CTO Blog

Well to start with you can actually now find places to play the search engines off against each other! To see the five most popular Search Engines in one place, (Google, Lycos, AltaVista, HotBot, AskJeeves), a visit to Freeality is an interesting way to spend some time.

However I guess if success is measured by columns of coverage then the battle has already been won by Google. It would probably still be won by Google if the number of online searches made is counted, so why am I not totally sure that Google is only answer? Well, for a start it doesn’t measure successful searches; in fact the number of searches made would be lower if each search produced a ‘right first time’ result. I suspect that many people are like me, and have to keep repeating searches with changes in the key word combinations to try to get what I want. Then if you get what you want, a perfect hit, how do you find it again? All the search engines use methods that lack deterministic results and almost triumphantly announce ‘different every time’ as if it’s a virtue. Sometimes it is, for dynamically changing information it makes sure that you get the most up to date result, but many times it’s not, so why can we not tag the search to ensure that it can return to the selected sites?

I have now found you can do this, its early days but it works, try del.icio.us to get the idea, not only can you use the tags to save in a structure manner around topics, but you can allow others to save tags into your topics too, or maybe you can access their topics. It’s different, and an interesting approach, its effectively adding the intelligence of people to do the sorting out of the mass returns of a keyword based search. This may be why I am increasingly drawn to using Wikipedia, it is effectively a cleaned up, and structured assembly of information with indexing and a search engine.

It's becoming clear that as I become more aware of how I want to use information, as opposed to perhaps content, (the original role of the web being to show content in a uniform manner), there two main strands to what I want; Searching/Locating – I basically know the answer in outline, but more information relating to this answer is required; Browsing/Discovery – the overall context, or relationship, is known, but the answer is not known so help is required to find the answer that feels right. What is interesting is that neither are particularly the role of a conventional web search engine, i.e.; keyword driven, perhaps that’s why Google is spending so much time and money with so many talented people to move Google to the next generation.

What are all those people doing?? They are looking for the answer!!