Draft XBRL and the Semantic Web Interest Group - Charter

Scope

Introduction

Today's global economy depends on an efficient flow of capital, and this in turn demands transparency in corporate reporting along with accurate and timely data. Paper-based reports are being replaced by electronic filing. HTML and PDF have paved the way as document formats, and we are now seeing the transition to a more rigorous approach to the semantics of business and financial data, where each reported fact has to be labeled with the financial (or other compliance requirement, such as tax, statutory, or statistical) reporting concept that applies to it. Taxonomies for generally accepted accounting principles such as US GAAP and IFRS are very detailed, with many thousands of concepts and relationships. XBRL (the extensible business reporting language) is an XML format for such taxonomies and has gained the support of governments world wide. The US Securities and Exchange Commission (SEC) has mandated that US public companies file reports in XBRL, starting with the largest companies in mid-2009. Other countries have similar plans, e.g. in the UK, thousands of companies already report in XBRL, which will be mandatory in 2011. In Asia, XBRL has gained early adoption in capital markets with Stock Exchanges in China, Japan, Singapore and South Korea all mandating the use of XBRL. Governments in Australia, the Netherlands and New Zealand, have made commitments to reduce corporate compliance burden using XBRL as part of Standard Business Reporting efforts. This makes it timely for W3C to take a close look at the potential relationship between XBRL and other Web technologies.

What is XBRL?

XBRL is an open standard for reporting business and financial data, together with the associated taxonomies of business and financial reporting principles. It has been developed by XBRL International, while the taxonomies have been developed by XBRL jurisdictions, governments and other efforts. XBRL offers major benefits at all stages of business reporting and analysis, for instance, eliminating the need for analysts to use error prone re-keying of financial data from company reports. XBRL makes use of several W3C technologies: XML, XML Namespaces, XML Schema, XPointer and XLink. Reports filed in XBRL consist of a number of components:

• Instance documents containing reported facts, footnotes, the dates and units for the facts, and further information on the context, e.g. when reporting by business segment and other accounting dimensions
• Schema documents defining XBRL concepts as well as the structure of different kinds of XBRL documents
• Linkbase documents defining the relationships between XBRL concepts, and expressed using XLink

XBRL linkbases cover a range of different kinds of relationships, e.g. defining concepts by reference to the financial reporting literature, human readable labels for concepts, calculations and formulae expressing constraints on facts, and the ordering and nesting of concepts for presentational purposes.

Much of the efforts on XBRL have gone into developing the core specifications and helping companies to prepare reports in XBRL. With growing interest in XBRL from governments world wide, the time has now come to focus on how to exploit the flood of semantically rich business and financial data that XBRL will make available.

What is the connection between XBRL and the Semantic Web?

Applying Semantic Web technologies to business and financial data has huge potential for searching, combining and analysing data at all stages of the reporting pipeline, including the potential for combining XBRL related data with other kinds of data already present on the Web, for example, the CIA World Factbook, DBpedia, geodata, current share prices, exchanges rates and much more. This in turn will unlock the creative potential for web-based services and make it easier for private and institutional investors to track all sizes of companies, and not just the largest, as now. Imagine being able to search for investment opportunities matching your custom risk profile, and to apply a variety of analytics, and to follow this up by locating and browsing through in-depth background information on markets and individual companies.

Facts, concepts and links in XBRL documents can be readily mapped into the Semantic Web. XBRL concepts can be identified via the XML namespace for the schema document defining them as elements together with the element name for each concept. A similar approach can be taken for XBRL instance documents and linkbases. XBRL Import is an open source solution for translating XBRL to RDF's Turtle syntax. Further study is needed for topics such as XBRL versioning, XBRL formulae, and the relationship between XBRL taxonomies and OWL.

Aims of the XBRL and the Semantic Web Interest Group

The XBRL and the Semantic Web Interest Group (XBRL IG) is designed as a forum to support researchers, developers, solution providers, and users of information services that use the Web as the delivery channel. The Interest Group will use email discussion, scheduled IRC topic chats and other tools as a forum to enable broader collaboration.

The following activities are within the scope of the XBRL and the Semantic Web Interest Group and two tasks forces will be formed to achieve the Group's mission:

Use cases and requirements for interactive access to semantically rich business and financial data

• Identify use cases, stakeholders and their requirements for interactive access to semantically rich business and financial data, e.g. within enterprises for more effective management as a result of improved transparency of business operations. Application to continuous auditing and real-time reporting. Support for private and institutional investors, and national reporting and tax authorities. The role of mashups of financial and other existing sources of data on the Web. The role of trust, security and authenticity in the publication and exchange of this information. Longer term possibilities for greater transparency of risk in financial markets.

XBRL and Semantic Web technologies

• Study of topics such as XBRL versioning, XBRL formulae, and the relationship between XBRL taxonomies and SKOS/OWL. The role of the closed world assumption for reasoning over financial data. The role of RIF and other rule languages, together with statistical techniques, e.g. for investment analytics. The means to support queries across disclosures made with different versions of taxonomies, or with taxonomies from different jurisdictions. What implications are there for XBRL and the Semantic Web? Is there a need for further standards work on existing or new specifications?

The XBRL IG may choose to broaden its study to include additional markup languages in closely related areas. Two such candidates include:

• RIXML, a markup language for categorizing, tagging and distributing global investment research.
• FPML, a markup language for swaps, derivatives and structured financial products.

Success Criteria

The XBRL and the Semantic Web Interest Group should be considered successful if the following conditions are fulfilled:

• An appropriate community is formed, including participants from different domains which will bring different perspectives on the topic of semantically rich business and financial data
• Interest Group Notes are developed as noted under Deliverables