Last week Twitter announced a change to their landing page. They now have a search tool and topic trend tracking widget that they hope will, “encourag[e] a sense of wonder and discovery,” in new or potential users.

The update by Twitter has proved rather convenient for me, coming just days after my last post posed the question, “Is human indexing really going to lead to better search results?” While I don’t yet have an answer I have had the time to play about with Twitter’s new more prominent search tool.

Unlike pages indexed by Google or Bing, Twitter results are based on time, that is real-time tweets, rather than relevance. This means you can see what people are saying about a topic right now.

For instance if I search “nokia n97 review” I get the very latest tweets that include these keywords, from as recently as a minute ago. If I search the same keyword phrase on Google.com I get Ubergizmo’s review as a first result, published 9 July 2009, and on Google.co.za I get my review of the device on TechSmart.co.za as a first result, published 22 July 2009.

While Twitter search is very basic at the moment, with the only advanced search options featuring location and time specific limitations, it opens up a new world of search possibilities. Twitter’s real-time results have led me to some really pointless tweets but also to some fantastic sources I would otherwise never have found on Google, or Bing for that matter. Human indexing is consequently a mixed-bag, a bit like search engines really.

But this search for better search now means that a Google, Bing and Twitter search is standard practice at the TechSmart office in an effort to find the very best results. With the use of Browsys’s Twoogle this isn’t a hassle either. With this useful search aggregator you can search a number of engines and networks simultaneously. For the time being this comparison of results will have to do, at least until some clever solution that mashes the results providing the most current and most relevant outputs comes along.