Web3 and Concepts for SharePoint 1.0 Releases Now Available
We are very pleased to announce the immediate availability of the 1.0 release of our Web3 Platform and Concepts for SharePoint products.
The Web3 Platform is our "next-generation" structured information server built on the foundation of topic maps but with support for Linked Data standards including RDF and the SPARQL query language. The Web3 Platform provides a fully-featured REST API making it possible to create applications in almost any programming language and to quickly and easily integrate structured information into existing applications and websites. The Web3 service provides full support for the topic maps data model and support for browsing content as Linked Data RDF as well as providing a SPARQL endpoint that implements support for querying topic map data as RDF as well as querying at the detailed topic map data model level.
Concepts for SharePoint is simply the best way to organise, search for and find content on a SharePoint 2010 server. It integrates with and extends the taxonomy features of SharePoint 2010, allowing users to create richly interconnected data models that reflect the business in which they work. Unlike the tags in a standard SharePoint 2010 taxonomy, the concepts in Concepts for SharePoint can be made available as published pages complete with links to other concepts and to related content across the whole SharePoint server.
Both products are available for evaluation. Download them now!
Using HTTP Verbs to describe Server Side Transactions
One of the challenges of putting together a REST service is how to let clients make transactional updates that involve two or more resources. One approach to this problem is to create a resource that is 'a collection of transactions'. A client can post a representation of a new transaction to this resource. The transaction becomes a 'child' of the 'transactions' resource and can be accessed via a returned URL.
Making the transaction explicit is good as it allows the transaction to be addressed and its status checked, it also allows for the service to return immediately and process the transaction in the background.
This basic pattern is good but it has a drawback in that for each different system you build you need to define the transaction language.
Our solution to this is to use HTTP verbs, URIs and resource representations as the building blocks of our transaction language. A transaction representation contains a number of operations. Each operation contains information about its Method, e.g. POST, PUT, DELETE, the resource URI to operate on and the request body.
<transaction>
<operation id="a1" method="POST" Resource="http://...../somecollection">
<body><!-- resource representation --></body>
</operation>
.. more operations
</transaction>
The server processes a transaction by looking at each operation construct and operating on it as though the request has come from a client. The only difference is that all of the operations listed occur in a single transaction.
Each operation has a local id. When an operation completes the return URL from the operation is bound to that identifier. The identifier can then be used later in the transaction. This is especially useful when you want to create two new resources and then associate them together all in the same transaction.
After a transaction is processed the representation can be retrieved. The representation contains the URLs of newly created resources, as well as an errors information that occurred when processing the transaction.
This is a very powerful model that removes the need to create a domain specific transaction language for different applications, can reuse existing resource processing code and is very descriptive.
For more information about this approach, download and review the REST service documentation for the Web3 Platform
Web3 Platform Beta Now Available!
The Networked Planet Web3 Platform is a complete solution for creating, organising and publishing structured semantic data. The Web3 platform stores and manages data in a schema less data store, allowing complete flexibility of the shape of the data stored.
This is the first public beta of the NetworkedPlanet Web3 Platform. This beta version is licensed for use until 31st July 2010. We are planning to refresh this Beta release on a regular basis up until the final 1.0 release is available. To keep track of update announcements please subscribe to the RSS feed of this blog.
Subj3ct.com Beta 2 Released
We have just refreshed Subj3ct.com with Beta 2. We have tried to incorporate the feedback we have had as well as adding some things we think are useful. The highlights are:
* Support for OpenID for signing in
* Ability to create and publish identifiers and indicator pages from with the Subj3ct portal
* A changes feed that can be used to aggregate whats happening in Subj3ct
* Embeddable snippets (RDFa, SKOS etc) on each subject record page that can be used to tag a page with a given subject.
* The start of a user community that currently supports user search
* Ability to process SKOS feeds
* Ability to register feeds without creating an account (this allows users to reference existing SKOS vocabularies that they don't own)
* Updated statistics on the home page and quick links to most recent subjects and some random subjects.
* Bug fixes and performance improvements
We now have 3180 identifiers in 23 feeds and we and others are adding to this all the time.
More stuff to come...
A small useful ontology and what it tells us about Web3
Implementing Topic Maps solutions for a variety of people we see many ontologies (information models, domain models, schemas) being used. Out of this we have tried to understand how different approaches are useful for different kinds of customers.
One approach, where the customer is a focused specialist group or small company/department who 'know their stuff' really benefit from having a rich ontology as it says more about what they are doing. These are also the people who want to encode knowledge as something explicit and then use that knowledge as part of the information displayed to users of the system.
At the other end of the scale large organisations tend intially to benefit from a less is more approach. This is typically becuase getting 'political' agreement on what the business does and where the focus should be is much harder at the large scale.
It is this second example I want to pick up on here. The most basic, useful ontology that we see in use consists of just three types; Content, Concept and Person. Concepts can be connected to other concepts as 'broader/narrower' and also 'as related'. Content 'is about' concepts, and 'related to' person. Person can be connected to concepts in a few ways that are often tweaked for the domain. e.g. expert in.
This ontology is not the ultimate base model for explaining everything, but more about being able to usefully cover enough about a domain to promote the things that are important in an information management system. (Things such as faceted search, concept centric, bi-directional linking, multiple names and identifiers. All the findability niceness of Topic Maps.)
This model works well because it has recognised that people and content are two fundamental pillars of the way we work with information. Making the users/ people part of the model is critical to be able to deliver targeted contextualised content. This is the kind of content and knowledge people want to see and have access to.
So now we are moving Topic Maps concepts and approaches from the enterprise to the web and in particular the web3 context how can these ontology lessons inform us about how web3 works and the impact it will have?
Well, the biggest observation is that while web1.0 provides content, web2.0 provides people and content there is an obvious gap and therefor need for the third pillar which is concepts. Having concepts on the web will enable people to group and organise the content around the subjects that are important to them.
Having concepts on the web will provide binding points around which content can be produced and grouped, it will facilitate social networking groups to be organised around concepts. Overall having concepts on the web will provide a new and vast dimension to the experience of using this content management system we call the web.
subj3ct.com Released
Today we are proud to have launched Subj3ct.
Subj3ct is a Subject Identity Resolution Service, it provides both a portal for humans to use and an API for developers and applications. We store subject identifiers, equivalent identifiers and also links to web resources that contain structured or unstructured information about a subject.
The resolution service is intended to provide services to:
a) people looking to re-use common identifiers in order to improve interchange
b) application developers who are building the next generation of web applications, where dynamically finding and merging linked data is key
We are starting out in gathering mode, trying to get collections of identifiers registered so that we have a useful set of subjects.
Once we have this we will be creating new features and writing more on how to use the service in building Web 3.0 applications.
In the meantime, we would really appreciate it if people can register subject feeds and just generally provide feedback on the portal and API.
Everything can be found at https://subj3ct.com
Finally, thanks for all the support and encouragement we have received in putting this exciting service together.
The Subj3ct Team
White Paper: Introduction and Vision for Web3
Linked here is a white paper that provides a vision and introduction to web3.
The summary below provides an introduction...
In the past 20 years, the Web has developed from a niche technology to a mass-media providing new forms of communication and interaction between people. Web 1.0 was a technical platform - a common set of protocols and formats that allowed machines to communicate and present information from a remote server to a local user. Web 2.0 used the technical platform of Web 1.0 to build more interactive web sites where users contribute and share content and become creators and owners of content rather than passive consumers.
Web 2.0 has reached the limits of what can be achieved on the technical platform of Web 1.0; new technologies must be put in place to provide a fundamentally new technical infrastructure, or platform, to enable the next generation of innovative web applications. Key to this Web 3.0 platform is a set of protocols and formats that allow the communication of subjects and people's perceptions of those subjects between computers, and that enable new applications to be built that allow users to create, share and integrate information and knowledge seamlessly...
A Vision for a Topic Maps World
Yesterday at the Topic Maps 2009 conference in Oslo I gave a presentation called 'A Vision for a Topic Maps World'. In this presentation I summarised some of the key successes of Topic Maps and then looked forward to how Topic Maps can work on a more global scale.
One of the key things I presented was a concept for a Subject Identity Resolution service, a kind of DNS for subjects and representations of statements about subjects.
The full presentation can be found here.
We'll be blogging more about the Subject Identity Resolution Service over the coming weeks.
Welcome to the Networked Planet Web3 Blog
Welcome to the Networked Planet web3 blog. This blog is dedicated to all things web3. For us at Networked Planet web3 is all about the publishing online of structured semantic data in a way that can be discovered and usefully processed by both humans and machines. This is not the semantic web, but a web of joined up structured information.
Semantic Technologies such as RDF and Topic Maps are great standards for the representation of statements about a subject, what we really are looking forward to are protocols and conventions that allow fragments of information to be published and connected.
As Networked Planet we have been doing a lot with the ISO13250 Topic Maps standard. This standard and our products have been used in many applications to improve the way in which information is organised and accessed. Our next goal is to take many of the ideas and concepts that have been so successful within enterprises into a more global space.
Over the coming months we will be releasing services, presentations, white papers and products that support the goal of a Web of linked data.
