- From: lisa.seeman <lisa.seeman@zoho.com>
- Date: Thu, 30 Nov 2017 18:49:17 +0200
- To: Michael Gower <michael.gower@ca.ibm.com>
- Cc: <Brooks.Newton@thomsonreuters.com>, <akirkpat@adobe.com>, <john.foliot@deque.com>, <w3c-wai-gl@w3.org>
- Message-Id: <1600dd34338.ff556c82261185.317759607889922134@zoho.com>
I think confirm got bit lost with trying to remove redundancies with submit it would be good to add it All the best Lisa Seeman LinkedIn, Twitter ---- On Thu, 30 Nov 2017 18:30:56 +0200 Michael Gower <michael.gower@ca.ibm.com> wrote ---- I can live with "forward" being on the list. i just think there are other more common controls like "confirm" which seem absent. For instance, almost every dialogue box has some kind of 'yes/no', 'confirm/reject', 'okay/no way' button set, I would think a "Confirm" purpose is a lot more important to capture than requiring "Forward". Maybe I'm overlooking that purpose and it's captured by something else? Michael Gower IBM Accessibility Research 1803 Douglas Street, Victoria, BC V8T 5C3 gowerm@ca.ibm.com voice: (250) 220-1146 * cel: (250) 661-0098 * fax: (250) 220-8034 From: <Brooks.Newton@thomsonreuters.com> To: <akirkpat@adobe.com>, <michael.gower@ca.ibm.com> Cc: <john.foliot@deque.com>, <w3c-wai-gl@w3.org> Date: 2017-11-30 08:10 AM Subject: RE: Purpose of Controls Mike, I agree with your observations regarding problems with the control types list. Thus, the reason why I brought up my objection to using a filtered version of the survey results on yesterday’s call as a basis for moving forward with normative text. I think we really want to present a highly refined, curated list of control categories (chunking terms) in the Purpose of Controls SC normative language, instead of the list of specific control names we currently have. These categories need to be added to the normative list by a process that assesses their beneficial impact to users who this SC is intended to serve. I’m not sure there is time to do that work, given the deadlines we face. Brooks From:Andrew Kirkpatrick [mailto:akirkpat@adobe.com] Sent: Thursday, November 30, 2017 9:48 AM To: Michael Gower Cc: John Foliot; WCAG Subject: Re: Purpose of Controls Mike, Part of the benefit is the ability for a tool to recognize the button purpose and add an icon or replace the text with an image. Of course, if that tool also looked at the button labels and responded to labels like “print” and “home” then the same result could be reached. I do think that “reply”, “forward”, and “print” are very commonly if not universally used in English-language sites, but am not sure if this is such a major issue. We could and may wind up removing some of these due to implementation results or comments on the draft. Thanks, AWK Andrew Kirkpatrick Group Product Manager, Accessibility Adobe akirkpat@adobe.com http://twitter.com/awkawk From: Michael Gower <michael.gower@ca.ibm.com> Date: Thursday, November 30, 2017 at 10:39 To: Andrew Kirkpatrick <akirkpat@adobe.com> Cc: John Foliot <john.foliot@deque.com>, WCAG <w3c-wai-gl@w3.org> Subject: Re: Purpose of Controls At the opposite end of the spectrum of considerations on this list, is that there are terms here which are so obviously the de facto term for a control function that I'm really having problems seeing what the value is having all authors forced to code the purpose. Can anyone think of a situationn where the button that forwards an email to someone else would be labelled as anything other than "forward"? It's a very specific term to a specific type of application. It just seems out of place to me compared to other more broadly used functions on this list which could have synonyms used in labels. Even some more globally used functions like "print" have a similar 'hard' function name. Tough to think of use cases where the word "print" would not be the label. I'm assuming there's a concern that a string like "Print document" is a synonym so that may not be understood, so the purpose needs to be clarified? Michael Gower IBM Accessibility Research 1803 Douglas Street, Victoria, BC V8T 5C3 gowerm@ca.ibm.com voice: (250) 220-1146 * cel: (250) 661-0098 * fax: (250) 220-8034 From: Michael Gower/CanWest/IBM To: Andrew Kirkpatrick <akirkpat@adobe.com> Cc: John Foliot <john.foliot@deque.com>, WCAG <w3c-wai-gl@w3.org> Date: 2017-11-30 07:20 AM Subject: Re: Purpose of Controls Here's the scenario I'm worried about: I'm filling out a complex form that gathers information on myself and my whole family. The form is divided into sections on me and on my spouse and on my offspring. So there are name fields all over the place, and those name fields my even be labelled the same (i.e., the section heading says "spouse" but the labels are the same on the inputs as on the ones for user). That would be crappy design, in my opinion, but I can see it happening. In such a scenario -- or even in ones less dire -- I think we have to ensure that our purpose designations don't actually make that experience more confusing. Michael Gower IBM Accessibility Research 1803 Douglas Street, Victoria, BC V8T 5C3 gowerm@ca.ibm.com voice: (250) 220-1146 * cel: (250) 661-0098 * fax: (250) 220-8034 From: Andrew Kirkpatrick <akirkpat@adobe.com> To: John Foliot <john.foliot@deque.com> Cc: WCAG <w3c-wai-gl@w3.org> Date: 2017-11-30 06:56 AM Subject: Re: Purpose of Controls John, Could the final note be used as a note on the SC and not indicate that the user is directly part of each of the items? So, for example, the name information that I encounter when filling out a form for a child or an employee would be marked up with name metadata and I could benefit from the metadata, but if I am using a form that requires that I enter information that includes my name and that of a family member only one should be used. Thanks, AWK Andrew Kirkpatrick Group Product Manager, Accessibility Adobe akirkpat@adobe.com http://twitter.com/awkawk From: John Foliot <john.foliot@deque.com> Date: Thursday, November 30, 2017 at 09:38 To: Andrew Kirkpatrick <akirkpat@adobe.com> Cc: WCAG <w3c-wai-gl@w3.org> Subject: Re: Purpose of Controls Hi Andrew, To expand upon the concern about *User* information, I've taken a quick pass and come up with this (I have cherry-picked the inputs of concern, the following is not the full list under consideration): * Name - Inputs used to handle information about a user’s name(s) (e.g. first name, family name, suffix, etc.) * Professional Title - The user's Job title (e.g., "Software Engineer", "Senior Vice President", "Deputy Managing Director") * Organization - Company name corresponding to the user's organization * Address - Inputs used to handle information about the user's address information, organization, or location (e.g. street address, city, region, postal code, etc.) * Country Code - The user's country abbreviation code * Country Name - Full name of the user's country * Credit Card - Inputs used to handle information about the user's credit card payments * Name on Credit Card - Full name of the user as given on the payment instrument * Language - The user's preferred language * Birthdate - The user's birthday information * Sex - The user's Gender identity * Photo - Photograph, icon, or other image corresponding to the user * Telephone Number - Inputs used to handle information about the user's telephone number(s) * Email Address - The user's Email address NOTE: For further clarity, input fields that collect identifying data should only be related to data associated to the end user. For example, on a form that collects multiple Names, Addresses, and Phone Numbers, content authors are only required to ensure that fields related to the actual user are addressed by this requirement. Thoughts? On Thu, Nov 30, 2017 at 8:31 AM, Andrew Kirkpatrick <akirkpat@adobe.com> wrote: AGWGer’s, Yesterday we spent a lot of time on Purpose of Controls and we need to minimize the time spent on this SC because while it is important it isn’t the only important SC. After the meeting a lot of work went into this SC, and we have an implemented version (two, actually) for people to review and think about. Things that have changed: 1. All listed items have definitions. 2. The lists are moved from an appendix to a section of the document, and are to be regarded as normative. 3. The term “user interface components” is used more consistently throughout the SC and additional section. a. The SC is retitled “Identify Purpose” as the previous “purpose of controls” mixes the User interface Component and “control” language and this seems clearer. 4. Within the lists, a few item groupings are collapsed into a single top-level item. Name, Address, and Telephone are examples of this. So if you build a form you need to make sure that the appropriate address field purposes are conveyed, and this helps with internationalization as well as to accommodate for different design decisions (e.g. one form uses “Full Street Address” and another has “address 1, “address 2”, etc – both need the purpose to be properly conveyed). One item to think about: A comment was raised at TPAC and since then as well that for the input control purposes we need to focus on the user’s information. So, instead of name inputs controls being about anyone, they are about the user directly. The concern is that if autofill attributes are used to satisfy this that a form with name fields for many people (e.g. HR system, booking a flight for a family) that there will be a lot of inaccurate information potentially automatically added to the form. If we agree that this is a problem, then we will need to adjust a handful of other input purposes (e.g. address, email, etc.). So, here’s the latest draft: With “AT RISK” items from yesterday: http://rawgit.com/w3c/wcag21/purpose_of_controls_changes2/guidelines/index.html#identify-purpose Without “AT RISK” items from yesterday: http://rawgit.com/w3c/wcag21/purpose_of_controls_changes3/guidelines/index.html#identify-purpose Thanks, AWK Andrew Kirkpatrick Group Product Manager, Accessibility Adobe akirkpat@adobe.com http://twitter.com/awkawk -- John Foliot Principal Accessibility Strategist Deque Systems Inc. john.foliot@deque.com Advancing the mission of digital accessibility and inclusion
Received on Thursday, 30 November 2017 16:49:56 UTC