The IKS (Interactive Knowledge Stack) Open Source Project, a European Commission-funded effort to bring semantic technologies to vendors' content management systems, is moving into its next stage this week. Helsinki will be the site of its Semantic Editor Hackathon, an event for testing some of the basic technologies the group has been working on for building a semantically-capable text editor that could be dropped in in place of common tools like TinyMCE and FCKEditor. The forum also is an opportunity for generating new ideas about how different tools can enter the picture at the editing phase for faster and smarter indexing and tagging of content.
Imagine a more deeply semantic experience of HTML 5 in-browser editing, as IKS Scientific Lead Wernher Behrendt of the Salzburg Research Forschungsgesellschaft Knowledge and Media Technologies Group describes it, as one where the system actually will understand that you are editing content that, for example, refers to a person. And, "just clicking on that person's name gets you into an editing mode where a vCard comes up, and if it's about a project you get a little semantic frame popping up that it knows the name of the project and who the coordinator is and when it was running," he says. "Imagine these relationships and objects being available at your fingertips through such editors, and then imagine the information that you enter is put in in a semantically reasonable fashion so that it could be used again."
Improving semantic annotation through such a rich text editor can lead to a substantial productivity boost, says John Pereira of the Salzburg Research Forschungsgesellschaft Information Society Research. This next phase builds off the work the IKS Project has undertaken so far, including the creation of the Furtwangen IKS Semantic Engine (FISE) during a previous hackathon in March. FISE, according to IKS, is an open source HTTP service meant to help content management system developers semi-automatically enhance unstructured content (text, images, and so on) with semantic annotations to be able to link documents with related entities and topics.
The total IKS project, which began last year, runs through 2012, and will encompass creating a new vision of what semantically-enhanced CMS technology stacks can look like in the future. "If we are able to improve the technology of CMS providers, they are able downstream to deliver better systems to hundreds of thousands of customers, and so millions of users get better productivity," says Behrendt. Among the CMS industry partners who've joined the seven research organizations involved in the IKS effort are some names familiar on this side of the ocean, as well as small-to-medium size technology providers and solutions developers that are lesser known here. They include Nuxeo, Alkacon Software, TXT Polymedia, Pisano, Day Software and Nemein.
It's been a balance between reaching for the stars and accommodating what industrial partners see as more practical issues to resolve. "We said if we had this stack idea where you start from an intelligent user interface, go through some knowledge ontology-based model of whatever task your user is doing, then you will be able to create content that is very well semantically described, and you can store and retrieve content that has very specific semantic annotations associated with it," says Behrendt. "What we see as a practical development is that some of the industrial partners said we have a pretty reasonable content management system that does a lot of the work our customers are asking for but is not so good at some things." Hence the focus on addressing some of the lower-hanging fruit concerns, which things like the FISE engine can accommodate.
"At the same time we're trying to still follow the advanced ideas where you can then actually visualize a semantically annotated object," he says. "The idea is to bring content objects up to the knowledge level, to interact with knowledge objects in your GUI."
Currently the IKS Project is running its latest survey for CMS vendors to evaluate their product's general coverage of features, in particular with respect to semantic capabilities. That will be published as a book, Pereira says, to provide a snapshot of where the industry stands in 2010. The survey is accessible here.
• Don't forget to propose your startup for our Semantic Web Impact Awards. The deadline is Sept. 15.