W3C home > Mailing lists > Public > public-wcag2ict-tf@w3.org > September 2012

Re: Contribution to ACTION-61: Propose notes for both 'user agent' and 'content' to clarify software usage + M376 harmonisation

From: Peter Korn <peter.korn@oracle.com>
Date: Tue, 18 Sep 2012 08:02:44 -0700
Message-ID: <50588D14.3030006@oracle.com>
To: Gregg Vanderheiden <gv@trace.wisc.edu>
CC: Michael Pluke <Mike.Pluke@castle-consult.com>, "public-wcag2ict-tf@w3.org" <public-wcag2ict-tf@w3.org>
Gregg, Mike,

PK: Mike - I think your notes are good as far as they go, but they don't 
go far enough.  The note(s) for content needs to make clear that there 
are (at least) two "flavors" of content: content that may be "embedded" 
within software/the user agent, and content that may be "separable" (or 
"stand-alone") from the content (e.g. "traditional web content" like 
HTML, media files - things that are "files").  Then I think we need a 
note on user agent to make clear that a user agent doesn't need to 
operate on "separable" (or "stand-alone") content to be a user agent.  
It will almost certainly also contain it's own "embedded" content.

I also think an expository discussion of this should be in the beginning 
portion of our document.  In many ways this is core to the nature of 
taking WCAG to non-web contexts, because in WCAG there is a clear 
distinction between what is and isn't content, what is and isn't a user 

Switching gears (and people/thoughts I'm responding to...)

PK: Gregg - If we aren't careful, it's "turtles all the way down". 
Comments in-line below.

On 9/18/2012 7:02 AM, Gregg Vanderheiden wrote:
> I really like this approach.
> My only concern is that I'm not sure what constitutes a "user agent". 
>   or "additional software in order for the user to view and interact 
> with it"
> Most all software requires software that is part of the Operating 
> System in order to run.  Also interpreted programs all have software 
> that must be used to play them.
> *Which of the following are non-web content? (in your opinion)  and 
> which are software  and which are both*
>  1. *html page on a desktop*

PK: non-web ("separable") content; and perhaps also software if the HTML 
also embeds non-HTML stuff

>  1. *html and javascript app (runs on desktop)*

PK: software (with "embedded" content) and perhaps also non-web 
("separable") content, depending upon the HTML (the HTML might perhaps 
not contain sufficient "stand alone content").

>  1. *flash app*
PK:  software (with "embedded" content) and perhaps also non-web 
("separable") content, depending upon the Flash (the Flash might perhaps 
not contain sufficient "separable" or "stand alone" content).
>  1. *app written in java*
>  2. *app writing in uncompiled BASIC  (interpreted)*
>  3. *app written in p-code **  (interpreted)*
>  4. *same BASIC app as 5, but compiled     (run by hardware but using
>     much software from Operating System)*
>  5. *app written in C++ **  (run by hardware - but using *much
>     software from Operating System*)*
>  6. *app writing in C++ (run on bare hardware - not requiring any
>     ** Operating System**)
>     *

PK: The programming language an "app" is written in is immaterial. 
Whether or not there is a "player" (interpreter) for a program that may 
be "in text" or "in p-code" or "in byte-code" or in "binary code" is 

A better example that didn't appear in your list, I think, is a PDF 
document with a form where some of the form elements contain a bit of 
Javascript code behind them (e.g. for validating that the data entered 
by the user into the form is clamped to certain values, or for 
propagating data from one field to another).  To me this content with a 
bit of software in it.  Still evaluate it as content.

> Also -- it would seem that we might run into a situation where two 
> things look exactly alike to a viewer but one is software and one is 
> content.
> Thoughts?

PK: What is important isn't whether we can 100% of the time separate 
everything we ever come across into clearly EITHER content OR software.  
What is important is that where there is a question about something, the 
outcome would be the same whichever way we view it. So long as we can do 
that, we should be fine.


> Gregg
> On Sep 18, 2012, at 6:13 AM, Michael Pluke 
> <Mike.Pluke@castle-consult.com <mailto:Mike.Pluke@castle-consult.com>> 
> wrote:
>> Hi
>> During the last meeting an ACTION-61was created  for Peter to 
>> “Propose notes for both 'user agent' and 'content' to clarify 
>> software usage; work with Gregg, Loic, and Mike”.
>> There has been a lively email thread on this. As part of this thread 
>> Loďc and I have tried to indicate how notes that we believe that we 
>> need to add to the M376 standard may largely meet the needs of 
>> ACTION-61. Peter has asked me to share some of these thoughts with 
>> the group.
>>     Content
>> The note for*content*should state the following (or similar):
>> *NOTE: Content exists as a separate entity that requires a user agent 
>> in order for it to be presented to users. Some examples of content 
>> are documents that have an associated document reader/editor, media 
>> files that are played in a media player, etc. See also “user agent”.*
>> **
>> *and/or*
>> **
>> *NOTE: What distinguishes content from software is that content 
>> requires some additional software (a user agent) in order for users 
>> to view and interact with it. Some examples of content are documents 
>> that have an associated document reader/editor, media files that are 
>> played in a media player, etc. See also “user agent”.*
>> **
>> *Note to WCAG2ICT only:*these definitions are based very firmly on a 
>> direct interpretation of both the WCAG 2.0 definition of content 
>> which is: “information and sensory experience to be*communicated to 
>> the user by means of a user agent*, including code or markup that 
>> defines the content's structure, presentation, and interactions”. 
>> Note also that “document” (interactive or otherwise) now only appears 
>> as an example of content.
>>     User agent
>> If the above note is included, and maybe a “see also” link to 
>> “content”,*I do not think that a note is needed for user agent*. The 
>> WCAG user agent definition still works fine outside the Web. Maybe a 
>> “see also” link back to “content” might help to reinforce the 
>> beneficial circularity between the two definitions.**
>> **
>>     Definition of terms
>> In WCAG2ICT we talk variously of “an electronic document or a 
>> software user interface” or “electronics and software” as combined 
>> grouping. However none of these terms are defined and we have 
>> received several comments in the review of the Working Draft that 
>> highlight this. This clustering of the two terms helps to avoid the 
>> need to precisely define these two or three concepts. However it does 
>> not hide the fact that we have, in reality, had many discussions 
>> where we say that “we can see how this works for documents but we 
>> don’t think it works (or we can’t understand it) for software”.
>> I think that the audience for WCAG2ICT will continue to be 
>> dissatisfied if we fail to define the terms that we use repeatedly 
>> throughout our work i.e. electronic document, software, software user 
>> interface.
>>     Harmonising between WCAG2ICT and M376
>> In M376 we wanted to divide between content (which we should now call 
>> “non-Web content”) and “software” when directing people to 
>> appropriate understandings of how to apply WCAG (in two separate 
>> clauses 10 and 11). We clearly totally failed to clarify the logic of 
>> our approach.
>> I believe that the WCAG2ICT agreed (WCAG) definition of content 
>> together with the above notes (or similar)*works perfectly for both 
>> WCAG2ICT AND for M376*.
>> When considering the application of WCAG 2.0 to software we in M376 
>> say that it should be applied to “*software that provides a user 
>> interface*”.
>> In M376 we  are thinking of adding a note to clarify exactly what we 
>> mean by the above term, and help to distinguish it from “content” in 
>> the other clause. A possible note is:
>> “Software that provides a user interface” includes both applications 
>> that act as a user agent for separate content and also applications 
>> where it is not possible to separate the content from the rest of the 
>> application.
>> Examples of software that provides a user interface include 
>> stand-alone self-running eBooks and most of the software user 
>> interfaces that are built into hardware ICT.”
>> Maybe we would also do better to refer to “non-Web content and 
>> software that provides a user interface” when saying how to apply SCs 
>> in WCAG2ICT.
>> Best regards
>> Mike

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
Received on Tuesday, 18 September 2012 15:03:30 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 20:17:46 UTC