I confess to having lost track of which post was Craig's original proposal;
there are too many threads crossing on my desk.
| Craig's arguments are very interesting, and his proposed solution
| promises to offer much greater capabilities and potential for the
| future. This potential carries a price though -- a more complex
| implementation including support for a semantics specification
| language (such as Java).
This part I have not understood at all. What needs standardization
is the semantics themselves, right?
| I like to keep things simple enough so that I can understand them.
| I have to admit that I do not fully understand Craig's solution.
Yes. Craig, would you point us to a summary (or post one) so we
can reorient?
| Maybe that's perfectly OK, and maybe Craig's solution should be
| more fully developed and deployed. Or, maybe we need to, simply,
| carry on from where Tim started.
|
| I'd like to hear from some others -- Dan, Terry, Lou, Dave, Tim,
| Eric(s), Tom, Stu, Yuri, Paul?
Never loath to pipe up, here. The point of departure here was HTML.
I believe Craig is proposing that one point off to link specs
somewhere outside the HTML doc. That's fine with me, but it seems
overly complex, and it may be confusing for users if one party's
ABC-NEXT is different from another DEF-NEXT (they're both NEXT,
so why should they differ?), or if I can redefine behavior
(TERRY-NEXT means PREV, for the document indicated by REV---
could I do that under your proposal, Craig?).
Back to defining the semantics. These are semantics of browser
behavior. I can't make my browser do something it can't do. What
I really want is access to the browser's API, in which NEXT,
PREV, and so on are standardized---outside HTML. Try thinking
of it not as "what does my browser do when I say LINK REL=NEXT"
but rather as "what do I have to send to the browser to make
it do LINK REL=NEXT"?
So I think an I-D on functionality of browsers is in order, in
which the behaviors they share in common are given standardized
labels. HTMLers could then use those labels as values for REL
and REV, and for well behaved browsers they might even get
results. Other behavior can be experimented with, then, too,
without having to limit the values of REL and REV in the DTD.
There may also be value in pointing off to link specs, but
that may just be a different issue.
-- Terry Allen (terry@ora.com) O'Reilly & Associates, Inc. Editor, Digital Media Group 101 Morris St. Sebastopol, Calif., 95472 occasional column at: http://gnn.com/meta/imedia/webworks/allen/A Davenport Group sponsor. For information on the Davenport Group see ftp://ftp.ora.com/pub/davenport/README.html or http://www.ora.com/davenport/README.html