The future of search…apparently

A little while back, a quality Sunday paper in the UK had a front page article about the next big thing in search engines. The product, Wolfram Alpha, developed by the mathematician from whom it gets its name, was touted as possibly being biggest thing to happen to search since Google. Its differentiator is that it will “provide a single source that can be relied on by everyone for definitive answers to factual queries”. Of course, you can type such ‘factual queries’ into current search engines and get the answer, but this may be in the midst of many less relevant results through which you have to trawl to get the most definitive answer.

 

I started to consider how this could be applied to our industry to good effect and this raised a number of questions. How would the revenue model work for the vendor? How would the answers be populated? What types of questions would job seekers ask? This was just the tip of the iceberg. Then I thought about the way that people search for information in general and came to the conclusion that this type of search engine may not be the way forward for recruiting. One of the net results of search engines is that users have become accustomed to shorthand methods of entering queries. Considering the transition to mobile devices for information retrieval, this is likely to become even more the case as input devices are more limited and time is of the essence. Hence, you are probably more likely to see people search using SMS style shorthand as opposed to fully constructed sentences. However, Wolfram doesn’t (currently) work well with abbreviations or incorrect spelling.

 

There are also technical and infrastructure elements to take into account. The computational power required to decimate more richly structured queries and provide fully structured answers is immense. Given the volume of searches that are performed daily on the major search engines, the hardware requirement alone would be prohibitive to most companies except, ironically, one like Google. You may remember when some ex-Google employees launched their own ‘Google killing’ search engine, ‘Cuil’ last year. As soon as it started being used in a real-world environment, it crashed and burned due to insufficient hardware resource. Now, it has lost the initial interest it created due to inability to deliver against its own hype when launched. Usage stats suggest that Google will have little to worry about.

 

Then, there is the need for population and maintenance of answers. It was this that made me think that perhaps there is an essential marriage of Wolfram and recruitment communications after all. They will need to find thousands of employees to keep their answer repository extensive and current. Allow me to offer our services!

TMP Worldwide
Written by TMP Worldwide

1 Comment

Leave a Reply

Your email address will not be published. Required fields are marked *