Presentation Notes

Notes by MCH on Jonathon MacDowell's presentation

SED Model:

This is a pre-Observation model so we can get on with SSA before the Observation model is ready. We need to make sure it converges with Observation rather than diverges as they develop.

Accuracy

To be expanded later. A placeholder for now.

For the SEDCoord, need pixellation, resolution and error (incl limits and systematic errors). Probably need to combine Flux Accuracy with SpectralCoord Accuracy (? If I have this right). Resolution is going to be complicated.

Quality

• For now, 0=good, 1-2=Something else
• To be set by data provider
• Extend later

Flux

(also includes background now)

Some 'brightness' types are not intended to be modelled; it's up to the data provider to provide some flux value (or a johnson magnitude or similarly convertable type) with appropriate accuracy.

[MCH: Could DataID be represented by VOResource?]

Corrections & Processes applied

Requests for some 'history' record of what processes have been applied in what order.

[MCH: Perhaps add a set of 'history' objects, eg De-Reddening. Needs to be applied to Flux, Coordinates and perhaps overall.]

-- MartinHill - 25 May 2004