Re: next chat - 29 May 2002

Charles McCathieNevile wrote:

> On the other hand, in talking to the folks at UsableNet they said that the
> location thing is going to be really pretty important for providing
> interoperability in the scenario wher a repair tool gets the information
> about what is wrong for a different tool. This is backed up by the fact that
> HiSoftware felt the need to have a tightly-specified way of referring to the
> location of a problem, and used their own namespace to do it.

Two quick comments.

If one purpose of EARL is to promote interoperability between different 
tools that do evaluations and/or repair (like reading reports done by 
two or more tools and show an integrated view of the status of a page to 
a user; or tool A that produces a report to be used by tool B for 
repair; or tool A that produces a report that is read by tool B and B 
wants to track the status of some problem across time), then I think 
EARL should specify how to represent the location of a problem. 
Therefore I would promote the location from the private namespace into a 
more 'public' namespace.

Secondly, as the location of a problem is already somehow represented 
within all the tools, EARL specification for it needs to encompass more 
than one possibility. I don't think it is viable to assume that EARL 
specifies the location as something (say # of line/column in the page) 
and then that every tool should adopt it.


Received on Thursday, 30 May 2002 12:53:14 UTC