Imperial Mar 06 Workshop - Feedback and Q+A

End workshop Issues

Issues picked up

TopologiesImperialGroupA

• used the NED system to find high Z gals
  • some of the NED votable types were un-readable by TopCat

StellarProjectImperialGroupB

• If AstroScope is searched via SIMBAD, need to know what position is used (can be ambiguious for large or moving objects).
• need an easier way of discovering coverage of survey data - searches fail when no data - so helpful to not have to do the search on areas where there is no data.
  • Better error catching/reporting in Colour Cutter needed
• standards cross survey for magnitude/flux info
• fitting of models to data - in the
  • Dave Clemens mention Herschel ESTEC team who have java clases for doing model/observation fits
  • also check JeremyYates code (Huntfit3)
• MySpace browsers do not always refresh easily

FIREllipticalsGroupC

• Astrogrid not right now able to search the relevant Spitzer data sets or the JCMT data (reduced data doesn't exist there yet).
• Need a list driven search of FIR data resources
  • missing some of the popular catalogues of nearby galaxies - and some that where there were missing columns (e.g. RC3 from HEASARC)
  • need to be able to do full ADQL search of the catalogues (currently need to do this in Vizier)
• Science here needed ISO and Spitzer images (but only very limited Spitzer data visible through the VO).
• BUT: workflow script did search SDSS and WFS and return images to MySpace - with input being a RA,Dec list

• Discussion on the visibility of well calibrated data. Some like Scuba and JCMT i shard to work with and/or raw. However, Spitzer data have good data products, could make thiis available via the UK if needs be.

• BenRogers: would like to have access to more spectral data resources: resources pop up in AstroScope but are not then visible through say the workflow or application launcher.
  • VOSpec should be available at any stage not just via AstroScope

• DaveMorris nice to have ability to take a search step out of AstroScope and create a for loop round this to say repeat search for multiple RA,Dec - in the workflow.

ExtremelyRedObjectsImperialGroupD

• wanted to search in SDSSDR4 to find more ERO's than seen in DR3 (could also use USNO-B for all sky)
  • issue with the size of the databases
  • source selection R-K > 5 - need to federate data from two databases.

• difficult to do an easy combined search just now - problem with row limits from ROE - 10000 column limit on 2MASS

• for known ERO's was possible to quickly get images back from a range of sources.

• issues:
  • error reporting
  • cut and paste (clip board stds for differing OSs)
  • database federation - what are the plans - DAVe would like to know how we progress in this area.

• see slides from Dave Clements

from NicholasWalton

• searching the registry from the workbench - want to search the results as well as you can get a lot of returns.
• speed of response to registry queries - could be useful to have some sort of progress indicator as some searches can take a while.
• cone search spec - looked at the RC3 catalogue - held at HESEARC. query with 0,0 on a 180 deg radious - returned all 23.000 odd rows, but only 8 cols. the original cat - see e.g. Vizier has 40 odd cols. where did these go in the cone search return. the registry entry for this resource does not state that you get part (only) of the catalogue. need to tighten up spec?
• the workbench windows - need to be able to 'cut+paste' fromt hese into other windows (e.g. XP)
• general speed issues - hard to quantify but c.f. speed of saving to MySpace with local downloads - need to get comparative or better performance
• cone searches - with astroscope you can hoit many cone services. however, you can't do a cone search in a workflow - need somesort of ability to write in cone searches to the hundreds of these services in the workflow system
• need to be able to upload a source list and then use this as an input to astroscope, also to a workflow which say queries a range of cats, images. EGS has a prototype script, needs packaging.

• possibility of wrapping ImpZ as a CEA app - need to check with TomBabbedge at Imperial

• need to sort out online presence and feedback mechanisms for astro users (also how do they input bugs, enhancement request).

from AnitaRichards

• Some projects need all-sky searches (e.g. Gould Belt, nearby galaxies)
  • Federate massive datasets like 2MASS, USNO-B?
  • More pre-selection using ADQL?
• SED construction (Topcat?)
  • Need a generic magnitude-flux converter for large heterogenous samples
    • 'Rough' for non-experts, based on one or a very few standard systems
    • Editable optionally for experts/high accuracy
  • Similar service for common X-ray instruments
  • Need to exchange columns and rows

from JonathanTedds

• usability
  • must fix completely non-obvious need to create blank file in which to subsequently save data in MySpace.
  • error reporting - why has workflow failed? e.g. may not be clear if because no data in resource at specified position.
  • change label of "ref" checkbox in e.g param workflows to something users find more obvious! They often assume it is a "use/don't use" binary checkbox rather than prompt to select from file - confusion results.
  • plasticise VOspec
• AstroScope
  • indicate offsets in arcsecs on visualisation page
  • ability to "capture" workflow version of search executed in order to repeat, loop and form basis of more advanced science script
  • HST available but can't download?
  • improve survey coverage metadata
• AstroGrid News
  • webpage/email needed as part of front end to users and to regularly update as new features added

Q+A

• put short tutorials on the web
• tutorial walk throughs for simple problems, text plus pictures
• look at CIAO threads
• format of the workshop was okay
• workshop seen as useful as it brought the scientists together, they could also discuss science
• location wise - all in one room was good - nice that people can move around a bit

• AG will get back to attendees re new developments in AG

-- NicholasWalton - 22 Mar 2006