On Tue, Jul 12, 2011 at 12:46 PM, Saašha Metsärantala <[log in to unmask]> wrote:
>> What's an edtf library supposed to do with URIs?
> That is an important question! I was also feeling that we need to clarify
> our way to work with URIs, and therefore I suggested namespaces. We need to
> discuss those questions.
But even this I think demonstrates my hunch that this adds a level of
complexity that is not worth the effort.
Namespaces don't solve the issue in the least: it's just a convention
for abbreviating URIs after all. At root, a namespec prefix + token is
If it were me, I'd cut this feature as out of scope.
>> I suggest that we take all the truly off-the-wall features and move them
>> to level 3:
> Four levels are quite a lot, but this could be a good way to gather elements
> within our specification. I suggest to create this fourth level (level 3)
> and then focus on the three first ones (level 0 to Level 2). When we are
> done and feel satisfied with these three ones, we could focus to the fourth
> level, having a stable and reliable ground (that is the three first levels).