- From: Jan Richards <jan.richards@utoronto.ca>
- Date: Fri, 5 Nov 2004 16:53:24 -0500
- To: Tim Boland <frederick.boland@nist.gov>
- Cc: w3c-wai-au@w3.org
Hi Tim, Thanks for the comments! I have integrated your comments into my new draft at: http://jan.rcat.utoronto.ca/public/auwg/guidelines.html (but: note success criteria 3 for 3.1 now appears as success criteria 2 of 4.1 as per a decision made on the last call) I will also respond to a few of your comments here (I'm removing the rest of the text for the sake of readability. ----- Quoting Tim Boland <frederick.boland@nist.gov>: > --accessibility(link to def?)-- > information(vague term?) JR: "Accessibility information" is a defined term currently meaning: Accessibility information is any information that is necessary and sufficient for undertaking an accessible authoring practice. This information may include, but is not limited to, equivalent alternatives. > 3. The --authoring tool(link to def?) --must always provide at least one > option to the --author(link to def?) (via alerts, prompts?) to "mark" (how > is this testable?) "less accessible" (how is this objectively measurable?) > "choices" (of what?), prior to the completion of all --authoring > actions(link to def?) --related to those choices. > > NOTE: I believe WCAG techniques documents are non-normative, so cannot use > non-normative example in a normative success criterion statement -- maybe > pick another example? JR: Yes, the WCAG techniques are non-normative but the actual act of having a WCAG techniques document for a format is a normative requirement - of ATAG 2.0, checkpoint 2.1. > 1. Any --authoring tool "process" (what is this? is it defined?) that > --"imposes" (meaning? > no author control?)--a sequencing of --authoring actions(link to def?)--, > must always integrate > (into what? ambiguous word? maybe use "provide" instead? > ) --accessibility(link to def?)-- > --prompting(link to def?)-- to the --author(link to def?)--prior to > completion of the last --authoring action (link to def?) -- of the > sequence. > > (NOTE: Why is accessibility prompting needed for the sequencing? What > purpose does it serve, > and how is it related to the sequencing? What if the author doesn't need > > accessibility prompting? I think this success criterion is unclear on > this point..) JR: The background here is that we would like prompting to occur as early in the authoring workflow as possible - so the author doesn't end up 5 minutes before a publishing deadline and suddenly be prompted about a bunch of problems. At the same time, we don't want to be overly constraining on developers and demand prompting at every hint of a problem (e.g. some tools may be designed to handle issues in batches) So...what would a success criteria for this look like? For tools like Dreamweaver that allow the author to perform any authoring action at any time it's difficult to see a clearly defined moment for requiring prompting other than the end of authoring. On the other hand, tools that force the user to follow a strict sequence (either for the whold process or just for part of the process - e.g. creating an image map to drop on to a page) it seems reasonable that we can require any relevant prompting to occur before that sequence ends. The reason I specified "earliest *completion point of the process*" is because developers may be tempted to make accessibility the last step in the sequence and then have a "Finish" button prior to that step (which is something we want to prevent). ------- I hope that helps! Cheers, Jan
Received on Friday, 5 November 2004 21:54:45 UTC