RE: Updated Testing document

Hi Tim,

I'll take a shot at answering:

> In the beginning of the document, under "decisions to make prior to testing", could additional guidance be given as to how to make the decisions indicated under points 1 and 2? 

JR: Sure. If you have specific ideas please send them over.


> For point 1, where would this info come from (authoring tool developer, tester, others)?  

JR: It depends on the context. If the developer advertises a compliance level that's a good place to start. If the evaluation is by a procuring organization (e.g. a govt) that has already set an expectation, then they would follow that.


> Also, what if a tool satisfies some but not all level A, but a lot of level AA, etc. - how would this fit into the decision making process - could be "all over the map" when it comes to level definition..)?  

JR: It depends on the purpose of the testing. Some testers might just be trying to (dis)confirm a conformance claim, others might be trying to take a more detailed audit of what's there and what needs improvement.


>For point 2, what is the meaning of "included" (included in the testing process.., or included from the defined subset mentioned later)?

JR: Both. Point 2 is asking the tester what technologies they want to consider to be in the subset for their testing. This will determine what parts of the tool are in and out of scope. For example, if I'm testing the Dreamweaver's production of HTML4.01(+CSS2.1,+JavaScript) I don't need to test anything to do with its production of ActionScript.


> Can the defined subset be null, or if only 1 of 10 web content technologies produced is tested for, is that sufficient?

JR: The "subset" refers to the list of technologies to be tested as a subset of all those that the tool can be used to produce.


> Also, is the meaning of "by default" not requiring any author interaction to specify a web content technology, or something else?

JR: (perhaps this also should go into the glossary) "default (by default)": refers to how an authoring tool operates prior to any modification of its settings by the author. In cases where there is no author setting to modify, the default behavior is the only behavior. In some cases, there is no default behavior other than waiting for author input.


Cheers,
Jan


> -----Original Message-----
> From: Boland Jr, Frederick E. [mailto:frederick.boland@nist.gov]
> Sent: March-20-13 10:58 AM
> To: Jeanne Spellman; AUWG
> Subject: RE: Updated Testing document
> 
> Thanks for updating this!   The document is shaping up nicely I think..
> 
> A few questions following..
> 
> In the beginning of the document, under "decisions to make prior to testing", could
> additional guidance be given as to how to make the decisions indicated under points
> 1 and 2?  For point 1, where would this info come from (authoring tool developer,
> tester, others)?  Also, what if a tool satisfies some but not all level A, but a lot of level
> AA, etc. - how would this fit into the decision making process - could be "all over the
> map" when it comes to level definition..)?  For point 2, what is the meaning of
> "included" (included in the testing process.., or included from the defined subset
> mentioned later)?    Can the defined subset be null, or if only 1 of 10 web content
> technologies produced is tested for, is that sufficient?   Also, is the meaning of "by
> default" not requiring any author interaction to specify a web content technology, or
> something else?
> 
> Thanks and best wishes
> Tim Boland NIST
> 
> -----Original Message-----
> From: Jeanne Spellman [mailto:jeanne@w3.org]
> Sent: Wednesday, March 20, 2013 10:37 AM
> To: AUWG
> Subject: Updated Testing document
> 
> Jan has updated the testing document:
> 
> http://www.w3.org/WAI/AU/2013/ATAG2-10April2012PublicWD-Tests
> --
> _______________________________
> Jeanne Spellman
> W3C Web Accessibility Initiative
> jeanne@w3.org
>

Received on Wednesday, 20 March 2013 17:32:26 UTC