Re: (Dis)Proving that 303s have a performance impact.

On 18 February 2013 05:03, Mo McRoberts <Mo.McRoberts@bbc.co.uk> wrote:

> On the other hand, somebody who knows 'enough', though not a lot of
> SemWeb, may well look at the examples, see no *obvious* reason for the
> fragids, see no explanation of why they're there, and knowing about them
> from HTML figure they're just a stylistic nicety. The fact is that in RDF
> fragment identifiers in URIs are a lot more important than people are used
> to outside of this realm.
>

::raises hand:: Yep, that's me! Or was, until recently. When I started
reviewing the WebID spec last year, I did not understand the reason for the
hashes, and found no explanation for them in any of the material. I was
looking at the spec as a way toward a more usable Internet-scale security
system and had pretty much *zero* experience with SemWeb concepts. I think
I'm starting to get it :-), but I still don't think it should be a
prerequisite, given that WebID can apply equally well outside that space.

And I also agree that some non-normative background information and
implementation or usage guidelines are welcome even in a spec, or in a
primer document that is linked to from the spec. However, if they make
recommendations, I would like to see the justification given, or referenced
from another accessible source, so I can understand the reasoning behind it.