[Esip-preserve] Identifiers

[Bruce Barkstrom]

Bravo - a real discussion that begins to identify the meanings
of a term.

Maybe one way of dealing with this diversity is to adopt the
model dictionaries use:
1.  Spell out the term, e.g. "data set".
2.  Put different definitions in as numbered aliases.  It looks like
we've got

Data set:
   1.  a data granule - meaning a file of data
   2.  a data granule with supporting information, including a file of
      data, as well as supporting metadata and documentation
   3.  a logical collection of data (which means, I guess, a collection
      of data with a written out organization, perhaps hierarchical).

This preliminary set of definitions hasn't been checked against the
full discourse.  However, the approach does help stabilize the data
structures of things we create, which should be a big help at avoiding
ambiguity.  I think this kind of list would be improved by also adopting
the OED approach of providing examples of the way the term is
employed in various contexts.  The notes on MODIS usage are
helpful.  I'll see if I can come up with some examples from ERBE
and CERES - and perhaps the precip data in the NOAA GHCN.

I think putting together a dictionary with this kind of familiar recognition
of aliases would be a better approach than looking for a rigid consensus
on a single definition.  The examples also allow other communities
to contribute their meanings and uses.

BTW, there's an interesting issue of the same nature about a couple
of other terms:
a.  What is a "datum"?  I think we usually expect it to be a numerical
value.  But what is the equivalent term for a biological specimen or
the stratigraphic name for a geologic feature in a rock core?
b.  What do we call the "measured quantity"?  The GUM (the ISO
standard on the Statement of Precision) calls it the "measurand".
Earlier, I was thinking of the term "parameter".  However, I'm a bit
uncomfortable with "parameter" as the "measurand" for a biological
specimen.

Thanks for the discussion.

Bruce B.

On Thu, Feb 16, 2012 at 12:28 PM, Curt Tilmes <Curt.Tilmes at nasa.gov> wrote:
> On 02/16/2012 11:58 AM, Mark A. Parsons wrote:
>>
>> My last thought on the intractable "what is a data set" question. FRBR may
>> help.
>>
>> In separate but related correspondence with Joe Hourcle, Joe writes: [...]
>
>
> Joe has been trying to define (or capture definitions of) a lot of
> terms.  See:
> http://vso1.nascom.nasa.gov/vso/misc/vocab_handout_HDMC2011.pdf
>
> Anyway, set aside data for now.
>
> What about all the 'other' stuff?
>
> People?  Organizations?  Sensors/Instruments?  Models?
> Systems?
>
> Is that a reasonable set of categories, and names for those
> categories?  What sort of identifiers should we use for things in
> those categories?
>
>
> --
> Curt Tilmes
> U.S. Global Change Research Program
> 1717 Pennsylvania Avenue NW, Suite 250
> Washington, D.C. 20006, USA
>
> +1 202-419-3479 (office)
> +1 443-987-6228 (cell)
>
> _______________________________________________
> Esip-preserve mailing list
> Esip-preserve at lists.esipfed.org
> http://www.lists.esipfed.org/mailman/listinfo/esip-preserve