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

Re: examples of sets of documents

From: Peter Korn <peter.korn@oracle.com>
Date: Thu, 13 Sep 2012 18:40:13 -0700
Message-ID: <50528AFD.9010305@oracle.com>
To: Gregg Vanderheiden <ez1testing@gmail.com>
CC: "Hoffman, Allen" <Allen.Hoffman@HQ.DHS.GOV>, Loďc Martínez Normand <loic@fi.upm.es>, Gregg Vanderheiden <gv@trace.wisc.edu>, "public-wcag2ict-tf@w3.org" <public-wcag2ict-tf@w3.org>
<PK2> Gregg,

I ship a product that contains 3 applications.  Are they a set?  If so, 
then by your reasoning, there must be multiple ways to navigate between 
those applications, yes?  And you suggest that if the applications have 
clear filenames, then they would meet the requirement given the ability 
of modern filesystems to support searching by filename and browsing by 
filename, right?

Or am I missing something?   My response to you was based on thinking 
the answers above were all "yes"...

Maybe we should start by examples of software application sets.  Can you 
provide examples (don't have to be real ones, at least not yet) of 
software applications that are and are not in sets?


On 9/13/2012 5:20 PM, Gregg Vanderheiden wrote:
> <GV2> Hi Peter
> not sure what you are trying to say
> What is the set of software that you are talking about here.   If it 
> isn't a set -- then this doesn’t apply.
> If you had a set -- it probably would already meet this success 
> criterion.   But can't tell without the set.
> thanks
> (PS the names in the start menu would be one method.  If they were a 
> set that is mentioned in one of the (a requirement for the set) I 
> would suggest that would probably be another -- so the file names 
> don't need to be involved.)
> Gregg
> On Sep 13, 2012, at 7:51 PM, Peter Korn <peter.korn@oracle.com 
> <mailto:peter.korn@oracle.com>> wrote:
>> <PK1> Gregg,
>> I really question this language/application for software.
>> On my home system, I use an application to copy the images from the 
>> SD card removed from my camera and inserted into my computer to the 
>> system's hard drive).  It's name is "MCULauncher.exe".  That isn't 
>> very descriptive.  But if it's name were the same text that appears 
>> in the shortcut - "Canon ZoomBrowser EX Memory Card Utility" - I'm 
>> still not so sure it would help me.  Because the name isn't "App that 
>> transfers images from the camera card".  And in order to find it, I 
>> need to traverse a VERY full directory tree in order to get to the 
>> directory that contains it: "C:\Program Files (x86)\Canon\ZoomBrowser 
>> EX MCU".  Maybe I'd know to look in "C:\Program Files (x86)" (or 
>> maybe I'd just give up after not finding it in "C:\Program Files"). 
>> But would I think to look in Canon?  I might not remember that my 
>> camera is a Canon.  And why in the world would I think "ZoomBrowser 
>> EX MCU" is the right subdirectory to find what I'm looking for?  Just 
>> solving the issue for the leaf name in the tree doesn't help with any 
>> poor names in the structure leading to that tree - which may not be 
>> under the control of the software developer.
>> And... when would I actually go this route to find and launch an 
>> application?  On Windows (7 and the last few), I'd go to the Start 
>> menu and hope I found it there.  Not finding it immediately, I'd 
>> troll through "All Programs" hoping to either find a name that made 
>> sense to /*me*/, or a file group name into which I'd look.
>> Of course, the importance of having a good name to help find 
>> something is hugely impacted by how many things I have to look 
>> through - the "web site" equivalent of which an individual software 
>> author has NO control.
>> If you were to say instead "it only has to have a logical name to 
>> distinguish it from other things in the set as the set is 
>> distributed", then again, I don't see the value.  I don't run these 
>> apps from the install CD.  I install them once from the Install CD, 
>> and don't care what they are named when they are on the CD.
>> So... while software application filename is a pretty logical analog 
>> to document filename, I don't see how that substitution actually 
>> helps the people this SC is intended to help.
>> And if we don't have filename, but we still think the unit of thing 
>> being looked for is the software app... then we get to the place 
>> where every software app in a set needs to be linked to each other?  
>> If my "set" is a collection of related apps (e.g. Adobe Creative 
>> Suite, or a collection of UNIX command line apps to run on Windows), 
>> must they all "link" to eachother directly somehow to meet this?  
>> ('ls' must have a GUI from which I can also run 'rm' and 'diff' and 
>> ...???).
>> Peter
>> On 9/13/2012 3:46 PM, Gregg Vanderheiden wrote:
>>> <GV1> I'm not sure why we are back on the topic of wanting to throw 
>>> some SC out.  Especially for this item.
>>>   * It DOES apply.
>>>   * It is a cognitive access issue but that should not make it less
>>>     important.
>>>   * It IS easy to meet -- but so is adding alt text to a picture
>>>     (remember that we don't require that alt text be good).
>>>   * It is no more easy for ICT than for Web content (it is pretty
>>>     easy there too).
>>>   * If you DON'T do what is required then things will and do fail -
>>>     and things are then less accessible.
>>>   * We have text that works.
>>> People challenged that any sets existed and examples were supplied.
>>> The samples can all be made to meet the SC by just giving the files 
>>> a reasonable name when they are downloaded or created or sent out. 
>>>  If you don't, then you have to add links.  If you don't then you 
>>> need to do something else to create a second method (search is 
>>> always a first and will work on any document that meets the other 
>>> success criteria )
>>> What am I missing?
>>> I think it is helpful to note that some of the reasons that it was 
>>> included in the WCAG 2.0, even though it was easy to meet, was to a) 
>>> bring attention to this issue of people with cognitive disabilities 
>>> having trouble navigating complex sites, b) to provide a place for 
>>> advisory techniques to live that addressed this general topic 
>>> (things that are important but can't be applied everywhere or in 
>>> testable form so they couldn’t qualify for required status, and c) 
>>> to be sure that the easy things were in fact done.
>>> I suggest that we replace  "web page" with software product as we 
>>> have elsewhere.  That make sense because many web pages are web 
>>> apps.  In fact many software products are indeed now also available 
>>> online as a web app (at a single URL making them a "web page". )
>>> This would mean that any software applications (that are produced 
>>> and sold as a set - and not bundled) could meet this success 
>>> criterion by being given a meaning filename (which most - BUT NOT 
>>> ALL - have).  OR they can provide links to the other apps in the set 
>>> from one of the apps if for some reason they don't want the apps to 
>>> have a recognizable file name.
>>> NOTE: A "set" would have to meet all the same criteria we talked 
>>> about before for a set -- so bundling etc would not create a "set" 
>>> (although most bundled software have a readable name so would meet 
>>> it anyway)
>>> I just added this to the Website.
>>> Does anyone have a real-world problem with this?  It is parallel 
>>> with WCAG interpretation and is easy to do, though not always done. 
>>>   And it leaves an anchor point for others writing advisory 
>>> techniques about APP navigation by people with cognitive 
>>> disabilities a good place to put them for those interested in this 
>>> real problem.
>>> Gregg
>>> On Sep 13, 2012, at 10:22 AM, Peter Korn <peter.korn@oracle.com 
>>> <mailto:peter.korn@oracle.com>> wrote:
>>>> Gregg,
>>>> We have been laboring under two critical constraints:
>>>>  1. That we must find a way to make all SCs apply
>>>>  2. That we cannot - in our NON-NORMATIVE document - re-cast the
>>>>     criteria based on the purpose & the significantly different
>>>>     world of non-web ICT to make it better apply
>>>> In this most thread we've been pushing against the first 
>>>> constraint.  But several of us have also suggested that we need to 
>>>> question the second constraint (with WCAG WG).
>>>> 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

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 Friday, 14 September 2012 01:41:13 UTC

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