
either for data collection or at least data exchange.  I would
guess that if we do this right with a newly revised database
that we could set what that standard was (at least for data
exchange).  Certainly we'll make ifeffit/athena/artemis read
these files easily.  But we should probably think about what is
needed for a more universal and long-term format. I know Bruce
has thought about this some and worked on a proposed format.  
Hopefully he'll post that.

--Matt