W3C home > Mailing lists > Public > www-xml-linking-comments@w3.org > January to March 2000

Fwd: more thoughts about show=embed...

From: Eve L. Maler <elm@east.sun.com>
Date: Wed, 23 Feb 2000 14:58:57 -0500
Message-Id: <4.2.0.58.20000223145841.00a8c1f0@abnaki.East.Sun.Com>
To: www-xml-linking-comments@w3.org

>Date: Fri, 11 Feb 2000 15:19:31 +1100
>From: Erik Wilde <netdret@dret.net>
>X-Mailer: Mozilla 4.7 [en] (Win95; I)
>X-Accept-Language: en-US,de-CH
>To: Steve DeRose <Steven_DeRose@brown.edu>, "Eve L. Maler" <elm@East.Sun.COM>,
>         dorchard@ca.ibm.com, bent@exemplary.net,
>         David Lowe <dbl@eng.uts.edu.au>
>Subject: more thoughts about show=embed...
>
>hello.
>
>consider the following scenario:
>
>- a simple xml-document (mostly text) named text.xml, containing a
>number of foreign words.
>
>- an xml dictionary of english translations named words.xml, looking
>like
>
><words>
>   ....
>   <word fword="kindergarten">
>     <eword>nursery</eword>
>   </word>
>   ....
></words>
>
>what i want to do now is create links that help me read a document
>containing foreign words by replacing them with their english
>translation. basically, there are two approaches to do this (with
>respect to the point that i am trying to make here):
>
>1. one locator, using a location set of all occurrences of the foreign
>words:
>
><extendendlink>
>   <locator role="fword"
>href="text.xml#xpointer(string-range(//text(),'kindergarten'))">
>   <locator role="eword"
>href="words.xml#xpointer(word[fword='kindergarten'])">
>   <go from="fword" to="eword" actuate="onLoad" show="embed">
></extendedlink>
>
>2. multiple locators, using a location set of all occurrences of the
>foreign words:
>
><extendendlink>
>   <locator role="fword"
>href="text.xml#xpointer(string-range(//text(),'kindergarten')[1])">
>   <locator role="fword"
>href="text.xml#xpointer(string-range(//text(),'kindergarten')[2])">
>   <locator role="fword"
>href="text.xml#xpointer(string-range(//text(),'kindergarten')[3])">
>   <locator role="fword"
>href="text.xml#xpointer(string-range(//text(),'kindergarten')[4])">
>   .... (as many locators as occurrences) ...
>   <locator role="eword"
>href="words.xml#xpointer(word[fword='kindergarten'])">
>   <go from="fword" to="eword" actuate="onLoad" show="embed">
></extendedlink>
>
>(i am pretty sure that i have made some syntax errors in the xpointers,
>but at least i think it has become clear what i wanted, the first
>example using an xpointer addressing one location set containing all
>occurrences of the foreign word in the document, the second example
>containing individual locators vor every occurrence).
>
>obviously, the second approach is a bit strange and voluminous, because
>it requires knowledge about the number of occurrences. but apart from
>that, how would the two different links differ in functionality? differ
>they at all? and if they do differ, why is that so? and is that
>intended? and what is the difference?
>
>to get more to the point that i really would like to make: the whole
>concept of show=embed in my view somehow implies the existence of a
>contiguous presentation (such as a rectangular window into which the
>embed has to be placed), which breaks down when location sets are being
>used, or when content is not rendered as linearly as text most often is.
>for example, consider a svg graphic, where some graphical structure
>(scattered all over the xml document in different and non-contiguous
>locations) is identified by a location set and should be replaced with
>another graphical structure, using xlink. in this case, it would be
>desirable to have the whole location set replaced by only one occurrence
>of the embedded link target, while in case of my dictionary example, it
>would be nicer to have each location in the location set replaced by the
>embedded link target. things get even more complicated when thinking
>about media types that do have more axes of presentation, such as smil
>or other time-based xml documents.
>
>to conclude all this: wouldn't it be a good idea to delve a bit more
>into this, into the implications of linking locations sets and the
>semantics of doing this? maybe this should even be reflected at the
>xlink level, such as two types of embeds, one for "embed once" and the
>other for "embed all locations". ok, now i have wasted enough of your
>time, and i am curious to hear what you are thinking about my example
>and my thoughts about it...
>
>cheers,
>
>erik wilde - mailto:netdret@dret.net - http://dret.net/
>     computer engineering and networks laboratory  (tik)
>     swiss federal institute of technology (eth zuerich)
>     etz d97.4 / fon: +41-1-6325132 / fax: +41-1-6321035

--
Eve Maler            Sun Microsystems
elm @ east.sun.com    +1 781 442 3190
Received on Wednesday, 23 February 2000 14:57:37 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 27 October 2009 08:39:40 GMT