W3C home > Mailing lists > Public > public-html@w3.org > May 2007

importance and emphasis [was Re: Cleaning House]

From: Gregory J. Rosmaita <oedipus@hicom.net>
Date: Fri, 4 May 2007 12:07:09 -0400
To: public-html@w3.org
Message-Id: <20070504155633.M222@hicom.net>

aloha, thomas!

consider yourself lucky that the presidential elections in france end 
sunday -- here we still have a good 18 months or so of 
electioneering...

while i am in violent agreement with your analysis, i did want to 
point out one thing:

Thomas Broyer wrote, quote:
By marking things up as being "important", you could ask your
screenreader to "extract" the important ideas from the page. If one
interests you, ask for more context (i.e. read the phrase or
paragraph).
This might be a new way of navigating documents with screenreaders
(because <strong> is already almost always presented in bold font, you
can already do this if you're not blind).
unquote

there already exists a "skim reading mode" in many screen readers
which allow a user to define what sort of text to skim:
1. read the first line of each paragraph
2. read the first sentence of each paragraph
3. text rules apply for reading (allows you to either search for a 
specific text string or read new content)
4. attributes and color rules apply for reading

so there is a mechanism which could allow one to quickly navigate 
documents with screenreaders, for the purpose of skim reading or 
searching for a particular portion of the text, but the problem with 
a skim reading function is that one must first know how the important 
and or emphasized text is styled.  none of the screen readers i have 
access to allow one to set switches for individual elements -- that 
burden is placed on the end user, who is forced to create his or her 
aural stylesheet to ensure that text marked EM is aurally presented 
in a manner that differentiates it from unmarked text, and STRONG is 
presented in a manner that aurally distinguishes it from all other 
text.

the same principle underlies such other accessibility features 
common to screen readers as a header list (which increases 
structure and provides an expeditious means for a blind user 
to move from header to header (or from level X header to level 
X header)

this is why support for aural cascading stylesheets is critical,
although the current available mechanisms to add a client side 
stylesheet is to hand-encode it and force one's browser to use 
it.  of course, this is predicated on the use of an ACSS aware 
text to speech engine.

so, i heartily ensdorse your conclusion that there IS a 
difference between STRONG and EM -- EM can, for instance, be 
used to generate air-quotes around terms that are not true 
quotes, but which use quotes to denote emphasis, sarcasm, or 
a nickname -- none of which should be marked-up as a Q, but 
which need to convey aurally the emphasis that one's brain 
attaches to a word in air-quotes.

as i have been stating for quite some time, STRONG does not equal
bold, nor EM italics;  this misconception is at the base of the 
STRONG and EM argument: most people share the erroneous impression 
that the only way to signal STRONG is to style text as bold and 
that the only way to signal EM is to use italics.  an author is 
bounded only by his or her imagination, rather than what browser 
X's decided to implement as a default stylesheet.

ostensively, all a screenreader would have to do is be able to expose 
to the user a list of defined classes used to add style to a 
document, and allow the user to decide what to do next - that is 
easier than querying an element directly, for that would require 
foreknowledge on the part of the user.

this is eminently do-able, as JAWS, for example, when it encounters 
an unsupported lang attribute, speaks the value of the attribute (i 
keep telling them that the 2 letter combos are NOT enough - they need 
to have a table of the standard 2 character codes for natural 
languages, so instead of identifying czech as "cz", as it currently 
does, it would voice the actual name of the language to which the 
abbreviation corresponds, if the user so desires, not merely to 
indicate clearly what natural language has been defined for an 
element, but in order that those who have knowledge of an unsupported 
language, or who has ready access to auto-translation software, can 
at least contextualize an otherwise indecipherable string of text

if JAWS can grab "lang" from an element, it should also be able to 
grab "class" and build a list of classes defined for the document (or 
in an external stylesheet) which could be integrated into either 
the "skim" mode or the "ReadAll" settinngs, without the user having 
to know what exactly the class defines; another feature that this 
would enable would be to trigger audible effects (change pitch, 
change voice, sound an earcon, etc.) for user-selected classes (those 
that are most important to the user or all or none)

hope your candidate wins -- mine never do...
gregory.
----------------------------------------------------------------
CONSERVATIVE, n.  A statesman who is enamored of existing evils,
as distinguished from the Liberal, who wishes to replace them 
with others.         -- Ambrose Bierce, _The Devil's Dictionary_
----------------------------------------------------------------
             Gregory J. Rosmaita, oedipus@hicom.net
  Camera Obscura: http://www.hicom.net/~oedipus/index.html
----------------------------------------------------------------

---------- Original Message -----------
From: "Thomas Broyer" <t.broyer@gmail.com>
To: public-html@w3.org
Sent: Fri, 4 May 2007 12:54:08 +0200
Subject: Re: Cleaning House

> 2007/5/4, Patrick H. Lauke:
> >
> > The difference being...what exactly? In other words: when do you 
want
> > to emphasise something that is not important, or conversely if
> > something is important, wouldn't it also be emphasised? Or is 
this to
> > allow for marking something up as important, but without any
> > distinction in how it's presented?
> 
> Ads or political programs (you might know we electing our 
> President in France ;-) ) generally mark things up in big 
> fonts, bold and/or a different colour so that without 
> reading at the whole thing you get the "important idea". 
> However, if you were to read the text (and speech it), you 
> wouldn't emphasize those words.
> 
> Words that need emphasis on the other hand are generally 
> marked up in italics which hasn't the same effect when you 
> just overlook without really reading.
> 
> By marking things up as being "important", you could ask your
> screenreader to "extract" the important ideas from the page. 
> If one interests you, ask for more context (i.e. read the 
> phrase or paragraph). This might be a new way of navigating 
> documents with screenreaders
> (because <strong> is already almost always presented in bold 
> font, you can already do this if you're not blind).
> 
> So yes, there *is* a difference between <strong> and <em>.
> 
> -- 
> Thomas Broyer
------- End of Original Message -------
Received on Friday, 4 May 2007 16:07:14 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Wednesday, 9 May 2012 00:15:58 GMT