W3C home > Mailing lists > Public > w3c-wai-ig@w3.org > April to June 1999

Re: wave file as alt tag?

From: Patrick Burke <burke@ucla.edu>
Date: Mon, 17 May 1999 13:17:30 -0700
Message-Id: <199905172017.NAA15804@serval.noc.ucla.edu>
To: w3c-wai-ig@w3.org
At 11:54 AM 5/17/99 , Peter Meijer wrote:
I suppose that perhaps most US citizens would 
>indeed have the association of the US flag with the particular piece
>of music that you mentioned, but in the rest of the world this may 
>already be different. Still, maybe my example was not a very good 
>one to make a point about generality. I sincerely sympathize with 
>using intuitive associations wherever possible, but I think it will
>be very hard to always find associations that are culturally neutral
>and unambiguous. Associations are known to work quite well with 
>limited image sets, such as the relatively small number of windows
>controls and event types in a computer GUI, but for arbitrary images
>I expect the scheme to break down for lack of suitable "multimedia 
>cliches" in the envisioned lexicographer's catalog.

Actually the case of a national flag is relatively easy. The national
anthem of
each country is not only an associated sound within popular culture, it is a
government-approved audio sign, a sonic icon equivalent of the flag. So an
abbreviated version (a few notes maximum) should serve this purpose as well as
anything. Even here, though, it wouldn't be easy to master the associations
for
200 or so countries. Still, this wouldn't be that much different from learning
the flags visually, I wouldn't think. Text giving the name of the country
still
necessary to solidify the association.

>This is why the attempt arose to define a completely general and
>unambiguous image sonification, being semantically neutral and
>amenable to automatic processing. Still, I admit it has its own 
>share of disadvantages, particularly the interpretation burden on
>the user. Learning a new language is never easy, so it had better
>be a really powerful language to justify all the effort involved.

These sonified images are certainly a strange experience for me (no vision
since 1.5 years of age). Conceptually I would say it is like learning to read,
learning that lines (or dots) on paper really can be associated with verbal
sound.

Another disadvantage of using natural sound, however, is that short sound
clips
(& they would have to be short to be used efficiently) can be easily confused.
To take a stupid example: a few crackles could be rain falling or bacon
frying.
... As for the Windows sound schemes, I can still only tell what half the
sounds are meant to be by looking at the filenames! :-)

Very interesting & fun topic,

Patrick 
Received on Monday, 17 May 1999 16:17:39 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 19 July 2011 18:13:43 GMT