Standards-based journalism in a semantic economy. Forget widgets and Twitter apps, this is the opportunity. Required reading.
WordPress, better cities, and everything else
Standards-based journalism in a semantic economy. Forget widgets and Twitter apps, this is the opportunity. Required reading.
4 Comments
Thank you very much. If you think these ideas are on to something, then I feel much more confident about them. For the record (and I tell this to everyone when the subject turns to RDF), my thinking on EVERYTHING turned when you suggested that I solve some of the problems I proposed via RDF. You did that in a comment on a post, so I also use that as my example when I talk about the value of comments.
So thank you, again.
Awesome, wonderful to hear. I agree that RDF is likely the best way to expose any semantic data on a HTML page. It’s cleaner than microformats and offers greater flexibility. Unfortunately, it’s not the best way to store structured data within a database because of the computational power required to parse it (especially if you’re parsing on every page load). Not to worry, databases are optimized for storing relational data. In my opinion, the killer application will be the editor that makes it simple and straightforward for the average journalist to compose a story in a structured manner. Camayak is working on this.
Something like ANT is your friend here- think about version control and generating a page only when there’s a change in the source data. This idea circulated early in XML history (think 1999-2001) but fell by the wayside when XSLT processing got much faster. My experiences with RDF suggest that going back to a structured build process is the way to go for retrieving documents- it’s not as clean, but storage has gotten pretty cheap in the past decade.
Oh, and btw: if you wanted to build a semantic directory out of your newsroom workflow and you didn’t want to do it with a triplestore, the easiest approximation would be an annotated directory of anchor URLs from your WinerLinks. It wouldn’t have the scalability and the other powers of RDF, but you could do some cool things with it.