Monthly Archive for July, 2007

So I’m reading an article I find via Digg. The article’s on the New York Times website. While reading, I run across a word I am not familiar with.

Automatically, I want to put this word into Google. Firefox (and I suppose IE) allows users to highlight text, right click, and choose “Search Google for” the selected text. If Wikipedia is the currently selected search engine in the toolbar search box, the right click allows the user to “Search Wikipedia for” and so on.

Of course, the easiest way to highlight a single word is by double-clicking on it. So that’s what I do. I double click on “abstemious.”

Suddenly a new window appears. While it’s loading, I think I must have done something wrong. Then the page loads. It’s a New York Times.com-powered dictionary lookup of “abstemious.” This is like the coolest thing ever. Well, maybe not ever.

Examining the URL of the newly opened window, we see a few things (with some line breaks to reduce ugliness).

http://query.nytimes.com/search/query?
srchst=ref&
query=more%20valuable%20to%20abstemious%20Martins%20than%20to&
fw=3

The query string includes, from the article, three words before the highlighted term and three words after the highlighted term. The field fw (focus word?) at the end of the query string indicates which word to pass to the dictionary, using a zero-based array. (I tried double-clicking other words in the article and changing the value of fw to test this.) I can only imagine that including surrounding words might provide context to generate an even more accurate definition. Why the target word is variable, however, eludes me.

Oh, and srchhst=ref sounds like search history = reference. Whatever that means.

Please, test this feature out for yourself. Head over to the New York Times site, click on any story, then double click on any word. (The feature doesn’t appear to work from the front page.) Awesome!