Time Domain presentation at CSP

[John Swinbank]

Hi Enrique et al,

> On 16 Jun 2015, at 16:04, esm <esm at cab.inta-csic.es> wrote:
> 
> FYI, attached you'll find a couple of documents that will be presented at the CSP meeting on Thursday

Thanks for these. It’s exciting to see folks exploring the possibilities for publishing time series data in the VO, and this is certainly the most concrete attempt to sum up the current situation that I’ve seen.

My own involvement with most of the VO’s work to date has been quite limited: my interests have really been focused exclusively on VOEvent, so I am not familiar with the detailed work in data modelling that have been undertaken elsewhere. Given that, my approach is naïve, and for that I apologise.

That said: what I wonder about, both when looking at Enrique’s document and when thinking about the issue in general, is what requirements we have for representing time series data. To take some examples:

- Enrique’s note (\S 2.6) points out that there’s no “DatasetType” parameter in SimpleTimeSeries. I’m sure that’s true, but one might also regard such a parameter as redundant (if I’m looking at a SimpleTimeSeries, there’s probably a pretty good chance that it involves time series data).

- At the recent Hotwired meeting, there was a fair bit of discussion about handling event notifications, but nothing much about the details of how to access time series data. The average astronomer regards it as “just” fetching a list of numbers from a database. Clearly that’s inadequate, but what’s _actually_ required to meet their needs? Why doesn’t the community regard this as a problem?

- Addressing the same problem from the other end: time series data is clearly not new. What prior work has been done by data centres in making it available to their consumers? It’s really awesome to see this work on the OMC Archive, but what approaches have others taken? — if none, why not? If SSAP was previously being used, what are its deficiencies?

To clarify, it’s not my intent to suggest that Enrique should attempt to cover all this in his presentation tomorrow. I’m also a bit worried that the above might sound as though I’m indicating this work isn’t important, which is absolutely not my intention — I just don’t myself feel like I have a good grasp of what we’re ultimately aiming for.

While idly looking around earlier, I came across http://wiki.ivoa.net/internal/IVOA/LightCurves/STSP.pdf ("Time Series Data in the VO”, an early draft by Tody, dated 2010). I glanced through the mailing list archives, and it doesn’t look as though it was discussed here at the time. Does anybody know the history? Why wasn’t this work taken forward? Is there something there we should be building on?

Anyway, I’ve probably exposed myself as a raving lunatic now, but perhaps this will give us some food for discussion tomorrow!

Cheers,

John