Re: Overview of testing in view of CR exit

On Tue, 14 May 2013 16:38:54 +0200, Robin Berjon <robin@w3.org> wrote:

> Hi,
>
> based on the discussion we had at the face to face, I've made a pass  
> over the ToC to reflect the notions we had about what is considered  
> stable on its own (as per exit criteria), what requires testing, and in  
> the latter set what has implementations and/or tests (I took a  
> conservative approach to flagging that and will be refining it to add  
> more).
>
> This provides the basic information from which to start planning the CR  
> exit. All the parts that are flagged as needing testing but not having  
> tests will need to have tests written, and once we have tests for all of  
> them they will need to be run through two implementations to produce an  
> implementation report.
>
> See:
>
>      http://www.w3.org/html/wg/tests-cr-exit.html
>
> Comments welcome,

It might be useful with a legend for "Section has no implementation  
requirements" so that it's easier to separate from "Considered  
interoperable".

I think it would be closer to the truth if the default is that any  
implementation requirement in the spec is *not* interoperable and needs  
tests.

For instance, section 2.5.6 Colors, I have an old test case:

http://simon.html5.org/test/html/rendering/color-attributes/no-quirks.html

This is probably not up-to-date with the current spec, but even so, I get  
different failures in Opera, Chrome and Firefox, showing it is not  
interoperable.

I could go on and demonstrate cases for sections marked "Considered  
interoperable" in fact aren't, but I don't think that's a good use of my  
time. Hopefully my point gets across anyway.

-- 
Simon Pieters
Opera Software

Received on Friday, 17 May 2013 14:29:24 UTC