Hi all,
Minutes of the DIWG weekly conference are here:
http:/
/www.w3.org/2006/03/15-diwg-minutes.html
Attached as text for your convenience:
[[
W3C
DIWG teleconference
15 Mar 2006
See also: IRC log
Attendees
Present
Rotan, Max, Cedric, Sailesh
Regrets
Rhys, Kevin
Chair
maxf
Scribe
cedric
Contents
* Topics
1. Publication of the DCO document as a WD
2. Access functions for DISelect
* Summary of Action Items
Publication of the DCO document as a WD
maxf: keep it a note or send it to WD b/c we expect to
change it many times?
... a Note is a document that's not expected to change any
more
... a WD can become a Rec or a Note
... (a Note doesn't need to be implemented)
Rotan: are we going to make statements how often the DI
group is going
to update the document?
maxf: I think we'll have to show some kind of roadmap
<Rotan> See this in status: "There are currently
no plans to amend this
document further."
sailesh: maybe we can add information when something
significant happens
in the field of DI
Rotan: I suggest that we change the statement in the Status
of the DCO
document
all: we agree to publish the DCO document as a working
draft.
<sailesh> The document will be updated when new
technologies related to
the document context emerges or matures
Access functions for DISelect
maxf: we discussed on the mailing list...
Rotan: I have concerns here. Everything that we do should be
in line
with CSS media queries
... we seem to be moving toward incorporating more and more
functions.
Maybe all this should rather be extensions
... I'm not sure what philosophy we chose here
maxf: I don't think we're trying to extend the CSS media
queries
... the real question is whether we want to have all CSS MQ
functions
Rotan: if CSS MQ is referring to a static info we should
assume that the
value is also static. Same for dynamic
... Our definition has to be tied closely to the definition
of CSS MQ
maxf: CSS doesn't really need to know if the property is
static or not
(work made inside the browser); unlike DISelect
Rotan: if you resize the window, the browser reflows but
does *not*
select another stylesheet
... all this happens when the document onLoad occurs
maxf: when the window is resized, I would expect the style
to be
reinterpreted
Rotan: the display area is the viewPort Bert talked about
<scribe> ACTION: Max to contact the CSS group
regarding the
interpretation of media query functions [recorded in
http://www.w3.org/2006/03/15-diwg-minutes.html#action01]
<trackbot> Created ACTION-163 - Contact the CSS group
regarding the
interpretation of media query functions [on Max Froumentin -
due
2006-03-22].
Rotan: the physical display width is static; the viewPort
(usable
display) is dynamic
... it's the display minus the scrollbars, the chrome, etc.
... typically you'll want elements not to overlap with the
scrollbars
... because any other interpretations won't be useful
maxf: according to what the CSS group is going to respond,
it won't
change the definition of the function but may change the way
we
implement it (static//dynamic)
Rotan: in some browsers the chrome component can also be
styled; we have
to take that into accound
... that's a paradox situation: I can use a stylesheet
depending on the
width, but the stylesheet can change the width (since the
scrollbar can
be resized)
... not sure how we would resolve that.
maxf: we're beyond this specification?
Rotan: not sure; that would be an useful piece of
information
... but we would maybe have paradoxes in the style
... maybe we could define the expected processing behaviour
... then the technologies in DISelect should mimic this
behaviour
maxf: my implementation was expected to do the adaptation
part on the server
... so the server should know what the viewport is
Rotan: adaptation is always a "best effort"
process
... with the best available info
... but the server may be relying on information stored in
the repository
... it's not always going to be perfect; it's driven by
different data
maxf: should we include that statement in the spec
Rotan: we'll have to express that the delivery context
might not be
perfectly replicated across different channels
... otherwise expectations of people will be that the end
result of the
presentation will be exactly the same. I think that's
impossible
maxf: I think that's the point of the "default
argument"
Rotan: the description says either you know the exact info,
or you know
nothing
... but in some circumstances you'll be looking up info
somewhere else
... where the information is not perfect.
<scribe> ACTION: Rotan to add some text to say that
you might use
alternative sources of information which reliability you
don't know
[recorded in http://www.w3.org/2006/03/15-diwg-minutes.html#action02]
<trackbot> Created ACTION-164 - Add some text to say
that you might use
alternative sources of information which reliability you
don\'t know [on
Rotan Hanrahan - due 2006-03-22].
maxf: does that conclude our discussion?
Rotan: OK
maxf: we might be able to wrap up
... it's only me who has open actions
... let's adjourn
... regrets for next week
Summary of Action Items
[NEW] ACTION: Max to contact the CSS group regarding the
interpretation
of media query functions [recorded in
http://www.w3.org/2006/03/15-diwg-minutes.html#action01]
[NEW] ACTION: Rotan to add some text to say that you might
use
alternative sources of information which reliability you
don't know
[recorded in http://www.w3.org/2006/03/15-diwg-minutes.html#action02]
[End of minutes]
Minutes formatted by David Booth's scribe.perl version
1.127 (CVS log)
$Date: 2006/03/15 16:02:45 $
]]
BR, Cedric
--
Cedric Kiss http://www.w3.org/Pe
ople/Cedric/
Device Independence Team Contact
mailto:cedric w3.org
The World Wide Web Consortium http://www.w3.org/
W3C/ERCIM - 2004 route des Lucioles, 06560 Sophia Antipolis,
France
|