Keen On…WordLift: a short story, how we plan to make WordPress a Semantic CMS

Hi Apache Stanbolers, It’s almost a year since we joined this wonderful community, yet it feels only yesterday when we started to get in contact with your ideas and great enthusiasm. That was Paris when we understood we were on the right track and with the right people. (have a look at Andy’s post to read more about it).

We, as InSideOut10, are willing to put our best efforts and to confront our ideas with the Web on a larger scale and that is why we targeted WordPress in the first place and developed WordLift 1.x as a front-end for the Apache Stanbol server enabling Schema.org support.We were extremely happy with this first dive in the open-source World and with our first release of the plug-in, showcased in Paris, and released to the public early in August. We’re now looking forward to address both back-end and front-end challenges and to increase user engagement on content-intensive websites and roughly these are our targets for WordLift 2.0.

1. building a front-end navigation pattern that allows the user to explore and discover contents via Semantic Entities thus enabling Semantic/Faceted Search and above all a better user experience;
2. improving the entities discovery rule of the refactor engine (adding places and events to the already existing rule that now discovers people and organizations);
3. enhancing the user-experience on the editor side by shifting processes to a background asynchronous task, hence providing a solid usage pattern to all of the suggestions we registered while testing version 1.x (from Enel and from the WordPress community) – this might take advantage of the recent STANBOL-343 and STANBOL-383 APIs for long term operations;
4. becoming polyglot with the IKS language support (needless to say starting from Italian, Danish and Spanish languages where our market is mostly focused nowadays);
5. creating a responsive theme to implement the navigation patterns ([1]) on laptops, tablet and smart-phones.

On the road towards WordLift 2.0, we recently took part in the CreateJS Hackathon led by Henri Bergius at LIIP premises in Zürich, Switzerland just few a weeks ago.

There we focused on the priority to shift the analysis tasks from synchronous to asynchronous by taking advantage of the Long-Term Operations Jobs recently released by Enrico Daga (CNR) and with the support of Alberto Musetti we integrated them with the Enhancer JobManager and developed the alpha support in WordPress.

And when Rupert Westenthaler joined the conversation this got even more fruitful with two fundamental contributions, the first being a major upgrade of the Enhancement Chains with support of different chain configurations that can be activated on behalf of the user requirements and the opportunity of running concurrent engines thus improving the overall enhancement execution. The second being a thorough explanation of the DBpedia indexing and caching features that we recently configured with Rupert’s support on our Apache Stanbol instance and that improved our response times ten-fold.

So now we have better performance, less waiting time for analysis tasks, initial support for asynchronous jobs in WordPress, and an excellent basis for the outstanding features that we will bring to the community soon.

We will keep you posted here and on the mailing-lists on our progress, and be sure not to miss our new WordLift Web Site.