10-Feb-2000
Attendees:
Johan Hjelm--W3C, Ericsson [JH]
Daniel Weitzner--W3C [DW]
Alain Grignac--WAP, Alcatel [AG]
Kevin Wagner--WAP, Motorola [KW]
Peter King--WAP, Phone.com [PK] (secretary)
Alastair Angwin--WAP, IBM [AA]
Philipp Hoschka--W3C [PH]
Lalitha Suryanarayana--WAP, SBC [LL]
Vincent Quint--W3C (chair) [VQ]
Kazunari Kubota--WAP, NTT DoCoMo [KK]
Daniel Dardallier--W3C, WAI [DD]
By invitation:
Scott Goldman--WAP CEO [SG]
Dave Raggett--W3C, HP Labs [DR]
Tetsuro Tachizawa--WAP, Nokia [TT]
Hidetake Ohto--W3C, Panasonic [HO]
Roberto Castagno--WAP, Nokia [RC]
<<ACTIONS and new ISSUES at end of document>>
[DR] XHTML 1.0 has become a recommendation. Modularization (XHTML 1.1) is in last call. Plan to move it forward to Proposed Recommendation pretty soon. Plan to do some other specs close to being done. XTHML Basic is entering last call next week. XHTML modularization is to split HTML 4.0 up into a number of different modules to meet the needs of a number of different people. The modules can be combined back into a specification of a document type and extended with new modules. XHTML Basic is an example of a very simple document type. There will be a new public draft of the document shortly. XHTML 1.1 provides another example of building a document type out of modules. The is a clean subset of the HTML 4 strict DTD. Talking about rechartering the HTML Activity. There is some interest in doing XHTML 2.0 (to fix some interoperability problems with the <object> element). Considering adding support for XLink into XHTML. XHTML 1.1 and XHTML Basic rely on DTDs. W3C's intention is to move forward on XML Schemas. There is a group that is looking at how good a fit there is between XML Schemas and how it could be used for XHTML modularization. There is also some work going on with extended forms which will be discussed later.
[LS] Question about the status of XML Schema. [DR] Expected to reach Proposed state in late March. May not meet all the needs of XHTML modularization.
[DR] There is some preliminary work on creating document profiles. A set of requirements have been published. Work on that will likely resume in the future.
[AA] XHTML is getting a lot of momentum, but there is a tremendous legacy of HTML 3.2 and HTML 4. People are only now starting to pick up HTML 4. How soon do you think the XHTML piece will be grabbed and made standard? [DR] There are a number of issues that touch on this. The whole point of moving over to XML is to get them to use more rigorous tools. The first step is to get people to use XHTML with there existing browsers. This is why we did not change the language. We have published some guidelines for XHTML publishing that will allow it to be presented on most existing guidelines.
[KW] Could you elaborate on the functionality of XHTML Basic. [DR] It contains the core functionality. Basic structure, basic list, link, a, basic tables. It is a bit less than a strict version of HTML 3.2.
[KW] Has there been any discussion on the use of XHTML Basic in the WAP Application work. [AA] There has been some initial discussions about using XHTML Basic as the basis of the future architecture, and examing how we can evolve the WAP features into the XHTML space.
[DR] If you want to extend XHTML with your own elements, you have to use namespaces for that. Originally, the HTML Working Group was looking at using the text/xml content types, but there were a number of issues that prevented the group from using that. Thus, the XHTML specification refers only to the text/html content type.
[VQ] Is there any dependancy between what the WAP Forum is doing, and the work the W3C is doing. [AA] The XHTML Basic document type. The two biggest problems areas we are going to get into are the concepts of (1) variables, and (2) event handling. [DR] We have a draft document called XHTML Events. It will reach last call in a few weeks. It is an extension of the DOM events model. [AA] DOM was looked at as something for our future, but there was some pushback.
[PK] There has been no objection within the WAP Forum to the proposition of using XHTML Basic as the foundation for the next generation of markup language. It has been suggested that the work to be done is to take the bits in WML that aren't available in XHTML Basic, identify which of those bits are candidates for evolution into W3C modules, and which should continue to be specified by the WAP Forum. Then we would take all those modules and roll them up into a document type that represents the continuation of WML.
[AA] There are three basic routes towards going to convergence: (1) taking what we have, throwing it away and going with vanilla XHTML Basic, or (2) a more seemless approach which allows an XHTML browser to understand what we have, or (3) taking pieces of WML and layering on top of XHTML.
[KK] Every time I hear about XHTML, I keep hearing about new things, such
as XHTML Profiles and XML Schemas. It appears that XHTML is continuing
to move on and on. [JH] This is not new information. The thing
that is happening is that we have not had a complete view from the start. [VQ]
I can understand your reaction. In our work, we are now having smaller
documents and specifications that are divided into modules. We publish
modules one after the other. When you look just at the documents, you do
not get the whole thing. If you want to understand the big picture, you
should look at the charter and the requirements documents. [KK] If
we are to coordinate, then we need access to the information. Some new
things that might have influence in the project should be raised to this
group. [PK] Are you asking how we can perform a comprehensive assessment
of the work that is going on in both organizations to see which work might
require coordination. [DR/JH] We have made presentations on
current work in progress to the WAP Forum. [PH] There is a list
that is coordinated by the IETF for the announcement of new initiatives.
[AA] The WAP Forum works differently. We go public when we produce
a specification.
[PH presented SMIL 1.0]
Started March 97
SMIL 1.0 Recommendation since July 98
Implementations:
Ongoing: SMIL Boston
Coordination Questions
[AA] We have the right people in the group to answer those questions. We've had a Multimedia Expert Group (MMEG) for about a year now. Recently, a Multimedia Drafting Committee has been created with some guidance from the Specification to look at how multimedia integrates into the WAP framework as a whole.
[Tetsuro Tachizawa, Nokia, presented the current state of the Multimedia Drafting Committee]
Multimedia&SMIL Activity
Summary
WAP Multimedia
Why SMIL?
[DD] There is another reason for using SMIL in that it is designed to be accessable.
Requirements
Requirements (2)
Requirements (3)
Process of Mobile SMIL
[HO] would like to act in the role of shepherd. Recommends that there be a greater degree of shared membership than just the one person.
[DW] Wondering what the goals of the SMIL shepherd are. [PK] Recommend following the model of the role Peter Stark has played in the HTML Working Group. Evangelized technologies and ideas in both directions between the groups.
[PH] Agreed with what has happened in the HTML space, but that was done within the context of WML already existing. Thinks we should consider whether it would be advantageous just to do all the work within the W3C, and have a Mobile SMIL come out of that group.
[LS] The CC/PP and UAProf working group relationship has been working well. The WAP Forum has at least three people with wireless subject matter expertise.
[AA] The WAP Forum is a body that understands the wireless space very well.
[DW] Is this is an issue of working group charters? [PK] Yes. It's an issue of charter. Moreoever, it's an issue of focus and working group resources.
[AA] While agree in the notion of having reviewing the charter and where the work should occur, we are on a tight schedule. We need to insure that we can keep to our schedules.
[Roberto Castagno] Our approach with our current work is simply to focus on SMIL 1.0 and avoid forward compatibility problems with SMIL 2.0. We want to take that work to the W3C as input to the SMIL 2.0 modularization and profile work. [PH] Given your time constraints, I can accept the current work going on in the WAP Forum, but I would not like to see the short term work turn into a long term divergence. [VQ] We can identify to separate activities: subset SMIL 1.0 for the short term and then start another activity within the W3C for the longer term Mobile SMIL. [DW] If the work is not moved into the W3C in the short term, we should at least agree (in the short term) to a longterm roadmap.
DECISION: No objections to work on SMIL in WAP specification planned for April, but should look into mechanism that would give W3C SYMM WG review opportunity.
[AA] Summarized the document lifecycle process. In the idea case, a
Draft goes into the Proposed stage, and then to the Accepted stage. The
Prototype stage is meant to give
[VQ] There is an issue with the way character encoding is mentioned when sending WML over whatever medium we use. The discussion seems to indicate that either the XML specification is not clear enough to state what has to be done, or there is some misunderstanding. [TT] The WAP Internationalization Expert Group is not addressing this problem. The discussion should be in the WAE team. My understanding is HTTP charset is ISO8859 and that applies to the WSP side. XML media type documents say the display of character is US-ASCII, but you must UTF-8 and UTF-16 encoding. There is a conflict between 8859 and UTF-8 which are not compatible.
[AA] Support for character sets is taking ages to resolve. [VQ] This issue is how to indicate which character set you are using for a single document. [AA] Tom Bridgeman has been driving the work from within the WAP Forum. [PK] Is it that the WAP Forum has specified it incorrectly, or is the world simply ambiguous and those disambiguities need to be resolved. [VQ] There is confusion and ambiquity in the specifications, especially in the relationship between the HTTP headers and the XML processing instruction.
[VQ] We should identify the people who should be involved in resolving this issue.
[DR] We need to insure that WMLScript and ECMAScript get resolved
into this as well.
[TT] Would like to exchange opinions with the W3C on how to handle
pictograms. We plan to publish a specification in the June
timeframe. [VQ] Now that we know each other we should continue the
discussion via email and report status at the next coordination meeting.
[DD] The WAI is working with several working groups to review the technology being produced by the W3C for accessibility issues. There are three guidelines published for accessibility: Guidelines for web content, authoring tool manufacturers, and for user agents (not yet Recommendation).
[AA] How far do you go. What kinds of impairment do you address? [DD] Visual, aural, congnitive, ...
[DD] Now working on XML requirements. I am interested in knowing what kind of guidelines is WAP working on? Something like the Mobile Access guidelines or like the web authoring guidelines. [JH] The WAP Forum guidelines are produced in the Developers Expert Group (WDEG). It was submitted by Telstra and SBC. It describes for content authors provides a minimum set of WML that can provide a basis for interoperability. They have been working on it and updating it for the new version of WML. There is a further need for a guidelines for WMLScript and transformations.
[DD] Is the model that people will publish WML documents, or take HTML and transform it for mobile devices. [AA] Both. [DD] Is there a working group that is working to define how to translate HTML into WML? [AA] Yes and no. Officially no, but there is a lot of expertise in the problem area, and there are some attempts to figure out the degree to which it would be appropriate to specify transformation.
[DD] It would benefit the mobile market if the web content was more amenable to transformation to whatever is used by the mobile device. The mobile community should promote this kind of accessibility guideline on the Internet. [AA] If you have any ideas on how we can collaborate in this area, we would like to hear them.
[DR] People today who care about their web site continue to use whatever mechanisms that work for them. If you can establish conventions for how to annotate those things, you aren't doing something prescriptive, but you are enabling the space for transformations. Examples include labels and classes.
[PK] How can we coordinate these efforts? [JH] There have been some discussions about this among the W3C staff. One idea is to have a "Device Independent Authoring" workshop to get some focus on this issue. [AA] I know some folks interested. [LL] The carriers would be interested. [JH] The Developers Group would be interested. [DD] We would like to have the gateway providers who have experience in this transformation space participate in this issue. [AA] IBM and I'm sure Phone.com (w/ PK agreeing) would be willing to participate, and I imagine there are a number of companies such as Oracle, etc. How early can we have a workshop? [8 weeks -Ed]. In eight weeks we have a WAP Forum meeting in Miami. We could put together a workshop there, say 1/2 to 1 day. [PK] I think this is a workshop that merits more like two days. [DD] I think the driver for this meeting should be content transformation and how you can use it to present on other devices. [VQ] You have to choose the focus carefully, otherwise you might get a lot of input from the XSLT crowd. [DD] Where do we scope this, on the WAP/W3C mailing list? [many] Yes, we have done it before on this list.
DECISION: The W3C will put together a two-day workshop.
Candidate meeting we could leverage: WAP Beverly Hills: June 19-23.
[Dave Raggett presentation on XForms and Voice Browsing]
Opportunities:
[AA] The mixed metaphor is the powerful point of this message.
[JH] The WAP Forum just had the kickoff its next generation architecture.
[DW] My hope is to sync up on our approach to privacy and security. Would like to gauge what's happening and the extent to which we need to be coordinated. There are two areas in the W3C: addressing Privacy (P3P) and a newer digital signature activity called XML Signatures. The goal of P3P is to develop the ability to build technology tools to address the privacy issues created by the web. The challenges we see are: (1) the global/cross-border nature of the issues (can't address them nationally or regionally), (2) there's a lot of data collected and used in complex ways on the web. P3P has gotten simpler as it was scoped out. It consists of two functions: (1) a schema for describing and disucussion a web site's data collection and use practices, and (2) a scheme for describing a person's preferences on how information will be used. Both of these are extensible. A couple of members have submitted a note to describe ECML in P3P.
[LL] Is this encoded in XML? [DW] P3P describes an XML syntax. There is currently a discussion about how whether P3P statements will be expressed in RDF. They will follow the RDF datamodel, though, so transformation from the XML to RDF will be straightforward. We are hoping to sync up with the simplified RDF schema about to be announced. The P3P specification will go to Candidate Recommendation in April, and it will be there for a few months while implementations are done.
Regarding XML signature, we believe the way to start with security is to create a simple structure to associate signatures, keys, certificates, etc. with documents. There are some issues to work out with regards to packaging and manifests. There is a semantic issue with the ISG regarding the use of symmetric algorithms. The signatures do not solve non-repudiation. Agreed with Espen to have a joint discussion to compare notes on where we are. [AA] The Wireless Security Group (WSG) has already taken a first pass at a mechanism to sign information. That work should be reviewed and commented on.
[AA] Regarding privacy, different countries have different requirements. Are you addressing the reconciliation of cross-border privacy rules? [DW] No. We are taking the approach of making the information available that will allow people to function in the face of all these varying policies.
[AA] Are you looking at any ways that a user could actually glean what a cookie is doing? [DW] There were people in the industry that were looking at P3P as a cookie replacement technology. For a whole lot of reasons, that is not on the current requirements list. IETF is trying to sort this out.
[PK] Can we look into how CC/PP and P3P relate to each other?
[KW presentation on WAP IPR] Summarized the requirements of the membership agreement.
[DW discussion of W3C IPR] At last count there have been over 18 Recommendations with no "problems". The current W3C policy is that we ask members to disclose as soon as they know what they reasonably know. The process does not state specific times. The current culture is represented in the W3C Staff and a bunch of the web development community. Whenever issues have come up, there has been a very strong sentiment against imposing licensing requirements on specifications.
There have been two pressures on reviewing and clarifying these policies: (1) the increase in convergence and collaboration, and (2) there are really significant problems raised by the way the USPTO grants software and business process patents.
Intermind had been a member of the P3P Working Group. At some point, they revealed to staff, the working group and the W3C membership that they had patent claims that read on P3P. There was concern that this patent might read on other W3C specifications, in addition to P3P. In response to the concern and confusion, there was a chilling effect on the P3P work. Some members felt the W3C were not handling the situation well enough. The W3C commissioned a legal analysis of the P3P patent. It was very careful about how the request to the lawyer was constructed so that it simply asked for an explanation of what the patent covered and an assessment of the validity of the patent. The W3C also put out a public call for prior art which ultimate yielded something useful. Several results. Intermind ultimately issued a statement that they didn't believe their patent infringed. Several members of the working group returned. Some implementations of P3P have begun. Another result is that a number of members have made a strong statement that the W3C needs to reconsider its approach to IPR.
The W3C Patent Policy Working Group has not yet come to grips with the
cultural issues related to intellectual policy. The group has rejected
the notion of setting royalty free licensing as a baseline. They are
discussing that on a group-by-group basis they determine the IPR policies for
that working group. [LS] Will the working group members agree to that,
or will it be their lawyers? [DW] That's a very good
question. What has been agreed on is that we need a very stable,
understandable, and predictable structure for managing these disclosures.
[JH] The Coordination Group will have access to both organization's site. I have been working with Bruce Martin to open up username/password mechanism on the WAP web site, but it does not appear to allow you to access the mail archives.
Setting up the web site. There is a current home page in
http://www.w3.org/mobile/wapw3c-coordination
[KW] One way we can handle the openness issue is to have the committee focus on the identification of issues but leave the resolution of the issues to a more public forum.
ACTION: PK to recover the input paper on WML design rationale and
differences between WML and XHTML.
ACTION: AA will arrange for the WAP Forum will provide feedback to the
HTML Working Group XHTML Events draft.
ACTION: VQ to go to the XML Core working group and ask them to get in touch
with the the WAP Forum members involved in working out this issue.
ACTION: DD to move forward the creation of a "Device Independant Authoring"
Working Group meeting targeting the WAP Beverly Hills meeting date.
ACTION: PK to go back to Esko to have Architecture Group arrange a dialog or
teleconference to discuss the multimode access issue and gauge the level of
interest.
ACTION: PK to instruct WSG to look at the working draft and requirements
document from the Digital Signatures group
ACTION: WAP to find the appropriate home for review of the P3P work.
ACTION: DW: Plan a joint WAP-W3C discussion or workshop relationship between
WPA & W3C security/privacy architectures.
ACTION: VQ to post the minutes of this meeting to the W3C web site.
ACTION: KW to set up replication of the group's home page onto the WAP Forum
web site.
ACTION TT: Send heads-up to WAP/W3C coordinatioon group as soon as decision
has been made whether April WAP draft will mention SMIL or not
New issues:
ISSUE: A mechanism for the communication of new activities.
ISSUE: A mechanism for bringing in external expertise. The WAP Forum
does not have an Invited Expert process (like the W3C's).
ISSUE: A mechanism for sharing internal private documents across working
groups.
ISSUE: With regards to convergence: once converged, how do we avoid subsequent
divergence.
ISSUE: How do we manage ownership of the specifications
ISSUE: Process by which people outside this committee can identify
coordination issues.
ISSUE: Impact of the following WAP Activities
STANDING ITEMS FOR STATUS REVIEW: