W3C

Linked Data Platform (LDP) Working Group Charter

The mission of the Linked Data Platform (LDP) Working Group is to produce a W3C Recommendation for HTTP-based (RESTful) application integration patterns using read/write Linked Data. This work will benefit both small-scale in-browser applications (WebApps) and large-scale Enterprise Application Integration (EAI) efforts. It will complement SPARQL and will be compatible with standards for publishing Linked Data, bringing the data integration features of RDF to RESTful, data-oriented software development.

Join the Linked Data Platform (LDP) Working Group.

End date 31 July 2015
Confidentiality Proceedings are public
Chair Arnaud Le Hors, IBM [updated]
Staff Contact
Eric Prud'hommeaux
Andrei Sambra
Yves Lafon
Teleconference Schedule One 60-90 minute call per week, plus task force calls as necessary
Face-to-Face Meetings Three expected during the course of the group, although the chairs may schedule or cancel meetings as needed to help the group reach its goals. These meetings will use teleconferencing facilities, but effective participation generally requires attending in person, so participants should budget for travel.

Introduction

This group is based on the idea of combining two Web-related concepts to help solve some of the long-standing challenges involved in building and combining software:

  1. RDF, the Resource Description Framework, is a W3C Recommended general technique for conveying information. It has a handful of syntaxes, including RDF/XML, RDFa, and Turtle, any of which can be used to transmit RDF statements. The items about which information is expressed in RDF documents are identified with URIs (eg, http://example.com/products/Widget-71) but the existing RDF specifications do not cover dereferencing them. RDF is the basis for Linked Data and the Semantic Web.
  2. With RESTful APIs and RESTful Web Services, clients use basic HTTP verbs, with their simple and direct meaning, to obtain and alter the state of objects on the server. In these APIs, the remote information objects are identified with URIs which are dereferenced in every operation. RESTful APIs can be defined independent of the formats used for conveying the state of the objects; typically services use custom XML and/or JSON encodings of state information.

The combination of RDF and RESTful APIs is therefore natural, with RDF providing a standard way to serialize information about things identified by URIs and REST providing a way to obtain and alter the state of those things. This approach has been proposed and explored for some time, in academia and industry, as shown by the items listed in References. Within W3C, the SPARQL Working Group developed a RESTful protocol for accessing data in SPARQL data stores and discussed its wider applicability. The participants in the Linked Enterprise Data Patterns Workshop expressed general support for the creation of a Working Group to define a way to use RDF with RESTful APIs in support of application integration.

The basic technique here is to expose application data objects ("resources") on the Web, allowing authorized clients to see and modify object state using HTTP operations (GET, PUT, etc) with an RDF data format. This RESTful approach leverages existing Web technology, including caching, linking, and indexing, and the use of RDF facilitates integration of data across systems and applications. This approach dovetails with SPARQL and is positioned for developers who want more direct access to the application data.

The Linked Data Platform is envisioned as an enterprise-ready collection of standard techniques and services based on using RESTful APIs and the W3C Semantic Web stack. Simple LDP applications can be developed and deployed using only RDF and conventional HTTP infrastructure. More extensive LDP applications can be built using other elements of the stack, including RDFS, SPARQL, OWL, RIF, and the PROV provenance vocabulary. Although expertise in these specialized elements may be helpful, it is not necessary for participation in this group and should not be required for using the Linked Data Platform.

Scope

The starting point for this group is the W3C Submission Linked Data Basic Profile 1.0. Based on this document and contributions from Working Group participants, the group is chartered to produce one or more W3C Recommendations that define a RESTful way to read and write Linked Data, suitable for use in application integration and the construction of interoperable and modular software systems.

The group will also produce supporting materials, such as a description of uses cases, a list of requirements, and a test suite and/or validation tools to help ensure interoperability and correct implementation.

Parts of this work may overlap with general Linked Data patterns and best practice. When they do, the group must take special care to coordinate with other stakeholders who might not otherwise be interested in the group's work. For example, the group may give advice on how to design URIs for long-term stability, as needed for its use cases. This issue is also in scope for the Government Linked Data (GLD) Working Group which is producing Best Practices advice for governments publishing Linked Data. On issues like this, the two groups must coordinate to make sure their advice is compatible. Also, wherever possible and practical, the LDP WG should develop semantics that align well with the SPARQL Graph Store Protocol and the RDF WG’s updates to RDF.

The Working Group will not produce a Recommendation specifying solutions for access control and authentication for Linked Data. However the Working Group may identify, based on a set of real world use cases, requirements for authentication and authorization technologies for use with Linked Data.

This work is to be based on existing HTTP specifications developed within the IETF, including RFC 2616 (HTTP/1.1), RFC 5988 (Web Linking), and RFC 5789 (PATCH Method for HTTP). The group may document requirements for additional HTTP standards work, but it is out of scope for this group to specify additions or modifications to HTTP, such as new headers or new verbs.

Technical Issues

To help explain the work expected of the group, here is a list of practical issues, many of which are addressed in Linked Data Basic Profile 1.0, that can arise in trying to use RDF and RESTful APIs for application integration. These issues and ones like them should be discussed by the group and guidance provided in the delivered Recommendation when practical.

  1. What is the minimum set of RDF datatypes that must be supported by clients and servers?
  2. Which RDF serialization syntax (eg Turtle or JSON-LD) must be supported by clients and servers?
  3. How to create a new resource, which might be a new collection?
  4. How to create a resource with information describing the resource, when the resource address isn't known at creation time?
  5. How to access resource state in reasonably-sized chunks (paging)?
  6. How to safely (predictably) modify a resource simultaneously from multiple clients? (concurrency)
  7. Can resource have multiple rdf:types? Can their rdf:types change?
  8. How can a client modify a resource without replacing the complete contents?
  9. How to get a list of all the resources in collection?
  10. How to move a resource in or out of a collection?
  11. How may a client determine whether/how it may modify a resource?
  12. How to operate on multiple resources at once, for efficiency or to group a set of changes into one atomic operation?
  13. How to integrate with SPARQL, including how to find a SPARQL endpoint serving data for a particular resource? (To be coordinated with SPARQL community/WG.)
  14. Which vocabulary terms are to be used for resource metadata, when necessary for interoperation?
  15. Are there situations where use of blank nodes is allowed/disallowed?
  16. How client software may determine at runtime which RDF vocabulary terms to use when reading/writing resource state? (This information may be used to support validation and vocabulary changes.)

Answering these questions may involve such work as:

Deliverables

Recommendation Track:

  • Linked Data Platform: W3C Recommendation(s) defining a platform based on RESTful read/write Linked Data, detailed in Scope.

Not Recommendation Track:

  • Use Case and Requirements: a collection of use cases and a derived list of requirements that gives a practical foundation with which to analyze proposed designs for elements of the platform. The group will chose the form of this deliverable, such as a set of wiki pages.
  • Test Suite and/or Validator: to help ensure interoperability and correct implementation. The group will chose the form of this deliverable, such as a git repository.
  • Access Control: Working Group Note on Use Cases and Requirements for access control and authentication mechanisms needed for this work.

Schedule

The group will document significant deviations from this schedule on its home page.

Date Event Description
2012-06 Start Group Launch, First Teleconferences
2012-07 UCR Release initial lists of proposed Use Cases and Requirements
2012-09 F2F1 Face-to-face meeting
2012-10 WD1 First public Working Drafts published
2013-01 WD2 Second public Working Drafts published
2013-03 F2F2 Face-to-face meeting
2013-05 LCWD Last Call Working Drafts published
2012-06 F2F3 Face-to-face meeting, if needed
2013-10 CR Candidate Recommendation published
2014-01 PR Proposed Recommendation published
2014-03 REC Recommendation published

Dependencies and Liaisons

W3C Groups

Many of groups listed below will complete their work before this Working Group is done, in which case the Working Group should liaise with the appropriate successor groups, or, if there is none, make a reasonable attempt to communicate with involved individuals and the larger community.

Liaisons

RDF Working Group
to identify those technical issues that may be better developed in the RDF Working Group and vice versa.
SPARQL Working Group
to identify the exact relationships of the planned deliverables with the SPARQL 1.1 Service Description, the SPARQL 1.1 Graph Store HTTP Protocol
JSON for Linking Data Community Group
to explore whether the HTTP response format used for the Linked Data RESTful API can be based on the  JSON-LD specification.
Semantic Web Coordination Group
to ensure a proper coordination between the work of this group and all the other Working Groups in the Semantic Web and eGov activities
Government Linked Data Working Group
the GLD Working Group is also working on advices on vocabulary selection and creation; there is therefore an overlap between the work mentioned for the LDP WG in section 2.2, and the ongoing work in the GLD Working Group. However, while the GLD Working Group concentrates on governmental data, the goal of the LDP WG is more general. The liaison between the groups should ensure a proper coordination between the two Working Groups.
Provenance Working Group
to ensure that the work on getting information about Linked Data resources is in line with the work on Provenance Access and Query
W3C Network Data Community Group
The use cases and requirement defined by that Community Group may be considered as input to the work of this Working Group.

Dependencies

RDF Working Group
This group is chartered to produce Recommendations for Turtle and a JSON syntax for RDF. One of these syntaxes might be chosen as the standard RDF serialization for LDP. The work of the RDF Working Group on graph identification is also relevant to the work of the LDP Working Group.

External Groups

Liaisons

IETF Hypertext Transfer Protocol (HTTP) Working Group
Although the LDP Working Group will not define any new HTTP headers or verbs, a liaison with the HTTP Working Group may be useful to help ensure proper use of existing HTTP mechanisms.

Participation

In general, people participate in this group as representatives of W3C member organizations. At least one representative from each participating organization is expected to devote significant time to this effort (about one day per week, or more, depending on duties), to accept and complete appropriate action items on a timely basis, and to travel to face-to-face meetings, as scheduled by the chairs in consultation with the group.

On a case-by-case basis, using the invited expert process, people may be allowed to participate as individuals, not representing an organization.

To be successful, the Working Group is expected to have between ten and thirty active participants for its duration.

Participants are reminded of the Good Standing requirements of the W3C Process.

Communication

This group primarily conducts its work on the mailing list public-ldp-wg@w3.org (public archives). The mailing list member-ldp-wg@w3.org (W3C member-access-only archives) may be used for administrative purposes, such as travel planning.

Information about the group (deliverables, participants, face-to-face meetings, teleconferences, etc.) will be available from the group's home page.

Decision Policy

As explained in the Process Document (section 3.3), this group will seek to make decisions when there is consensus. When the Chair puts a question and observes dissent, after due consideration of different opinions, the Chair should record a decision (possibly after a formal vote) and any objections, and move on.

Patent Policy

This Working Group operates under the W3C Patent Policy (5 February 2004 Version). To promote the widest adoption of Web standards, W3C seeks to issue Recommendations that can be implemented, according to this policy, on a Royalty-Free basis.

For more information about disclosure obligations for this group, please see the W3C Patent Policy Implementation.

About this Charter

This charter for the Linked Data Platform Working Group has been created according to section 6.2 of the Process Document. In the event of a conflict between this document or the provisions of any charter and the W3C Process, the W3C Process shall take precedence.

Modifications

  1. 25 June 2014: charter extended from 1 June, 2014 until 1 December, 2014
  2. 17 December 2014: charter extended from 1 December, 2014 until 31 July, 2015

References

The following items should be included as background material by the Working Group, documenting some of the work in this field:

[1] Annotea Protocols, Ralph Swick et al. 2002.

[2] Read-Write Linked Data, Tim Berners-Lee. 2009.

[3] REST and RDF Granularity, Eric Wilde. 2009.

[4] A Developers' Guide to the Linked Data APIs, Jeni Tennison. 2010.

[5] Toward a Basic Profile for Linked Data , Martin Nally and Steve Speicher. 2011.

[6] Callimachus API, David Wood and James Leigh. Last updated Feb 29, 2012.


EDIT: 23 April 2013, removing the name of Erik Wilde as co-chair, since he stepped down. (diff).

Alexandre Bertails (bertails@w3.org), Sandro Hawke (sandro@w3.org), and Ivan Herman (ivan@w3.org), editors

$Id: charter.html,v 1.48 2015/08/11 19:19:22 sandro Exp $