What is Vertical Search?

A vertical search engine differentiates from a regular search engine due to it’s focus on a given subject or content type.

Topic focused vertical search engine
A great example of topic based search is Trip Advisor, a website with great information for people who are looking to travel.

Content focused vertical search engine
Media Type search can be found on sites such as Youtube, Vimeo, Flickr and others, where you search within specific media file types.
For example when I’m looking to fix CSS bugs my basic instinct makes me search on Google but if the first page doesn’t return the best answer for my problem I use Delicious, for books I use Amazon, etc.

The advantage of using a vertical search engine is that it will narrow the amount of information indexed which will result in more relevant search results.

What is Semantic Web

Semantic web can be interpreted as an intelligent, self-learning web. The future of semantic web will belong to services that will understand the user and deliver them the best answers for it’s need.

While it’s not perfectly clear for users, semantics have been implemented on several websites for a while now. Amazon was, as far as I know, the first site to implement a semantic user experience when it started suggesting further products based on users preferences. When a user bought a book, Amazon’s software would recommend other books based on purchases made by users who have bought that very same book. The result was a massive increase of sales and profit.
Google also is improving it’s search engine by reordering search results based on user navigation history.

So it’s fair to say a semantic web, or Web 3.0 as some call it, is each more a reality and not a trend. This brings me to this posts topic:

Can Vertical Search Engines coexist in a semantic web?

Personally I see vertical search engines as a the first step into the semantic web concept. I believe some people will say there completely separate things and I’d love to hear a different opinion on the subject.
If I want to find a needle inside a huge barn and know the needle is inside a haystack, I’ll search only inside the haystack. The haystack is my vertical search engine.

Using the same metaphor with semantic web once I’m inside the barn I’ll get a suggestion the needle is probably on the haystack, along with a picture of how the needle might look like and also where in the haystack it will probably be.
Eventually, once I find the needle, related subjects such as strings, other kinds of needles, groups, etc would be suggested as a complement to my interest on the needle.

Being the first time I searched for something I would also get suggestions related to hay and the haystack but if I show no interest in that, further similar searches would not retrieve those suggestions.

What do you think about the future of vertical search? How can it grow in order to survive on a Web 3.0 environment?