Re: QA Tips: Make readable URIs

Le jeu 18/09/2003 à 18:18, Alex Rousskov a écrit :
> 
> On Thu, 18 Sep 2003, Dominique [ISO-8859-1] Hazaël-Massieux wrote:
> 
> > - one of the principle behind the design of the URIs is that they are
> > opaque, which means that nothing/nobody should infer anything from the
> > characters used in the URI; this should be reminded at the very top of
> > the tip, and maybe the title of the tip should be changed to make that
> > clearer.
> 
> Hmm... Will W3C be changing its URI to http://unforgettableuri.net --
> easy to remember, but otherwise as meaningless (opaque) as you can get
> with current TLDs? Or does the above design principle apply only to
> newcomers that were too late to grab meaningful and easy to remember
> domain names?

The principle that URIs are opaque says what I said it does: do not
infer anything from the URI. For HTTP, that's true for the path
component, eg
- http://example.org/ThisPageIsSuperCool doesn't tell you anything about
the content of the resource you can GET from it; esp. it doesn't tell
you if the content is cool or not, and nobody should infer anything
because of that
but also for the DNS part of the URI, eg
http://www.homepageofasuperguy.net/myPicture doesn't tell you anything
(valuable) about the owner of the resource served there, the purpose of
the resource, etc.

As per the RFC 2396, the non-opaque parts of a URI are:
- its scheme
- some of the reserved characters used in the non-scheme part (such as /
at least in hierarchical schemes)

Now, that architectural fact doesn't mean that you should make your URIs
as unreadable as possible; there are some usability pros to have
readable URIs (as I tried to explain in my reply), but I think their
scope has also some limits.

Dom
-- 
Dominique Hazaël-Massieux - http://www.w3.org/People/Dom/
W3C/ERCIM
mailto:dom@w3.org