[csp] Any update re: VO sprint?

Mon Apr 8 19:03:34 CEST 2019

Thanks very much for this Tom,

Do you think you could share these ideas with the astropy people to see which kind of project would they be more interested on eventually? By the way, who are they?

To your list I would add one that has to do with the CSP goal of having a truly global portal to all data, something that would allow users to request e.g. HST data and would serve it from the server nearest to them without the user having to know where each datasets are stored. As a plus, such a module could have a taxonomic description of the services it connects to such that users could search for all available data of certain type in certain wavelength range..

Cheers

Bruno

----- Original Message -----
From: Tom Donaldson <tdonaldson at stsci.edu>
To: Bruno Merin <bmerin at sciops.esa.int>, Patrick Dowler <pdowler.cadc at gmail.com>
Cc: Mark Allen <mark.allen at astro.unistra.fr>, Janet Evans <janet at cfa.harvard.edu>, Baptiste Cecconi <baptiste.cecconi at obspm.fr>, Markus Demleitner <msdemlei at ari.uni-heidelberg.de>, AdaNebot <ada.nebot at astro.unistra.fr>
Sent: Mon, 08 Apr 2019 18:36:41 +0200 (CEST)
Subject: Re: Any update re: VO sprint?

Hi Bruno,

I forwarded the confirmation to the Astropy folks on Friday, so I believe they are starting their travel arrangements.

I’m CC-ing Ada and Markus as they have interest in these topics. Here is a non-comprehensive list of possible sprint topics that I’ve heard of or thought of:

  *   Generalized TAP utility in Astroquery that supports upload and authentication, and fixes the long-standing issue where queries hang with services that don’t support auto-start (https://github.com/astropy/astroquery/issues/1160).
     *   This topic could resolve/coordinate the work of CADC and ESA where there are some stalled pull requests where both are trying to address these and other issues.
        *   https://github.com/astropy/astroquery/pull/1240
        *   https://github.com/astropy/astroquery/pull/1266

  *   VO Table parser items (the first two or three I have looked at a little, and may make progress on prior to the Interop):

     *   Astropy votable parser discards non-FIELD metadata (https://github.com/astropy/astropy/issues/7759)
        *   This will be important if we want to utilize TIMESYS or COOSYS during the sprint or after!

     *   Improve Astropy SkyCoord.guess_from_table
        *   SkyCoord.guess_from_table looks at the column names in an astropy table, then tries to determine which columns define the position from the given coordinate frame (defaults to ICRS).  See if the guesswork can be improved when the astropy Table was created from a VOTable, because the VOTable contains more metadata about the columns.
           *   It turns out that finding RA, Dec, RA proper motion and Dec PM are fairly straightforward.  More difficult is determining the correct coordinate frame.

     *   Support VOTable 1.4.
        *   Using this will require the metadata handling mentioned above.

     *   Have char array table data be instantiated as strings instead of byte strings.

  *   NAVO pull request to add a registry query capability to Astroquery, and back it with generalized Image and Cone queries, while integrating with the existing mostly-generalized TAP and Cone queries already there (https://github.com/astropy/astroquery/pull/1266)

  *   Possible use cases with time domain data.  I’m relying on the experts to fill in the details here, but figuring out how to map from actual time domain data to Astropy data structures could inform:
     *   Our data modelling
     *   VOTable 1.4 use cases and implementation
     *   VO-DML mapping of our models in VOTable and Astropy
     *   Likely related, this item showed up on STScI Slack this morning:
        *   *HELP WANTED* -- Astropy Project is planning to move `timeseries` package from its own space (https://astropy-timeseries.readthedocs.io) into `astropy` core for its upcoming 3.2 release. If you deal with time series data (exoplanet ppl?), you might be interest. What we want is feedback on API design and whether the doc is sufficient before it is officially released. Please leave your feedback at https://github.com/astropy/astropy/pull/8540

Regards,
Tom

From: Bruno Merin <bmerin at sciops.esa.int>
Date: Monday, April 8, 2019 at 3:15 AM
To: Patrick Dowler <pdowler.cadc at gmail.com>
Cc: Tom Donaldson <tdonaldson at stsci.edu>, Mark Allen <mark.allen at astro.unistra.fr>, Janet Evans <janet at cfa.harvard.edu>, Baptiste Cecconi <baptiste.cecconi at obspm.fr>
Subject: Re: Any update re: VO sprint?

Thanks Pat!

Great to hear that we have a room for this and that it will be small. We should communicate this to them asap to maximize the chances that they do come in person. I will figure out who might be the person form ESAC to join that session and let you know before the meeting (it depends slightly on what the topic of the sprint might be). Since Mark is already interacting with them, should you continue to be our spoke person to coordinate this Mark?

Regarding the topics to address in these sessions, I will certainly propose to discuss it at the next CSP telecon this Thursday at 4 pm CEST where I will invite Tom Robitaille, Erik Tollerund and Adam Ginsburg in case they want to join. If there would not be available for any reason, we have another TCG and another CSP telecon before the Paris meeting to do the same. Otherwise, I guess we would have to discuss the goals of the sprint there with them directly, which is also fine of course.

Cheers,

Bruno

On Apr 5, 2019, at 9:53 PM, Patrick Dowler <pdowler.cadc at gmail.com<mailto:pdowler.cadc at gmail.com>> wrote:

Some extra details about what has evolved...  (added Bruno since this email evolved from room organisation to include other details)

- there will be a small room for 8-12 hackers (astropy folk won't be outnumbered :-)
- we will have to somehow limit who can participate so I am proposing that we set a good example by having organisations (eg CADC, CDS, etc) put forward a single hacker each; that should help ease any possible tension from someone  having to pick and say yes or no (who decides? TBD)
- there will be 4 x 1.5 hours on Wed and Thursday AM for this, plus whatever additional time people want to spend on the side

I ran these ideas past Adrian (the lead python dev from CADC that is coming to IVOA mainly for this and who already does astropy work) and he agreed it is a good amount of time to get stuff done.

Janet and I will try to minimise the impact of this running in parallel with two other sessions, but since it effects ~5 ivoa hackers that should not be a problem. I am still interested in contributing to the goals and topics that could be included... not sure when/how that gets discussed and how to involve astropy people in that... TCG and/or CSP telecons? face to face?

PS-Bruno: can you talk to others at ESAC and figure out who that single hacker from ESAC will be?

PS-Tom: can you pass along that astropy people need to register for the meeting (and the optional banquet if they are interested in that).

cheers,

--
Patrick Dowler
Canadian Astronomy Data Centre
Victoria, BC, Canada

On Fri, 5 Apr 2019 at 12:29, Tom Donaldson <tdonaldson at stsci.edu<mailto:tdonaldson at stsci.edu>> wrote:
Thank you, Mark.  That's great news.  I've forwarded this and the meeting information to Erik who will notify the rest of the group.

Cheers,
Tom

On 4/5/19, 2:50 PM, "Mark Allen" <mark.allen at astro.unistra.fr<mailto:mark.allen at astro.unistra.fr>> wrote:

    Dear Tom,

    The Astropy/VO sprint sessions are confirmed for Wednesday and Thursday.

    Please tell Erik.

    best regards,
    Mark

    On 4 Apr 2019, at 23:58, Tom Donaldson <tdonaldson at stsci.edu<mailto:tdonaldson at stsci.edu>> wrote:

    Hi Mark and Pat,

    Any chance we can confirm the meeting for the Astropy folks.  I sense concern on their side, and it’s late in the game for some US travel procedures.  An email from Mark would be ideal.

    Thanks,
    Tom
    ________________________________________
    From: Erik Tollerud
    Sent: Thursday, April 4, 2019 4:02:48 PM
    To: Tom Donaldson
    Subject: Any update re: VO sprint?

    Hi Tom,

    Should I have heard something about the ivoa sprint space yet? And if so, from who?

    ---

    Erik Tollerud

    Mark Allen
    mark.allen at astro.unistra.fr<mailto:mark.allen at astro.unistra.fr>

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).

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).