From Decision XG
What is a Decision Markup Language (DecisionML)?
We're all decision-makers and every day on the web thousands of decisions are being made. Everything from which web page to visit, which search to perform, which article to read, which item to purchase, which trip to take ... all of these are decisions. The proposed DecisionML will provide a choice of formats for representing decisions. What is a decision? A choice among reasonable options, a question to be answered, an assessment of choices based on a set of metrics, and more. There are pros and cons to each option. There are stages or states of the decision process, from information gathering to assessment to decision-making to communicating the decision to others. A Decision standard would include these components.
Why is DecisionML needed?
Our decisions, made every day on the web, represent a valuable resource which should be documented, managed, and shared appropriately. Currently, most of this information is lost or locked into proprietary formats by individual companies. Even the decision-maker, the user, does not have ready access to their own decisions. We cannot easily share our decisions across systems and tools. To effectively document, archive, store, search, share and manage decisions, a standardized format is needed.
Who would use this standard?
The standard would be used by everyone. Commercial vendors, social networks, individual users, libraries, historical sites, hospitals, emergency managers, all would benefit from capturing and managing decisions in a standard format.
What is the mission of the W3C?
The W3C mission is "to lead the World Wide Web to its full potential" by developing standards that "ensure the long-term growth of the Web". This includes seeing and enabling a "Web of Data and Services". For example, RDF and OWL enable expressive and flexible representation and lightweight merging to create this web of data. GRDDL supports converting traditional XML and HTML data into these more expressive formats. SPARQL enables querying across the web of data, while the linked open data initiative utilizes these standards to enable interconnecting and exploring the web of data. What's next?
How does the proposed DecisionML promote the W3C mission?
The proposed DecisionML format would take this work to the next level, by allowing the linked web of data to be used not just for awareness, but for assessment for a purpose, a decision, and to allow that higher level of usage to become itself part of the web of linked data. In short, today decisions are being made using the web, but those decisions aren't "on" the web. This Decision standard would both literally and figuratively "expand" the web by integrating decisions into the fabric of the web. In the future, decisions would appear on the web, decisions would include assessments automatically generated from the web, and collaborative decisions would be be managed and shared across the web. For these reasons, this work directly supports the W3C mission.
What is the status of DecisionML?
The term "DecisionML" is a proposed name for a standard format for representing decisions. The usage scenarios, component requirements, and potential formats have been explored in the Decisions and Decision-Making Incubator in 2010 and documented currently in the Draft Final Report. The recommendation of the members and participants of the incubator is to transition the current work to a new Decisions and Decision-Making Working Group for the development of a DecisionML W3C Recommendation.
What are some of the key usage scenarios demonstrating the importance of this DecisionML?
The key DecisionML usage scenarios are documented as part of the Decisions Incubator Final Report and can be reviewed here. They include
- Selecting Among Alternatives
- Capturing Expertise
- Collaborative Decisions
- Historical Decisions
- Emergency Management
What are the Core Elements of DecisionML?
The Core Components of DecisionML are documented in the Draft Final Report. The components include a question, answer, options, metrics, metric weights, assessment, transition states, context/relationship, and events (who, what, where, when, how, why).
What will DecisionML do for me?
DecisionML will allow you to capture and share in the digital social web your choices, your options, your reasons and your digital life in terms of the choices you make and your understanding of those choices. If you represent an organization, DecisionML will allow you to more effectively document, understand, organize and chart your organizational knowledge.
Why are there multiple formats for DecisionML?
To support a wide range of users and to encourage wide adoption, multiple formats are recommended for DecisionML. A simple traditional XML format is recommended to support the wide range of users who utilize XML tools and capabilities. This format is particularly useful for gathering the basic data of decisions. An RDF format is recommended, and a GRDDL conversion capability, to allow decisions to participate in and leverage the linked open datasets and SPARQL querying. The Web Ontology Language (OWL) format is recommended to take advantage of the mathematically-sound inferencing and support from a decision ontology and from decision subject matter domain ontologies. With a variety of aligned formats, the DecisionML will enable both ease-of-use through simplicity for basic data collection but also expressive machine inferencing through supporting ontologies for advanced applications.
Would there be any intellectual property restrictions for using a DecisionML?
Where can I get more information on DecisionML?
A nice overview of the Decision Incubator purpose is available by reviewing the Charter and the position paper: "A Decision-Making Format for the Semantic Web" E. Blomqvist, M. Ceruti, J. Waters, and D. McGarry (WOP 2010). More detail is available in the Draft Final Report of the Decisions Incubator.
How can I get involved in the development of a DecisionML?
You are welcome to join. Please send an email to the three proposed co-chairs: Jeff Waters (Jeff.Waters@navy.mil), Don McGarry (email@example.com), and Eva Blomqvist (firstname.lastname@example.org) and we can discuss the various options for participation.