W3C home > Mailing lists > Public > public-cognitive-a11y-tf@w3.org > October 2015

RE: Last set of comments for the aria proposal

From: Michael Pluke <Mike.Pluke@castle-consult.com>
Date: Sun, 18 Oct 2015 21:49:35 +0000
To: lisa.seeman <lisa.seeman@zoho.com>
CC: public-cognitive-a11y-tf <public-cognitive-a11y-tf@w3.org>
Message-ID: <A48C91EB13E45544B16FBC94C9298D8D43F66C@S11MAILD013N1.sh11.lan>
Hi Lisa

The demo is very nice and well executed, however it does illustrate the difficulty of the use of global assessments of “importance”. If the user of the clothing store is a single male who is very keen on sports, the second stage of simplification loses the “Sports” option and leaves him with three options – “Men”, “Women”, “Children”. Only one of these is of any interest to this user and he will have completely lost the one option that is of most interest to him.

This simplification is probably exactly what you would get if it was based on overall frequency of use. However, if people were able to choose how they would like the website simplified almost nobody would select these three options – they would all be happy to lose one or two of the three options they are given. Also, when anyone has a question or a problem, what they desperately need is the “contact us” option – it is the only option that can help them. However, because it is almost certainly used a lot less than the main shopping categories it is the first to go once the frequency of use rule behind aria-importance gets to work.

Maybe I am overly pessimistic, but it seems to me that this way of handling “importance”, “reminders” and “distractions” sounds attractive  on the surface but I fear that these might (if widely applied) produce unintended negative consequences in practice.

I can see that they might produce universal benefits if site creators actually implemented them with finesse, but I fear that in the real world there would be too many barriers to their deployment. For example, a site owner that rated the advertisements that pay for the running of the site as of “low importance” and “distracting” would very soon lose those advertisers to a rival site that doesn’t and they would go out of business!

Ad blocking is pretty controversial at the moment (see http://uk.businessinsider.com/been-choice-founder-on-apples-ad-blocking-conundrum-2015-10?r=US&IR=T for example). These aria attributes, if used, would make life a lot easier for ad-blocking software and a lot harder for businesses that rely heavily on advertising for their existence.

Don’t get me wrong, I am an advertisement hater – but I’m also very aware of the advertisement driven economy that fuels the modern Web and modern apps!

Best regards

Mike

From: lisa.seeman [mailto:lisa.seeman@zoho.com]
Sent: 18 October 2015 21:29
To: Michael Pluke <Mike.Pluke@castle-consult.com>
Cc: public-cognitive-a11y-tf <public-cognitive-a11y-tf@w3.org>
Subject: RE: Last set of comments for the aria proposal

Hi Mike

I agree that the wording could be changed (as per my last email) but did you see the demo at https://rawgit.com/ayelet-seeman/coga.personalisation/demo/conactUs.html


you can really see how the personalisation settings such as importance can turn a complex page into a simple easy to use one.

At the end of the day, this is information that the personalisation settings can use or not as per the user preferences. No content is bared, but just put under a "more" option. It reduces the cognitive load when that load is too high.

if people are using the other aria attributes then they can also allow a user to overide the importance. It is just useful information.

All the best

Lisa Seeman

Athena ICT Accessibility Projects <http://accessibility.athena-ict.com>
LinkedIn<http://il.linkedin.com/in/lisaseeman/>, Twitter<https://twitter.com/SeemanLisa>



---- On Sun, 18 Oct 2015 21:18:39 +0300 Michael Pluke<Mike.Pluke@castle-consult.com> wrote ----

Hi



I have a few comments.



Air Importance

I repeat my concern over the “aria-importance” attribute. “Importance” is a very subjective concept and can differ significantly from person. I think that the naming of the aria-importanceattribute and the wording used to describe it are very misleading. I think that two sentences give the wrong impression, these are:



-          “These (critical, medium and less important features) need to be defined as important from a user perspective.” Taken at face value this sounds like true personalization i.e. the value of importance reflects their own preferences.

o   What is actually proposed is that these values are assigned based on a statistical analysis of actual usage of the features. Given that these are assigned by the page author, this is obviously the best that can be done.

-          “Based on personalization setting the user can see less options.” Again this sounds like true personalization i.e. the user can see fewer of the features that are less important to them. This is not what they get! If a user has minority interests the features that they really value will be removed as the majority of users are not interested in the same features.
(NOTE: It will not only be ASD users who may be frustrated and/or confused. In the example, it is suggested that the “archive” feature is of “low” “importance”. Whereas the majority of people leave a lot of email in their inboxes and never archive it, followers of the Inbox-zero way of managing email should use “delete”, “archive” and “reply” or “forward” in roughly that order of importance. If these users attempt to further streamline their lives by simplifying their email user interface using “aria-importance” they will immediately lose their second most “critical” function!)



In many instances I agree that the use of “aria importance” may provide a helpful simplification of a site that will benefit many users with cognitive impairments. However I can see that this capability might confuse some users who will wonder why a feature that they see as important has disappeared. I can also see that this could really harm some ASD users who will find that all the features that related to their, possibly atypical, interests suddenly disappear and they are left in with a site that suits everyone but them!



Summary

This feature, in itself, has nothing to do with personalization (although the way that the various levels of “importance” are dealt with can be personalized – just like everything else).



I think the name “importance” is very misleading as it has nothing to do with an individual’s rating of importance.



Perhaps this would be better named “aria-usage” – which is consistent with the description of how it should be assigned.



Values of aria-action

In general, the values are simple action verbs – which is what I would expect. But there are two exceptions that I find confusing.

-          One is “received” – I do not see what a “received” button would do. Should this be “receive”?

-          Another is “unexpanded” – similarly I’m not sure what would happen if I pressed an “unexpanded” button. Surely the action should be “unexpand” (the reverse of “expand”).

I’m also not sure what the “critical” action “compose” is meant to be. Is this related to the compose key that is present on a few keyboards that is used to generate composed characters? Clearly this name would have little meaning to end-users, but as long as its intention is clear to developers I guess that this will be OK.

I think that items like “calendar”, “settings”, “my profile”, etc. that don’t follow the action verb model are fine as I assume that the functionality that they are intended to activate is to display those items (for editing as well as viewing I assume).



Reminders

Aria-reminder relies on the same type of usage statistics proposed for aria-importance to categorize which reminders might be seen as important (or, in contrast, distracting). In practice the value of reminders, just like importance, will be a very personal thing and the usage statistics could mask some major differences.



One reminder that only 5 - 20% of the population use may be one that I find really important but to get it I have to have a personal preference to have “extra” reminders even though 95% of those reminders are things that drive me to distraction”. I’m just not convinced that trying to categorize things as important or having to accept whole groups of reminders to get one that I want will really be beneficial.



Distractions

This is getting repetitive, but working out what is “non-essential” distracting content is another dangerous business. To me, status updates from most social media sources would be a major distraction that I choose to live without. To others this is their life!



I think that it would be very valuable for most people and essential for others to be able to eliminate blinking adverts. However, WCAG 2.0 already has success criterion 2.2.2 Pause, Stop, Hide that should provide users with a way to take control of whether they see such adverts. I’m not certain that aria-distraction adds a lot to that well-established success criterion. Anyone who does not design their website to meet 2.2.2 is very unlikely to add the “aria-distraction” attribute to the offending content!



Best regards



Mike



From: lisa.seeman [mailto:lisa.seeman@zoho.com]
Sent: 18 October 2015 14:18
To: public-cognitive-a11y-tf <public-cognitive-a11y-tf@w3.org<mailto:public-cognitive-a11y-tf@w3.org>>
Subject: Last set of comments for the aria proposal



Hi Folks



This is the last chance to rewrite  the aria proposal on https://rawgit.com/w3c/coga/master/issue-papers/links-buttons.html before the joint meeting next week. Please send in your comments as soon as possible - preferably before tomorow call





All the best

Lisa Seeman

Athena ICT Accessibility Projects<http://accessibility.athena-ict.com>
LinkedIn<http://il.linkedin.com/in/lisaseeman/>, Twitter<https://twitter.com/SeemanLisa>


Received on Sunday, 18 October 2015 21:50:09 UTC

This archive was generated by hypermail 2.3.1 : Sunday, 18 October 2015 21:50:09 UTC