RE: 1.2 feedback: Streaming

Robin,
I'm going to reflect your comments right back at you with questions. 

What streaming features do you think is needed in SVG 1.2 (versus what
might be needed in SVG 2.0 or SVG 3.0)? Do we need push as well as pull?
Do we need skip forward or jump backwards?

Incidentally, my assumption has been that anything SVG needs in terms of
streaming would also be needed by SMIL. Therefore, my thinking is that
any new features in terms of time-based content, such as streaming,
should show up first in SMIL -- for time-based content SMIL is the
leader and SVG is the following. Because of this thinking, I have been
assuming that the SVG working group would look at SMIL 2.0 and see what
streaming-related features from SMIL might get incorporated into SVG
1.2. For example, perhaps the 'syncBehavior' attribute from SMIL might
be used to deal with the vagaries of network delivery of streaming
content. Does this approach make sense to you?

It is important to ask what are the use-cases for streaming. Long
cartoons? Or just allowing the user to interact immediately with a
document while more detail is getting downloaded? Something else? Some
people in the printing community want "streamable SVG" because they
don't want to have to create a DOM inside of a low-end printer.
Personally, I don't think long cartoons are the sweet spot for SVG in
the marketplace.

Jon


-----Original Message-----
From: www-svg-request@w3.org [mailto:www-svg-request@w3.org] On Behalf
Of Robin Berjon
Sent: Monday, November 18, 2002 10:09 AM
To: www-svg@w3.org
Subject: 1.2 feedback: Streaming


Hi,

the description of the streaming functionality is somewhat short. One
thing it
doesn't state is whether the WG is only considering push streaming, or
if pull
streaming (load on demand) is also being investigated. Streaming also
tends to
pull in other features such as the ability to control the stream (skip
forward,
jump backwards...), and generally some form of (possibly minimal)
indexing
facility to cut up the document automatically into several small
components.

Has this been scoped or is this more in the "we'll think about it later"
stage?

-- 
Robin Berjon <robin.berjon@expway.fr>
Research Engineer, Expway
7FC0 6F5F D864 EFB8 08CE  8E74 58E6 D5DB 4889 2488

Received on Monday, 18 November 2002 17:35:49 UTC