July 11, 2001
Attending: Chris O'Dea, Cathy Imhoff, Ed Hopkins, Megan Donahue,
Niall Gaffney, Tim Kimball, Stefano Casertano, Rick White
Notes from Megan Donahue
Distillation of discussion (not minutes)
Group members had been asked to review and comment on a
few archive services, not necessarily HST. Those comments
have been collected and posted on the FASST website.
- Many sites, but difficult to find and difficult
to navigate. AstroBrowse may be a partial solution.
- Sites were usually accessible but not always very
- The time required to investigate even a single
moderate-sized site is long.
Sites of note (please check these out if you haven't, group
members recommended them as good sites):
Things we liked:
- Simplicity of use
- Usefulness without a lot of frills (practical output)
- Things whose use was quite obvious (i.e. *why* we would
want to use this service was immediately apparent.)
- I would like to search more than one catalog at once.
- I would like to be able to take the results of one
query and effortlessly make a query to another service.
- I would like to be able to make basic plots with the
catalogs before or after filtering them.
Some partial solutions may already exist but they are not well-known
- Astronomical Markup Language /XML
A wish list is emerging based on the limitations listed above
- A service with access to both observations and catalogs.
- A way to plot catalog columns without downloading a file.
- A way to take the results of one query and create the
query of the next service.
- A portal of archive data services with...
- ... a information "map" of archive services (with wavelength,
data type, sky coverage, service types, access protocols...)
- A customizable portal (myastronomy.edu or something like
that) where astroph lists are ordered by keyword preference,
observation or public availability status are reported, favorite
A data portal emerged as the kind of project that we might be able
to do short term and build on long term, because finding these services
can be as difficult as inventing them yourself. We might consider
doing this under the ADCCC (Astrophysics Data Centers Coordinating
Council) web site to encourage participation from the ADCCC members.
Ed Hopkins suggested we look at an example of an NIH portal he
thought was well done:
Niall will find and send oceanography/earth science portals (if they're
Next meeting: 7-18-2001. The wish list is being developed. We'll
very quickly (<15 min) review where we are now and what is in the
queue for the next 3-6 months. We'll complete the wish list, perhaps
taking a cue from the science cases discussed in the first two meetings.
Review the current wish list, add to it as desired. One subject we
discussed in the early meeting was integration of "user data" - data
either from space or from the ground, from individual observers. Short
term: how should we serve HST "user data" (or should we?)
Proposed future agenda items:
- Rate the wish list on technical grounds: been done, been tried, has
partial solution (where), easy to do, hard to do.
- Rate the wish list on scientific usefulness: how much time saved,
how often would it be used by a given individual, by whom and
how many would it be used, scientific payoff, scientific necessity.
- Identify technical dependencies of the wish list.
- Prioritize list based on a combination of technical and scientific
- Make recommendations.
Once we have completed this task, we have largely completed our
goal as a working group. We will put our review and recommendations
in a white paper for the archive and the data systems team,
as well as ACDSD and ESS divisions to share at the MC level.