- From: Richard Light <richard@light.demon.co.uk>
- Date: Mon, 9 May 2011 18:31:34 +0100
- To: Karen Coyle <kcoyle@kcoyle.net>
- Cc: public-lld@w3.org
In message <20110509102059.157827xtf8ezcf3v@kcoyle.net>, Karen Coyle <kcoyle@kcoyle.net> writes >It reads ok to me. I want to point out that this paragraph: > >"However web developers are sometimes turned off Semantic Web (Linked >Data) technologies because they feel like they would need to throw >away their current application, to swap their database for a >triplestore, and their database query language for SPARQL. This is >simply not the case, since RDF serializations can be generated on the >fly just as web application frameworks do fo HTML, XML and JSON >representations. The use of http URIs to identify and link together >resources in RDF's data model make it a natural choice for serializing >and sharing entity state in a database neutral way--which has >traditionally been of great interest to cultural heritage >organizations and the digital preservation community. " > >totally parallels a discussion we just had in the LLD recommendations >group, and which perhaps should also be included in the benefits area. >Our discussion in LLD-R was that we should emphasize what some library >practices, like authority control and controlled terms, and LD with >URIs have in common -- for identification, for sharing (and the >subsequent efficiencies), and for re-use. If we can present those >concepts as being a new way to do what libraries already have in their >metadata toolkit, we make it much less foreign and frightening. In that context, has the Linked Data API work [1] been mentioned yet? While it is primarily designed to put a friendlier REST-style face on a triple store, it can equally be used as a design pattern for creating URI-based interfaces to non-triple-stores. Richard [1] http://code.google.com/p/linked-data-api/ -- Richard Light
Received on Monday, 9 May 2011 17:35:02 UTC