- From: Mary Jo Mueller <maryjom@us.ibm.com>
- Date: Thu, 16 Jul 2015 18:09:39 -0500
- To: "lisa.seeman" <lisa.seeman@zoho.com>
- Cc: "public-cognitive-a11y-tf" <public-cognitive-a11y-tf@w3.org>
- Message-ID: <OF0B27D2BA.787D506F-ON86257E84.0052DD65-86257E84.007F39FB@us.ibm.com>
Added in my comments below. Best regards, HA&AC Logo Mary Jo Mueller IBM Accessibility Standards Program Manager, IBM Accessibility Phone: 1-512-286-9698 | Tie-Line: 363-9698 E-mail: maryjom@us.ibm.com HA&AC Able Website: w3.ibm.com/able Follow the Human Ability and Accessibility Center on: and within IBM on: “If your actions inspire others to dream more, learn more, do more and become more, you are a leader.” ~ John Quincy Adams From: "lisa.seeman" <lisa.seeman@zoho.com> To: "public-cognitive-a11y-tf" <public-cognitive-a11y-tf@w3.org> Date: 07/15/2015 01:21 PM Subject: Action item to review missing semantic support for personalization criteria and graded help - in page Proposed semantic support for personalization criteria and graded help - on page (see meta-date proposal for alternative content version) Task: we need to enable the personalization discussed in 1. https://www.w3.org/WAI/PF/cognitive-a11y-tf/wiki/Personalization_criteria and 2. https://www.w3.org/WAI/PF/cognitive-a11y-tf/wiki/Providing_graded_help Language type support (in page) Proposal for aria attributes : aria-hasAlt and aria-alttype <mjm> Should we add in a description in here what each of these attributes would add? e.g. aria-hasAlt would provide easier to read text as an alternative to more detailed or difficult to understand content on a page. </mjm> aria-alttype supported values: On text: vocab500, vocab1000, vocab2000, active, literal, numberfree, smallsentences, chunks, General: lessoptions. Whitespace, largefont <mjm>Fixed 'General' misspelling</mjm> Images: symbolic , realistic po, p1, p1, p2, p3 Where po are images with true content, p1 is a symbol or illustration that is typically shown with alt=””, P2 are extra illustration that most people will not mind such as at the start of a paragraph or option. p3 are for disability centric illustrations that can disturb other people and may be mid sentence. <mjm>Since all others have a number, shouldn't 'po' be 'p0'? With a zero rather than the letter 'o'?</mjm> Example syntax in a webpage Option 1 <span aria-hasAlt=”uri1” > some convoluted obtuce text</span> Uri can be rdf by third party, or a hidden text elswere Such as: Example 1 <mjm>For consistency, added 'Example 1' since there is an Example 2 below. </mjm> <span is=”uri1” aria-alttype=”vocab5000 Activevoicing”> text that is hard to read</span> <mjm>There is no vocab5000 defined in the above list of supported values. In addition, the above proposed values has 'active' not 'Activevoicing' so we need to make sure the listed attributes above match these examples.</mjm> Example 2 <span is=”uri2” aria-alttype=”vocab1000 ActiveVoicing literal”>press the submit button when you are finished</span> <mjm>Fixed spelling of 'literal' attribute and same problem with ActiveVoicing not matching the proposed attribute 'Active' above.</mjm> Option 2 Proposal for aria attribute aria-easylang aria-easylang supported value: text <span aria-easylang=” some text that is hard to read”> some convoluted obtuse text</span> <mjm>Wouldn't the aria-easylang contain the easier to read and understand language? If so, the example text contained for that attribute should say 'some easier text to read' rather than 'some text that is hard to read'.</mjm> Where aria-easylang should use as simple well-known words as possible, and active, literal text, small simple sentences. <mjm>Should probably develop an example of what is considered active, literal text. To me it is subjective unless you specify a certain reading level is needed. By 'active' do you mean a certain verb tense should be used? I would also suggest that acronyms and abbreviations be avoided. </mjm> Support for small chunks Use the section tags to divide things that could be in separate pages with a next button. Sentences can not be reliably divided. For that use aria-easylang or aria-hasalt as described above Types of images Icons on items with aria-function set can have symbol sets in the json and aria-concept Images in the content <img aria-hasAlt=”uri1” scr=”img1.gif” aria-alttype=”iconic p1” /> <img id=”uri1” scr=”img2.gif” aria-alttype=”realistic p1” /> Extra help Proposal for aria attributes: aria-tooltip, aria-moreinfo and aria-extrahelp <mjm>ARIA already has a tooltip, so how is aria-tooltip different from the currently available attribute? Is it to be something that a screen reader doesn't announce so it can be more descriptive? If this is added, we'd have to make sure that in the user agent implementation that the text remains visible long enough for the user to read it - persistent till the mouse is moved away from the object. Otherwise there would be no way for the user to read all of the information before it disappears. In addition, perhaps there should also be a keyshelp. It would be nice if there was keystroke help info readily available for complex widgets and interactions.</mjm> aria-moreinfo value: URL aria-tooltip value: text string aria-extrahelp supported values: tooltip/humanhelp/morehelp/glossary/dictionary/thesaurus example <button type="button" aria-tooltip=”remove everything I have typed on this page” aria-moreinfo=”uri2 uri3” >undo</button> URI 2 may read <mjm>Example below is URI 2, so changed '3' to '2'.</mjm> <div id=”uri2” aria-extrahelp=”morehelp” aria-hidden=”true”> pressing the undo button will erase all your work on this page. Use this button with care. If you press it by mistake, press control and y at the same time and your answers will come back. </div> <mjm> Fixed misspellings. </mjm> Do we want to use aria-tooltip or title attribute? <a href=”functiongethelp()” aria-extrahelp=”humanhelp” aria-hidden=”true”>I want a person to help me</a> <mjm>Fixed misspelling of 'humanhelp' attribute </mjm> <div id=”uri3” aria-extrahelp=”glossary” aria-hidden=”true”> <dl>///</dl></div> Reminders Proposal for aria attributes : aria-reminder aria-reminder supported values are: normal that over 90% of users would want, more would be used by over 20% of the population, extra would be used by over 5% of the population, and maximum would be good for 5% of the population or less. example <div role=”alert” aria-reminder=”basic”> It is your daughter’s birthday</div> <mjm>'basic' isn't one of the values proposed above.</mjm> Proposals for aria attribute aria-distraction example Aria-distraction=”blinking, add” <mjm>By 'add' do you mean 'advertisement', or that the user has attention deficit disorder? If it is advertisement, it should be 'ad'. Blinking isn't the only distraction that is undesirable. Ongoing animations, auto-starting multimedia, constantly changing content such as carousels, ticker tape display, constantly auto-updating content are also problematic. I'm wondering about the </mjm> Or is this obvious as nobody wants them? All types should be encodable in the JSON as well as <mjm>Seems like an incomplete sentence here and text-wise it appears as if it is a heading, so I'm not sure what you intend here.</mjm> options -one of the following: -Use a given symbol/picture set as preference (can include use my families or personal symbols/pictures wherever possible) -Use a given symbol/picture set strong preference <mjm>This is a repetition of the info on the previous line.</mjm> -Hide content without symbol/picture support <mjm>I could use more description on this. I'm assuming these would be user settings that would have to be either supported in the user agent or in the web application settings?</mjm> 1. I want tooltips / I do not want tooltips when an item has focus 2. I want side panel that with show the “morehelp” 3. I want side panel that can be opened when I click on my help icon 4. I want extra help in a popup window 5. Show “more info” link 6. I want links to "more information" when ever possible 7. I want definitions in the side panel 8. I want link to human help to be shown/not shown enlarged cursor screen reading when available a view bar for isolating text while reading, Speak text and symbols Highlighting: off, by word, phrase, sentence (do we need spans to identify divisions?) <mjm>Misspelling in question posed. I'm not sure I understand the question. Divisions of what, the content that would be read? Wouldn't that simply follow normal document/information structure (e.g. paragraph, line of info being displayed, sentence, etc.)</mjm> aria-reminder values are: normal that over 90% of users would want, more would be used by over 20% of the population, extra would be used by over 5% of the population, and maximum would be good for 5% of the population or less. Disable background audio Color and background color preference <mjm>How about persistent user settings for captioning or audio description or for filtering background noise in media?</mjm> Open Issues: breadcrumbs, step tracking (step 2 out of 5) recoverable history, calendar, timers/alerts/alarms, , - doesn’t everyone want them? Left out for now https://www.w3.org/WAI/PF/cognitive-a11y-tf/wiki/Personalization_criteria#Learning_Specific And planning items such as goal tracking , Goal tracking, calendar //// Can we do anything about showing borders?? <mjm>Borders of what content?</mjm> <mjm>In addition, is there any need for reminders or bookmarks - for a user to set them for themselves (to remember where to start the next day, for example). </mjm> All the best Lisa and Ayelet Athena ICT Accessibility Projects LinkedIn, Twitter
Attachments
- image/gif attachment: ecblank.gif
- image/gif attachment: 2C799076.gif
- image/gif attachment: 2C253331.gif
- image/gif attachment: 2C966124.gif
- image/gif attachment: 2C794374.gif
- image/gif attachment: 2C583001.gif
- image/gif attachment: graycol.gif
Received on Thursday, 16 July 2015 23:10:31 UTC