At 17:20 04/04/2012, Stracke, Christian wrote:
Dear all,
I'm sorry bit "jfc" is mixing up different standards and committees
(what is easy):
Sorry for the confusion. The problem with ISO and ISO documents is
that they are not documented on (i.e. interested in? i.e. interesting
for?) Wikipedia.
1. 19788-1, can be found for free at:
http://www.sis.se/PageFiles/2140/MLR-utkast%20arbetsmaterial.pdf
2. The Wikidata issue, as I see it, is that we do not start from an
architectural framework based on a business plan, charter or TOR. It
seems that the current target is to implement a W3C semantic web
current-Wikimedia general data-store. This does not consider Wikidata
as a major contributor to the future-Wikimedia development.
Agreeing on the role Wikidata is to play in the Wikimedia adaptation
to the Internet future should be our first point of consensus. Otherwise :
1. Wikidata is of no direct interest to Internet Users, only to
Wikipedia contributors;
2. another Wikimedia project is to be considered as a Wikidata back-end.
Certainly, at a time, there will be a need for a datamodel. But first
we need to locate Wikidata in the Wikimedia strategy as a
data-collector and as a data-desseminator, not only as a data-store,
in a real world where there are five main conceptual channels : (1)
Business World diversity (Search Engines, etc.), (2) JTC1, (3) W3C,
(4) emergent IUsers [Intelligent lead users: FLOSS, IUCG], (5)
Wikimedia. Possibly we have to consider Big Data, and the border with
Big Data as it will emerge.
As being on the IUse side, I know that we need a convergence of these
channels if we want to attain a good degree of (meta/syllo) data
interchange and not multiply costs, complication and lack of progress
everywhere (not only in sciences). Wikidata is not only about
millions of lonely contributors/authors as Wikipedia is, it also is about
* ISO 11179 conformant very large sources contributions
* diktyologies (from greek diktyos, network), i.e. dynamically
auto-maintained intelligent ontologic spaces. This is a part we have
to explore and support with new concepts (such as syllodata: data
between metadata), cloud architecture, programming languages.
Otherwise Wikimedia will soon be a story of the past, its free
commons being copied by many (like
http://wikipedia.orange.fr) and
distributed and automatically enhanced by powerfull diktyologies with
integrated big (scientific, etc.) data servers.
jfc