Re: Focus-appearance testing

Hi Jon,

Thanks for taking a look. On your points:

If you think of (or find) a page which has a better set of examples, please let me know.

> we should be sure we are comparing what we think is the indicator

One of the instructions is to record any uncertain aspects in the notes, so I think we’ll capture the assumptions. Most of them are straightforward, but there are a few people might need to make a note about.

> I had trouble finding the dotted indicator style on the Microsoft site in dev tools…

Did you see the notes at the top with some instructions on that? The screenshot is of the MS dotted style indicator in dev tools.

> So it appears in my case a 2px border is sometimes not 2px and sometimes it is!

I also put some notes about that before the examples, it is important to be at 100% zoom on a 1:1 screen or the browser will adapt the view and you aren’t getting the authored styles.

Kind regards,

-Alastair


From: Jon Avila <jon.avila@levelaccess.com>
Date: Tuesday, 3 January 2023 at 15:14
To: WCAG list (w3c-wai-gl@w3.org) <w3c-wai-gl@w3.org>
Subject: RE: Focus-appearance testing
Hi Alastair, yes, I believe the exercise is important.  The examples seem to include different types of situations such as dotted indicators, changes in background/foreground color, solid indicators, etc.  I’m not sure if the examples are inclusive of everything.  One thing I think it might be good for testers to indicate is what they considered the indicator.  In some cases you could have multiple things as the indicator – although some of them may not be sufficient.  For example, the text color change, the background color change and an underline appearance.  So if we are comparing results we should be sure we are comparing what we think is the indicator and if we believe that passes or fails.

I had trouble finding the dotted indicator style on the Microsoft site in dev tools and while I could take a screenshot- I was hoping that we could provide more info on how to find different types of indicators in developer tools.  When I did take a screenshot I found it hard to tell what was a pixel – I used the gridlines feature in MS Paint to show what I believe are a box per pixel.  Although it appears that some content is ant aliased which could cause some confusion.  Another challenge I discovered with some other indicators was trying to determine if the border change was actually on the border or outside the border. The browser (Chrome in my case) indicates  in dev tools a 2px border – but when I take a screenshot the border only appears to be 1 pixel.  Other times the border says 2px and appears to be 2px.  So it appears in my case a 2px border is sometimes not 2px and sometimes it is!  We may need more guidance for non-technical people.  Other tools such as an overlay line of 1px might be useful to some as well.

Jonathan

From: Alastair Campbell <acampbell@nomensa.com>
Sent: Tuesday, January 3, 2023 6:06 AM
To: WCAG list (w3c-wai-gl@w3.org) <w3c-wai-gl@w3.org>
Subject: Re: Focus-appearance testing

I’d like to bring this up again, post holidays, to see if anyone can take a look at this…

From: Alastair Campbell
Date: Wednesday, 21 December 2022 at 12:57
Hi everyone,

Something that came out of a previous meeting was to test the complexity / reliability of focus-appearance in practice.

I’m trying to set that up, but before people start filling it in, I’d like to check:

  1.  It makes sense as an exercise;
  2.  The examples are reasonable.

If you are interested and have a few minutes, please take a look at the instructions:
https://docs.google.com/document/d/1lE7jN_BVvGz48StHODhvvRxRp0XMhGtL0P0Pp8oM-ms/edit#<https://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fdocs.google.com%2Fdocument%2Fd%2F1lE7jN_BVvGz48StHODhvvRxRp0XMhGtL0P0Pp8oM-ms%2Fedit%23&data=05%7C01%7Cacampbell%40nomensa.com%7Ca56325c3d6f642fa376208daed9d2c1c%7Cebea4ad6fbbf43bd8449c56e26692c35%7C0%7C0%7C638083556576615601%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000%7C%7C%7C&sdata=6PIGW%2FqeyKGFUrCHQdhqJQzher%2BCnuYqIWdYEUxWcCk%3D&reserved=0>

If you have minor improvements, just reply to me and I’ll do them.

If you have a better (and feasible) approach, or have better examples to use, please reply to this thread.

Kind regards,

-Alastair

--

@alastc / www.nomensa.com<https://eur01.safelinks.protection.outlook.com/?url=http%3A%2F%2Fwww.nomensa.com%2F&data=05%7C01%7Cacampbell%40nomensa.com%7Ca56325c3d6f642fa376208daed9d2c1c%7Cebea4ad6fbbf43bd8449c56e26692c35%7C0%7C0%7C638083556576615601%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000%7C%7C%7C&sdata=Wxi0zDsmOvlzzEMV7j91PKgmXaAZAXKlhWH6nB0VucU%3D&reserved=0>