Re: is it possible to handle an XML/HTML elements attribute via the URI?

Interesting comment Dr. Hausenblas.

I came across three examples of what I think are the same unhelpfullness you mention, all have a root in Federalization of Types.

England (UK) - Several of the Subdivisions (Towns) have English and Welsh versions of the place name.  In the UK system, they are given different codes.  This solution is used elsewhere too.

Ireland (IE) - There are 8 Regions for statistical purposes [1].  I doubt anyone outside of Government knows them, but oddly, nobody outside seems to have any trouble remembering the 17th Century list.

Canada (Quebec) - Quebec has Federal (Voting) Counties and "Natural Counties".  Both describe the same Quebec.  The rest of the Provinces and Territories have just the Federal system.

--Gannon


--- On Sun, 10/31/10, Michael Hausenblas <michael.hausenblas@deri.org> wrote:

From: Michael Hausenblas <michael.hausenblas@deri.org>
Subject: Re: is it possible to handle an XML/HTML elements attribute via the     URI?
To: "Julian Reschke" <julian.reschke@gmx.de>, "Matthew Millar" <mattmill30@hotmail.com>
Cc: uri@w3.org
Date: Sunday, October 31, 2010, 3:45 AM

 
>> I apologise if I've mis-understood how a fragment works, i thought it
>> was used to instruct the browser, how to behave (i.e. where to start
>> displaying the visible page from).
> 
> That *is* how a fragment works.

My 2c:

Discussing the fragment's semantics without taking the representation's
media type into account isn't really helpful (cf also my write-up at [1]).

Cheers,
      Michael

[1] http://www.w3.org/2008/WebVideo/Fragments/wiki/Semantics

-- 
Dr. Michael Hausenblas
LiDRC - Linked Data Research Centre
DERI - Digital Enterprise Research Institute
NUIG - National University of Ireland, Galway
Ireland, Europe
Tel. +353 91 495730
http://linkeddata.deri.ie/
http://sw-app.org/about.html



> From: Julian Reschke <julian.reschke@gmx.de>
> Date: Sun, 31 Oct 2010 10:38:34 +0100
> To: Matthew Millar <mattmill30@hotmail.com>
> Cc: <uri@w3.org>
> Subject: Re: is it possible to handle an XML/HTML elements attribute via the
> URI?
> Resent-From: <uri@w3.org>
> Resent-Date: Sun, 31 Oct 2010 09:39:13 +0000
> 
> On 31.10.2010 02:29, Matthew Millar wrote:
>> Hi Julian,
>> 
>> Thanks for the tip, I've had a look at the target-pseudo, and it seems a
>> really useful feature.
>> 
>> Its not quite what i was getting at, as the second example of an
>> embedded video, is perhaps a better example of the usefulness of my idea.
>> 
>> I apologise if I've mis-understood how a fragment works, i thought it
>> was used to instruct the browser, how to behave (i.e. where to start
>> displaying the visible page from).
> 
> That *is* how a fragment works.
> 
>> I thought this idea, may have been an extension of the fragment feature,
>> as it would be instructing the browser how to behave, but would be
>> useful, for third-parties who want to harmlessly manipulate a page for
>> their own benefit. e.g. changing the quality of video playback.
>> 
>> If the URI isn't the correct place for handling element manipulation
>> (and won't be considered in future developments), could somebody explain
>> in lame-mans terms, why?
> 
> You're trying to overload URIs with something they haven't been designed
> for.
> 
> Also: you say "manipulate", "third-party", and "harmless" in one
> sentence :-). You'd need to design this in a way so no harm can be done.
> That sounds very hard, considering that you want to essentially rewrite
> the page.
> 
>> Or better yet, could they recommend where i should look/ask for a
>> feature that will allow users to easily, and safely manipulate a
>> web-page to behave in a user tailored manner to the original developers
>> generic design - XSLT perhaps? I'll have to read up about it (Thanks Claus).
>> 
>> Thanks,
> 
> When you say "user" you apparently mean "programmer" or "web author".
> The average user doesn't understand URIs or HTML.
> 
> If you want to change the behavior of sites with modifying the sites,
> bookmarklets and browser extensions seem to be what you should be
> looking at.
> 
> Best regards, Julian
> 
>