Re: Comments on draft-fielding-uri-rfc2396bis-07

On Sun November 7 2004 16:43, Roy T. Fielding wrote:
> 
> One would have to go through leaps and bounds of ignoring parts of
> the specification and overzealously interpreting other parts to come
> to the conclusion that rfc2396bis specifies that mailto URIs must
> (or must not) percent-encode a "@" in path.

The discussion has never been about "rfc2396bis specifies that
mailto URIs [...]"; obviously the draft under discussion does not
specify mailto URI syntax directly. Rather the discussion has
been about (possibly unintended) implications for mailto URI
and other specifications caused by the massive changes in the
generic URI syntax draft, one of which is the change of '@'
from being explicitly not "reserved" in a path component to
its being unconditionally "reserved".

Recent discussion, including the message to which your message
was in-reply-to, mentioned specific text in the draft which can
charitably be described as "unclear" and "ambiguous", and/or
"redundant". It also raised the question of whether or not the
implications of the changes on existing URI schemes and other
specifications which are based on the generic URI syntax had
been duly considered.  I note that rather than address that
issue, you have chosen to change the subject.

> That decision is specifically left to the mailto
> specification to define

The mailto specification does indeed define behavior of
implementations, in terms of "reserved" (and "unsafe")
characters.  The problem is that the underlying definitions
which are part of the generic syntax have changed,
apparently with no consideration given to the effect of those
changes on the external specifications which make use of
those definitions.  [To date, I have not addressed "unsafe"
per se, although there was brief discussion of two particular
characters that resulted from a specific real-world example;
"unsafe" was changed in 2396 to "excluded", and the matter
is not addressed at all in the draft under discussion, leaving
a gaping hole in at least one URI scheme specification.]

> I suggest you find some other
> outlet for your energy, such as converting the mailto draft to XML
> and fixing the bits you want fixed so that it can be revised.

Your suggestion has been noted and has been given due
consideration.  As far as I know, there is no "mailto draft".
If there were, and if deemed desirable by its author(s), I would
be happy to put its source into RFC 2223 source format.
I have already communicated concerns based on the earlier
changes in generic URI syntax (1738 -> 2396) to the mailto
RFC authors who could be reached, as I have done for
(unrelated) RFC 2396 issues in the past to its authors, and
in fact as I have been doing regarding the present draft here
in the forum suggested for such issues.

Received on Tuesday, 9 November 2004 17:21:17 UTC