Topsy: a Twitter search engine that’s a good idea in theory only

Topsy is a new Twitter search engine/ meme tracker that offers Twitter search results ranked by the number of times the result has been retweeted.

As an idea, it’s a fairly good one. Using tweets as a measure of influence in our now Twitter obsessed society makes perfect sense, and in many regards this is a logical progression from a traditional meme tracker that in theory ranks influence by incoming links.

However it’s a good idea in theory only. Louis Gray rightly points out that Topsy can never give the best result because of the undue influence of Twitter (the company) over follower counts. Sites such as TechCrunch and Mashable have over 500,000 followers because they are recommended on Twitter’s “Suggested Users List.” That isn’t to blame either site for their follower count, but it does call into question Topsy’s results, because those sites with the most followers will always have the most retweets, and that is a reflection on volume, not quality.

There’s also one other major flaw with Topsy, and that’s its ability to resolve short URL’s. Twitter relies nearly exclusively on short URL’s due to the character limit, with each short URL then resolving to an actual page URL. You can draw resolving data from major services such as Bit.ly, and TinyURL, but not all others. The result is that any short URL service Topsy can’t resolve isn’t counted in a total for the end link. Yes: use of Bit.ly and TinyURL would make up the majority of short URL’s used on Twitter, but more and more people are using different services, or even their own. For example, we run our own Inquisitr.com short URL service for our links (it’s as easy as a WordPress plugin.)

The result looks like this

results

I should note that the first result links to the UK Newspaper The Sun.

I won’t pretend to fully understand the technical difficulties involved in correctly resolving short URL services, but I’d guess it’s not something that is necessarily easy. The difficulty aside though, the inability to correctly group links to actual end results means that any pretense that Topsy is truly tracking the most popular retweets is deeply flawed.

Perhaps without some irony, Topsy takes its name from the phrase Topsy Turvy, which by one dictionary definition means “In or into a state of utter disorder or confusion.” I don’t doubt the idea, only the implementation. Given time, it could become better, but in the mean time, if you want to read the latest stories from TechCrunch and Mashable, I’d suggest that visiting them directly would be quicker than using Topsy.

Comments