Concluding Remarks Paolo Padovani Head, Virtual Observatory Project Office, ESO, Garching bei München, Germany & EURO-VO Facility Centre Scientist & Multi-wavelength Astronomer

Disclaimer This is a personal, biased view of the Workshop, with no claim of completeness (I have done my best!) All views expressed here are the author’s only and do not reflect the views of the EURO-VO, IVOA, ESO, etc.

What the VO is NOT responsible for Producing data  data providers’ job Data processing  data providers’ job Data quality  data providers’ job (but needs to check metadata quality!) Producing specific tools  tools writers’ job (but can help in porting to the VO)

Multi-wavelength Astronomy: VO needs VO Science (whatever that means) is a reality Basic VO need: easy access to all available data at all wavelengths in calibrated format! Still work to do: access so far only “simple” (somewhat hard to pre-select); relevant meta-data not always searchable: e.g., texp for spectral tools, observing date [mostly VO problem, but meta-data need to be there!] still very few “science-ready” data [data providers’ problem] yet, minimal raw data access (needed, e.g., for large [re-]processing/archival projects) [data providers’ problem] Main message from multi-wavelength astronomers: more data in the VO! Bright future: ASTRONET Roadmap report and questionnaire

there will never be a “killer” application VO Tools and Standards VO tools have evolved a lot in the past few years One of the main recent developments (thanks to EURO-VO!): very easy “click of a mouse” interoperability between tools (e.g., Aladin can “send” a catalogue to TOPCAT or a spectrum to VOSpec) Problems/still missing things: too many/too complicated VO tools VO searches for lists of sources, massive scaling: scripting for, e.g, cutout services or access to thousand of spectra at the same time “quick” check on all available data (raw data access!) standard for filter/response curve + zero point: magnitude to flux conversion  vital for Spectral Energy Distribution (SED) building! SED “on-the-fly” from spectra and catalogues (at least!) Table Access Protocol (soon to come: Noddle, p.c.) top level “VO portal” there will never be a “killer” application

Surveys: VO needs 1 Medium/large surveys (e.g., COSMOS) have less pressing “standard” VO needs, as all their data are local; moreover, no VO tool will ever be able to compete with their own tools VO tools “nice to have” to maximize exploitation: e.g., to allow easy match between imaging and spectral data Large surveys are also on the “provider” side of the VO: need tools for data publishing [available]

Surveys: VO needs 2 Top three needs from round table discussion: Data coverage: was this position ever observed and, if so, is there an X- detection or not? catalogue available: easy reduced data available: difficult (on-the-fly Sextraction?) raw data: extremely hard [impossible?] (on-the-fly reduction first!) Quantitative quality information: distinction needed metadata quality: VO should check that relevant information is provided and make an effort to check that it is correct data quality: data providers’ task! This is NOT a VO issue but a fact of life: there are (and there have always been!) good data and bad data. We have refereed papers for that Table Access Protocol: very important, should be done soon

More information on available VO tools SEDs, SEDs, SEDs! VO near Future VO tools should take advantage of all available metadata (if more needed, they should be included); quality control on metadata information More information on available VO tools SEDs, SEDs, SEDs! Catalogues, catalogues, catalogues! Searches for lists of sources, scaling, scripting Raw data should be included VO Portal, VO Portal, VO Portal!

(My) Vision for the (Petabyte) Future Larger telescopes and more complex instruments (e.g., ELTs, LSST) Data reduction will (almost) become a thing of the past for the average astronomer (see VISTA now!) Astronomers will deal mostly with catalogues and science-ready data All of the above better be well-described and interoperable, otherwise impossible to make sense of Deviants will be weeded out and go extinct (do not underestimate the power of natural selection) Same for the many tools: usage will decide which will survive Same for astronomers: “dinosaurs” will disappear Nobody will remember what the VO was, although everybody will be using it!

On behalf of the EURO-VO and of the Science and Technical Organizing Committee a big Thank you to all participants for their input and contributions