- From: Kelly Ford <kford@windows.microsoft.com>
- Date: Sun, 19 Oct 2008 20:49:04 -0700
- To: "w3c-wai-ua@w3.org" <w3c-wai-ua@w3.org>
- Message-ID: <2828BDE8DC61004E8104C78E82A0B39710ADDCBB4A@NA-EXMSG-W601.wingroup.windeploy.ntd>
Hello, Below are my notes after reviewing for my assigned items of clarity, testability and conciseness. Note, I didn't spend as much time on principles 4 and 5 yet. Clarity, Conciseness and testability 1. Abstract presently mentions it is important to communicate with assistive technology. Should we mention that it is important to support accessibility APIs where they exist in the intro as well? 2. What's the point of the following sentence in the abstract Technologies not addressed directly by this document (e.g., technologies for Braille rendering) will be essential to ensuring Web access for some users with disabilities. I think of a Braille display much like a monitor. It is the job of the assistive technology in the case of the Braille display to determine what's displayed. This sentence just seems kind of out of place. 3. The abstract needs to be strengthened in terms of indicating how web content is important. Right now I'm left with the impression that this is more about user agent user interface. 4. The intro section gives two definitions for user agent. Should we have two definitions, even if one is what the document indicates we primarily use. 5. 1.1.1, dealing with following operating system conventions, has a specific point talking about installation being accessible yet earlier in the document we have an entire section saying in part: This document does not include a success criteria requiring that installation procedures be accessible. 6. There are many references to telling the user agent developer to follow operating system conventions, not to interfere with standard items and such. From a strictly pass/fail perspective this makes things a bit tough to test because you are depending on the developer to tell you which conventions were followed and such. We can list some of these in supporting documents and such but to some degree this is a risk for testability. 7. I wonder if principle 2 would be better renamed programmatic access or something similar. This currently talks about facilitating access by AT but is really about programmatic access. 8. I think we need to reduce principle 2 down to a smaller number of guidelines. Why for example do we separate HTML from XML? Is this really necessary? 9. 2.10 dealing with timely exchanges is not testable. 10. Principle 3, in particular 3.1 needs to be more clear. I've mentioned on a few calls how the user interface does a lot visually to convey meaning. I don't think these guidelines give enough detail on what needs to happen or the expectation around accessibility. User agents today do a lot to convey state from security to page functionality with unique UI elements. AT jumps around loads of hoops today to convey this info. 11. 3.4 Isn't clear enough on what of this has to be done by the user agent, versus just ensuring the info is available programmatically. 3.4.2 For example says the user can access info on different relationships. Is it enough that this is available programmatically and then the user's AT conveys this or do I have to do more in my user agent. Further, for testing purposes, the exactly requirements that need to be exposes need to be more explicit. 12. The list of items in 3.6.1 needs to be looked at. Visited links for examples seems like something that can be removed. But much like we've done with keyboard, we should be asking if this is the right list in general. 1. 13 I wonder if 3.10.1 and 3.10.2 can be combined. Reading these today for example there's no requirement that the user be able to specify a stylesheet, just that if the user has done so, the user agent allow it to be used. What these items are really collectively saying is that the user agent needs to allow the user to select from any author defined style sheets or those that the user may have created. 14 We should review the list of requirements in 3.14.1 and 3.14.2. Are these all really needed for accessibility?
Received on Monday, 20 October 2008 03:49:32 UTC