WorkingTogetherDiscussion

 

Aus-VO Workshop 2003

Working Together Discussion

chair: Brian Boyle

IVOA perspective - Bob Hanisch

  • everyone should go to the IVOA web page
  • IVOA wants to avoid duplication of effort, but capitalise on diversity of priorities and resources
  • recognise difference of national priorities, and exploit strengths of different approaches
  • develop standards: data access transparent across national boundaries
  • open environ. for s/w exchange; OpenSrc licensing
  • there is an IVOA roadmap: key technology areas needed for coming year -> set of working groups (8) to address these areas
  • participation in WGs is critical otherwise standards may not reflect needs of "us" -> Australian participation essential.

discussion

  • Neil asks if some top-down direction / architecture needs to be "enforced"? on partners
  • Bob points out that WGs have to overlap, and wants to avoid a top-heavy or dominant partner effect
  • Chris Tinney asks if their can be a centralized structure. Finds it all confusing, and perhaps wants some "central point of presence" (? check).
  • Bob responds that much education will be needed, giving eg. of tutorial at ADASS. Outreach just starting... Sess this as a respons. of IVOA.
  • Brian - most projects will down-select technology at some point. Should we "schedule" this for IVOA or "let it happen"?
  • Bob reiterates "iterative development" scheme, FITS is the example we don't want to follow.
  • Chris T. wonders when first product will roll out. Bob says US intends to have a product by next summer (2004 I gather). Get the grad students involved.
  • Chris T. is training needed - does that mean it's tough to use? Bob promotes a "jump start" (perhaps bootstrap) approach.
  • Brian notes that beyond 2005 roadmap to be released mid 2004, so now's the time for Australia to give input, via IVOA exec. or the WGs.

CVO perspective - David Schade

  • limited interactions within Canada. Lot's of empty space in Canada! Mgmt takes place at CADC.
  • get ppt slide from David summarising CVO.
  • funding $5M CAN over 5 yrs
  • 4 FTE going to 5 FTE start 2004
  • work with GAVO, Aus-VO
  • substantial h/w infrastructure.

discussion

  • Brian endorses the CVO product from the point of view of incorporating 2QZ data.

Aus-VO structure - Rachel Webster

  • funding structure 2003 and 2004.

discussion

  • Brian points out as we grow we need to be managed. eg. Science WG to oversee, how do we meet targets, review etc.
  • Rachel - agrees

Aus-VO collaborations - David Barnes

  • ausvo2003_wtcopy.ppt: barnes 3 page ppt file on collaborations
  • lists extant collaborations within and outwards from Aus-VO
  • level of collaboration very healthy for less than one yr operation

Aus-VO outcomes 2003 - Neil Killeen

  • endorses strength of collaborations
  • Web page
  • WiKi - Use It.
  • Melbourne: networking
      • HICAT & SUMSS online
      • integration of TOPCAT with skycat
      • portal for MHD studies
      • visualisation - volren + vrml
  • ATNF: ATCA on-line archive (Lamb, Power, Wark, Epping)
      • is online, prototype interface (ICT), reverse
      • engineering metadata, -> public soon, data policy etc.
      • Data processing pipeling - just starting work
      • Visualisation (Chandra) - generic framework, prototype tool, "two weeks away" says Anil!
      • WCS
  • Sydney: work contracted out to combine SUMSS image and catalogue databases
      • information system for SKAMP
      • work on correlator for SKAMP (Tim Adams)
      • generic infrastructure for Molonglo -> combine with new correlator
  • AAO: 2dFGRS, 2QZ, 6dF-EDR in VOTable format spectra via NED
      • RAVE design work to ensure VO compliance

discussion

  • David S. remarks that this is the most impressive effort he has seen!

Next year targets - ie. what we'll actually do - Rachel Webster

  • ppt slides

discussion

  • Brian notes that the range of 2004 work packages is diverse and are we moving to a distributed set of nodes?
  • Rachel suggests one approach is to think about an Austn. data centre.
  • Brian - we gain a lot by sending our fabulous data out into the world, in terms of exposure and usage
  • Peter Lamb - metadata looks like RPFITS headers! This is the easy path, but it's not a long term solution as it is pretty ATCA-specific. Working in metadata WG to do a radio-telescope data model, suitable for synthesis, mb single dish, phased array, etc. Maintenance an extremely important issue.
  • Rachel mentions "residual expertise".

General disc.

  • Michael D. asks when is a data set considered published? - answer: when in a registry
  • Bob H. - some comments:
    • funding agencies worry about bricks and mortar
    • nevertheless having a data mgmt strategy for Austn. data would be extremely useful.
    • four themes for Aus-VO:
      1. visualisation - only group doing signif. work.
      2. radio astro data - vacuum to be filled, software tlescope observations on demand, ...
      3. theory - impressed by Sarah's talk. One of two or three groups in the world.
      4. Australian registry for data products to be replicated overseas; data coming from original source rather than NED, ...

-- DavidBarnes - 19 Nov 2003