- From: Kingsley Idehen <kidehen@openlinksw.com>
- Date: Mon, 08 Mar 2010 09:44:44 -0500
- To: Norman Gray <norman@astro.gla.ac.uk>
- CC: Linked Data community <public-lod@w3.org>
Norman Gray wrote: > Greetings. > > On 2010 Mar 7, at 22:17, Aldo Bucchi wrote: > > >> I found this guy, and just by looking at his passport I can craft a >> URI. Then I can access all his personal details and viceversa, others >> will know that I found him. >> >> This NIN scheme should be taken seriously... >> > > The discussion is moving away from the initial question, into a more specifically Linked-Data one, but (at the risk of spoiling that party) it might be worth pointing out that there appears to be a mechanism already in place, and standardised, for generating globally unique person IDs. > > If I look at the machine-readable strip at the bottom of my passport, I can see my passport number and a country code, along with my birth date, and the passport's expiry date. > > There's already an international standard for this <http://en.wikipedia.org/wiki/Machine-readable_passport>. > > Thus http://global-id-num.org/<X> (where <X> is characters 1-13 of row 2 of the passport) would work and be a URI generatable directly from a passport. > > That isn't a complete NIN, of course. This pair of lines may include a NIN, and there is doubtless a similar standard for national identity cards, especially given that, in the EU at least, they may be used in place of a passport. Even though the UK doesn't have ID cards (yet), everyone has a National Insurance number, and I would lay quite a lot of money that there are protocols in place to map these to ID card numbers in an internationally standardised way; similarly for SSNs in the US. > > The questions about the universal legality of repurposing this ID still remain, of course. > > My only point is that this wouldn't need to be reinvented. > > Best wishes, > > Norman > Norman, The original question also made a reference to DBpedia namespace, which is what triggered the Linked Data discussion etc.. In this scenario the canonical records are best served with URNs rather than HTTP URIs at the core DBMS level. They can we published to HTTP networks in line with HTTP based Linked Data best practices (which doesn't define Linked Data the concept general, just how you apply the concept to HTTP Networks such as the World Wide Web). -- Regards, Kingsley Idehen President & CEO OpenLink Software Web: http://www.openlinksw.com Weblog: http://www.openlinksw.com/blog/~kidehen Twitter/Identi.ca: kidehen > >
Received on Monday, 8 March 2010 14:45:14 UTC