Everybody loves lists. Whether it’s a grocery list, the New York Times Best Seller Lists, or, my favorite, the RSS list, lists are cherished and used by many. Everybody loves popularity contests, too. Imagine a world with no Ubuntu popcon, no Ugly Dog contest, or no Superhero Hall of Fame. Pretty boring.
Recently the team in our Laboratory of Applied API Sciences donned our lab coats again and started experimenting with combining a popularity contest with a list. After several accidental explosions, we ended up with a new API, the Most Popular API.
The first release consists of three categories of popularity measurement: the most e-mailed articles and blog posts, the most viewed articles and blog posts, and the most shared articles and blog posts. The first two are pretty self-explanatory. The third, most shared, is an aggregate of Times items that were posted to Facebook, Twitter and a few other notable sites. For each type of list, you can get aggregate results for all of NYTimes.com or filter results by section. Along with standard article data (URL, title, abstract, section, byline), we are also returning all the Article Search API facets (keywords).
One of our primary objectives with the Most Popular API was to enable ways to interact with other Times APIs through standard “join fields.” Fields like url, title, section, published_date, source, des_facet, per_facet, org_facet,
and geo_facet
are included in responses for this purpose. For example, using the url
field, you could make a mashup of the Most Popular API and the Movie Reviews API.
And by using the des_facet, per_facet, org_facet,
and geo_facet
fields, you could easily mash up the Most Popular API and the Article Search API. As we build more and more APIs, fields
like these will become even more valuable for interactivity and interoperability.
Doing mashups with APIs built elsewhere is a bit more challenging, but that is where all the fun is. The idea behind the Most Popular API is that it is an entry point API, rather than an end point API. Using this API as the foundation of an app offers limitless mashup possibilities: How about an app that presents the Most Popular lists with data mashed in from Digg or Reddit? Or how about an app that presents the lists as a Google gadget for your iGoogle page? Those are just a couple of ideas.
We have a few more enhancements planned for this API, so be sure to come back and check for updates. Have fun!
Comments are no longer being accepted.