[volaski]

I think the difference is that the times potentially has the ability to execute. Semantic web was a cool idea but failed to catch on even with all the grassroots efforts. NYT already has tons of content with tons of metadata and they have the readership. Maybe they can pull it off.

[sparkzilla]

I don't think they can put it into production easily. They would have to rework their entire site, their entire workflow, and the way their journalists work. What they are describing also doesn't work well on existing articles, because those articles, which they have some meta data, are not in a format that works well as a database. For example, each article has a lot of repetition with earlier articles. They say that it would require a lot of manpower. We solve this issue by crowdsourcing the events and their meta data.

[ftwynn]

The article is published by the same group that's building tools to help with this exact problem. [1]

[1] http://nytlabs.com/projects/editor.html