- From: Jeni Tennison <jeni@jenitennison.com>
- Date: Fri, 4 Oct 2002 09:33:08 +0100
- To: www-tag@w3.org, Norman Walsh <Norman.Walsh@sun.com>
Hi Norm
Nice message :)
> I surrender all hopes and aspirations for a common linking
> vocabulary as I conceived it. Perhaps my conceptions will change
> over time, but I think a more realistic outcome is that (1) HLink
> will be no more widely deployed in the future than XLink is today
> and (2) the fact that XHTML, *the* hypertext vocabulary for the web,
> isn't going to use XLink kills it. Dead. Finito. The proverbial
> cherubic soprano has begun her song.
You know, I think that you can be slightly more positive than this. As
I see it, the likely outcome is that:
(1) The basic XLink concepts become relevant outside XLink
itself; when someone designs a language that includes links,
they think of it in terms of "simple" and "extended" links, of
"arcs" and "locators".
(2) Linkbases continue to use XLink for out-of-line links.
(3) Document-oriented languages (e.g. XHTML, DocBook...) use XLink
if they want, another linking solution if they can't, but it
doesn't really matter anyway because most applications that
*present* them are hardwired to know what is a link and what
isn't in any case, based on the natural language spec.
(4) Most markup languages continue to use their own terms for their
links and use some transformation technology to turn that into
XHTML or XSL-FO when they need to present it.
(5) Robots and spiders try to dereference any URL that they find
within a page, just in case, and continue to gather every
mailto: so that someone can send 50 messages a day offering
substantial financial inducement for assisting someone in Africa
whose relative has recently passed away.
OK, maybe that isn't positive, especially for the Semantic Web and all
that...
But I think that if we want HLink (or whatever it turns into --
NeXLink ;) to be more widely deployed than XLink, we have to look
seriously at the reasons that XLink *isn't* more widely deployed. I
suspect that it's because for users, the cost of incorporating XLink
into their markup language is greater than the benefit of application
support they get from doing so, and that on the vendor side there's
simply no user demand. Getting something that's easy for users to
include -- in their schema, in their CSS or wherever -- will be the
deciding factor.
Cheers,
Jeni
---
Jeni Tennison
http://www.jenitennison.com/
Received on Friday, 4 October 2002 04:40:30 UTC