A while back Frederick emailed to ask if I would like to configure Revyu.com to Ping The Semantic Web. Naturally I said yes, and he blogged about it here. Well, the job is done.

The biggest issue for me was deciding whether to send pings at review-time, or whether to run a batch job every x minutes. In the end I decided that adding any ping time to the review process was gonna annoy users (who probably don’t care too much about what’s going on behind the scenes), and went for the batch job approach. So, once every x minutes, Revyu now pings PTSW with the details of all new reviews.

At the moment, pings are only sent with the URIs of new Reviews (and PTSW handles the HTTP303 redirect seamlessly – nice). The updated FOAF information about the reviewer and the updated Tags associated with the resource being reviewed aren’t currently included pings. This will be added in the next version.

So, in the meantime Frederick, how about a new heading on PTSW for Recently updated Review documents?

Login to Revyu.com, write at least one review that has some tags, and get your own personal tag cloud on the Tags page. Obviously in true tag-cloud style, the more times you use a tag, the bigger it grows in your tag cloud. Hooray. All credit to domenico79 for the suggestion.

I’ll add the same functionality to the People page in due course, and maybe even create a People Cloud showing all of you folks who are creating reviews, rather than just a plain old list.

One reason for developing Revyu.com was to help unlock reviews from proprietary backend databases, and make them available to the world in a structured, machine-readable form. So far there are at least 2 ways of doing this: by requesting and parsing the RDF/XML version of reviews (that looks like this), or by querying the SPARQL endpoint. The SPARQL endpoint really is the jewel in the crown of reusing data from Revyu.com on other sites, and drewp is the first to do this. Nice one Drew.

On his homepage at bigasterisk.com, he now displays his 10 most recent reviews from Revyu.com, retrieved from the SPARQL endpoint using a chunk of Python code. We’ll have a look at getting the Python code packaged up and released into the wild asap. In the meantime there are other mechanism about to be released that will allow you to show your reviews on your site, without having to do more than copy and paste. I’ll talk more about that in Part II of this series.

For the record, if you’re interested in the query drewp is using it looks something like this:

PREFIX rev:
PREFIX rdfs:
SELECT ?thing ?name ?review ?createdOn
WHERE
{
?thing rdfs:label ?name .
?thing rev:hasReview ?review .
?review rev:reviewer .
?review rev:createdOn ?createdOn
}
ORDER BY DESC(?createdOn)
LIMIT 10

And if you want to get some background on SPARQL, have a look at the resources listed under What is SPARQL? on the Semantic Web FAQ.