A big thanks to Ran at Omgili.com for his recent blog post about turning Google into a near real-time search engine. You can do it too! Here’s how it works.
If you’ve been reading ResearchBuzz for a while you know about hacking URLs — changing parameters so that the pages display different kinds of results. Ran hacked a Google search result URL and changed the search results so that they were being returned only for items that had been indexed in a very small period of time — like a few seconds!
Here is a basic Google search for the word fred.
Ran hacked the tbs parameter, which possibly stands for Time Being Searched? The tbs parameter, added on to the end of a query, can look like this:
&tbs=qdr:d (searches a day’s worth of indexed content)
&tbs=qdr:h (searches an hour’s worth of indexed content)
&tbs=qdr:n (searches a minute’s worth of indexed content)
&tbs=qdr:s (searches a second’s worth of indexed content)
You just add the parameter to the end of your query. So if you were searching for content containing the word fred that was indexed in the last minute, the query would look like this:
You can add a number to the end of the parameter to indicate a span. For example, &tbs=qdr:h3 at the end of a Google query will find content indexed within the last three hours.
Note that this parameter seems to work only in Web search — I didn’t get any results when I tried to use it in news, images, etc.
So what do you DO with this? You can find content that Google has indexed really really really recently. :-> You can also use these switches to dig around in Twitter. For example, say I want to search Twitter for Tweet accounts that describe themselves as the “official” blah-de-blah. I can’t do that with Twitter’s search page because I can’t search in just the biography text. But you can with Google using the following query:
site:twitter.com “Bio * official”
By adding &tbs=qdr:h3 to the end of a search URL for the above query, I could search Google for recently-indexed Twitter accounts/updates in a way that Twitter’s search page doesn’t allow me to. Now, will this get you everything? My assumption is no; I go by the rule of thumb that an external search engine applied to a source will never get everything that that source’s internal search engine will. But I got enough that I wanted to look around some more.
Or you can use Twitter as a Deliciousesque search using Google! The following search will look just in Twitter users’ favorited lists for the word “tool”:
site:twitter.com inurl:favorites tool
I have been trying to use Twitter as an information trapping resource to find links that people announce in tweets, etc, but with the explosion of spam in Twitter it’s gotten just about impossible. But maybe by using recently-indexed Google content and Google’s syntax to severely restrict WHERE I search, maybe Twitter can become, in the information trapping sense, useful again…