On 3/2/09 02:58, Kingsley Idehen wrote:
On 2/2/09 8:13 PM, Hugh Glaser wrote:
Wow. A couple of great messages.
Interestingly (for me) I read Dan's message as not being antagonistic
to the minting of URIs; rather as an excellent discussion of some of
the issues.
(Re-reading it, I find I may have not given sufficient importance to
the statements about "avoid creating artificial URIs ".)
Anyway, whatever Dan's opinion, I think I am in great agreement with
Richard.
I understand the expanse of Dan's concern perimeter, but I do need
clarification about what constitutes an "artificial URI" if we hold true
to "Open World" assumptions :-)
Fair point. I guess all URIs are artifacts, and begin their life unknown
to the rest of the world. A better name than "artificial URI" might be
"Alice in Wonderland URIs", by which I mean, URIs whose meaning isn't
yet widely shared throughout the Web community:
'''“When I use a word,” Humpty Dumpty said, in a rather scornful tone,
“it means just what I choose it to mean — neither more nor less.”'''
If URIs are to be used as terms ('words'), they need to be richly
embedded in multiple sites, services, datasets. If I merely declare that
http://id.danbri.org/kingsley is an identifier for you, it isn't yet a
Semantic Web "word". But if I get this URI embedded in software,
external databases, interesting RDF documents etc., it starts to become
one. My main point was that progressing along this scale is easier if
the URI-minting party sticks to some (as yet undocumented) practices
around persistence, privacy, data etiquette ...
Quick reply also to Richard (I'm in a meeting), ... yup, I agree on
"“Don't trust and re-use every random identifier you find.” ". My post
wasn't anti-minting, but a suggestion firstly that additional metadata
and documented best practice about identifiers (and identifier sets)
will be useful, and a plea for special care and clear practice when
identifying people.
cheers,
Dan
''
http://danbri.org/
