WebAIM - Web Accessibility In Mind

E-mail List Archives

Re: SC 3.1.2 lang exception: proper names

for

From: John Foliot
Date: May 6, 2019 2:29PM


Wolfgang writes:

> According to Patrick's interpretation, pronunciation issues of proper
names like *Jean* in French / English would be an usability issue and not a
problem of conformance to 3.1.2.

From a "legally conforming" perspective, evaluators do not need to call
this as a compliance failure. But again, they "legally" cannot fail this
code sample either:

<img src="" alt="logo">.

Putting aside the exemptions of SC 1.1.1, it still only states: "*All
non-text content that is presented to the user has a text alternative that
serves the equivalent purpose, except for the situations listed below*."
Equivalency is and will always remain a subjective call.

In other words, the "quality" of the web output is not something measured
in "legal compliance reports", although it is surely something we care
about at a higher level.

Based upon the above, and looking at what the spec states related to SC
3.1.2, it states: "*The human language of each passage or phrase in the
content can be [JF: read, "must be"] programmatically determined except for
proper names, technical terms, words of indeterminate language, and words
or phrases that have become part of the vernacular of the immediately
surrounding text." *- which outlines what *MUST* be marked up, but does not
specifically forbid authors from adding @lang on proper names, technical
terms, words of indeterminate language, or words or phrases that have
become part of the vernacular of the immediately surrounding text, only
that it's not mandated. (i.e. you could NOT fail this - <span
lang-"fr">Jean</span>)

In response to JF I'd be interested in criteria for subjective applyment.


Well, the "*Proper names need no lang-attribution, if the pronunciation
between habitual and synthesized listening only differs in nuances of
accentuation or intonation*" quote isn't in the normative text, but it's a
'hint' as to how to apply subjective decisions; the key to that quote (to
my mind) is the qualifier: IF

So, subject matter expertise, experience, and "context" all need to apply.
If, in your professional opinion, the user-experience is improved by adding
the lang attribute to a proper name, then do it. Else, don't - you won't
fail "legal conformance" either way.

JF

On Mon, May 6, 2019 at 12:57 PM Wolfgang Berndorfer <
<EMAIL REMOVED> > wrote:

> Hi all,
> According to Patrick's interpretation, pronunciation issues of proper
> names like *Jean* in French / English would be an usability issue and not a
> problem of conformance to 3.1.2.
> In response to JF I'd be interested in criteria for subjective applyment.
> The SC itself only mentions *human language* to be technically determined
> and takes off proper names. Human language is defined as *language that is
> spoken, written or signed (through visual or tactile means) to communicate
> with humans*.
> Does this include listened language?
> Pronunciation, accentuation and intonation are just mentioned in the
> understanding section. And the pronunciation is classified as an AAA
> conformance SC in 3.1.6.
> Seems we have to reflect the normativity of the understanding section.
> So:
> 1) Should we consider the exception of proper names as strict and
> universal? (listening apprehension sounds substantially as requirement for
> *communicate with humans*.)
> 2) Should we judge the understanding section of the WCAG generally as
> normative or as substational interpretation or just as informative?
> From a normative conformance point of view, I fear that Patrick is right
> with his interpretation. But from AX point of view, pronunciation often
> causes irritation also with proper names, especially if you are interested
> in politics or geography (not only with German as default language).
> Regards!
> Wolfgang
>
> -----Ursprüngliche Nachricht-----
> Von: WebAIM-Forum [mailto: <EMAIL REMOVED> ] Im
> Auftrag von John Foliot
> Gesendet: Freitag, 03. Mai 2019 21:25
> An: WebAIM Discussion List
> Betreff: Re: [WebAIM] SC 3.1.2 lang exception: proper names
>
> Hi All,
>
> I think, in the end, this comes down to a matter of subjective
> determination.
>
> For example, Wolfgang's name may or may not be pronounced with an emphasis
> on the "W" sound, but perhaps a more "V" sound (Volfgang), that may or may
> not matter in some contexts, but in others may have a significant impact.
> (For example, in a literary context).
>
> In other circumstances, some "names" may be more problematic than others:
> for example, my name in French is Jean (J E A N), which *should* be
> pronounced as (something approximating) "jzawn" (with a very soft N), and /
> but certainly not "Geen". In the context of an email, no harm, no foul; but
> in other contexts it may make a significant difference.
>
> The exception in WCAG states: *Proper names need no lang-attribution, if
> the pronunciation between habitual and synthesized listening only differs
> in nuances of accentuation or intonation ,* and I've offered 2 different
> scenarios where the first (Wolfgang) likely meets this exception, but the
> French variant of my name (Jean) would likely require the lang attribute to
> ensure proper pronunciation.
>
> Finally, the exception is offering times when adding the lang attribute is
> *optional*, but that does not mean that you cannot do so. So (again)
> depending on the context, it remains a subjective call on the content
> author: if you feel it is correct to do so, do so.
>
> HTH
>
> JF
>
> On Fri, May 3, 2019 at 2:07 PM < <EMAIL REMOVED> > wrote:
>
> > No. I am thinking of situations where the Web developer does not have
> > background information about the name Perhaps they can do a Web search to
> > determine the relevant nation and language.
> >
> > -----Original Message-----
> > From: WebAIM-Forum < <EMAIL REMOVED> > On Behalf Of
> > Wolfgang Berndorfer
> > Sent: Friday, May 3, 2019 2:13 PM
> > To: 'WebAIM Discussion List' < <EMAIL REMOVED> >
> > Subject: Re: [WebAIM] SC 3.1.2 lang exception: proper names
> >
> > Hi Jeff,
> > I don't know of any online resources about how proper names are
> pronounced
> > or should be pronounced. How they are pronounced by a speech engine, you
> > can
> > test with a screen reader.
> > But: How they should be pronounced, is my question. And technically that
> > means, when should a proper name get a lang-attribute, if it's not a
> proper
> > name of the default language.
> > You mentioned an older guide for newspapers. Note, that my issue is about
> > acoustic representation of proper names.
> > Did this help?
> > Wolfgang
> >
> > -----Ursprüngliche Nachricht-----
> > Von: WebAIM-Forum [mailto: <EMAIL REMOVED> ] Im
> Auftrag
> > von <EMAIL REMOVED>
> > Gesendet: Freitag, 03. Mai 2019 17:41
> > An: 'WebAIM Discussion List'
> > Betreff: Re: [WebAIM] SC 3.1.2 lang exception: proper names
> >
> > Hello Wolfgang,
> > Are there any online resources for identifying how proper names are
> > pronounced? Content developers are bound to encounter names where they do
> > not know what language to use.
> > Years ago, I helped maintain a newspaper style guide that included some
> > pronunciation rules. I think this was done to help reporters be informed
> > when dealing with the public. We did not give this guide out.
> > Jeff Gutsell
> >
> >
> > -----Original Message-----
> > From: WebAIM-Forum < <EMAIL REMOVED> > On Behalf Of
> > Wolfgang Berndorfer
> > Sent: Thursday, May 2, 2019 4:35 PM
> > To: 'WebAIM Discussion List' < <EMAIL REMOVED> >
> > Subject: [WebAIM] SC 3.1.2 lang exception: proper names
> >
> > Hi especially non-English native speakers,
> >
> >
> >
> > SC 3.1.2 takes out *proper names* from the obligation to attribute
> phrases,
> > when phrase language differs from the language of the page. But why and
> > when?
> >
> >
> >
> > I didn't find any further information in the W3C understanding section or
> > in
> > the archives of this discussion list, except for following small hint in
> > the
> > W3C understanding section:
> >
> >
> >
> https://www.w3.org/TR/UNDERSTANDING-WCAG20/meaning-other-lang-id.html#meanin
> > g-other-lang-id-188-head
> > <
> https://www.w3.org/TR/UNDERSTANDING-WCAG20/meaning-other-lang-id.html#meaning-other-lang-id-188-head
> >
> >
> > "Providing language markup on proper names to facilitate correct
> > pronunciation by screen readers (future link)"
> >
> >
> >
> > So I ask You for correction and comments to my following thoughts:
> >
> >
> >
> > 1. The pronunciation of the proper name through synthesized speech should
> > sound similar to the pronunciation of the name in broadcast media in the
> > default language of the document.
> >
> >
> >
> > German language radio stations commonly pronounce the name of English or
> > French persons in an English or French way. French radio stations on the
> > other hand pronounce any name usually in a French way. This is what
> > listeners are used to and should hear from their speech engine.
> >
> >
> >
> > For example the name of the former French president, François Hollande,
> on
> > German stations always was pronounced like it would have been in French.
> So
> > no screen reader user would know, whom it's about, if "Hollande" was
> > pronounced via German sythesizers. And for the English community on the
> > other hand: Who is the man who is pronounced in German like Shwurzenega?
> >
> >
> >
> > So what stays from the exception for proper names in 3.1.2?:
> >
> >
> >
> > 2. Proper names need no lang-attribution, if the pronunciation between
> > habitual and synthesized listening only differs in nuances of
> accentuation
> > or intonation.
> >
> >
> >
> > Thanks for input!
> >
> >
> >
> > Wolfgang
> >
> > > > > > at
> > http://webaim.org/discussion/archives
> > > >
> > > > > > at
> > http://webaim.org/discussion/archives
> > > >
> > > > > > at
> > http://webaim.org/discussion/archives
> > > >
> > > > > > > > > >
>
>
> --
> *​John Foliot* | Principal Accessibility Strategist | W3C AC Representative
> Deque Systems - Accessibility for Good
> deque.com
> > > > >
> > > > >


--
*​John Foliot* | Principal Accessibility Strategist | W3C AC Representative
Deque Systems - Accessibility for Good
deque.com