2012-01-05. Frozen archive - links may not resolve - see directory of files at MoinMoin wiki archive

> AccessForAllDrafts

Three documents have been posted by ISO for public consideration. They are a Framework document, and two parts (expected to be the first of many).

It would really good if these documents could be read and commented on by a wide range of people. In the ISO context, where they are currently being processed, only countries can vote on them but all comments are welcome here.

The docs are available for public comment at:

If you wish to comment, please add you name. Instructions for editing a page to add your comments are available below. Please try to make comments in an orderly fashion - we have found it is a good practice to add a comment and then a horizontal line.

Lisa Seeman

I have not finished reviewing it, but I thought I should submit some comments as time is running out. The use case and the comments on Personal Needs Statements may be the most important.

Frame works comments:

Page 7 Access strategies can also be automatically learned and inferred from user behavior. This may become more important as adaptive interfaces adapt to transient user needs as users skills improve determinate, or they just become tiered….

The extent to which elearning systems can adapt resources is a bit limited(to supplement , reformat and replace). Although these our two core elements of an adaptive interfaces, the main point is to tailor the content to the users access strategies

Page 8

Some Metadata – such as RDF can also be used to enable a resource to be adapted to the users access strategy. Any Metadata can actively increase the accessibility of a resource. For example, adding a DC: description to a image file such as SVG can be reused to provide alternate text - That may be out of the scope of this document but may be valid to mention

Further, beyond the discovery of appropriate alternative or supplementary representations, one can also use metadata to discover other metadata or resources that enable content adaptation of the current resource.

Therefore (in page 9) I think we also need a resource to relate to other resources that enable adaptations., The combination , with the right software could be more accessible then the initial content.

Further, with this model if the resources has not been mapped to a new or less popular Access for all element, then the model is no longer useful.

I would suggest building in another step, Knowledge about the resource, and it is that knowledge that is mapped to the access for all profile. Then that knowledge can be reused to map to a new definitions as the are created.

I also found the yes / no of the UML diagram ( whether to display a resource) a bit crude. It does not address the preferred value of Personal Needs Statements. How will that work in the case of learning disabilities, where what you want is the best match to the user needs, but not necessarily a perfect match? Is that what is happening as the recourses augment?

Let me give a use case,

A student is a new immigrant from rural Ethiopia. They are not used to computers or textbooks. A resource may have a page narrative, that explains things like bullet points, and how the page is structured. As the student is a midway through the resource the student will need less narrative to understand and follow the material.

I would say that ideally one would look for a recourse with page natives, that have been tagged for different types, or an application that works with knowledge based content tags , so they can be turned off as the user skill improve.

Personal needs statements: diactric marks for Hebrew and Arabic, are not mentioned This is the key barrier to access in these languages for many people, Partial diactric marks are also useful.

How well is the page encoded – Is a visual architecture used, have bulleted lists been marked up as such so it can be adapted for my scenario? I also this more and more applications will be looking for more explicit knowledge in the markup, such as use of roles.

Adaptation type may have an underdeveloped value space, for example concept coding as a pose to plain BLISS as a pose to other symbolic languages.

Content does not address at all the cognitive skills needed to use a site – For example, do you need a good visual memory to deal with the constant reuse of confusing and similar acronyms?

How would the content address the use case given above? Low education level? That is not necessary true. It is, maybe, a low western education level.

Should there be space for emotional disabilities, are there warning available for potentially traumatic material.

Lisa Seeman on Digital Resource Description

I may have missed how each attribute in this work (Digital Resource Description ) maps to each user preferences as described in the framework. I think that mapping would make a good appendix and may identify some holes

I think we have fallen a bit short on dealing with the learning disabilities spectrum, For example, in making content for someone with an attention related mismatch (Yes that is a great way to describe disabilities…) such content would be well marked up with main point highlighted. How can I let the user know that this content is tailored to them? Again what about cognitive skill required to access a page beyond a curable the affect of a disability such as reading age?

Again, look at the user scenario above and see if we can encode it using the current model. How would the content adequately address the use case of content support for the new immigrant?

Quote “with W3C WAI priority 1 and 2 ensure that the presentation and control of text is transformable.”

I think we meant that text is transformable to brail or speech in some languages. (That does not mean that the text is transformable to symbolic languages ( sign or BLISS), or is correctly read in Hebrew or Arabic.)

I am not sure how dependent this work is on WCAG. WCAG have been concerned about the need of legislations in their conformance architecture. As a result they demoted or removed success criteria things that were not widely applicable (sometimes because of due burden) . As a result Success criteria related to barriers of understanding tend to be demoted or removed. Hence conformance of WCAG 2.0, P1 or P2 may be less concerned with what to educational content may be the main point: Barriers to understanding.

Unfortunately current accessibility conformance statements are not build around severity inside a type of violation. For example, take the violation to have accessibility alt tags, filled in, Now were as some may be important, if a resource has left out an alt tag on a pixel spacer image it is not a huge problem.