Minutes of TT WG Meeting on March 4-5, 2004, Mandelieu, France

Minutes of TT WG Meeting on March 4-5, 2004, Mandelieu, France

Attendees

  Glenn Adams (XFSI, Chair, Scribe) [GA]
  Geoff Freed (WGBH/NCAM) [GF] - IRC only (day 2 partial)
  Sean Hayes (MSFT) [SH]
  Dave Kirby (BBC) [DK]
  Thierry Michel (W3C) [TM]
  Dave Singer (Apple) [DS]

Observers

  Bert Bos (W3C) [BB] - day 2
  Philip Hoschka (W3C) [PH] - day 1 partial

Regrets

  Mike Dolan (Invited Expert)
  Erik Hodge (REAL) [EH]

************************************************************************
Agenda
************************************************************************

Day 1 (Thursday, March 4, 2004)

  09:00 - 10:30 Agenda Planning
		Review and Acceptance of Prior Minutes
		Consideration of SYMM WG Merger
		Consideration of transition from CPP to Patent Policy
		Planning for Next F2F

  10:30 - 11:00 Break

  11:00 - 12:30 Review Layered Document Structure Assumptions
		Consider supporting standard style/timing at LF layer;
		in particular, moving applicative style/timing to LF
		layer, leaving referential/inline style/timing at PF/NF
		layers.

  12:30 - 13:30 Lunch

  13:30 - 15:00 Continue review of architecture/assumptions
		Review xpath subset proposal [SH]

  15:30 - 16:00 Break

  16:00 - 17:30 Review style draft from [GF]

Day 2 (Friday, March 5, 2004)

  09:00 - 10:30 Example walk throughs

  10:30 - 11:00 Break

  11:00 - 12:30 Review timing and animation details

  12:30 - 13:30 Lunch

  13:30 - 15:00 Descriptive markup vocabulary
		Standardized metadata vocabulary

  15:30 - 16:00 Break

  16:00 - 17:30 Content selection

++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Day 1 (Thursday, March 4, 2004)
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

+ AGENDA PLANNING

WG: accepted as shown above.

+ REVIEW AND ACCEPTANCE OF PRIOR MINUTES

WG: accept minutes up to but not including previous F2F (Feb 4-6, 2004)

Action: [GA] Post edited draft minutes for previous F2F for review.

+ CONSIDERATION OF SYMM WG MERGER

[GA] Reviews how we got to current point of considering merger. Notes
the following:

* W3T/M wants informal input from TTWG;

* if we need to consider question formally, then this would involve
[TM] sending formal requests to AC representatives of current TTWG
members;

* current charter for TTWG is good up until 12/31/04;

* if TTWG doesn't need to change (add) deliverables, then it is
possible to extend existing charter for TTWG;

Consensus: if merging with SYMM WG would bring new members who are
interested and already up to speed, then there would be general
support for merger;

[DK] When would SYMM WG be ready to contribute to TT?

[TM] Probably wouldn't have time even if they wanted to do so in the
current charter under consideration.

[TM] Current work under consideration for SYMM WG charter:

(1) SMIL 2.0 Revision 1 (incorporates errata)
(2) SMIL 2.1 - a mobile profile
(3) SMIL 3.0 - professional profile ++

[TM] Most likely scenario - SYMM WG work would not have time to
consider any TT work until SMIL 3.0 work at earliest.

Consensus: Do not wish to formally consider merger until (1) it
becomes necessary to renew charter to add deliverables AND (2) it is
clear that there is an advantage in terms of cross-participation
among SYMM WG and current TT WG members. To the extent that
co-location of SYMM WG and TT WG meetings enables and encourages
cross-participation, then such co-location should be considered on an
ongoing basis.

Rationale: (1) concerns that premature merger would entail
additionals delays in current TT WG work as a result of need to bring
up to speed new members and re-establish consensus as necessary; (2)
requiring change to new patent policy and increased scope would need
to be covered by new participation agreements which might not be
obtainable from current members; (3) in order to meet obligations of
having to sign participation agreement covering all of SYMM WG
scope/charter would entail increased work load on part of TTWG
participants that may effectively be outside of their field of
interest.

SIDE-BAR

Action: [TM] consult with W3C Legal to determine whether requirements
exist to explicitly obtain disclosure from past and current public
comments that have resulted in requirements, etc., specifically,
requirement R307. Note that this question is in regard to current
TTWG status under CPP.

+ CONSIDERATION OF TRANSITION FROM CPP TO PATENT POLICY

[GA] Do the current participants have an opinion on voluntarily
transitioning to new patent policy prior to being necessary as a
result of rechartering with new deliverables?

[SH] Doesn't want to do this.

[DS] Would rather not.

[DK] Can't see advantage.

[SH] Huge amount of work on my part without seeing any advantage.

Consensus: Don't transition voluntarily to new patent policy until
forced to do so by rechartering with new deliverables or other
changes to formal W3C policy.

+ PLANNING FOR NEXT F2F

[GA] Had previously planned informally a meeting in June in Japan,
host and date TBD.  Note that NCRD (National Rehabilitation Center
for Persons with Disabilities).

[GA] SYMM WG has tentative plans for F2F at CWI Amsterdam for week of
May 10; would like us to consider co-locating.

Tentative Dates:

* May 10-12 in Amsterdam, CWI; meeting on 12th up to noon
* Jun 22-24 in Cupertino or Mt. View, CA (either Apple or MSFT)

Actions: [TM] Coordinate with Dick Bultermann for logistic details.

* Coffee Break *

+ REVIEW LAYERED DOCUMENT STRUCTURE ASSUMPTIONS

[GA] Consider supporting standard style/timing at LF layer;
in particular, moving applicative style/timing to LF layer,
leaving referential/inline style/timing at PF/NF layers.

[GA] Reviews previous discussions and consensus positions
from last F2F.

[DS] Draws diagram:

		LF		PF		NF

Content		X|I		I(C)		I(G)
Style		A|I		R|I		R|I
Timing		A^I		I		I

Key:	A = applicative
	R = referential
	X = external
	I = inline
	I(c) = inline characters
	I(g) = inline glyphs
	| = inclusive or
	^ = exclusive or

Notionally, PF is closest to DFX (distribution format transform)
profile.

[DK] Why not define DFX to be identical to PF? *or* PF could be DFX.

[DS] If we don't do something

[SH+GA] Let's define identity transform from DFX to "TT-DF-NS-1-0"
(NS = non-streaming), i.e., this implies a non-streaming DF will drop
out of TT-AF-1-0 directly.

[PH] Agrees.

[TM] Wants to discuss testing/implementation requirements for
developing test suites for CR process.

[SH] Should have normative testable assertions in spec.

[GA] Does QA group define standard assertion format?

[SH] Thinks it is specific to WGs.

Action: [TM] review test suites developed by RDF WG, which focused on
authorial intent, to see if some concepts and methodology can be
reused in TTWG context.

[TM] Recent experience with XForms to write assertions and develop
tests within WG.

[SH] Good practice.

[TM] Test suites allow common understanding of functionality; also
helps avoid huge specs where some functionality won't be implemented.

[TM] What will be the CR exit criteria?

Action: [TM] Will research CR exit criteria and make proposal.

Possible Model for Test Methodology

 LF <--> DF(lf), e.g., AML

 |
 | xfrm(x)
 |
 v

 PF  --> DF(pf), e.g., 3GPP TT, RealText, SAMI

 |
 | xfrm(y)
 |
 v

 NF  --> DF(nf), e.g., SVG

1. define example model for xfrm(x)
2. define example model for xfrm(y)
3. define example model for PF->DF(3gpp)
4. define example model for NF->DF(svg)
5. use DF(svg) output from #4 as visual presentation semantics

Principles

1. must be able to implement arbitrary xfrm(x), xfrm(y);

2. must define in spec at least one xfrm(x), xfrm(y);

3. define mechanism on LF and PF that allows author to express that
specific xfrm() must be used to produce PF and NF, respectively;

4. define mechanism on PF and NF that allows author to express that
instance(PF) and instance(NF) was produced by specific xfrm()

5. specify two classes of tools: C0 and C1, where C0 is tool that can
perform xfrm(x) and C1 is tool that can perform xfrm(y)

6. C0 is such that, given standard input(LF), then it can produce
standard output(PF)

7. C1 is such that, given standard input(PF), then it can produce
standard output(NF)

8. there must be 2 independently developed tools in sets C0 and C1
(N.B. this may be relaxed to operate on a feature by feature basis;
i.e., process doc only requires 2 independent implementations for
individual features)

+ CONTINUE REVIEW OF ARCHITECTURE/ASSUMPTIONS

Summary of Consensus from Review of Architecture/Assumptions

1. LF now has standard style/timing, both of which are applicative,
but also support inline; N.B. this is a change from previous F2F;

2. PF now limited to referential/inline style/timing; N.B. this is a
change from previous F2F;

3. must define at least standard transform between each of LF->PF and
PF->NF;

4. must define mechanism to indicate necessity of using specific
transform and prior use of transform;

5. need documents test(LF) and test(PF) and test(NF) in order to test
standard transforms; intention is to build tests from normative
examples in documentation;

* Coffee Break *

+ REVIEW XPATH SUBSET PROPOSAL [SH]

[SH] Shows examples of how one might use XPath 2.0 tools to perform
space delimited token selection.

Resolution: Use subset of xquery [probably drawn entirely from xpath
2.0] to perform selection for both applying styling and timing at LF
layer.

Action: [SH] to draft standard ready text for subset of xquery.

+ REVIEW STYLE DRAFT FROM [GF]

Resolution: For normative definitions of style parameters, reference
first the normative definition in XSL FO, and normatively augment
this with SVG representations of desired results to the extent this
is possible. Where possible, do not re-express prose of referenced
definitions; however, in all cases, specify at least one normative
example using NF/SVG, possibly may need PF and/or LF depending on
abstractness of property.

Rationale: in case SVG WG says SVG 1.2 does everything we want, then
why do TTWG; possible answer: we're providing a workflow chain going
from a more abstract source document with domain specific semantics
through a series of more concrete layers to end up with a subset of
SVG 1.2.

Rationale: in case CSS WG asks why we aren't using CSS syntax, the
answer is that CSS has no infoset, can't be XQueried, can't be
XSLTed, isn't DOMable, etc. we want full XML based infoset workflow...

[DS] Il voudrait une compilation des resolutions/diagrammes...

++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Day 2 (Friday, March 5, 2004)
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

[GA] While waiting for folks to show up, finds and reads Goon Show
Scripts from [1].

[1] http://www.thegoonshow.co.uk/

[GA] Suggestion: use Goon Show script as example TT content in REC.

+ EXAMPLE WALK THROUGH - DEFINITION OF EXAMPLE DOC FROM TOP-TO-BOTTOM

See 20040304-ex1.xml.

Action: [GA] post example file to member reflector.

+ REVIEW TIMING AND ANIMATION DETAILS

Working Assumption: Adopt following time semantics model for PF layer:

Element    Container    Timeable    Default Action  Other Actions

flows      par          no          none            none
flow       par          yes         visibility      none
block      par          yes         display         visibility
inline     par          yes         display         visibility

To specify other actions will require timeAction attribute or
equivalent.  For other elements that above, no time semantics apply.

[DS] Concerned that above model may not handle all needed scrolling
semantics.

Resolution: Flow in PF layer is not associated with region; rather,
top-level blocks are associated with regions via style.

Resolution: Inline timing is permitted. [See above working
assumptions.]

Open Issue: Whether to allow block as immediate child of inline? N.B.
XSL-FO does allow this. [GA] showed example of renmoji (horizontal
block in vertical Japanese lines).

Action: [DS] Based on email of 01/14, integrate with fillBehavior as
described in example.

[BB] What additional style properties on region?

[DS] review R306 as possible candidates: e.g., background-color,
border-*, visibility

[GA] region is defined as a reference/viewport area pair; where
block/inline- scroll offset are used to locate origin of reference
with respect to viewport; animation of these parameters make
scrolling operation possible, such as smooth or jump scrolling.

[SH] suggests considering not supporting float related properties in
R306; either remove or change shall to should or may in requirement.

Action: [SH] make proposal regarding keeping or removing float style
parameters.

+ DESCRIPTIVE MARKUP VOCABULARY

Insufficient time to discuss; however, see example file [?] for
references to agents.

+ STANDARDIZED METADATA VOCABULARY

Insufficient time to discuss; however, see example file [?] for some
attempts regarding agents, performers, etc.

+ CONTENT SELECTION

[GA] describes DISelect WD from DIWG [3].

[3] http://www.w3.org/2001/di/Group/di-selection/

Provided this mechanism is well-defined and validatable as XML, then
it could be used to satisfy R207 [and could provide a few useful
features beyond R207 such as the ability to match all conditionally
selected elements].

++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
START SUMMARY
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

*** RESOLUTIONS ***

Consensus: Do not wish to formally consider merger until (1) it
becomes necessary to renew charter to add deliverables AND (2) it is
clear that there is an advantage in terms of cross-participation
among SYMM WG and current TT WG members. To the extent that
co-location of SYMM WG and TT WG meetings enables and encourages
cross-participation, then such co-location should be considered on an
ongoing basis.  Rationale: (1) concerns that premature merger would
entail additionals delays in current TT WG work as a result of need
to bring up to speed new members and re-establish consensus as
necessary; (2) requiring change to new patent policy and increased
scope would need to be covered by new participation agreements which
might not be obtainable from current members; (3) in order to meet
obligations of having to sign participation agreement covering all of
SYMM WG scope/charter would entail increased work load on part of
TTWG participants that may effectively be outside of their field of
interest.

Resolution: Use subset of xquery [probably drawn entirely from xpath
2.0] to perform selection for both applying styling and timing at LF
layer.

Resolution: For normative definitions of style parameters, reference
first the normative definition in XSL FO, and normatively augment
this with SVG representations of desired results to the extent this
is possible. Where possible, do not re-express prose of referenced
definitions; however, in all cases, specify at least one normative
example using NF/SVG, possibly may need PF and/or LF depending on
abstractness of property.

Resolution: Flow in PF layer is not associated with region; rather,
top-level blocks are associated with regions via style.

Resolution: Inline timing is permitted. [See above working
assumptions.]

*** OPEN ACTION ITEMS ***

Action: [SH] Will investigate use of media queries in this context and
report back.

Action: [DS with help of Paul Nelson and Peter Lofting] Write RFC to
register appropriate opentype/truetype font types as MIME media types,
suggest model of "application/font-<font-type-name>", e.g.,
"application/font-truetype".

Action: [GA] Make proposal regarding use of Xlink vocabulary or
"src" attribute.

Action: [GF] Investigate whether to use IRIs instead of URIs?
Note: XPointer and Namespaces in 1.1 use IRIs?

Action: [SH] Investigate use of "role" vs "class" attribute.

Action: [GF] Investigate mechanism for cascading
semantics and whether to support cascading on either or both
logical and presentation flowed vocabularies.

Action: [GA] Draft new requirement on "Integrability"
in general terms that should not impact testing or implementation
requirements.

Action: [GA] incorporate agreed changes into TT-AF-1-0-REQ in
preparation for publishing final W3C Note.

Action: [SH] will review and propose subset of aural parameters
(see R305).

Action: [GA] Add figure showing logical structure anticipated
by requirements.

Action: [GA] Add note to R217 and R219 that shows use of data: URI
scheme.

Action: [TM] Find out if xpointer() scheme WD [1] is still being
progressed forward.

[1] http://www.w3.org/TR/xptr-xpointer/#document-order-notation

Action: [SH] to propose subset with extensions for use in CS
Profile.

Action: [GA]/[GF] investigate syntax for regions vis-a-vis style.

Action: [DK] To write up short paragraph on uses of
role tokens. Suggest removing or adding as he
progresses.

Action: [TM] to propose and define standard MD attributes.

Action: [*] Think about standard MD items (and write something
down and send to list so we can think about it too).

Action: [GA] Need to start planning for June meeting in Japan!

Action: [GA] Post edited draft minutes for previous F2F (02/04)
for review.

Action: [TM] consult with W3C Legal to determine whether requirements
exist to explicitly obtain disclosure from past and current public
comments that have resulted in requirements, etc., specifically,
requirement R307. Note that this question is in regard to current
TTWG status under CPP.

Action: [TM] review test suites developed by RDF WG, which focused on
authorial intent, to see if some concepts and methodology can be
reused in TTWG context.

Action: [TM] Will research CR exit criteria and make proposal.

Action: [SH] to draft standard ready text for subset of xquery.

Action: [GA] post example file to member reflector.

Action: [DS] Based on email of 01/14, integrate with fillBehavior as
described in example.

Action: [SH] make proposal regarding keeping or removing float style
parameters.

*** OPEN ISSUES ***

Issue: Whether to use XLink vocabulary, e.g., as used consistently by
SVG, or use "src" attribute as apparently will be done in XHTML2?

Issue: Whether to use IRIs instead of URIs? Note: XPointer and
Namespaces in 1.1 use IRIs?

Issue: Should we use "class" instead of "role"?

Issue: Probably want to permit in logical content mode the selection of
content based on generic XML features of non-TT namespace descriptive
markup, e.g., for applying style and timing semantics, in which case an
appropriate TT container element shall be implied based on nearest
ancestor TT namespace element.

Issue: Need to think about cascading semantics; how to express, how to
apply, etc. Possibly use CSS semantics here as well.

Issue (2004-03-05): Whether to allow block as immediate child of
inline? N.B. XSL-FO does allow this. [GA] showed example of renmoji
(horizontal block in vertical Japanese lines).

*** URLs ***

[1] http://www.w3.org/TR/xptr-xpointer/#document-order-notation

*** NEXT MEETING DATES ***

* Mar 11 - commence regular telecon schedule
* May 10-12 in Amsterdam, CWI; start on 10th at 1030h
* Jun 22-24 in Mt. View, CA or Cupertino (either MSFT or Apple)

++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
END SUMMARY
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Thierry MICHEL
W3C/ERCIM

Received on Monday, 26 April 2004 10:12:42 UTC