RE: What is at the end of the namespace?

> -----Original Message-----
> From: ext Roy T. Fielding [mailto:fielding@ebuilt.com]
> Sent: 16 November, 2001 23:05
> To: Stickler Patrick (NRC/Tampere)
> Cc: a.powell@ukoln.ac.uk; www-talk@w3.org; uri@w3.org
> Subject: Re: What is at the end of the namespace?
> 
> 
> On Fri, Nov 16, 2001 at 12:30:56PM +0200, 
> Patrick.Stickler@nokia.com wrote:
> > > The scheme name is completely irrelevant to a URI's capacity for
> > > identification -- it merely indicates the syntax for that 
> > > namespace and,
> > > when used in the context of a user action, some hint to 
> the software
> > > responsible for that action as to how it should go about 
> handling the
> > > identifier.  
> > 
> > I really can't agree with that. 
> 
> That's okay, it's a (mostly) free world.

One often gets another impression...  ;-)
 
> > That's like saying that, because a 'mailto:' URI is a URI and
> > URI's can identify anything, I can use a 'mailto:' URI to 
> > denote an abstract concept ...
> 
> Yes, you can.  It is just an identifier.  A variable.  A 
> mathematical symbol
> described by a sequence of characters in a syntax defined by the first
> part of that string leading up to the colon character.

Why have any syntax, since it's so opaque, so meaningless?!

Why not then just use UUIDs instead of URIs for *everything*.

Yeah! What a great breakthrough! All we need are UUIDs! Fantastic.

We can then just describe their interpretation with RDF, or leave
it to the "context"!

We don't even have to worry about syntax!

Wonderful... problem solved!

> > and software should *know* that
> > it means the abstract concept and not a way to send some
> > content to a particular mailbox.
> 
> Nonsense.  Software doesn't *know* squat -- it has no 
> intelligence, not
> even AI.  It is merely used for various purposes.

Obviously, you missed my point. The URI scheme provides
clues to its intended interpretation, such that e.g. it
should be meaningful to an agent that implements a particular
protocol, and the remainder of the URI is structured in a
fashion to be parsed and manipulated by that agent in some
useful, and predictable manner, as defined by that protocol.

> > The dilution of semantics of URLs and URNs into just URI
> > creates just too much confusion about the nature of specific
> > URI schemes which invites abuse which results in chaos.
> 
> The only chaos I have seen is in the writings of more recent 
> specifications
> that ignore the research and experience of the Web developers 
> in favor of
> their own personal view of an ideal world.  When they 
> implement something
> that works and has the same expressive power as the Web 
> itself, then I will
> take their writings seriously.

With all due respect, (most of) the web today works because 
an 'http' URI is interpreted according to the HTTP protocol, and a
'mailto:' URI is interpreted according to common mail
protocols (e.g. SMTP, etc.), etc. etc., and my hat is off to
the founding fathers and mothers of the web.

However, the "hot" issues of the day are due to the
fact that new extensions based on previous mechanisms
*don't* work reliably and there are
few clear practices, only hunches and hacks, such as using
HTTP URIs for namespaces and putting something like RDDL
there. Fine, but that's a hack. Not a solution. Because
you can't put RDDL at the end of *any* URI because not 
all URIs have interpretations which allow that.

> > The URI scheme *should* say something about the nature and
> > general semantics of the identifiers grounded in that
> > scheme. To say that the scheme identity is irrelevant and
> > everything goes is ludicrous.
> > 
> > Why not then toss out *all* URI schemes, and just call
> > everything 'uri:'?!
> 
> That is explained in Tim's original Web design notes and 1993 papers
> on the Web.  The notion that everyone is going to agree on the same
> namespace, let alone the same syntax, just isn't practical, 
> and with the
> proper design it isn't necessary.  So why waste time trying 
> to convince
> everyone to use a single namespace?  Tim's recent argument isn't about
> making everyone use "http" as a namespace -- it is about not wanting
> everyone to create their own namespace just because they can.  Sure,
> it is possible, but it creates additional work and complexity for
> everyone else.

Well, I was presuming that the founding architects of the web
appreciated the benefit of a functional taxonomy of URI schemes
but perhaps they weren't.

Patrick

Received on Monday, 19 November 2001 07:18:37 UTC