W3C home > Mailing lists > Public > w3c-wai-er-ig@w3.org > October 2001

Re: Validation as EARL

From: Nick Kew <nick@webthing.com>
Date: Wed, 24 Oct 2001 01:36:35 +0100 (BST)
To: Charles McCathieNevile <charles@w3.org>
cc: <w3c-wai-er-ig@w3.org>
Message-ID: <20011024012159.A991-100000@fenris.webthing.com>

On Tue, 23 Oct 2001, Charles McCathieNevile wrote:

>   Ahem - did you read the page?  The validator is generating XML to start
>   with: the EARL is generated by filtering it through XSLT!
>
> Yep, but I meant RDF XML syntax instead of N3 syntax. (there are online
> conversion services available to do this - Sean probably has a URI to hand).

Hmmm - fair enough, but is there any advantage to using EARL as your
starting point over using XML?

>
>   >	 and the second is to store some data on test cases,
>   > fix them and see how we go about comparing two csequential reports.
>
>   I'll be interested to see what that looks like.  It sounds like
>   precisely the kind of thing to help judge whether the EARL reports
>   are going to be useful in the real world.
>
> I think it looks like two sets of reports that have some conflicting
> statements about the same objects. They would also differ in author and date.

- and presumably the acid test is whether you can really gain anything
by combining those reports?

>   > The other second thing I would like to do is to hook it up to the Annotea
>   > system for storing the information, and then play with the threading of
>   > reports in Annotea.
>
>   Another one I'll be interested in ...
>
> This is really a mechanism for doing the previous task, but it is in
> development - I believe the next version of Amaya will have a working user
> interface to play with it.

Where does this get discussed?  Is there any reading material?

>   > The other second thing is to collect other information that has some overlap,
>   > and then to squish the data together and se what we can do.
>
>   - and that's a direction I'm looking to go in myself!
>
>   OK, if you are pursuing these projects, just let me know when you are
>   about to reach the point where something will break if my stuff changes,
>   and we'll see to it that you can have something stable to work with.
>   In the meantime, I still need your comments on the reports I'm generating,
>   as the format and contents of these will very likely be subject to
>   change as I get to understand the things better.
>
> Well, since I am looking to chain online services together, changing the URI
> will break my bookmarkelts I am assuming that (modulo temporary bugs) the
> output will continue to be available as RDF in N3, so I can use a service to
> convert from (say) EARL 0.95 to EARL 1.0

hmmm, well,

I'll post advnce notice of any change to this list, so you get the
chance to shout if it looks likely to break something.  If you
absolutely want an immutable URL, I'll run it (also) at some
path away from any namespace I might want to re-use.  Maybe
something like valet.webthing.com/Chaals/ ?

-- 
Nick Kew

Site Valet - the essential service for anyone with a website.
<URL:http://valet.webthing.com/>
Received on Tuesday, 23 October 2001 20:45:40 GMT

This archive was generated by hypermail 2.2.0 + w3c-0.30 : Thursday, 9 June 2005 12:10:39 GMT