W3C home > Mailing lists > Public > w3c-wai-gl@w3.org > January to March 2002

RE: Multiple versions of a page

From: phoenixl <phoenixl@sonic.net>
Date: Sun, 17 Mar 2002 20:02:04 -0800
Message-Id: <200203180402.g2I424TR015517@newbolt.sonic.net>
To: w3c-wai-gl@w3.org
Hi,

I think Jason has a point about supporting a greater range of complexity.
However, I think it would be helpful if there was better understanding
of three or four basic pages in order that users don't have to make
a lot of choices.  A reasonable approach might be where the user can
choose some basic model and then fine tune the model if desired.

Scott

> I think the scenarios being discussed in this thread are rather more
> limited than those which current and developing technologies are
> intended to support. In particular, instead of considering what would
> be the consequences of having, for example, three or four "different
> versions" of a page, it would be more realistic to conceive of a page
> generation mechanism which customizes the output sent to the user
> agent, based on a profile. This profile may include details of user
> agent characteristics (e.g., the kind of device on which the content
> is to be rendered) as well as expressions of user preference. It may
> be supplied manually by the user or automatically with CC/PP or other
> protocols.
> 
> Under these conditions, the output can be constructed to take account
> of multiple needs, and a fortiori multiple disabilities, by including
> all of the required features in accordance with the profile, and the
> problem which is being discussed on this thread is solved, provided
> the server supports all of the necessary adaptations. Of course, it is
> likely that there will be some adaptations which won't be widely
> supported by web servers. For example, braille coding and formatting
> can only be carried out adequately by specialized software, which, I
> expect, most web content developers won't obtain and install on their
> servers (of course the vendors of such software would no doubt prefer
> otherwise). Rather, this adaptation will take place at the "user
> agent" end, or perhaps in a specialized proxy server under the control
> of the user or of a third-party organization.
> 
> To allow for this possibility, it can be argued that content
> developers should provide one options (that is, one profile) whereby a
> semantically rich version of the content, suitable for adaptation by
> other software further along the delivery chain. In fact, there is
> already interest in this area, including specifications that allow the
> originating server to impose constraints, for intellectual property or
> other reasons, on the types of adaptations to which the content may be
> subjected as it traverses the delivery path from the server, through any
> intermediate proxies, to the user agent.
Received on Sunday, 17 March 2002 23:02:13 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:47:18 GMT