I’ve been spending the afternoon working on my Temporal Topic Explorer, and wow, I think I’m going to be using this a *lot*. I believe in sharing my toys, so if you want to try it too it’s at https://searchtweaks.com/tte/ .

TTE takes advantage of dates in URL patterns. Many blogs and news sites publish with the date in the URL, like this:

/2024/04/12/example.html

If you use inurl: to search for those strings (and do some secondary filtering to make sure you’re getting the exact date strings you need and not false positives) you can search by individual days in the month without relying on a search engine’s date search mechanism. Further, because you’re looking for dates in the URL, you are much more likely to get results from news articles or blogs as opposed to more static pages.

(Another nice bonus: searching by URL strings also helps prevent dynamic content false positives.)

TTE, using the Stract search engine API, uses inurl: syntax to search the month and year of your choice (I’ve found results from 1994 but 2000 onwards is probably your best bet) and then displays the most popular title words for each day in that month. Click on a word to display the articles from that day containing that word. You can also view all the articles in a particular day.

It is wild to browse news like this. You can explore the media coverage around an event and all you have to know is the date. Your topic modifiers are generated by the program and they’re all relevant because popularity. And since you’re looking at a particular keyword on a particular *day*, you tend to get small lists of articles. It’s also giving me hints about how to shape my searches; I was doing test searches for “Wuhan pneumonia” (what covid was called before it was recognized) and discovered that coronavirus was showing up a lot earlier as a viable keyword than covid.