Re: XHTML & hyperlinking opinions (long, sorry)

Tim Bray writes:
> >><main-theses>
> >>1. If you want to extend XHTML to do any of the three things 
> >>that XLink  claims to be designed to do, then XLink is a good way to
do
> >> it.
> > > I'm curious as to why you believe that so strongly.
> 
> I reviewed the XLink spec, and I thought about how I'd go about
> designing markup for multi-ended and out-of-band links, and I thought
> XLink presented a pretty compelling design for how you'd do those
> things.

XLink looks an awful lot like the objects I would use to implement such
a thing in Java, but I'm hardly convinced that my Java model represents
the best possible solution for use in markup.  (Among other things,
namespace issues in Java are real, but they tend to be solved by saying
"here's a list of things we're talking about today", not by putting
extra identifiers on every name used.)

I'm quite content in saying that even multi-ended/out-of-band/metadata-
loaded links are more easily expressed as part of a vocabulary than as
an alien module glued on to the side of a vocabulary.  

Even if XLink actually satisfied all of my needs for linkbases, I'd
still have to construct a host element vocabulary for it, which feels
kind of odd to say the least, and since these are global attributes, we
have to have namespace prefixes on _all_ of them, and... how many times
do I have to be reminded that this isn't my markup design?  (Better,
perhaps, than the explicit XLink elements in earlier drafts which might
simply have amplified the integration mismatch.)

> I think disagreement should be accompanied by examples: "here's a
> better way to do a multi-ended/out-of-band/metadata-loaded hyperlink,
> and here's why it's better."

XArc was an interesting approach to the same set of problems in answer
to similar requests from the XLink WG long ago:
http://www.jfinity.com/xarc/spec-current/

XArc doesn't solve every problem out there - it doesn't try - but it's
remarkably easy to figure out and makes out-of-band linking pretty
simple.  It does seem to make sense to a lot of people, though I've had
people ask why "from" is necessary at all. (Sure, it could use some
additional development, but it looks pretty good for almost four years
old.)

> > Why is it more important for XHTML to migrate in the direction of
> > generic XML rather than for generic XML to incorporate the linking
> > mechanism of XHTML 1.x?
> 
> I don't accept that reformulation of what I said.  

You don't have to accept Mike's reformulation; I have a very hard time
seeing what you propose as anything other than Mike's description
regardless of your acceptance, and others will likely make up their own
minds depending on their background.

> I observe all sorts 
> of people faking multi-ended hyperlinks with all sorts of non-obvious 
> non-portable javascript hacks, and I deduce from this that if they 
> there, they would be put to good use right away.

I've shown XLink extended links to a variety of audiences, from
programmers to Web developers to XML Geek Cruisers, even shown a simple
demo with multiple connections between pictures using pop-up windows,
and I deduce from their responses that there isn't a groundswell of
demand waiting for the XLink answer.  It's been a year, admittedly, but
the most excitement I've gotten is "that's really cool, and it would be
neat if you could do that, but I'm not sure I'd ever want to do that."
(That was at XML Connections, a heavily Microsoft-oriented show, much to
my surprise.)

We did have a very nice panel discussion at Software Development on
XLink's capabilities and implications a few years ago.  The audience was
very small, though maybe that was just the result of a huge room and
bright lights...

> I also have bitter personal experience in maintaining HTML linkage
> networks in the presence of updates and think that if I had an
> out-of-band linkbase capability it would make those problems way
> easier.

I've had the same problems, and that's a lot of why I was excited about
XLink in the first place.  Maybe all those years of waiting have taken
the edge off my excitement, or maybe it's just not precisely clear how
this technology should work with HTML, as even the XLink folks seem to
have trouble describing how multi-ended links and browsers should
cooperate.  

(Throw in an XSLT transform or two, and then your XPointers are tough to
find, and then... well, what's left?  Or should the XSLT transform
process or otherwise preserve XPointers too?)


-------------
Simon St.Laurent - SSL is my TLA
http://simonstl.com may be my URI
http://monasticxml.org may be my ascetic URI
urn:oid:1.3.6.1.4.1.6320 is another possibility altogether

Received on Friday, 4 October 2002 17:24:33 UTC