Talk:Moved permanently 10

• Step 2:
  • I think we need an additional manual step to confirm that the adjacent images really form a group of images that need only one alt.
  • Agree with you suggestion in the second "open question" box. The appropriateness of the alt attribute of the group should be checked by a separate test. Mainly because the report must be connected to the image that actually contains the alt. By adding a test "img-group-alt" (or similar) we also avoid that the same image group is presented multiple times to the user.
    • Agreed Wilco Fiers (talk) 14:16, 30 October 2014 (UTC)
• Step 5: "Check if the <img> element is part of an a element containing additional text." Too vague. Does this mean only parent elements or other ancestors as well? Can the text be contained in sibling elements? Maybe we need a restriction on the type of element that can contain the additional text.
• Step 6: Add instruction what to present as "corresponding alternative text".

--Annika Nietzio (talk) 13:48, 30 October 2014 (UTC)

• This test involves a lot of steps repeated in other tests as well about finding which attribute provides the text alternative. Should we instead move these steps into an algorithm which different tests can reuse, so we can keep our tests DRY?
• Should we create a template for user questions? I'd like to force a little more consistency into this.
• I think the way attributes are indicated is confusing ( <title>, <aria-label>, etc.) this makes them look like elements. Please use title="" instead. Do we need to define a style guide for this?

Wilco Fiers (talk) 14:16, 30 October 2014 (UTC)

• Step 6 may also pass if the image is decorative (step 7), we should test if it is decorative everywhere before we fail

--Hanno Lans (talk) 15:03, 30 October 2014 (UTC)