Time Serie: model vs mapping

[Laurent Michel]

Hi,

It seems to me that there is a confusion between the model and the mapping in the Jiri's proposal, posted on 18/07/2017 
09:01CEST (http://mail.ivoa.net/pipermail/dm/2017-July/005602.html - see enclosed diagram).
The Yellow part of the diagram denotes the observable part of the TS. The problem is that the observable can be anything, and we 
cannot have a model for "anything".

To sort this out, the TS model uses references to table columns with the class "ColumnRef". In other terms, the model is 
complete only with the assumption that it is applied to tabular data and that all requested metadata are available in that 
table. Thus, a part of the model is described out of the model, of any model. This is inconsistent somewhere.

Normally, a model describes a set of quantities and the way they are linked together. The links between the model leafs and the 
data are in charge of the mapping mechanism, but here, both are mixed.

LM

-- 
jesuischarlie/Tunis/Paris/Bruxelles/Berlin

Laurent Michel
SSC XMM-Newton
Tél : +33 (0)3 68 85 24 37
Fax : +33 (0)3 )3 68 85 24 32
Université de Strasbourg <http://www.unistra.fr>
Observatoire Astronomique
11 Rue de l'Université
F - 67200 Strasbourg

-------------- next part --------------
A non-text attachment was scrubbed...
Name: TimeSeriesCube DM detail.png
Type: image/png
Size: 48848 bytes
Desc: not available
URL: <http://mail.ivoa.net/pipermail/dal/attachments/20170720/7b76a815/attachment-0001.png>