Re: URI-protocol mapping (was Re: How to add new "protocols" ?)

[Dan Connolly <[e-mail]>]

Ron Daniel Jr. wrote:

> I'll admit that one *could* resolve current URLs in a location-independent
> fashion. However, one cannot do it in a manner conformant to existing
> standards. We can do anything if we change the definitions of our terms.

Let's just agree on that and declare victory, huh?

The bottom line is: we don't have to change all the
addresses in all the documents in the world in order
to increase the quality of service (availability, persistance,
etc.) of resource access on the web.

We just need to design and deploy some better resolution
mechanisms (and we have to be especially careful to
"follow the money" while we do it: business considerations
like IPR, trademarks and brands will dominate technical
considerations, I suspect).

Moreover: a change in some of the basic syntactic rules
(like going from slash to colon for hierarchical separators)
has significant cost, with no increase in expressive power.

You could make a human-factors argument that people traffic
in colons more reliably than they traffic in slashes, and
since faults at the human-machine interface have one of
the most dramatic impacts on the reliability of the web,
the change is merited. But I would want to see a significant
body of emperical evidence.

I'd be more receptive to a similar argument for a switch
to case-preserving-but-case-insensitive from the contemporary
case-sensisive specs. I can't tell you how many times
URLs have been mistranscribed and then printed in a
zillion newspapers! Thank goodness for redirects!


Dan