RE: New test for review: Bidi misc

> From: Martin Duerst [mailto:duerst@w3.org] 
> Sent: 07 October 2004 23:51
> To: Simon Montagu; Richard Ishida
> Cc: GEO; www-international@w3.org
> Subject: Re: New test for review: Bidi misc
> 
<snip/>

> Some comments for Richard while I'm looking at this:
> 
> - Given that the page says "You will need an Arabic font and Arabic
>    rendering capabilities to compare the glyphs in these 
> tests.", I think
>    it would indeed be a very good idea to have a Hebrew test, too, so
>    that people interested in Hebrew can test the browsers they are
>    interested in without having to install fonts,...

Given some other mail I received, I think it might be good to use words as
follows:

French Arabic Hebrew German 

This should not affect the expected ordering of directional runs, but having
two RTL directional runs side by side improves the test - see the results
for Firefox at 
http://w3.org/International/tests/results/bidi-misc


> 
> - Wording nit: "Check the window title bar contains" ->
>    "Check that the window title bar contains"

Done.

> 
> - "The html element contains dir="rtl"." -> "The html element of
>    this page...".
> 

Done.

> - "Mouse over the graphic below": I think there are other ways
>    than just 'mousing' to activate a 'tool tip'. Also, I think that
>    'tool tip' is a Windows-specific term, and different browsers may
>    have different ways to show the alt or title attribute on an image.
>    (do we know which one is shown? do we know whether there are
>     differences between the two?)

This test is specifically for tooltip rendering, so I'm content to produce a
tooltip by any means. Note that if you display attribute text using CSS
:before, say, it will be rendered appropriately because it is rendered with
the rest of the text.

> 
> - It seems inconsistent that <title> has its own dir='rtl', but
>    the image with the alt/title attributes doesn't.


Thanks for pointing that out.  I forgot to remove it after some
experimentation. Now gone.

> 
> Regards,   Martin.
>

Received on Friday, 8 October 2004 10:49:08 UTC