[fitswcs] DCF proposals

[Frank Valdes]

Hi Mark,

I don't argue against having a maximum index and I think 999 as the
maximum is fine.  We should come up with a scheme where the orders
associated with keywords are data and not part of the index.  This was
a good idea of yours.  This eliminates the sparseness issue between
different forms of polynomials at the expense of X more data pieces (X
being the number of variables) for each coefficient.  We are keen on
the idea of single strings encoding the orders and coefficient.  I
suspect this will be an area we will have to work on together for
agreement.

I will look forward to seeing your draft.  I am excited by the possiblities
of the DCF to address the situations we encounter in our work with optical
data.

I think we should make sure we are collaborating on this properly.  I
hope the idea is that you write the first draft (as you and Eric
suggested), then we would take the draft source and modify it, and you would
modify our draft, and so on.  Others might either contribute comments
and suggestions or actually add to the draft.  At the end we will have
a proposal with all our names in some order yet to be determined.
Certainly you and me and others who work on the drafts will be at the
top of the list.  Does this sound right to you?

Best regards,
Frank