[Fwd: [Fwd: Small, scruffy provenance profile?]]

[This is my THIRD attempt to send these comments to the prov working
group.  The primer at http://www.w3.org/TR/prov-primer/ says to send
comments to public-prov-wg@w3.org , but since that has not worked I'm
now trying public-prov-wg@w3.org ]

-------- Forwarded Message --------
From: David Booth <david@dbooth.org>
To: public-prov-wg <public-prov-wg@w3.org>
Subject: [Fwd: Small, scruffy provenance profile?]
Date: Fri, 14 Sep 2012 12:01:09 -0400

FYI, this is a comment that I sent to the working group last February,
though apparently it never got there, as I don't see it in the archives.
I'm not sure that it makes sense now -- seven months later -- to
consider this as an actual request, since the working group is much
farther along now, but I think it is still worth reading.  The primer at
http://www.w3.org/TR/2012/WD-prov-primer-20120724/
is definitely a good step in this direction.  Thanks!

-------- Forwarded Message --------
From: David Booth <david@dbooth.org>
To: public-prov-wg@w3.org
Subject: Small, scruffy provenance profile?
Date: Thu, 23 Feb 2012 17:01:49 -0500

I find myself intimidated by the size and complexity of the current
provenance work -- to the point that I am tempted to invent my own small
ontology instead, which would be very unfortunate.

Could we please have a small, scruffy provenance profile that could be
used in the many situations where great detail and precise semantics are
not needed?  I'm sure the existing detail and precision are important to
some applications.  I appreciate the effort involved in figuring them
out, and I'm not suggesting discarding this work.  But a small profile
might allow both needs to be addressed.

As a comparative example, the big success of SKOS is due to its
simplicity: it is un-intimidating.  It can be used by the masses for
lots of simple things.  (And as Jim Hendler famously said, "A little bit
of semantics goes a long way".)

So . . . how about a small, simple profile that does not require a
tutorial, does not require learning a new abstract syntax or data model,
and does not require the user to study its formal semantics (for fear of
using it wrong)?




-- 
David Booth, Ph.D.
http://dbooth.org/

Opinions expressed herein are those of the author and do not necessarily
reflect those of his employer.

Received on Thursday, 27 September 2012 15:33:36 UTC