An introduction to Quatro

Quatro Newsletter 2 (21 April 2006) (0.6Mb, PDF)

Project partners

Pira International

ICRA - The Internet Content Rating Association

IQUA - The Internet Quality Agency

Web Mèdica Acreditada

ERCIM - European Research Consortium for Informatics and Mathematics (the European host of the W3C)

NCSR "Demokritos"

ECP.NL - eCommerce Platform for the Netherlands

University of Milan, Dipartimento di Informatica e Comunicazione (DICo)


Quatro is co-funded by the European Union's Safer Internet Programme

The Quality Assurance and Content Description project (Quatro) aims to help internet users find what they want, trust what they find and avoid material that, for whatever reason, they choose not to see. This will be achieved by devising a common method that can be used by labelling schemes, privacy and security policies, classification and discovery metadata, permissioning and rights management systems. In short, one system with many uses.

The common approach will enable content providers, and especially end users, to benefit from familiarity and interoperability. An end user might, for example, be able to see easily that an article on breast cancer is a recommended reference for a school biology project, its images can be used in schools and private study but may not be republished without permission, and that the article contains depictions of breasts in a medical and educational context.

As well as supporting as many different types of information as possible, Quatro will be applicable across as many forms of digital media as possible, potentially all.

Isn't this possible now?


Systems are already established for carrying content descriptions such as the PICS system as used by ICRA. Quality labels, such as those offered by Internet Quality Agency and Web Mèdica Acreditada operate by a logo being placed on a website. Click the logo and a database entry is shown confirming the logo's validity as appropriate. Privacy policies can be made machine readable using P3P, there are a growing number of digital rights management systems and so on.

The problem

Whilst these systems remain disparate, none can benefit from the take up of the others. Content providers must go through separate processes to apply each kind of metadata. In practice few content providers will apply a full range of labels and metadata simply because it is so time consuming. Moreover, tools designed to do one thing can't do another. The demand for a tool that just adds, say, a privacy policy using P3P without also adding DRM information for example, is likely to be of limited value and therefore of limited commercial viability.

From a user's point of view, separate tools are required to read the different types of data, even if they know it is there and what it means.

Finally, a significant amount of the data that is available is unused. Many webmasters will include metadata describing the content of their site. Some will use standards such as Dublin Core to encode things like authorship and subject matter but the lack of a common approach makes it difficult, if not impossible, for search engines to locate and make use of this data. As a result it is often discarded. If a quality label were machine readable and machine verifiable, search engines might indicate their presence in the search results. Users would then know that a site had some sort of recommendation before they visited it, not once they got there and saw a logo.

The Quatro solution

The Quatro project rests on two foundation stones: policy and technology. Users, content providers, labelling scheme operators, eCommerce practitioners and relevant industry organisations will be consulted on what they wish to see the system achieve and how it fits in with current operations and practices. On the technology side, academics and engineers from different fields will review the existing standards and practices.

A combination of these two reviews will lead to the definition of common platform to support the labels and metadata.

The detail of the platform is a substantial part of the Quatro project, however, it is clear already that Semantic Web Activity at the World Wide Web Consortium will be a central point of reference.

W3C LAyer Cake Model of the web of trust

The semantic web as visualised by the W3C.

XML is ubiquitous across digital media, RDF is the metadata platform that works best with XML and in turn, OWL is the language that allows machines to interpret the RDF data found. The desired goal is a "web of trust" - a way to find what you are looking for and trust what you find.

Wider context

Quatro is not an isolated project. The Internet Association of Japan and Keio University have begun a project to apply content rating to mobile phones using RDF. Others, such as mobile operators, have looked at expressing ratings and other metadata in RDF. Since the project group formed, interest has come from spam filtering companies, mobile technologists, fixed internet service providers and major search engines.

Additional outcomes

As well as the platform definition, Quatro will have at least three further outcomes:

  1. The expression of partner labelling schemes using the new system
  2. A tool for end users that will provide a simple graphical interface for the metadata found. The presence of metadata will be signalled by an icon that can be clicked for more information.
  3. A wrapper for the Open Directory project's database. This will enable the many search engines and portals that use this database to flag results with appropriate (clickable) icons.

Verified metadata

An important aspect of Quatro is that metadata should be verifiable. This will be achieved either through direct database look-up with things like quality labels, through analysis of a resource by artificial intelligence, or a combination of the two.

Future vision

As a result of the Quatro project and the other work with which it interacts, content providers will be given a compelling reason and simple methods for adding metadata and labels to their resources. This might be done through a variety of methods depending on the provider's preferred methods, but alongside XML code and server configurations, one can imagine other encoding methods such as digital watermarks playing an important role.

However the data is encoded, users will be able to make use of it either through versatile and easy to use tools or simple icons alongside search engine results. Whether on the fixed internet, digital TV, games consoles or mobile phones the single system can be used and readily understood.

Project links

Home | Contact | Sitemap | Associate Members | Members