[ImageDM]: Provenance

[Arnold Rots]

I am not sure I get the full implications of what you are suggesting.
If it means that Provenance is something that comes to us through
Observation, then I think it is the wrong path.
Provenance is something that is, inalienably, a property of a Dataset.
It may include information that is inherited from an Observation, but
that is information that characterizes the Observation - not its
Provenance.
If you want to take this to a more abstract level, one could argue that
the Observation is part of the Provenance contents of a Dataset, not
the other way around.

Cheers,

  - Arnold

-------------------------------------------------------------------------------------------------------------
Arnold H. Rots                                          Chandra X-ray
Science Center
Smithsonian Astrophysical Observatory                   tel:  +1 617 496
7701
60 Garden Street, MS 67                                      fax:  +1 617
495 7356
Cambridge, MA 02138
arots at cfa.harvard.edu
USA
http://hea-www.harvard.edu/~arots/
--------------------------------------------------------------------------------------------------------------



On Fri, Nov 15, 2013 at 3:42 PM, CresitelloDittmar, Mark <
mdittmar at cfa.harvard.edu> wrote:

>
> All,
>
> While we aren't trying to define the content of the Provenance package in
> this thread, this input has been really helpful to me.
> I'm not cutting off discussion on the thread, but I do want to avoid this
> going off into an attempt to define Provenance.
>
> For the present question, what I'm getting is that the Provenance package
> doesn't have a logical 'head'.  Where the expectation is that users will
> include "Provenance" and get everything they need.  Instead, it is (will
> be) a collection of classes and groupings which users (like Observation )
> might choose from as needed, perhaps grouping them locally.  Even this is
> speculation, so I would hesitate to add organization to something which is
> still very undefined.
>
> So I think adding a "Provenance" class to the ObsCore Provenance package
> would be inappropriate.
>
> It *may* be appropriate to add a Provenance container at the Observation
> level, but I see no compelling reason to do so for the "as is" picture that
> we are working.  It may be that during the further discussion of the
> ImageDM and Observation/Dataset issues, that we want to add some
> organization to components, but it isn't clear that a single 'Provenance'
> node would be helpful.
>
> Again, I am not closing discussion on the thread, but am leaning heavily
> in the direction of option 'b', and will be updating my diagram accordingly
> unless I hear a strong objection.
> I expect the next drop of the image to be to the twiki (either directly or
> via link) so as to not pollute mailboxes with outdated images.
>
> Mark
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.ivoa.net/pipermail/dm/attachments/20131119/11bb03c3/attachment.html>