Chacha - Social search Engine
December 29th, 2006 by Minas Gjoka
The Internet has become a great source of information and knowledge nowadays. Traditional printed content such as newspapers, magazines, books, encyclopedias, dictionaries are now available online. People may still prefer to read the printed version but there are features which can only be provided on the digital content. Some examples include searching capabilities and interactiveness.
The problem is no longer about generating data but about locating specific data. Google’s Pagerank algorithm in 1998 was a good step towards indexing and ranking web sites in a more meaningful way. It comes as no surprise to me hearing the news this week that Google displaced Yahoo as the 2nd most popular web site worldwide. Microsoft leads the list. It is interesting to notice here that Google’s searching business part accounts for more than 80% (Youtube is probably not included yet since it was recently purchased) of the generated traffic. On the other hand Yahoo’s traffic comes largely from other services provided on it’s portal-like website. Google has been so successful that the word google is now identified with searching (the word googling is included in the online version of Merriam-Webster). But still something is missing. Machine learning and other sophisticated algorithms used in Google and other similar search engines, such as Yahoo and Windows Live!, are unable to provide satisfactory searching results sometimes. Irrelevant to the searched keywords websites are not uncommon. Persistence is necessary to retrieve better searching results. Sometimes you may have to rephrase the keyword to get what you want or use special search features (operators, specific filetype or site etc). But not all people like fiddling around with computers. Getting the work done is more important.
The other day I fell upon a new search engine called Chacha. It’s not one of those meta search engines which combines results from all different popular search engines. It’s a social search engine. A human guide receives the searching keyword submitted and using his/her experience filters the results obtained from traditional search engines and provides more accurate results. Reading further in Wikipedia and other websites I found that the people providing this service are getting payed starting from $5/hour. Everyone can apply but they have to pass through a training procedure and qualify to be able to start getting paid. Each guide declares up to three categories of interest to which he/she will provide service to users. The information produced each time is stored somewhere to be used for non-guided searches from other users. It’s interesting to see whether this type of search engine will hold.