Re: Web Accessibility Tutorials: Images and Tables

Hi Eric,

thanks for taking the time to look into this, much appreciated!

Let me try to explain why I think your explanation [1] does not align with the rules explained on the "Images Concepts" tutorial overview, and furthermore, even if you adjusted that overview page to accommodate the way visuals are used on the table tutorial overview page, [2] why I think it is still inadequate.

[0] Some background:
- the visualisations for the table types are actually quite useful, the duplicate some (but not all!) text based info, so they create redundancy (which is good) by repeating essential information using different mechanisms (text plus simple graphic)
- the visuals help users who are struggling with the text to complete their understanding of what each bullet point is trying to say about a given type of tables
- the visuals are valuable as independently as for example a diagram is in a case where numerical information is provided on a textual basis (because it makes certain aspects more prominent than others, thus making consumption and interpretation of numerical information easier)
- it is relatively straightforward to create pretty good standalone descriptive text for each visual (it can be much harder in other cases)
- the way alt text is used for some visuals in the table tutorial overview page essentially turns it into text that becomes part of the text flow for those readers that consume alt text

[1]
- functional images as explained in the Images Concepts tutorial overview page and the detailed page about it would not make anybody expect that it could also be turned into a portion of a text flow, so I have difficulty following your reasoning here
- the visuals are definitely too strong for being considered decorative images
- personally I think they are just informative images - actually, very informative, so they deserve a standalone descriptive alt text

[2]
Now let me explain why this whole issue bothers me:
I can't rid myself of the impression the approach as executed here is highly screenreader centric: 
- the current state of the table tutorial overview page could work quite well for screen reader users (the only thing that will probably confuse screen reader users is the fact that their AT will announce the presence of an image, and if they wanted to understand which piece of the text flow that is being presented matches the image portion, text like "typically have one header row and/or one header column" .. hmm… is interesting)
- other users (low vision, dyslexic, essentially anyone who can see but not necessarily easily, or who can read, but not necessarily easily, or who can see well and wishes to scan the page for visuals but has limited motor capabilities, … ) are not served well here by the current alt texts, and not served at all by the absent alt texts.
- my position would probably be a little bit different if the visuals were less strong and more decorative - but they aren't. Having done some research into tables I do know it is quite difficult to make others understand the various table concepts, and getting the visuals right is not easy (yes, I do think the visuals are really well crafted, they are very good abstractions of each of the types)

[3]
The last bit goes back to my limited familiarity with details in HTML:
- is 
	alt
really equivalent to
	alt=""
?
- if so, why do the examples in W3C WAI related content I have run into (and definitely those in the image tutorial)  use
	alt=""
?


Olaf



On 12 Sep 2014, at 08:29, "Eric Eggert" <ee@w3.org> wrote:

> Hi Olaf,
> 
> the intent is that the alternative texts are used where having the image has an additional value to determine if the page is what the user is looking for. It is – depending on the context – a functional or a decorative image. The alt text does complete sentences. For example:
> 
> Basic Tables [typically have one header row and/or one header column]: For simple tables with content that is easy to distinguish, mark up header cells with <th> and data cells with <td> elements.
> In this case, Basic Tables is a good heading, but the image (as well as the alt text) help to get a better understanding on what basic tables actually are.
> 
> In the Irregular Tables example, the description is completely in the text after the colon, as I felt the image is only representing a certain example of irregular tables and the text has more (or the same) info in it.
> 
> I also think that captions are pretty self-describing and therefor don’t need an alternative text to further describe the redundant image.
> 
> I’m looking forward to your feedback on this.
> 
> Best, Eric
> 
> in the tutorial overview about tables the images illustrating the types of tables are tagged in a somehow weird and inconsistent fashion.
> 
> For the basic table image, the alt text is
> typically have one header row and/or one header column
> For the multi-directional table image, the alt text is
> usually have headers for rows and columns
> 
> For the irregular table image, the alt text entry does not have a value
> 
> For the multi-level table image, the alt text is
> have multiple header cells associated per data cell
> 
> For the Caption & Summary image, the alt text entry does not have a value
> 
> Apart from being a tad inconsistent, I consider the ones that are filled out to be inadequate - I have difficulty seeing how they serve as alternate text…. - they don't fit any of the descriptions in the image tutorial.
> 
> What am I missing?
> 
> Olaf
> 
> --
> 
> Eric Eggert, Web Accessibility Specialist
> WAI-ACT Project
> 
> I’m yatil on IRC.
>