SED data model v0.92

[Jonathan McDowell]

Hi Ed,

> Jonathan,
>     I am just beginning to look at theSED document.  I do not see where 
> one can place an error on each individual point.  Have I missed it?   

Yes. 
The Flux object has an accuracy, and there's one for each Point.
This should be clear in the diagram and in the serialization examples,
but it isn't clear in the text in section 4.7. Sorry about that.

> 
> Secondly, you have for upper limit information:
> 
> "If the data provider has only upper limit information, it should be 
> represented by setting the flux value and the lower error value equal to 
> the limit. "
> 
> But what about the postiiveError limit?  Is it then set to zero?

Yes, I'll add such text.
This is one point I'd particularly encourage feedback on - is this
an OK way to go for upper limits? I don't want to add an extra field
that's usually not used.

> Finally, for now, I think redshiftError and redshiftConfidence should be 
> children of redshift (in the XML Schema), as all good quantities should.

Agreed.
 Jonathan