Re: trusting providers Re: Accessible web content services.

For "I looked at this" you could use EARL evaluation results directly -  
since they record who made the claims in an evaluation. Asking humans to  
sit down and code EARL for this purpose strikes me as ridiculous. You  
might like to look at Hera - a tool in development, whose user intereface  
still has some problems, but which works through a form and different  
style sheets applied to the page and then generates the EARL (or a plain  
HTML report).

Other tools generating some flavour of EARL include AccessValet,  
AccVerify, the development version of the Wave, the W3C validator, and so  
on.

As for trust vocabularies, I think the FOAF people are a good mob to ask.  
You want to be able to specify who you trust, something about how much,  
and very importantly in what context. (I trust Tim Berners-Lee's judgement  
on Web Architecture as a rule. As a rule, I don't trust his judgement on  
restaurants in Melbourne...)

cheers

Chaals

On Mon, 2 Aug 2004 07:54:30 +0100, <David.Pawson@rnib.org.uk> wrote:

Chaals said
>     the authors/creators, e.g. from Dublin Core, FOAF, etc. The
>     cnfusion is over what exactly someone did - whether they
>     are listed as author because they created the website (as
>     in the generator value for the meta element that many tools
>     put in - there's a quick way to get a sense of how good
>     tools usually are in practice) or because they wrote the content.

> dp. Just addressing this part, I can get author, and with dc:author,
> I'd know it was the person who 'wrote', i.e. generated, the site.
> What about, I checked this site out in detail, Charles, I think you've
> done that on occasion? I won't find such properties as that in foaf
> or dc? That's where I think this group could help.

> Equally, I recommend this person or organisation, again I won't find
> that in foaf.

>   Without some indicator of vocabulary, any such set of assertions
> would soon fall apart. Protege allows us to generate an ontology.

> I'm less sure about asking people to write RDF assertions about
> authors or sites, the foaf 'form' idea seems much more reliable.
> I think protege does this too, but I haven't checked it out yet.

> re tools to search such a collection of assertions, I think that's
> what makes it worthwhile! Especially if the source of the assertion
> can be ascertained.



-- 
Charles McCathieNevile     charles@sidar.org
Fundación Sidar             http://www.sidar.org

Received on Monday, 2 August 2004 03:23:01 UTC