From: "Houghton,Andrew" <[log in to unmask]>
> The point being that:
>
> urn:doi:*
> info:doi:*
>
> provide no advantages over:
>
> http://doi.org/*
I think they do.
I realize this is pretty much a dead-end debate as everyone has dug
themselves into a position and nobody is going to change their mind. It is a
philosophical debate and there isn't a right answer. But in my opinion ....
I won't use the doi example because it's overloaded. Let's talk about the
hypothetical sudoc. I think info:sudoc/xyz provides an advantages over:
http://sudoc.org/xyz if the latter is not going to resolve.
Why? Because it drives me nuts to see http URIs everywhere that give all
appearances of resolvability - browsers, editors, etc. turn them into
clickable links. Now, if you are setting up a resolution service where you
get the document that the sudoc identifies when you click on the URI, then
http is appropriate. The *actual document*. Not a description of it in
lieu of the document. And the so-called architectural justification that
it's ok to return metadata instead of the resource (representation) -- I
don't buy it.
--Ray
|