Welcome to Talis Platform News

Welcome to the first issue of Talis Platform News, a new way for you to find out how we are continuing to develop the Talis Platform, and to share your own stories on putting it to use in powering the Web applications that matter to you.

The result of several years of investment here at Talis, the Talis Platform is now becoming visible to the outside world as comprising a powerful combination of Semantic Web, information retrieval, collective intelligence and behavioural mining technologies that are founded deeply in the most profound and challenging phenomena of our time: the connectivity disruption. It provides the underlying data management, organisation and analysis components that can learn and understand patterns of behaviour and present them in a form fit for embedding within your application. These capabilities are exposed via consistent and accessible APIs, in order that the benefits of the Talis Platform are available to applications across the Web in a flexible manner that supports many potential use cases.

Talis Platform News will be released each month, in full form on the Talis Web site and as an email alert to draw your attention to the main topics in each issue. I invite you to subscribe to our automated email alert, and can assure you that we will not pass your details on, or otherwise subject you to a deluge of unwanted spam.

If you have stories of your own to share, please don't hesitate to get in touch with me, Paul Miller, Danny Ayers or any of your existing contacts at Talis.

As well as being visible in the wider development community, many of the team tend to congregate on Talis' IRC channel on freenode.net, #talis. Feel free to join us there.

Ian Davis, Chief Technology Officer and Director, Talis

Danny Ayers joins Talis

Talis has recently recruited Danny Ayers to act as Platform Community Manager, interfacing with developers and users of the Talis Platform in the Web community at large. Professionally, Danny has in recent years been a freelance developer and technical writer, having co-authored ten books from "Professional Java Server Programming" to "Professional Web 2.0 Programming". However, he's best known in the development community as a long-term advocate of Semantic Web technologies, through his blog and numerous community initiatives. Danny is an Invited Expert on the W3C's Semantic Web Education and Outreach Interest Group and GRDDL Working Group. He also writes a regular column on advances in Web technology for the IEEE Internet Computing magazine.

Danny explains "I've long had high regard for the innovative work done by Talis, so was delighted to have the opportunity to join the company. From a personal perspective, the role I'll be taking on puts me in a position which perfectly captures my own ambitions. I'd like to see the potential of the Semantic Web realised sooner rather than later, but this is dependent on raising awareness in the developer community at large. I can't think of a better way of achieving that than through building a community around an environment like the Talis Platform that takes the hard work out of Semantic Web development."

This month Danny had an article featured in IEEE Distributed Systems Online, "Evolving the Link".

Harnessing Sophisticated Mass Collaboration on a Global Scale

In Talis’ latest white paper, Justin Leavesley and Ian Davis talk about the rationale behind the Talis Platform, and describe the set of real-world circumstances to which it represents our considered reaction.

Justin and Ian explore the ‘connectivity disruption’ with which so many of us are currently grappling. They discuss the societal implications of ubiquitous access to peers and resources via the Web, the technological trend toward the ‘Internet inside’ modern applications, and the economic opportunities presented as the falling cost of communication and participation undermines the trend toward ever-larger organisations. All of this, and more, has influenced the thinking that has shaped the way in which we are designing and building a robust set of Platform capabilities.

GRDDL Primer released by W3C

GRDDL - Gleaning Resource Descriptions from Dialects of Languages - is a technique for obtaining RDF data from XML documents and XHTML web pages. Ahead of a formal release of the GRDDL specification, the GRDDL working group has released a primer in which a number of examples from the GRDDL Use Cases document are used to illustrate the techniques GRDDL provides for associating documents with appropriate instructions for extracting any embedded data. Talis' Ian Davis and Danny Ayers serve on the GRDDL working group, and Ian is one of the editors of this latest document.

What is the Talis Platform?

Ian Davis, Chief Technology Officer and Director, Talis

The Talis Platform is a new technology environment for building human-centric, information-rich applications. It is provided as Software as a Service (SaaS) and is designed to hide the complexity of storing, analysing and managing large amounts of semi-structured data. Applications can use the Platform API, a series of RESTful web services, to create, update and search all kinds of data managed by the Platform.

The Platform provides a generic capability to manage and organise content and its metadata. Any type of content such as images, documents or files can be added to the Platform and described using RDF. Each piece of content is automatically assigned a URI and so is made addressable and linkable in the web. This allows assertions about the data, i.e. more metadata, to be made by anyone, whether managed by the Platform or not.

Content is organised into regions called stores which can be separately configured and controlled. The Platform encourages sharing and remixing of content so stores may be combined in many different ways for searching. Additionally stores can augment the content in other stores. This augmentation relates content managed by one store with content in others by analysing each store's metadata and determining the best way to combine them. For example, in the case of a store containing book information, other stores containing images of book jackets, author biographies, reviews or tags can be used to augment items discovered by searching for books. The result is a very simple mechanism for building applications that combine data from many sources.

The Platform supports a rich set of services for accessing and remixing data such as Sparql and automatic faceting. The default query interface provides full-text searching over indexed content and metadata supporting a rich search syntax and sorting by platform-determined relevance factors or by specified metadata fields. The Platform aggregates, anonymises and analyses patterns of usage of the content managed by the Platform to calculate its relevance factors. These patterns of usage will also be available as individual services so that applications can incorporate suggestions and improved relevancy of information to their users.

The Talis Platform is designed to facilitate sophisticated mass collaboration. Multiple applications supporting many thousands of users can share content such as reviews either by agreeing on a standardised schema and using a single store, or by using independent formats in distinct stores relying on the Platform's capabilities for relating different schemas together.

A number of existing Platform capabilities, described in the context of their use within library applications, may be found here.

Those interested in exploring ways of leveraging the Platform's capabilities within their own applications should contact Danny Ayers and Paul Miller.

Conference Round-Up

May and early June was a busy time for conferences, and various Talis staff were to be seen at WWW2007, XTech, Semantic Technology and the European Semantic Web Conference. Danny Ayers chaired the Developer track at WWW2007 this year, and Paul Miller and Rob Styles participated in a panel session on Linked Data. At XTech, Ian Davis, Rob Styles and Paul Miller delivered presentations, and Chris Clarke offered a lightning talk. Danny Ayers gave an invited talk at the Scripting for the Semantic Web workshop at ESWC on the use of the agent paradigm in the Web environment.

As usual, Talis bloggers were on hand to provide a running commentary of the unfolding event, with comprehensive coverage of WWW2007 beginning here and XTech beginning here on our Nodalities blog.

Open and Linked Data

A strong theme running through recent conferences, from Tim Berners-Lee’s keynote presentation in Banff onward, was that of Linked or Open Data. As Paul Miller discussed on the Nodalities blog recently, Linked Data is rapidly emerging as a compelling and intelligible driver for business adoption of the long-discussed Semantic Web. Recent Talis podcasts, including those with Revyu's Tom Heath and Metaweb's Jamie Taylor delve further into this topic, illustrating some of the opportunities for realising the value locked up in your data by exposing it for linking to complementary resources elsewhere.

Tim Berners-Lee further emphasised a number of these trends in a short video interview with ZDNet's David Berlind.

As Talis' Rob Styles argued at WWW2007 in Banff, effective licensing of data is of great significance, even if your intention is simply to make the data available for permissive reuse. Whilst license regimes such as those from Creative Commons are effective in extending the utility of copyright, their legal provisions do not stretch to cover the factual nature of many databases, despite their frequent use in this way by well-meaning data owners. European countries benefit from the notion of a database right, and work at Talis has extended this right via our draft Talis Community License in a very similar way to that in which Creative Commons extends the copyright provisions. We are now looking to undertake further work in this area, and welcome approaches from those willing to work with us on tackling the problem of creating mechanisms with which data owners can encourage use and reuse without waiving all of their rights.

Chemist Peter Murray-Rust from the University of Cambridge picks up on some of these issues in his contribution to the ongoing Talis podcast series when he discusses the importance of 'Open Data' to the conduct of scientific research.

Why a Platform?

The rise of collaborative commons from open source to open science is being made possible because everything and everyone is increasingly connected together by the Web.

Many complex problems no longer need to be solved by a single company taking a big risk and making a large investment. A whole range can now be solved by mass collaboration and a commons approach. Effort or knowledge is aggregated from a large number of participants to solve a complex problem with very little risk or cost to the individuals. With no individual taking a large risk or investment it is natural that the output is at least shared with each participant if not the whole world.

Wikipedia and open source have shown quite clearly that aggregating a small effort from large numbers of participants can solve complex problems that would costs millions of dollars for a single company. What is less obvious is that this creates a huge amount of wealth and increases the productivity of the whole economy.

Linux is a great example of this. 'Sunday afternoon' effort aggregated from thousands of developers can create - for free - a product that might cost tens or hundreds of millions of dollars for a single company to produce. Open source operating systems and databases are directly saving businesses a fortune in license costs. Importantly, it isn’t just individuals that can collaborate to solve common problems; companies can too.

Today’s Web is based on human readable documents. We can connect millions of humans together with text and video, but we cannot write applications that draw on incredible information in the millions of databases around the world. Imagine trying to run Dell’s sophisticated supply chain system with just wiki’s, blogs and Web pages. It would take weeks for people to organise the parts for a single computer. Software greatly amplifies our productivity and allows complex activities to be automated.

To support ever more complex mass collaboration between individuals and companies it is necessary for the Web to move from a Web of human documents to a Web of machine data, the Semantic Web. The Semantic Web is not an end in its self but is inevitable as we seek efficient collaboration.

The need for sophisticated mass collaboration is the reason for the Talis Platform. It is an applications platform for harnessing sophisticated mass collaboration built on Semantic Web technologies but aimed at making it incredible easy for developers or scripters to use.

To receive notification of new issues of the Talis Platform News please remember to SUBSCRIBE.