This Incubator Activity (XG) has published its final report and is now closed. The work undertaken under the XG provides an important input into the work of the Protocol for Web Description Resources Working Group (POWDER).
The public and member mailing lists are also closed and are archived.
March 2006: The group has begun its work and is focusing on use cases. The charter includes 4 use cases which are being expanded upon. Additional use cases are also being written by the group but we welcome even more. If you have a use case, please let us know through the public mailing list. All use cases will be published in full, following which, detailed requirements will be drawn up.
June 2006. Following a presentation of WCL-XG’s work to the Mobile Web Best Practices Group at their face to face meeting on 22 June 2006, it was resolved that the output of WCL-XG would be used as the basis for a straw man for the proposed "mobileOK trustmark."
February 2007: Following publication of this XG's final report, the Protocol for Web Description Resources Working Group has been chartered. This will allow the work done in this activity to transfer to the Recommendations Track.
Meetings are held every other week. The timing alternates between convenience for participants in East Asia and West Coast USA. A single Face to face meeting is planned and will be co-located with the WWW2006 conference in Edinburgh.
The inaugural meeting was held on 24th February 2006 during which participants were introduced to each other and the time line for the XG set. In particular, it is noted that the Mobile Web Best Practices Group is due to consider what mechanism should be used for the proposed mobileOK trustmark in mid June 2006. As a consequence, the group decided to work towards a target of completing its work by early June.
10th March 2006: The group reviewed the use cases in the charter. Each will be expanded upon and additional use cases written. These will be published here as soon as possible. Further use cases are being written and non-participants are invited to submit yet more.
23rd March 2006. This was a more difficult meeting as we tackled the question “what is the group for?” in more detail. It was resolved that in order to maintain our focus and avoid setting over-ambitious and ultimately unreachable targets, the use cases in the charter, plus a fifth that can be summarized as being able to say “this resource is a good example of something that is bad,” would set the boundaries for the work. In addition to completing work on reviewing the 4 use cases in the charter, group members took actions to begin to derive requirements. An evolving new document setting out the use cases and requirements will be made available on the public website.
7th April 2006: The group was pleased to be joined by Akio Kokubu and Taniguchi Kunikazu, as well as Noboru Shimizu from I A Japan. Their project on content filtering for mobile has now been completed with high political success. However, there remains a great deal of technical work to be done, hence their support for the XG.
As the use cases are now more or less complete, the meeting turned to discussing the requirements. Several requirements have been suggested within the group and some of these were discussed in the meeting, with several formally adopted by resolution. Further requirements will be added following e-mail discussion. See the full list of agreed and draft requirements.
20th April 2006: The meeting began with a welcome for Diana Pentecost from AOL. The participants now include representatives from 16 organizations: AOL, Asemantics, AT&T, Center for Democracy and Technology, Centre Virtuel de la Connaissance sur l'Europe, IA Japan, ICRA, Institute of Informatics & Telecommunications (IIT) at NCSR Demokritos, Maryland Information and Network Dynamics Lab at the University of Maryland, Opera Software, RuleSpace, Segala, T-Online, University of Bristol, Vodafone, Yahoo!
After a discussion about the time and date of the group’s face to face meeting (now set for Saturday 20 May in Edinburgh) attention turned to the use cases and requirements. Resolutions on the remaining draft requirements will be taken on the member mailing list so as to avoid taking up more of the very limited time available on the fortnightly teleconferences.
Most of the meeting was spent discussing whether an RDF-based approach is the correct way to proceed. The group is resolved that it is. However, it is also recognized that steps must be taken to show that the underlying system can be used in a non-RDF environment. AT&T has offered to review the use cases and direction of the group from a non-RDF perspective.
5th May 2006: The meeting focused on discussion of the requirements. This resulted in some being formally adopted and others being removed. The major issue though was what data should be included in the labels and what data should be made available about the labels. In other words, the label describes the content but what are the requirements for the group for defining how data such as who issued the label, when it was last reviewed and by whom, should be made available. See the Use Cases and Requirements document.
Work will begin now on creating the framework of the final report as a guide for the face to face meeting in Edinburgh on 20 May.
20th May 2006: The group’s face to face meeting was held at the Scotch Malt Whiskey Society in Edinburgh on Saturday 20th May. The meeting used an early draft of the final report as a focus for discussion.
A series of resolutions were taken on the detail of what should and should not be included, how various statements should be worded and what clarification was necessary. These resolutions are included in the substantially updated version of the report.
The key decisions were:
2nd June 2006: The group continued to discuss the report with significant attention being given to the definitions of relevant terms. A new section is to be written that sets out the scope and general semantics of a content label. There was also debate, unresolved, about whether the cLabel system should make speicific provision for making assertions that a particular thing is not true. In some cases this has a practical advantage “these resources are meet all mobileOK criteria except x, y and z.” The problem is that it then becomes very easy to write self denying statements such as “these resources met all mobileOK criteria except a…z (i.e. all of them).”
9th June 9th June The group discussed the recently published glossary, in particular the issue of whether we needed to have different constructs for saying that a resource has a property whose value is X and that a resource has a particular property. It was resolved “no” – i.e. to ensure that all assertions can be processed in the same way, we will define content labels in such a way that there will always be a value for a given property even if the value is true|false.
There was then a discussion about how the overall scope of a label. This was couched in policy terms – to what extent can a content provider make statements about content? Ownership of a domain is the logical starting point since ownership is readily identifiable. But should we make it a requirement that the scope of all labels is defined at an outer level in terms of domain names? The resolution was that “The scope of labels must be definable in a flexible way based on string patterns that allows, but does not mandate, detailed lists of domains; and that allows exceptions to be made.”
16th June 2006 A short meeting this time to reflect on recent discussions concerning the proposed abstract model and system architecture. The group is preparing for a presentation at a meeting of the Mobile Web Best Practices Group in New York where a primary use case for WCL, mobileOK, will be under discussion.
30 June This majority of this meeting was taken up by a discussion about a model for the grouping of URIs. This is a largely abstract model that maps readily to XML and is also being mapped to the RDF Model for Content Labels.
Also this meeting took a resolution to seek transition to the Recommendations Track.
7 July A relatively brief meeting that discussed how label data should link to a Labeling Authority (and what information an LA should provide); the steps to be taken to move to the Recommendations Track and the current status of the Report.
21st July The group held a further discussion on the report, particularly with respect to the grouping of resources issue and how the core vocabulary fits into the abstract model. It was resolved that the abstract nature of he WCL model would be emphasized by removing the encoding section into a separate document. This will be done by next week when the group hopes to conclude the bulk of its work as an XG before seeking the WG charter.
August 2006 The incubator activity is now actively seeking a charter as a Working Group.
The diversity of material on the Web continues to grow to encompass audio, video, games and all manner of data services alongside traditional documents. The Content Label Incubator Group [WCL-XG] aims to foster ideas for how content providers can inform search engines, aggregators and other data systems, that their content is of a certain type, fulfils certain criteria or meets given requirements.
Labels will need to be applicable to a resource or a group of resources in a way that is convenient for content providers. Importantly, it should be possible to build systems that in some way show the labels to be trustworthy. It is anticipated that the solutions discussed will be centered on Semantic Web technologies.
The charter includes a number of detailed use cases and highlights relevant work already done in this area.
The Label XG discusses its work both through public and member-only mailing lists that are archived.