- From: Robert Rothenburg Walking-Owl <wlkngowl@unix.asb.com>
- Date: Thu, 17 Apr 1997 00:41:05 -0400
- To: www-html-editor@w3.org
- CC: connolly@w3.org
- Message-ID: <3355A9E0.E4D@unix.asb.com>
Attached is a post from the www-html list. It would be a good thing to expand the use of <link> relationships or somehow associate id/name attributes in links with their own URLs. Two uses of this: (1) When other sites (beyond the maintainer's control) maintain links to areas with a document that have expanded enough to merit their own pages, it would be useful to have a way of redirecting these old links. For instance, a there is link from another site to document foo.html#bar however the section at id=bar/name=bar has expanded to merit it's own file (bar.html). It would be convenient to specify a URL that the id signifies, say (in the head element) <link name=bar href="bar.html"> or perhaps in the document (for older browsers) <a name=bar refer-href="bar.html">bar</a> <p>This section has been moved to <a href="bar.htm">bar.html</a></p> (2) This would also allow one to maintain 'logical' names for links (aliases) for an entire site. When pages are moved or reorganized, only the initial definition in the documents needs to be changed. Or (better yet) an index of aliases can be kept in a separate file shared on a site, referenced by <link rel=aliases href="aliases.html"> or something equivalent. For example, <link name=Appendix href="notes/Appendix.html"> <link name=Notes href="notes/Notes.html"> <link name=Table-of-Contents href="/toc.html"> This would also help maintenance in situations where there are multiple links within a document to one site. A common example when pages contain icon-links and text links in a menu at the bottom (and/or top) of the page: <link name=Purchase href="purchase-form-1.html"> .. <a href=#Purchase><img src="purchase-icon.gif"></a> .. <a href=#Purchase>[Purchase]</a> --Rob Rothenburg (wlkngowl@unix.asb.com) or (mutradio@wusb.org)
Date forwarded: Wed, 16 Apr 1997 15:32:30 -0400 Date sent: Wed, 16 Apr 1997 14:31:52 -0500 From: Dan Connolly <connolly@w3.org> Organization: World Wide Web Consortium To: "Robert Rothenburg 'Walking-Owl'" <WlkngOwl@unix.asb.com> Copies to: www-html@w3.org Subject: Re: Handling relocated anchors... Forwarded by: www-html@w3.org Please consider sending a review comment on the recent working draft on linking[1] to the editors as per [2]. [1] http://www.w3.org/pub/WWW/TR/WD-htmllink [2] http://www.w3.org/pub/WWW/MarkUp/Cougar/ Robert Rothenburg 'Walking-Owl' wrote: > > There's a variety of choices to handle moved pages, but what if > someone puts a link to http://www.any-site.com/foo.html#bar where > foo.html is still there, but the anchor "bar" has moved (say when > that section is large enough to merit it's own page at bar.html)? > > A simple kluge is to put a <a name="bar">Moved</a> message for pages > that link to that section, but it would be nice to mark a referral > for an anchor that functions like Refresh in HTTPD headers: > > <a name="bar" refer-href="bar.html"> > > So that any links to foo.html#bar are referred to bar.html. I suggest you consider using link relationships, ala: <a name="bar" href="bar.html" rel="redirect">moved</a> -- Dan Connolly, W3C Architecture Domain Lead <connolly@w3.org> +1 512 310-2971 http://www.w3.org/People/Connolly/ PGP:EDF8 A8E4 F3BB 0F3C FD1B 7BE0 716C FF21
Received on Thursday, 17 April 1997 00:41:25 UTC