Re: updated timed-text document

At 10:45 -0500 3/11/02, geoff freed wrote:
>Everyone:
>
>The timed-text requirements document has been updated; the new 
>version is available at
>
>http://www.w3.org/AudioVideo/timetext.html
>

Hi Geoff, some brief comments:

a) I'm not sure what the opposition to fonts is;  what we said in 
3GPP was that the font request was a comma-separated list of 
"suggestions", which might include the "meta-names" serif, 
sans-serif, and monospace.  The interaction of fonts, styles, and 
character sets needs careful exploration.

b) I'm not sure what point (2) "usable in all character sets" means? 
Does this mean "allow all Unicode characters"?

c) 3 says to have a default Unicode font, but Unicode isn't a font at 
all.  I'm a little lost here.

d) Adding the URL <http://www.w3.org/TR/ruby> to point 5 would help.

e) Point 6 should say "markup OF the language", i.e. identify the 
language the text is in.  This is needed for text-to-speech, sorting, 
and various other operations.

f) 9 asks for motion, right, and IV.1 says no motion.  I'd say that 
"marquee" and "credits" are both pretty key requirements.

g) 10 appears to be asking for custom or non unicode glyphs.  I guess 
a major question is whether we use only Unicode as an encoding 
standard.

h) III.2 should probably say "Have a valid XML representation" as we 
may want to permit compact binary format(s) (e.g. in an RTP packet) 
as well.


In general, the document desperately needs a paragraph or two on the 
usage modes and expected needs, for example:

sub-titling
karaoke
credits
ticker tapes
closed captioning
text overlay
hyperlinks and interactivity
type-in (user-edited text)???

-- 
David Singer
Apple Computer/QuickTime

Received on Tuesday, 12 March 2002 16:42:37 UTC