W3C home > Mailing lists > Public > www-talk@w3.org > January to February 2002

Re: Searchable Reader-side URL Anchors ...

From: Mirsad Todorovac <mtodorov@alu.hr>
Date: Thu, 24 Jan 2002 14:52:21 +0100 (MET)
To: Reinier Post <rp@win.tue.nl>
cc: <www-talk@w3.org>
Message-ID: <Pine.OSF.4.33.0201241352520.689-100000@domac.alu.hr>
On Thu, 24 Jan 2002, Reinier Post wrote:

> > Back then I proposed syntax for new type of URL with new type of anchor:
> >
> > 	http://{...}/.../.../rfc1738.txt##Second%20Chapter
> >
> > ... to distinguish it from #Second%20Chapter, which would in turn refer to
> > a <A NAME="Second Chapter"> in HTML file (but in plaintext, as you have
> > already spotted - there are no anchors).
> >
> > (But I do not insist on same syntax, that didn't pass earlier, and was
> > blocked - I'm promoting the issue of Reader-side anchors, not the syntax)
>
> WN has supported somehing like this ('fragments') for a long time, see
>
>   http://hopf.math.nwu.edu/
>
> However, yours seem to be client-side.  How do you propose to determine
> where the anchors are?

In multiple ways:

1. By byte offset in document's (HTML, txt or other) source
	(this might work for audio and video too)

	Example syntax:
		http://www.internic.org/rfc/rfc1738.txt##+12387b
			== open rfc1728.txt at page which starts with this
			   byte offset
		(now the syntax which will be used is irrelevant,
		 let's try to look at it with pragmatism)

	This way is the simplest, but suffers from a serious problem
	-- if destination document is changed, all byte offsets change

2. By search string in source

	Example syntax:
	http://www.internic.org/rfc/rfc1738.html##/%3CH1%3EChapter%20Three/

		(searches for "<H1>Chapter Three")
		-- I know this looks ugly, but it's only an example of
		additional functionality - please do not stumble on syntax

	NOTE: You can insert additinal paragraph in "Chapter Two" in
	      destination document, but URL pointing at "Chapter Three"
	      remains the same

3. By Nth occurance of search string

	Find 3rd occurance of <H2>:
	(example syntax)
	http://www.internic.org/rfc/rfc1738.html##3,/%2CH2%3E/

> > You must remember many situations when you were refered third chapter in
> > larger page, but have been diverted your attention in first or second
> > chapter and started to red that, forgetting what you came for in the first
> > place?
>
> Is the third chapter the third <H1>?  Or the third <H2>?  What if
> the document is not HTML?

At this pointer source author points at a point in destination document,
that point being described either by byte offset, or by search string (or
simplified regular expression), or by Nth occurance of search string.

Browser needs not to have a basic idea of what is chapter in destination
document, and this allows extension to other media types except for HTML,
txt and XML in future without modifications -- all wisdom of reader-side
anchors is in browser and URL extensions; HTTP server needs not to be
aware of it at this point, except if we extend this feature to the point
where only desired fragments of destination document would be fetched.

That about covers the basic concept I guess,
Mirsad

--
This message has been made up using recycled ideas and language constructs.
No tree has been cut nor animal harmed in process of making it.

Visit http://www.thehungersite.com/, save one CHILD from DYING!!!
Received on Thursday, 24 January 2002 08:52:40 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Wednesday, 27 October 2010 18:14:27 GMT