On 21 May 2005, at 12:50, Eric Scheid wrote:
On 21/5/05 7:47 PM, "Henry Story" <[EMAIL PROTECTED]> wrote:
But perhaps it is not such a problem after all, because perhaps we
can have
dereferenceable ids that allow us to move entries around. I can
think of a
few:
1. if an entry moves the old entry position redirects to the new one.
not always possible.
2. one creates a special redirect service to redirect ids to their
current location
Such things already exist in various forms. There is for example a
service
which resolves DOI URIs.
3. one creates a new URI to URL mapping protocol
blech.
e.
I agree, but the question is whether they do their work better than
the one we are trying to give to a non de-refrenceable atom:id.
I like the intent behind atom:id. But I don't think one can just
hope
that things work out the way we would want them to. For the moment
there
is a worry that the atom:id gear is just decoration, a magical place
holder
for properties we would like something to have, but have no idea how
to achieve.
If there are other solutions, it would be nice to hear of them in
more detail.
I have heard that signatures could help resolve the problem. Perhaps
we should
get a security expert to give some advice on a solution like that.
Henry Story
