Dear All, Here are the minutes from today's teleconf: http://www.w3.org/2012/10/09-hcls-minutes.html
Cheers, Scott On Tue, Oct 9, 2012 at 4:53 PM, Eric Prud'hommeaux <[email protected]> wrote: > I was asked to introduce Linked Data, so here goes: > > Linked Data, please meet the HCLS WG. HCLS WG, this is Linked Data. > You all get to know each other; I'll be over by the keg. > > If that wasn't what you had in mind, here's an introductory coupld of > paragraphs: > > = Linked Data = > > All data requires structure to provide useful information. > This structure establishes codifies specific linkages within the data. > Using a graph language such as RDF helps both to clarify those internal links > and to generalize the mechanism for connecting these data to other data. > Use of an appropriate model must be complemented by social protocol to > establish common terms for things [@@in the domain of discourse]. > RDF exploits the World Wide Web to define and discover such terms. > [http://linkeddata.org/ Linked Data] is a set of principles to optimizing > term re-use by ensuring that terms are backed by the Web, i.e. they can be > pasted into the location bar in a browser. > > While the interest of optimizing communications pressures everyone to use the > same terms, this is sometimes in tension with legacy identifiers some social > pressures (such as preserving a site's user experience by funneling > identifier resolution through that site). > As in the conventional relational database world, this imposes a requirement > of a "lookup table", which maps one code set to another. > This "Mapped Data" is managed and queried like "ideally" linked data, with > the additional minor complexity of incorporating the mapping when connecting > pieces of the graph which use one code set vs. another. > -- > -ericP -- M. Scott Marshall, PhD MAASTRO clinic, http://www.maastro.nl/en/1/ http://eurecaproject.eu/ https://plus.google.com/u/0/114642613065018821852/posts http://www.linkedin.com/pub/m-scott-marshall/5/464/a22
