RE: Two variants for the redefinition of "accessibility services of software"

Hi Peter and Gregg

Peter had picked up on exactly the same point as me - that it is software that is actively doing things, not documents. However, I totally agree with you Gregg when you say that "most ALL non-web documents that have user interface components in them expose them through user agent services".

I think that the thing that is causing the problem is that adding documents into the definition in the way that you say implies that it is the document that is ACTIVELY USING the user agent services. I think that my view (and perhaps Peter's) is that in reality:


-          the user agent (the active entity) parses the document (the passive entity) and it uses its own (or lower level) services to ensure that the document is presented in the way the encoding in the document describes.

The net result is the same, but the latter interpretation does not stretch (my) credulity in seeing a basically passive thing like a document actively using the services of its user agent.

I'm wondering if we can slightly adapt the wording to overcome this small (but to me very confusing) shift in perception? I'll give it some thought.

Best regards

Mike

From: Gregg Vanderheiden [mailto:gv@trace.wisc.edu]
Sent: 05 June 2013 16:32
To: Peter Korn
Cc: Michael Pluke; public-wcag2ict-tf@w3.org
Subject: Re: Two variants for the redefinition of "accessibility services of software"

Hi Peter

Not ALL non-web documents do  (and not ALL Software makes use of platform services).   But since SOME do , it needs to be in the definition - No? .

 If you want to put SOME in front of  non-web documents  and MOST in front of software that is fine. But not necessary.

Actually don't most ALL non-web documents that have user interface components in them expose them through user agent services?   Doesn't all AT access the content via the user agent ?  (or can they access content on non-web documents that are not opened in a user agent?)


Gregg
--------------------------------------------------------
Gregg Vanderheiden Ph.D.
Director Trace R&D Center
Professor Industrial & Systems Engineering
and Biomedical Engineering University of Wisconsin-Madison
Technical Director - Cloud4all Project - http://Cloud4all.info
Co-Director, Raising the Floor - International - http://Raisingthefloor.org
and the Global Public Inclusive Infrastructure Project -  http://GPII.net

On Jun 5, 2013, at 11:26 AM, Peter Korn <peter.korn@oracle.com<mailto:peter.korn@oracle.com>> wrote:


Wading in...

While I see many (though not all) user agents as being platforms (hence Note 1 in platform software), I don't see all (or even most) documents as utilizing "a set of software services".  Since software services are APIs, and it is programming code that invokes APIs, documents that don't contain programming code (e.g. a simple text document) by definition cannot use those APIs, and so by definition don't use software services.

Recall the WCAG2ICT definition of user agent<http://www.w3.org/TR/wcag2ict/#keyterms_ua> - it is the thing that "retrieves and presents documents".  That thing clearly parses the documents - gets whatever markup is in them, etc. - and then utilizes the accessibility services of the platform underneath it.  Where that user agent is also a platform (Adobe Flash, Microsoft Silverlight, a web browser running Javascript code, a Java runtime), it also is a platform.  But Notepad and Wordpad aren't platforms.  They are, however, by our definition, user agents.


Make sense?


Given that, I would not insert the text "non-Web documents" as Gregg is proposing.


Peter
On 6/5/2013 7:50 AM, Michael Pluke wrote:
I guess conceptually from a WCAG point of view that is the case.

It seems that I have a persistent problem seeing how lines of code in document (e.g. Web page, word doc) can, in reality, do anything like "expose information". To me it is clear that it is the user agent that takes the web page/document and "exposes information about the user interface (as encoded in the page/document) to assistive technologies." Although conceptually the user agent may offer its services to the document, I still struggle to see what a document, or anything else that is not software, can do with this offer. Surely only software can actually do things - and that is why all documents need a user agent to do things.

But I guess I will have to learn to live with this conceptual myopia (if that is what it is) - as long as everyone else is comfortable with what you have written. Certainly your text is simple and clear.

I would still prefer to see the notes in their original order.

Best regards

Mike

From: Gregg Vanderheiden [mailto:gv@trace.wisc.edu]
Sent: 05 June 2013 15:14
To: Michael Pluke
Cc: Peter Korn; public-wcag2ict-tf@w3.org<mailto:public-wcag2ict-tf@w3.org>
Subject: Re: Two variants for the redefinition of "accessibility services of software"


On Jun 5, 2013, at 10:07 AM, Michael Pluke <Mike.Pluke@castle-consult.com<mailto:Mike.Pluke@castle-consult.com>> wrote:



I'm not so certain whether this addition is needed. In my mind it is always software that actually uses the services that the platform provides. In the case of non-web documents I see it as being the user agent that uses the services to "expose information about the user interface to assistive technologies". So I do not see that it is necessary to add non-web documents to the first definition. For the second it is more complex as I see the user agent using the services to expose information about the user interface of both the user agent AND the document to assistive technologies. In this case it might be OK to stick with Peter's original wording or it might be necessary to craft something much more complex.

Did you not see that USER AGENT is an example of platform?

All browsers are platforms.






I realise that I am far less experienced at interpreting the underlying WCAG 2.0 model of content and user agents, so I accept that my interpretation may be wrong - but I think that expert eyes need to look again at Peter's original definitions and Gregg's amendments.

In either case I do not think that reversing the notes as Gregg has done adds clarity to the original (it either has no effect or, in my view, makes it marginally less good).

In constructing the survey I will point to the place where Peter has written the original proposals. If we can resolve some alternative text before the survey is sent out, then this text needs to be changed (preferably by Peter or Gregg who are adept with editing the wiki).

Best regards

Mike

From: Gregg Vanderheiden [mailto:gv@trace.wisc.edu<http://trace.wisc.edu/>]
Sent: 05 June 2013 04:27
To: Peter Korn
Cc: public-wcag2ict-tf@w3.org<mailto:public-wcag2ict-tf@w3.org>
Subject: Re: Two variants for the redefinition of "accessibility services of software"

very nice

only one thing I think needs to be fixed.

You discuss user agents as an example but don't have  non-web documents anywhere in either.

also

Below are the same text with NON WEB DOCUMENTS in the correct places

Because both notes contain User agents and virtual machines -- I think it reads better to reverse them (as shown below) (I didn't fix the note numbering so you can see the switch)


Very nice

gregg





platform software

The term platform software, as used in WCAG2ICT, has the meaning below:
platform software

collection of software components that run on an underlying software or hardware layer, and that provides a set of software services to applications OR NON-WEB DOCUMENTS that allow them to be isolated from the underlying software or hardware layer

Note 2: Sometimes platform software is also a software application (e.g. a user agent or a virtual machine).

Note 1: Examples of platform software include operating systems, user agents, and virtual machines.





accessibility services of platform software

The term accessibility services of platform software, as used in WCAG2ICT, has the meaning below:
accessibility services of platform software

services provided by platform software that are used by software OR NON-WEB DOCUMENTS to expose information about the user interface to assistive technologies

Note 1: These services are commonly provided in the form of accessibility APIs (application programming interfaces), and they provide two-way communication with assistive technologies, including exposing information about objects and events.

Note 2: Platform software that is also an application may simply expose the accessibility services of the underlying platform layer, rather expose its own set of accessibility services.  Alternately it may translate between the set it exposes and those of the underlying platform layer.

Gregg
--------------------------------------------------------
Gregg Vanderheiden Ph.D.
Director Trace R&D Center
Professor Industrial & Systems Engineering
and Biomedical Engineering University of Wisconsin-Madison
Technical Director - Cloud4all Project - http://Cloud4all.info<http://cloud4all.info/>
Co-Director, Raising the Floor - International - http://Raisingthefloor.org<http://raisingthefloor.org/>
and the Global Public Inclusive Infrastructure Project -  http://GPII.net<http://gpii.net/>

On Jun 4, 2013, at 8:46 PM, Peter Korn <peter.korn@oracle.com<mailto:peter.korn@oracle.com>> wrote:




Hi gang,

Coming out of our last meeting on 31June13, I have taken a whack at redefining "accessibility services of software" to make more central the concept that this is about platform software, and not all software generally.

Please see Proposal #3 at New glossary term "accessibility services of software and assistive technology"<https://sites.google.com/site/wcag2ict/edits-for-michael-post-2nd-public-draft/new-glossary-term-accessibility-services-of-software-and-assistive-technology>

In particular, please see both Variant #3a in which I keep our existing definition text, but simply change the title of the term to "accessibility services of platform software"; and then see Variant #3b in which I introduce yet another new term: "platform software", when I then leverage in next text for the retitled term "accessibility services of platform software".

Fundamentally Variant #3a is the more minimal / less invasive change, while Variant #3b makes fuller use of the "teachable moment" that our Technical Report affords us.  Please also note the section For reference, from ISO 13066-1 at the bottom of that wiki page, from which I draw on (but do not expressly mimic) that ISO text.  While it is somewhat tempting to lift definitions word for word from ISO 13066-1, those definitions leverage terms & concentps that have slightly different existing definitions in WCAG 2.0 (e.g. AT), and I am also unclear on whether such copying is of a copyright ISO standard is OK in a non-ISO document such as our TR.

Below both variants on the wiki page please see "Edits to other terms common to both Variants #3a and #3b" where I show show how the new term "accessibility services of platform software" would impact our two glossary terms "programmatically set" and "programmatically determined", as well as Principal 4 and Guideline 4.1 (the change is the same under both variants).


I personally don't have a strong preference between Variant #3a and Variant #3b - different things attract me to each of them.  I solicit comments / feedback on them, ahead of a formal survey (perhaps tomorrow?) ahead of our Friday meeting.  I suggest we survey both approaches (as well as the follow-on edits to those two terms, the principal, and the guideline).



Peter
--
<oracle_sig_logo.gif><http://www.oracle.com/>
Peter Korn | Accessibility Principal
Phone: +1 650 5069522<tel:+1%20650%205069522>
500 Oracle Parkway | Redwood City, CA 94065
<green-for-email-sig_0.gif><http://www.oracle.com/commitment> Oracle is committed to developing practices and products that help protect the environment


--

Peter Korn | Accessibility Principal
Phone: +1 650 5069522<tel:+1%20650%205069522>
500 Oracle Parkway | Redwood City, CA 94065
Oracle is committed to developing practices and products that help protect the environment
<oracle_sig_logo.gif><green-for-email-sig_0.gif>