Fw: Help on EARL

----- Original Message ----- 
From: "Sean B. Palmer" <sean@mysterylights.com>
To: "Venkat Injapury" <ivenkatn@hotmail.com>
Sent: Wednesday, March 06, 2002 6:48 PM
Subject: Re: Help on EARL


> Hi there Venkat,
> 
> > Is there any EARL validator on the market like RDF
> > validator. What is use of EARL in Accessibility Evaluation
> > and Repair Tool. Can i have any real time example of
> > EARL report? [...] How can i approach to build the EARL
> > for the given html page. Please advise me in these regard.
> 
> I'm very happy to help you, but you may also want to forward your
> query to w3c-wai-er-ig@w3.org - a publically archived [1] mailing list
> to which I, and many other ERT members, are subscribed. Those people
> should be able to help you as much as I can.
> 
> As to your question about an "EARL validator", there isn't such a
> thing as an EARL validator as such. In RDF, you define general data
> models, and so validation is a matter of specifying where you find
> inconsistencies. In other words, it's different from RDF validation
> (presuming that you refer to the W3C's online validator at [2])
> because it's model validation, rather than syntax validation.
> 
> As for work on model validation, I did some work of my own with RDF
> Lint [3], and TimBL has gone a lot further with his own validator.n3
> [4]. I've run these tools against the EARL schema (they are mainly
> schema checking tools), and they work well. Instance data "validation"
> is a similar sort of thing, I presume, but I haven't run any
> experiments so far.
> 
> Another idea that we had was to produce a canonicalized version of
> EARL [5]. In other words, this would be a strict subset of the RDF XML
> syntax, just for writing EARL documents. Then, we should be able to
> come up with an XML Schema or RELAX NG grammar against which people
> could validate. However, it is not evident as to whether or not one
> could guarantee that a syntactically conformant canonicalized EARL
> document contains a consistent model. Indeed, I'm not even sure that
> producing a syntactic grammar is possible, and so that project has
> been pretty much suspended.
> 
> Your next question is "What is use of EARL in Accessibility Evaluation
> and Repair Tool". The ERT group have been working on various scenarios
> for EARL ever since we started. One of the most important features of
> EARL is that it allows reports to be merged, filtered, and queried
> very quickly. You can think of it as being a huge decentralized
> database: it doesn't matter so much who does the assertions etc.,
> because as long as the data is fairly consistent, you should be able
> to merge/split evaluation reports at a later date.
> 
> As an example of where this may be useful, consider the following
> scenario:-
> 
> * The WCAG guidelines group (producing a specification on
> accessibility) need to ground their guidelines and checkpoints in
> specific cases on the Web. Since there are many members, and many
> checkpoints, the tests need to be compiled with an incremental effort:
> with a few people working separately, and then merging their results
> at the end. EARL will make it easier to manage such a project.
> 
> There are other scenarios as well: for example, ones associated with
> storing the EARL reports in an annotations server (a hot topic on the
> lists at the moment), but you'll have to ask the ERT members for
> these. I'm just the schema lackey... :-)
> 
> Next Q, "Can i have any real time example of EARL report?". Try the
> "EARL 0.95 by example" page at:-
> 
> <http://lists.w3.org/Archives/Public/w3c-wai-er-ig/2002Feb/att-0031/01> -0.95byExample>
> 
> At the moment, there aren't all that many examples floating about for
> EARL 0.95. This is partly due to the fact that design changes since
> the 0.95 release (some driven by the current implementations of EARL)
> are moving us towards a 1.0 version of EARL, which the ERT group seem
> to be hoping to formalize soon. The discussions on this matter are
> ongoing. In any case, we will be publishing lots of examples of the
> next version of EARL for your persual.
> 
> If OTOH you're looking for some examples that are actually produced by
> applications, you can try one or more of the implementations listed
> under [6].
> 
> Onto the next question, "How can i approach to build the EARL for the
> given html page." Actually, I'm not too sure what you mean by this
> question. Possibly one of: how do I make evaluations about a given
> HTML page; or how can I screen scrape some XHTML into an EARL report.
> I'll assume you meant the former.
> 
> For now, you can use the bookmarklet listed in the implementations at
> [6], or Jim's bookmarklet hack outlined at [7]. Hmm... these are
> really just implementations of EARL. If you meant the latter (screen
> scraping XHTML into EARL), I've not really considered it before, but
> it should be possible with XSLT.
> 
> P.S. If you don't mind, could I possibly forward this mail to another
> publically archived list? I'd like people to be able to refer to this
> message if they have similar questions to your own.
> 
> [1] http://lists.w3.org/Archives/Public/w3c-wai-er-ig/
> [2] http://www.w3.org/RDF/Validator/
> [3] http://infomesh.net/2001/05/rdflint/
> [4] http://www.w3.org/2000/10/swap/util/validate.n3
> [5] http://lists.w3.org/Archives/Public/w3c-wai-er-ig/2001Jul/0003
> [6] http://www.w3.org/WAI/ER/#earl-implementations
> [7] http://jibbering.com/2002/2/earlassertions.html
> 
> --
> Kindest Regards,
> Sean B. Palmer
> @prefix : <http://purl.org/net/swn#> .
> :Sean :homepage <http://purl.org/net/sbp/> .
> 
>

Received on Thursday, 7 March 2002 10:31:47 UTC