- From: Shadi Abou-Zahra <shadi@w3.org>
- Date: Fri, 06 Apr 2012 03:31:22 +0200
- To: Peter Korn <peter.korn@oracle.com>
- CC: Eric Velleman <evelleman@bartimeus.nl>, Eval TF <public-wai-evaltf@w3.org>
Hi Peter, Good example, thank you. I'm still unclear how generalizable it is, especially with respect to demonstrating non-interference, but it does help to have a concrete example to think about. Regards, Shadi On 6.4.2012 02:58, Peter Korn wrote: > Hi Shadi, > > To your last question about amount of work to make something that is duplicated > elsewhere accessible in both instances vs. verifying non-interference... a real > life example is setting tab stops and margins in an editor. We dealt with this > in an office suite I was helping make accessible. > > A mouse user can do this by clicking first on the image of the tab stop type > (left align, right align, centered, decimal-point centered) and then > clicking/dragging along a graphic of a ruler above the text, perhaps with a > light visible vertical bar overlaid over the text for aid in alignment. Using > essentially the same UI, a mouse user might adjust the margins. A keyboard user > can bring up the paragraph's tab& margins dialog box, and use a combo box or a > new button to choose an existing / new tab or otherwise adjust an existing > margin, and then enter the position along the rule in whatever the document's > measurement units are. > > We looked at making the ruler keyboard accessible, and extending the > accessibility API to cover it. We decided that the functionality of setting tab > stops and margins was fully duplicated with the pre-existing dialog box, which > was very straightforward to make accessible since it was based on the stock UI > components that were already accessible. Verifying non-interference was also > straightforward. And... since setting/resetting the margins and tab stops is > done a relatively small percentage of the time (compared to simply entering and > editing text), the fact that a keyboard only user - including of course a screen > reader user - has to take significantly longer at the task than a mouse user > still has an overall very small impact on the productivity overall - again > another argument for not investing a significant amount of time to make this > type of - duplicated - user interaction accessible. > > > Regards, > > Peter > > On 4/5/2012 5:45 PM, Shadi Abou-Zahra wrote: >> Hi Peter, All, >> >> Good point about the functionality of individual page elements versus the page >> as a whole. I would personally interpret the latter from the WCAG 2 >> conformance requirement 5 on "non-interference": >> -<http://www.w3.org/TR/WCAG20/#cc5> >> >> "If technologies are used in a way that is not accessibility supported, or if >> they are used in a non-conforming way, then they do not block the ability of >> users to access the rest of the page [...]" >> >> So, in the example outlined below, one could argue that if a duplicate of a >> functionality is accessible then the functionality is effectively usable. >> However, one would also need to demonstrate the non-conforming functionality >> does not interfere with the overall accessibility of the page. For example, >> users may get stuck on the inaccessible piece when it is not apparent to them >> that there is an accessible alternative, in which case the page would not be >> usable altogether. >> >> In conclusion, the mere existence of an accessible alternative for a >> functionality is probably not sufficient to demonstrate conformance with WCAG >> 2; one would need to demonstrate non-interference too. >> >> Question out of curiosity: how frequent are there situations in which the >> effort needed for all the necessary work-arounds to ensure that inaccessible >> components do not cause interference for the users would actually outweigh >> making these components directly accessible? It may be good to have real >> examples as a basis for the discussion. >> >> Best, >> Shadi >> >> >> On 5.4.2012 18:46, Peter Korn wrote: >>> Eric, >>> >>> As a start, it might be worth noting in introductory text the range of things >>> being covered by this methodology. In step 3 on selecting a representative >>> sample, we might again note that for complex web sites& for large web >>> applications, there may "Exemplar functions" of a web app which should >>> definitely included, as distinct from "rarely used functions" - some of which >>> should be perhaps be included anyway as part of the sampling process (to not >>> ignore them entirely). >>> >>> By the way, I thought of what might be a better example to use: a configuration >>> page/dialog for setting per-instance-overridable defaults (e.g. whether the >>> default currency is expressed in Dollars vs. Euros vs. Yen, which can be >>> expressly set each time by the user when they enter the currency in the >>> spreadsheet web-app). If that config page/dialog has an accessibility error >>> (e.g. an unlabeled combo-box), but the per-instance setting has no error (e.g. >>> the "set currency for this field" config page/dialog) - then... there is a good >>> argument to be made that the importance of the accessibility of the default >>> configuration setting isn't so great. >>> >>> Ummm.... this raises another question... In Section 508 among other places we >>> have a notion that all functionality must be accessible, not necessarily all >>> ways of achieving all functionality. The example I made in the paragraph above >>> also connects to this question. It would be a clear WCAG conformance failure if >>> one part of a page failed one of the checkpoints. BUT... what about the >>> situation in which the failed part of the page was fully duplicated elsewhere. >>> This is a contrived edge-case for a single web page, but not at all unusual for >>> a complex web site or web application (after all, we already have the notion of >>> multiple-ways in WCAG). Is there any conformance distinction to be made between >>> a feature/aspect failure when it is the sole way of doing something vs. when it >>> is one of multiple ways and the other ways are accessible? >>> >>> >>> Regards, >>> >>> Peter >>> >>> On 4/5/2012 8:45 AM, Velleman, Eric wrote: >>>> Dear all, >>>> >>>> What shall we say about auditing complex applications that are very large >>>> and offer enormous amounts of features (some important, some less >>>> important)? Example would be online document editors, online mail, online >>>> photo applications. Or should we point to ATAG and UAAG? >>>> Kindest regards, >>>> >>>> Eric >>>> >>> >>> -- >>> Oracle<http://www.oracle.com> >>> Peter Korn | Accessibility Principal >>> Phone: +1 650 5069522<tel:+1%20650%205069522> >>> 500 Oracle Parkway | Redwood City, CA 94065 >>> Green Oracle<http://www.oracle.com/commitment> Oracle is committed to >>> developing practices and products that help protect the environment >> > > -- > Oracle<http://www.oracle.com> > Peter Korn | Accessibility Principal > Phone: +1 650 5069522<tel:+1%20650%205069522> > 500 Oracle Parkway | Redwood City, CA 94065 > Green Oracle<http://www.oracle.com/commitment> Oracle is committed to > developing practices and products that help protect the environment -- Shadi Abou-Zahra - http://www.w3.org/People/shadi/ Activity Lead, W3C/WAI International Program Office Evaluation and Repair Tools Working Group (ERT WG) Research and Development Working Group (RDWG)
Received on Friday, 6 April 2012 01:31:31 UTC