- From: Steve Harris <steve.harris@garlik.com>
- Date: Sat, 26 Mar 2011 09:18:38 +0000
- To: David Booth <david@dbooth.org>
- Cc: Sandro Hawke <sandro@w3.org>, Pat Hayes <phayes@ihmc.us>, semantic-web@w3.org
On 2011-03-26, at 03:44, David Booth wrote: > On Fri, 2011-03-25 at 23:04 -0400, Sandro Hawke wrote: >> On Fri, 2011-03-25 at 13:23 -0500, Pat Hayes wrote: >>> BTW, I think that skolem URIs should NOT be dereferencable, as a >> matter of design. If you want to put a 'real' URI name in there, you >> always have that option: but then (for example) changing it will make >> a non-equivalent graph (as it should). > > I am stunned that folks are even discussing new URI schemes for this, > when linked data has clearly -- once again -- demonstrated the benefits > of *dereferenceable* URIs. That's one take. I would say that the use of bNodes in linked data indicates that sometimes you want to mint identifiers that aren't "cool URIs". Pragmatically, some pieces of information are transitory. > Please, at *least* make it dereferenceable to *some* kind of useful > information. In the very least, it could be information about how > bnodes are skolomized. Nobody is required to dereference a URI. But it > is helpful to have the *option* of deferencing an identifier to learn > more about it. You always have the option to just use an HTTP URI. - Steve -- Steve Harris, CTO, Garlik Limited 1-3 Halford Road, Richmond, TW10 6AW, UK +44 20 8439 8203 http://www.garlik.com/ Registered in England and Wales 535 7233 VAT # 849 0517 11 Registered office: Thames House, Portsmouth Road, Esher, Surrey, KT10 9AD
Received on Saturday, 26 March 2011 09:19:14 UTC