[Esip-preserve] Some Thoughts on ACCESS Presentations

[Ruth Duerr]

Hi Bruce,

That may be the case with the one project we've heard a little about; but, do we know that that is true for the other 4?

- Ruth

On Sep 9, 2010, at 1:06 PM, alicebarkstrom at verizon.net wrote:

> Perhaps.  However, we have already had an indication that
> they are not considering high-throughput production, and
> are apparently interested in ad hoc or exploratory production.
> Since production involves input data, processes, and output
> data, we already know that much about what this group
> wants to do.  The questions seem to me to be sensible ways
> of improving our knowledge of what they're going to have to
> deal with in creating production history provenance.  Why
> not ask the question and see what kinds of responses
> we get?
>  
> Bruce B.
> 
> 
> Sep 9, 2010 01:37:17 PM, rduerr at nsidc.org wrote:
> >Hi Bruce,
> >
> >I think having a set of common items presented, if we manage to get the ACCESS 
> >projects all to come, is a good one. However, it isn't obvious to me that we 
> >know enough about the goals of each of the projects to know what an applicable set 
> >of topics would be.
> >
> >Ruth
> >
> >On Sep 9, 2010, at 8:54 AM, alicebarkstrom at verizon.net wrote:
> >
> >> I think it would be helpful to request some
> >> common information from ACCESS presentations
> >> on Provenance. For example
> >> 
> >> 1. For what kind of data is the provenance being reported:
> >> images, vertical temperature or composition profiles,
> >> point in situ measurements, or something else?
> >> 
> >> 2. How much data is being input - or, perhaps more
> >> usefully, how many files are being ingested?
> >> 
> >> 3. How much data is being output - or, again, perhaps
> >> more usefully, how many entries are there in the
> >> results sets for a session?
> >> 
> >> 4. How are the workflows put into the provenance
> >> tracker: graphical workflow session, automated
> >> generation (and if so how), text file, or some other
> >> means?
> >> 
> >> 5. How are the processes in the workflow made
> >> understandable: "code" listings, reference to particular
> >> documentation in the tool, or some other means?
> >> 
> >> 6. Can the workflow from a session be recorded and
> >> edited for later use? Could they be converted to
> >> high-throughput use?
> >> 
> >> 7. Are provenance sessions used as a stand-alone
> >> interactions with data, or are they used in some ordered
> >> fashion in fulfilling a larger role in a project, for example,
> >> as part of validation?
> >> 
> >> I think if we can agree on some common items to
> >> have presented in the January meeting, we will find
> >> it much easier to understand what these projects
> >> are doing and save everyone time later when we
> >> might need to write up a comparison of the end
> >> results.
> >> 
> >> Bruce B.
> >
> >

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.lists.esipfed.org/pipermail/esip-preserve/attachments/20100909/9085a9bb/attachment.html>