So is there anything wrong with having both that http-based PURL URI
available, AND an info uri? Not only available, but in common use?
It gets complicated thinking about these things. There are potentially
several things wrong with it.
Jonathan
Ross Singer wrote:
On Mon, Mar 30, 2009 at 10:12 AM, Ray Denenberg, Library of Congress
<r...@loc.gov> wrote:
Leaving aside religious issues I just want to be sure we're clear on one
point: the work required for the info URI process is exactly the amount of
work required, no more no less. It forces you to specify clear syntax and
semantics, normalization (if applicable), etc. If you go a different route
because it's less work, then you're probably avoiding doing work that needs
to be done.
Avoiding the religious debate that I *think* Ray is referring to (http
vs. info URIs) and instead raising a different religious debate...
I don't have a problem with going through this process to formalize an
info URI once a domain has been thoroughly evaluated and worked out,
but it throws any and all sense of 'agility' out the window and in
many cases, kills any potential hope of actually seeing these
identifiers at all. The upfront costs are just too high, the details
too arcane and the payoff too low for somebody like Jonathan to solve
an immediate problem.
I'm not saying we shouldn't think these things out beforehand;
recklessness, of course, is not the answer. Perfection, however,
being the enemy of the good makes me think the info:uri process isn't
a particularly good or efficient one for working with real world
problems.
Add to it that nobody gives a damn about info:uris outside of
libraries, it seems like a total waste of energy.
Although I suppose that strays back into the original religious debate.
-Ross.
