- From: Roger Menday <roger.menday@uk.fujitsu.com>
- Date: Wed, 18 Dec 2013 14:23:46 +0000
- To: Kingsley Idehen <kidehen@openlinksw.com>
- CC: "public-ldp-wg@w3.org" <public-ldp-wg@w3.org>
- Message-ID: <7B3AE960-2D6D-4B1B-AFBB-CD5058BC0596@uk.fujitsu.com>
hello Kingsley, Thanks. Some of the examples in the Primer could be adjusted based on that. The examples in the Spec seem to use relative URIs, although I suppose they fall a little short for you in other ways. Roger p.s. noticed that all the @base specifications in the Spec examples are all commented ... On 18 Dec 2013, at 13:01, Kingsley Idehen wrote: > On 12/18/13 5:11 AM, Roger Menday wrote: >> Well, I think there isn't a clear cut answer on this one. I appreciate the answers from Kingsley, but, when appealing LDP to a broader audience, a lot of people will probably appreciate the simplicity that comes from allowing a URI entity to also be directly de-referencable ?? >> >> My opinion is that we are 'good' in the Spec, use both ways in the Primer, and (as Steve suggested) outline the pros-and-cons of both in the BP&G. >> >> Roger > > Roger, > > If you use relative HTTP URIs in the examples, all the glory of Linked Data manifests as part of the reading and learning experience. Simple rule, stay away from prefixes, use relative URIs, leaving use of absolute URIs for properties and classes defined by shared vocabularies. > > Users should be able (circa. 2013) to copy and paste examples from specs to documents stored in their own personal data spaces e.g., those provided by the likes of Dropbox, Google Drive, SkyDrive etc.. > > I know of nothing simpler, in the context of RDF based Linked Data demonstration and tutorial material, than the pattern I am describing here. It just works :-) > > > Kingsley > >> >> >> On 16 Dec 2013, at 20:37, Kingsley Idehen wrote: >> >>> On 12/16/13 1:20 PM, Wilde, Erik wrote: >>>> On 2013-12-16, 10:11 , "Kingsley Idehen" <kidehen@openlinksw.com> wrote: >>>>> On 12/16/13 12:31 PM, Roger Menday >>>>> wrote: >>>>> Maybe we should be good in the Spec, and be naughty in the >>>>> Primer ... (?) >>>>> No, once its naughty it just gets naughtier and harder to rectify. >>>>> History (e.g., RDF and Web) shows, it can even take 13 or so years >>>>> to fix the ensuing misconceptions. >>>> agreed that it's different for linked data, but on the web, having >>>> identifiers that do not resolve is perfectly acceptable. that's why >>>> hypermedia links are typed: you follow the ones where the type implies >>>> they're dereferencable, and you only use them as identifiers where the >>>> type implies they are identifiers only. >>>> >>>> i am not sure if what roger suggests is to point out that this is what's >>>> natural for the larger web and REST in general. i agree that we should be >>>> careful to promote/show patterns that are not exactly the way how things >>>> should be done in a certain context, but then again, if that demonstrates >>>> how things are done in practice (even though it may not be the ideal way >>>> of doing them), then there might be value in describing those examples as >>>> well. >>>> >>>> cheers, >>>> >>>> dret. >>>> >>>> >>>> >>> As far as I can understand, Roger is seeking clarification about entity (thing) denotation using HTTP URIs. Right now, we have examples in the LDP specs that inaccurately denote entities (that aren't Web accessible Resources) using HTTP URLs. >>> >>> As I said, an HTTP URL is an HTTP URI that denotes a Web Resource. >>> A WebID is an HTTP URI that denotes an Agent. >>> HTTP URIs can be used to denote anything i.e., any kind of entity. >>> >>> Another way to look at this is through the difference between "words" and "terms" . >>> A "term" is a specialization of a "word" i.e., a "term" is a "word" that embodies denotation (naming) and reference such that a name resolves to the description of its referent. [1] >>> >>> The World Wide Web is comprised of webby words i.e., words link to documents (using controls like <a/> in HTML , no referent description is implied this particular association. >>> >>> The RDF based Linked Open Data Web is comprised of webby terms i.e., you have denotation and reference at the very core. >>> >>> [1] http://www.wikihow.com/Differentiate-Between-a-Term-and-a-Word >>> >>> -- >>> >>> Regards, >>> >>> Kingsley Idehen >>> Founder & CEO >>> OpenLink Software >>> Company Web: http://www.openlinksw.com >>> Personal Weblog: http://www.openlinksw.com/blog/~kidehen >>> Twitter Profile: https://twitter.com/kidehen >>> Google+ Profile: https://plus.google.com/+KingsleyIdehen/about >>> LinkedIn Profile: http://www.linkedin.com/in/kidehen >>> >>> >>> >>> >>> > > > -- > > Regards, > > Kingsley Idehen > Founder & CEO > OpenLink Software > Company Web: http://www.openlinksw.com > Personal Weblog: http://www.openlinksw.com/blog/~kidehen > Twitter Profile: https://twitter.com/kidehen > Google+ Profile: https://plus.google.com/+KingsleyIdehen/about > LinkedIn Profile: http://www.linkedin.com/in/kidehen > > > > >
Attachments
- application/pkcs7-signature attachment: smime.p7s
Received on Wednesday, 18 December 2013 14:24:12 UTC