Re: ERB call on addressing

In case it wasn't already clear, I think the ERB call is pretty good.

At 3:28 AM +0000 3/25/97, Gavin Nicol wrote:
>I think it was Lee who also wrote:
>>Drop the ?/&/; thing and you'll be fine.  Simply allow them without
>>saying what they mean.
>
>I agree with this quite strongly. My main objection to the
>ERB proposal is using queries as the *standard* sub-document
>addressing scheme, which I believe to be wrong. Give me the
>freedom to build the system the way I want, and I'll be happy.

    Here Gavin has a point. I don't think we can successfully try to tell
servers what query syntax to use, nor should we. On the other hand, I think
that a well-defined common query format might be of use to people. Since
XML does not have to talk about servers at all can't we just say "Servers
that wish to implement sub-document addressing rather than leaving it to
clients may choose to use query strings to embed XML locators in URLs".

   The problem with the "|" syggestion is that as an implementor I simply
_can't_ tell how to process it unless we also define a protocol for
determining that I'm talking to a "magic" server -- so I think we should
say nothing. Clients are already free to engage in whatever URL mangling
they want if they know what server they are talking to. We might want to
suggest the "|" syntax and semantics to the URL group, but I don't think we
should extend the RFC like this, especially since we can't tell anyone how
to implement the new feature.

  -- David

_________________________________________
David Durand              dgd@cs.bu.edu  \  david@dynamicDiagrams.com
Boston University Computer Science        \  Sr. Analyst
http://www.cs.bu.edu/students/grads/dgd/   \  Dynamic Diagrams
--------------------------------------------\  http://dynamicDiagrams.com/
MAPA: mapping for the WWW                    \__________________________

Received on Tuesday, 25 March 1997 17:09:55 UTC