W3C home > Mailing lists > Public > w3c-wai-ua@w3.org > January to March 2000

Profile definition in UAAG10-20000121

From: Harvey Bingham <hbingham@acm.org>
Date: Tue, 25 Jan 2000 20:07:06 -0500
Message-Id: <4.2.2.20000125194212.00ae83d0@pop.tiac.net>
To: w3c-wai-ua@w3.org

The definition of profile there is:

<quote>A "profile" is a named and persistent representation of user 
preferences that may be used to configure a user agent. Preferences include 
input configurations, style preferences, etc. On systems with distinct user 
accounts, profiles enable users to reconfigure software quickly when they 
log on. Profiles may be shared with other users. Platform-independent 
profiles are useful for those who use the same user agent on different 
platforms.</quote>

Different UA may have different, possibly incompatible sets of preferences, 
and
may respond differently to some preferences. The quote suggests that profile is
open-ended. For a user to understand all that can go into a profile, a means
should be provided to allow full specification, to avoid surprises of an
author-supplied change to a default UA behavior, possibly augmented by some
user-supplied profile values. If the user provides a full specification, 
such surprise can be avoided. But for a user to provide a full stylesheet is an
unlikely and burdensome requirement. For example how does one provide style
differentiation for the presentation of recursively defined element contexts.

Different combinations of browsers and AT may require different sets of
preferences.

Where is the set of controls that may be defined for platform-independent
profiles? They may apply for a particular closed application: strict HTML 4.0
rendered on a particular user agent that runs on more than one OS. The
mapping of (presumed same-function, same-side-effect) control across OS is
chancy. Consistency is unlikely across UA on the same OS. Interaction of AT
with different UA on the same OS is also chancy. We haven't been so 
presumptuous
to try to standardize these. Anarchy is likely to win.

Regards/Harvey Bingham
Received on Tuesday, 25 January 2000 20:15:05 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 27 October 2009 06:49:51 GMT