Re: [rdawds_workflows_wg] [rdawds_workflows_wg] AW: [rdawds_workflows_wg] Development...

09 Nov 2015

I found a copy
at: file:///C:/Users/Gary%20Bergcross/Downloads/Sacchi_ASIST2010.pdf
FINDINGS (they note some agreements but variations are noted since things
like purpose and relatedness may different between groups there will be
Our examination of some explicit definitions of dataset in scientific and
technical literature reveals that:
1. There are common themes in dataset definitions, suggesting that there is
some degree of agreement and shared understanding, at least at a high level
of generality.
2. More specifically, these definitions usually exhibit in some form these
four characterizing features: *grouping, content, relatedness, and purpose*.
3. Although definitions of dataset do appear to fit a common pattern, with
recurring phrases and semantically similar terms, it is clear that there is
no single well-defined concept of dataset. The variations in individual
terms are significant, the terms themselves are often used in different
senses, and critical characteristics are left underdetermined.
4. In particular there is uncertainty as to the ontological status of
datasets, and considerable ambiguity and conflation with respect to level
of abstraction of dataset contents.
5. It is clear from the forgoing that the notion of “dataset” found in the
literature cannot itself be provided with a precise formal definition, but
that this general notion must be replaced by an interrelated family of more
specific concepts.
Gary Berg-Cross, Ph.D.
SOCoP Executive Secretary
Independent Consultant
Potomac, MD