At the call: Raffaele, Mark Lacy, Mark Allen, Janet, Chao Liu, Pepi, Gregory, Bruno

Agenda:
1.- Welcome

2.- Debriefing from College Park interop

MA: reported that the IVOA is concerned on whether the CSP if following a standard procedure 
w.r.t. TCG and the processes.

ML: also thinks that the CSP is slightly disconnected from the TCG and the other technical 
groups.

PF: It might be useful to have a review of the current work of the TCG and science priorities as
seen by the TCG and report the science priorities to the TCG and to the exec. It might pay off
to be a bit more formal in that respect.

JE: The technical groups would need use-cases to be presented, beyond just the science 
priorities, which are a lot more useful  

3.- Status on portal concept (data discoverability)

MA: I think we should clarify what are we talking about data discoverability. The example
of the multi-dimensional search (which can be considered mostly closed now) is a good one
to follow for providing new use cases, priorities and minimum requirements. This helped
interacting with the TCG a lot and should be used.

JE: maybe we can declare the Multi-D as accomplished indeed, but that statement should 
be made explicitly to the TCG such that it can be understood what is the final goal of
the CSP and TCG interactions.

MA: I could help drafting some small document explaining how all this was done and what
was the outcome of the process for the TCG.

PF: It would be also useful to mention that as an article for the newsletter.

MA: the time-domain interest group produced a set of minimum requirements for time-series
with use cases that motivated the TIMESYS proposals with the IVOA note and the presentation
in College Park. The basic goal is to be able to compare two time-series of different
origins. It would have been useful for the CSP to interact more with the time-series.

JE: I would second MA in saying that we need both the vision arguments as much as the
day to day use-cases for implementation.

4.- Who does what for Paris interop

	4.1.- Focus session

		MA: two focus session have been organized earlier, focussing on the priority
			areas of the IVOA, multi-D and survey science. We contacted external
			projects to get their engagement. The approach was to highlight our
			priorities and engage with big projects. It is not clear that we have
			a similar approach for Paris.

	4.2.- Python hackathon

		PF: the idea for this might be to contact with interested people who want
			to prototype some python workflow in time-domain data from IVOA resources
		RD: I think we should try to obtain to participation from people external 
			to the IVOA in Paris to join to the hackathon.
		MA: It was not clear to me whether we needed to organize two things or just
			the hackathon, but it seems that there is more interest in the hackathon.
			Erik Tollerund seems to be OK about the hackathon but asked about having 
			a “smallish” hackathon so we should be careful not to overwhelm them.
			He might be thinking about a 5 to 10 people.
		MA: Tim ‘Jenness’ participates in the astropy so he could be someone who could
			contribute to this discussion.
		GDF: Tom Donaldson and others organized a session on the AAS meetings on how
			to access all NASA archives from python and they have interest 
		GDF: It might work better if it arises as a something organic that we negotiate
			with them instead of something that the IVOA proposes initially.
		GDF: I have e.g. noticed that there is little python support for SIAP 2.0 and
			I could report to the CSP by email in the upcoming weeks.	
		GDF: Maybe for the second hackathon I would like to propose two use-cases, one
			associated with TESS and another one. People put new services for astronomical
			data not following the IVOA patterns that we have worked out. And now
			the question is how can we make “lower friction” paths to those
			services to be part of the VO. I was looking at packages available to 
			assist people with that as described in the “publishing data in the VO
			twice page” but some of them might not be working and there is no obvious
			one for python so it might be useful to define such a thing for python, 
			since so many people seem to be using python today. In particular for
			any DALI service, and people supporting DataLink it would be very useful
			for people. In LSST we are planning to serve a SODA service in python
			and by summer we might have something to share with others. This would be
	 		marvelous new addition to the IVOA-python collaboration, essentially the
			datalink.

5.- Next meeting

BM: I will circulate a doodle for the mid-January including everyone related to the
time-domain hackathon.
MA: I would like to clarify whether we are organizing a focus session and hackathon
or just the latter.

6.- AOB