W3C home > Mailing lists > Public > w3c-wai-gl@w3.org > January to March 2018

Re: Finding agreement on common purpose

From: Andrew Kirkpatrick <akirkpat@adobe.com>
Date: Mon, 15 Jan 2018 21:16:07 +0000
To: Marc Johlic <marc.johlic@gmail.com>
CC: "w3c-waI-gl@w3. org" <w3c-wai-gl@w3.org>
Message-ID: <97573850-78BC-498A-9C65-6C10B182630B@adobe.com>
I’m worried about people feeling that this says that all technologies need to have all of these properties.

If I’m using a PDF form with “first name” as the meaning, I won’t have a way to include the meaning since there is no analog to the autofill properties.  That’s the reason for the “In content implemented using technologies with support for identifying the expected meaning for form input data”.


Andrew Kirkpatrick
Group Product Manager, Accessibility


From: Marc Johlic <marc.johlic@gmail.com>
Date: Monday, January 15, 2018 at 15:14
To: Andrew Kirkpatrick <akirkpat@adobe.com>
Cc: WCAG <w3c-wai-gl@w3.org>
Subject: Re: Finding agreement on common purpose

Hi Andrew,

OK, I could +1 this - and your bullet #3 makes sense.  But if possible I would still like to be able to simplify the language and get it somewhere closer to what Jake was proposing.  Would something like this be possible, or do you think too much is lost by dropping the extra wording?

Proposed:  The meaning of user-specific input fields that map to any of the [link]HTML 5.2 Autofill field names can be programmatically determined.

I think any questions around technologies that support identifying meaning could be sorted out in the Understanding doc.  I would just like the SC(s) to be as easy to parse as possible.


On Mon, Jan 15, 2018 at 12:50 PM, Andrew Kirkpatrick <akirkpat@adobe.com<mailto:akirkpat@adobe.com>> wrote:
What I’m hearing is that we are in general agreement that:

  *   The HTML 5.2 autofill values for the common input purposes are ok and we will not include the purposes that are not input-related
  *   The list can’t change over time
  *   The purposes need to relate to the user directly, not inputs related to someone else

Other points that we don’t have general agreement on:

  *   Limit this to markup
  *   Limit this to HTML
  *   Put the list into WCAG 2.1 vs referencing the list in a date-specific HTML version (e.g. 5.2)
  *   Include the “for the user” aspect in the SC text vs the list

My best attempt on this this that I like is:
In content implemented using technologies with support for identifying the expected meaning for form input data, for each user-specific input field that has a purpose that maps to any of the [link]HTML 5.2 Autofill field names,  the meaning of the input field can be programmatically determined.

The reasons I like this relate to the items that we don’t have general agreement on:

  1.  Not limited to any technology, as technologies like PDF and mobile frameworks and software evolve this can still be applied.
  2.  Same as #1 above
  3.  If we put it into our spec, we are adding to our localization burden, understanding burden, editorial burden, and will deal with the “add this to WCAG’s list” questions. Referencing the external list, stable and date-bound, makes sense to me. External standards (read EN 301 549) that incorporate WCAG 2.1 SC won’t need to worry about also adding the list because it is referenced.
  4.  “the user” is part of the SC and it is not going to get lost as part of a separate list. And, since I’m linking to the external list, we need to add it.


Andrew Kirkpatrick
Group Product Manager, Accessibility


From: David MacDonald <david100@sympatico.ca<mailto:david100@sympatico.ca>>
Date: Monday, January 15, 2018 at 12:25
To: Alastair Campbell <acampbell@nomensa.com<mailto:acampbell@nomensa.com>>
Cc: CAE-Vanderhe <gregg@raisingthefloor.org<mailto:gregg@raisingthefloor.org>>, "Abma, J.D. (Jake)" <Jake.Abma@ing.nl<mailto:Jake.Abma@ing.nl>>, Andrew Kirkpatrick <akirkpat@adobe.com<mailto:akirkpat@adobe.com>>, "lisa.seeman@zoho.com<mailto:lisa.seeman@zoho.com>" <lisa.seeman@zoho.com<mailto:lisa.seeman@zoho.com>>, WCAG <w3c-wai-gl@w3.org<mailto:w3c-wai-gl@w3.org>>, Amihai Miron <amihai@user1st.com<mailto:amihai@user1st.com>>
Subject: Re: Dealy in "Common Purposes"

Hi Alastair

> With it focusing on HTML’s autofill attributes, there has been widespread browser support for years

Yes absolutely... further in
 email I suggested that we consider limiting the SC to HTML.
​With each of Gregg's questions the only clear answer I was able to come up with was HTML autofill.​ However, Léonie is making a good case against referencing HTML directly and sticking with our list in the spec... I think Lisa would rather also prefer our list instead of referencing HMTL 5.2 ... so ...


I would like to see a more robust answer to Gregg's questions other than implementations are in place and coming... so far I haven't seen an explanation of how this will work, and the implementations I've seen seem to be general personalization widgets rather than an implementation of a set of form fields with a mapping functionality back to our common purposes...

Here are Gregg's questions:


 how are different languages and different taxonomies being handled?

how does the AT find the mapping of new terms back to the definitions in WCAG?
•  how does AT know the format of the map?
•  it is machine readable?
•  how does the AT find that map?

Are you saying
 COGA attributes will all map back to our numbered list
​ in these mapping documents that sit in the AT? If there are currently no implementations of this, is it reasonable to at least provide a step by step description of how it will work once implemented.

David MacDonald

CanAdapt Solutions Inc.

Tel:  613.235.4902<tel:(613)%20235-4902>





  Adapting the web to all users
            Including those with disabilities

If you are not the intended recipient, please review our privacy policy<https://na01.safelinks.protection.outlook.com/?url=http%3A%2F%2Fwww.davidmacd.com%2Fdisclaimer.html&data=02%7C01%7Cakirkpat%40adobe.com%7C0592f609a08942953cf808d55c3ceb76%7Cfa7b1b5a7b34438794aed2c178decee1%7C0%7C0%7C636516339077263984&sdata=Yfel9lmu92Kav5LOuwSxD410uDtvAJGbku%2F%2FuLM2tIg%3D&reserved=0>

On Mon, Jan 15, 2018 at 4:41 AM, Alastair Campbell <acampbell@nomensa.com<mailto:acampbell@nomensa.com>> wrote:
> I share Gregg's concerns about the speculative nature of an SC that has no existing AT to make use of it

Huh? With it focusing on HTML’s autofill attributes, there has been widespread browser support for years:

Lisa also posted about a couple user-agent side implementations of the meta-data aspects, and 5 sites that are or will be using the more extended set now.

Microdata is also standardised, but we seem to have dropped the non-autofil purposes, so I’ll stop there.


Received on Monday, 15 January 2018 21:16:33 UTC

This archive was generated by hypermail 2.4.0 : Thursday, 24 March 2022 21:08:21 UTC