Cube/ImageDM - Comments on Observation-Dataset Overview diagram/ composition , references

[CresitelloDittmar, Mark]

Mireille,

I hope I haven't confused things.. in this round, I combined two previous
diagrams
  + the Experiment Overview and Dataset Overview.
This diagram shows the 'inter-relationship' of these two models.


On Wed, May 14, 2014 at 8:13 AM, Louys Mireille
<mireille.louys at unistra.fr>wrote:

>  Hi Mark , hi all,
>
> More on the Observation/Dataset Overview diagram.
>
> I noticed that the current version of Observation has a bunch of
> composition relations to
> Target, ObsConfig, Proposal.
>

These were in the Experiment diagram.. the Observation 'owns' those
elements, and the Dataset references them (as 'provenance').  I notice that
I lost one, the composition of Dataset by Observation (1:*).. will take
care of that.



> I think a simple association with Proposal would be better. I guess n..m
> relationships here too, because one Proposal may need several Observations
> and One Observation may match  the requirements of several proposals.
>

I wasn't making much attempt to create a 'true' Observation model, merely
show that we expect these elements to come from that model domain.
Creating an actual Observation model would/will be a huge effort on it's
own.



> The 3 red arrows on the diagram duplicate the paths between ObsDataset and
> Proposal, Obsconfig, Target .
> I think there references are already designed by the Observation-->
> ObsDataset relation .
>

I'm not sure I understand... these are the same paths as before.



> In the Experiment package: what are DataSource and BandPass?
> BandPass is close to the concept of Filter isn't it?
>
>   BandPass == Filter and comes from the Spectral document.. I seem to
recall some discussion about this
   before, but am not sure right now.  I chose "BandPass" over "Filter"
because the spectral model has
   references to BandPass coming from the Resource Metadata document, and
"Filter" didn't.

 DataSource: from Spectral document.. == "original source of the data (e.g.
'survey')"
   One could see that folding into the observation model as some sort of
'classification' of the Observation,
   but perhaps a tenuous connection.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.ivoa.net/pipermail/dm/attachments/20140514/cccbae20/attachment-0001.html>