Spectra DM for theoretical spectra?

Carlos Rodrigo Blanco crb at laeff.inta.es
Wed Jun 3 06:28:09 PDT 2009


> PLEASE also look into the SimDB data model.
> It already allows you to describe the provenance/history of
> synthetic/simulation data products.
> Why is that not good for you?

Sure, sure, I will do. I just find it too hard for my week brain and still 
haven't been able to understand it ;-) (actually I have to say that I 
don't find the time to do more than "take a look to it" and SimDB is not 
simple enough so that it's enough with that )

In fact, I _think_ (I underline it because it's only my first impression) 
that, as you say, SimDB makes a lot of effort to describe a lot of things 
about the provenance of the model (and I guess it can be most useful for 
that in the case of theoretical spectra or other kind of 
"microsimulations", I don't dare to say if better or worse that he ObsDM 
provenance because I honestly don't know).

But all the parameters that I am interested now are just "properties" that 
each model developer has to define. If that's so, it doesn't help much to 
solve the problem that I'm talking about. Actually I think that this is 
not such a difficult problem and it can be addressed in different ways 
(from simple to complicated, inside or outside simdb, inside or outside 
some data model... who knows), but I just wanted to highlight that it 
exists (or that's what I think)

Carlos

  > >
>> The summary would be:
>>
>> - This is not part of the spectrumDM.
>> - The provenance DM (where it seems to belong conceptually)
>> does not consider these kind of information right now and it
>> is difficult because, in real life, each model will use
>> different parameters.
>> - It is not either the purpose of SimDB going to such detail
>> level (although it could be extended, but should be discussed
>> if it's worth to do so)
>>
>> But, at the same time, we are told that the case of
>> theoretical spectra is already considered in SSAP
>> (+SpectrumDM + ObsDM) and that we shouldn't discuss much about it...
>>
> This is the old story indeed. But this is what we are now discussing,
> whether that answer is still correct.
> So don't give up yet!
>
> Gerard
>
>
>
>



More information about the dm mailing list