[csp] Astropy - IVOA - pyVO follow-up online discussion in May

Wed May 19 21:48:25 CEST 2021

To echo Brigitta, I agree the Paris workshop went great, and I think trying to replicate that to whatever extent is possible in the virtual format is a great idea!

Maybe we should go into it with a list of questions we want to try to answer?  Ideally that provides a bit more concreteness to a space where in a virtual format it's rather easy to drift off into doing separate things on our computers, defeating the point of a meeting. The obvious ones I see so far are:

  *   Where TAP/TAP+ belongs between astroquery and pyvo (might need more refining to be a productive discussion!)
  *    VOTable metadata
  *   (maybe) Coordinated status or not

Perhaps some of those "other smaller things" Brigitta mentioned can be done more in hack-day format? I realize that's not on your schedule Bruno, but to replicate what we saw as success of the Paris event, perhaps we should do that? Perhaps then the discussion segment of the 60 min on the schedule can be discussion of the above items + planning for "hack day" activities to be done async?

On the question of coordinated, I think the question is kind of the other way around: there is *no* extra bureaucracy that I can think of to having it be coordinated given the repo is already in the astropy org. It's really more whether it solves any problems the current package has.  From the Astropy Project perspective, having something be coordinated means the Project is a "backstop" against the current maintainer team disappearing - TBH I'm not sure we could commit to that now anyway since most of the Astropy expertise on the VO lives in pyvo already. In return the *users* of that package get some trust that something bigger than an individual is supporting it.  In PyVO's case, the IVOA might be just as big/trustworthy of an organization as Astropy would be. There are other smaller things like you get your names on the Astropy team page and get invites to some of the core team events, although in most cases those are in principle open anyway if you're interested enough.

So it might make sense to consider the coordinated pathway if we think having it be listed on a web site as coordinated will help, say, some of the people on this email chain convince their development team to have a more open process, or have some funding source be more likely to fund developers doing work in an open way.  Or if we think the user community is more likely to trust it if it's "coordinated" (which I doubt because I think most non-insiders don't know the distinction).  But if not, I agree it's not clear anyone is served by making it coordinated.  But if the pyvo maintainers want it, that opinion can change!

---
Erik Tollerud
Assistant Astronomer
Space Telescope Science Institute
________________________________
From: Brigitta Sipocz <bsipocz at gmail.com>
Sent: Wednesday, May 19, 2021 2:28:48 PM
To: Bruno Merin
Cc: Tom Donaldson; Erik Tollerud; Thomas Robitaille; Adam Ginsburg; TCG IVOA; Baptiste Cecconi; AdaNebot; Damian, Adrian; Christine Banek; Javier Duran; AstroPy Coordinators; csp
Subject: Re: Astropy - IVOA - pyVO follow-up online discussion in May

External Email - Use Caution
Dear All,

This is an excellent idea to follow-up on our previous meeting, and I'm planning to join, too. I personally feel the Paris workshop was a success, we saw really good momentum and more collaboration than we had before, but of course there are still a number of items that we can work on going forward.

As to address one of the points below, pyvo currently is an astropy affiliated package. And I feel the maintainer group is very well set up balancing between multiple stakeholders, at least from the sidelines it seems to me that it is working rather well. At the moment I don't see advantages just more bureaucracy of bringing in as a "coordinated" package, there isn't any extra offered for the coordinated ones compared to what pyvo has. We can discuss the nuances, bottom line is that currently bureaucracy what keeps the gates closed.

I agree with Tom that one of the bigger outstanding questions is TAP. What features should get to be included in the pyvo version (or in the standard) to be able to deprecate the parallel implementation from astroquery. I'm also aware of smaller things that could go into pyvo from either astroquery or astropy, but I think those are more straightforward cases that don't need much coordination from this larger group.

Cheers,
 Brigitta

On Thu, 22 Apr 2021 at 10:40, Bruno Merin <bruno.merin at sciops.esa.int<mailto:bruno.merin at sciops.esa.int>> wrote:
Dear all,

I hope my email finds you all and your families well and with good health.

It is now roughly two years since some of us got together at a nice hackathon in Paris between astropy, astroquery and IVOA software engineers to identify possible synergies and to try to put a collaboration on motion.

In preparation for the upcoming IVOA online interop meeting from the 25th to the 28th of May, we would like to ask whether you could be available for a follow-up discussion of approx. 1.5 hours, on what worked well, what is ongoing, and what could still do to better for the benefit of the astronomical community as a whole. Among other things, we would like to understand what is pending for pyvo to become an astropy affiliated package and to get your advice on what could be the best use of people’s times in this collaboration or by contributing to astropy.

Please share your thoughts about this conversation and this collaboration via email and let me know whether you could participate in a session, whose exact date and time is still TBD.

Many thanks in advance for your time and I look forward seeing you soon,

Bruno

ESA - European Space Agency

Dr. Bruno Merín
Head of the ESAC Science Data Centre (SCI-SAA)

European Space Astronomy Centre (ESAC)
Camino Bajo del Castillo s/n
Villanueva de la Cañada, 28692 Madrid, Spain

Bruno.Merin at esa.int <mailto:bruno.merin at esa.int> | cosmos.esa.int/esdc<https://urldefense.com/v3/__http://cosmos.esa.int/esdc__;!!CrWY41Z8OgsX0i-WU-0LuAcUu2o!jUyxIG7M2mwZ4C4j1U5-eDCgxNfrOFOLuiBbINDnWPEIVfgPlU0r_8zymN-8Lp8ClQ$>
Phone: +34 91 813 1456 | Fax: +34 91 813 1218

On 20 Mar 2019, at 18:32, Bruno Merin <bmerin at sciops.esa.int<mailto:bmerin at sciops.esa.int>> wrote:

Dear Tom, Erik, Thomas, Brigitta and Adam,

We are planning to organize a hackathon on May 15th 2019 in the afternoon at our Paris IVOA interop meeting<https://urldefense.com/v3/__https://indico.obspm.fr/event/64/overview__;!!CrWY41Z8OgsX0i-WU-0LuAcUu2o!jUyxIG7M2mwZ4C4j1U5-eDCgxNfrOFOLuiBbINDnWPEIVfgPlU0r_8zymN_R4-8UXA$> where we would like to invite people to do hacks that explore the links between astropy/astroquery and the VO infrastructure, in particular focussing possibly on time-domain use-cases.

Ada Nebot, who is coordinating the IVOA time domain interest group (TDIG), she proposed the following two areas of possible interest / overlap although there might be others:

1 - VOTable <—> astropy: Conversions between VOTables with the TIMESYS element and the astropy.time class.
VOTable: http://www.ivoa.net/documents/VOTable/20190318/<https://urldefense.com/v3/__http://www.ivoa.net/documents/VOTable/20190318/__;!!CrWY41Z8OgsX0i-WU-0LuAcUu2o!jUyxIG7M2mwZ4C4j1U5-eDCgxNfrOFOLuiBbINDnWPEIVfgPlU0r_8zymN_eeZoJUg$>
astropy.time http://docs.astropy.org/en/stable/time/index.html<https://urldefense.com/v3/__http://docs.astropy.org/en/stable/time/index.html__;!!CrWY41Z8OgsX0i-WU-0LuAcUu2o!jUyxIG7M2mwZ4C4j1U5-eDCgxNfrOFOLuiBbINDnWPEIVfgPlU0r_8zymN_4ujnxvA$>

2 - T-MOC & astropy : The T-MOC format is a time-dependent Multi-Order Coverage structure, e.g. a healpix structure that indexes spatially the existing data in the sky (that is the MOC), but chopped off in time intervals to allow easy visualization and access to data in both space and time (that is the T-MOC).
https://wiki.ivoa.net/internal/IVOA/InterOpMay2018TDIG/TDIG-TBoch-TMOC-MOCPy.pdf<https://urldefense.com/v3/__https://wiki.ivoa.net/internal/IVOA/InterOpMay2018TDIG/TDIG-TBoch-TMOC-MOCPy.pdf__;!!CrWY41Z8OgsX0i-WU-0LuAcUu2o!jUyxIG7M2mwZ4C4j1U5-eDCgxNfrOFOLuiBbINDnWPEIVfgPlU0r_8zymN_YuF9fmw$>
https://github.com/cds-astro/mocpy/blob/master/notebooks/tmoc.ipynb<https://urldefense.com/v3/__https://github.com/cds-astro/mocpy/blob/master/notebooks/tmoc.ipynb__;!!CrWY41Z8OgsX0i-WU-0LuAcUu2o!jUyxIG7M2mwZ4C4j1U5-eDCgxNfrOFOLuiBbINDnWPEIVfgPlU0r_8zymN_RitrISg$>

The reason of my message is to inform you about this hackathon, ask for your feedback on whether this would be of interest or not for some of the astropy developers and to ask for advice on how do you think this event could be more useful to both the astropy users and IVOA users. It would also be great if you would know some astropy contributor who might be based in Paris and might want to come in person to the session.

Finally, should you be interested, you are all of course invited to join any of the two telecons described below where we will discuss what to prepare and how to organize the hackathon.

Thanks in advance for your time,

Bruno

This message is intended only for the recipient(s) named above. It may contain proprietary information and/or
protected content. Any unauthorised disclosure, use, retention or dissemination is prohibited. If you have received
this e-mail in error, please notify the sender immediately. ESA applies appropriate organisational measures to protect
personal data, in case of data privacy queries, please contact the ESA Data Protection Officer (dpo at esa.int<mailto:dpo at esa.int>).

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.ivoa.net/pipermail/csp/attachments/20210519/3bf9633d/attachment-0001.html>