« July 2007 | Main | September 2007 »
This is an argument that's been going on among search bloggers - what's more important, the Results Visualization UI or the core Algorithm? Which one should a small search engine startup focus on, to help capture market share away from the mainstream search engines?
The latest Software Abstractions survey (in the left sidebar) asks this question. Do you have a strong opinion? Vote now and let us know!
Thank you to everyone who participated in the last Software Abstractions survey! We asked: which features do you see as the most important ones for Web Search in the future? The results were interesting.
Out of a total of 33 votes, the top votes were closely split between a variety of answers.
For search engines with advanced linguistic parsing capabilities, it's reasonable to assume that semantic processing will be applied to both the query and to the indexed content as a whole. If you combine those two answers, then Semantic Processing is the clear winner with 33% of the votes!
The high number of votes for the "Trusted Sources" answer was a surprise - it's clear that a stronger focus on quality of the results in the future (and their being spam-free) weighs heavily on users.
The complete picture of results is given below:
What is the exit strategy for low-traffic Internet Search Engines? This is a question I've been secretly wondering about for the past few months as I study the growing number of companies in this popular category.
This already-crowded space is getting saturated. My friend Charles Knight of Alt Search Engines keeps a running list of the Top 100 engines; the overall number easily exceeds 1000.
Photo Source: Funny Hub
There is no doubt that as the amount of online content explodes, driven by easy low-cost publishing and the popularity of social networks, Search is becoming increasingly important as a strategic solution - both within an Enterprise for tying together all the Web 2.0 tools, and on the Internet, for making relevant content accessible.
On the Internet at large, Search is currently dominated by the 5 top-tier Search Engines: Google, Yahoo!, Live Search, AOL and Ask. There is also a second tier of engines that have captured enough buzz that they are likely to be sustainable for the medium-term: Hakia, Quintura, the yet-to-be-launched Powerset, and others. There are also specific Market segments where niche players are likely to thrive - Shopping, Jobs, Travel, Audio, Video and so on. Apart from these top tier and vertical segment players - what about all the rest?
I fully expected that many of these smaller, innovative search engines would get absorbed by the larger ones for their technology [for example, Microsoft acquired Medstory, and there are ongoing rumors of a simplyHired acquisition by Google]; but recently, a pattern has emerged that suggests a different possible outcome.
Search has always been a critical feature for large content providers; conventional wisdom until now for these sites was to implement this feature in one of two ways: either (i) using a site search widget from one of the mainstream search engines (as this blog does), or (ii) by creating a custom search engine based on Google, Yahoo!, Rollyo, Eurekster or others.
Increasingly however, large content providers want to harness captive search engines to improve the user experience. Here are some of the indicators of this trend:
It's not difficult to envision a future where every major provider of content implements a powerful search capability optimized for their particular set of content. It will be interesting to watch how the major search engines leverage these capabilities to enhance findability and the user experience. For example, should they continue to directly index the actual site content, or is it more effective to delegate search tasks for each of these sites to their particular search engines, to enhance relevance of search results? More important - does this trend somehow lead us back towards walled gardens?
----
Alternative vision
Charles Knight of the Alt Search Engines blog (mentioned above) has been working tirelessly to promote an alternative vision: to band together a bunch of alts to create a Universal Interface. This seems the best strategy for the group as a whole - although they bring to the table innovative approaches, interesting new user interfaces, visionary algorithms and specific data sources, the one thing that most Alternative Search Engines lack is a significant amount of traffic. By working together towards a common interface, they could improve that situation. Or a larger company could acquire several of these engines and put them together, achieving a similar effect.
If this vision gets traction, then Google had better watch out!
I've been busy and away from this blog for a while. But I'm back now!
Over on Alt Search Engines, the 3-part series ran as planned. Here are the links:
Charles and Kaila are knowledgeable and experienced bloggers in this field. The series provides an interesting blend of perspectives.
[On that note, also worth checking out are the winners of the recent Essay Competition on Alt Search Engines.]
