W3C home > Mailing lists > Public > public-html@w3.org > January to March 2007

Re: pronunciation and other guides RE: Deprecate <accronym>

From: Colin Lieberman <colin@cactusflower.org>
Date: Sun, 25 Mar 2007 12:04:44 -0700
Message-ID: <4606C7CC.7080405@cactusflower.org>
To: "Dailey, David P." <david.dailey@sru.edu>
CC: Noah Slater <nslater@gmail.com>, Rimantas Liubertas <rimantas@gmail.com>, public-html@w3.org

Two things -

A lot of folks have expressed interest in letting this thread die until 
we've got a better idea of the direction the working group is going in. 
I'd really like to suggest we do that.

Second - pronunciation is a tricky thing. I think I've mentioned 
previously that I've asked the makers of JAWS (the market leader screen 
reader) why they don't support aural css. The answer is that they don't 
trust HTML authors to understand and meet the needs of their users.

I think that we'd be wasting our times discussing pronunciation rules, 
because screen readers will (with very good reason) ignore those 
suggestions.

So let's really let this thread die this time, and revisit the issue 
when it's more relevant to the group's work.

Dailey, David P. wrote:
> Noah Slater wrote:
> 
> "Acronyms, initialisms. and abbreviations all need to be spoken differently."
> 
> I have no strong feelings about this thread and rather concur with Dan Connolly's advice of a couple of days ago to get busy and make some specific use cases and so forth.
>  
> In the meantime, the topic of pronunciation guides, generally, makes me question just how far our scope should go.
>  
> Concerning pronunciation, there are many categories of text where pronunciation guides would lead us beyond the textual or graphemic utterance:
>  
> 1. "suprasegmental morphemes" as they used to be called in linguistics: characterisics of intonation, pitch, tonality, etc. which, though not phonemic, convey sense or meaning.
> 2. while Japanese may have use cases of things resembling abbreviations, how about a ideographic script? I suspect there are similar mechanisms at work (in terms of the morphological and human factors influences) in such writings.
> 3. English and many inflected languages I suspect, allow for contractions and for the dynamic construction of new contractions according to rules. Should we have a <contraction> tag to help guide the pronunciation of those, or simply hope that our pronunciation guide covers those which are already premapped to a pronunciation?
> 4. How about echoic utterances: I'm told that Pig-Latin-like languages exist in languages other than English but that the morphology is rather different. Spanish has a similar system and I think Tamil does to? 
> 5. <jibberish> wo ofasdof awfj sdafl owiej iofsdf </jibberish> Should we occasionally be able to alert our browser that certain strings do not need to be pronounced?
> 6. Rhyming in poetry. Occasionally forced rhymes are inherited from the meter of text. Words can inherit new pronunciation from the words they are rhymed with. Should authors have a way to indicate that? 
>  
> A solution?: If in addition to each graphemic utterance, an author might stipulate an alternative IPA pronuciation tag, then we might solve several issues at once.
>  
> If, in addition to the pronunciation geometry attached to our writing, we considered the projection or our writing into a variety of other spaces, we could end up with a most intriguing topology. 
>  
> That is, we might consider the peculiar juxtaposition of glyphs which sometimes make up the visual pun :
>  
>   L
> E  EVATE  (In case my email messes this up it is the word elevate with the L doubly superscripted)
>  
> Or in the broader sense "Word Art"...
>  
> Should markup for that be a part of the next HTML? I don't think I saw it in WHATWG's current draft. (Though there is a lot there that I am only now discovering.)
>  
> We might wish to allow a variety of parallel semantic markups to convey authors' intentions:
>  
> connotative markup in some sort of Osgoodian Semantic Differential space
> thesaurus-based disambiguation of polysemy
> THEREFORE/BUT  inferential diagrams
> OWL-ish stuff
> causal chaining
> dependency hierarchies
> etc.
>  
> I guess the particular case in hand of <Acronym> and <ABBR> gets me to thinking about just how far the WG wants to go.
>  
> David Dailey
> 
> 
Received on Sunday, 25 March 2007 19:05:19 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Sunday, 25 March 2007 19:05:21 GMT