- From: Robin Berjon <robin.berjon@expway.fr>
- Date: Wed, 25 Jan 2006 11:40:44 +1100
- To: Bjoern Hoehrmann <derhoermi@gmx.net>
- Cc: www-svg@w3.org
Dear Björn,
On Jan 23, 2006, at 22:15, Bjoern Hoehrmann wrote:
> * Robin Berjon wrote:
>> The primary goal of discard is to reclaim memory for content that is
>> no longer needed. Given Full content sent to a Tiny UA (presumably
>> running on a constrained device), and given that the choice is "keep
>> forever" vs "discard now", the best option indeed would seem to be
>> "discard now". It might break the content but then so might the other
>> option, and it's more in line with the goals of discard.
>
> The proposed change makes it considerably more likely that the viewer
> will preserve the author's intent and makes it easier to construct
> content that degrades gracefully, which is an important accessibility
> requitement.
There are two aspects to facilitating graceful degradation: either it
is there thanks to the author having thought about the issue, or it
is serendipitously provided by the language.
For the former, any author who stops and thinks about graceful
degradation will not use values that are not Tiny compatible —
they'll use seconds.
For the latter keeping the element around or removing it have
identical degradation pros and cons. Removing at 0s will sometimes
cause useful content to be gone, never removing will sometimes cause
useless content to obscure or invalidate the rest.
So all other considerations being equal, optimising for the viewers
becomes a differentiator. That is what the current option does.
--
Robin Berjon
Senior Research Scientist
Expway, http://expway.com/
Received on Wednesday, 25 January 2006 00:41:00 UTC