We are launching a demo of our search engine Metahint. You can try it on the front page for exploring some of our favorite blogs.
Metahint is a tool for searching within a website. It collects and analyzes the contents of a website and comes up with suggestions as you type. Try the search boxes on the front page and you will see what we mean. As you start typing in any of the boxes, you will get suggestions that originate from the contents of the given website. These phrases are extracted automatically and ordered by how well they describe a given page, according to Metahint.
The search boxes are intended to be placed on the blogs themselves as embeddable widgets. We will soon start sending out beta invites, so you can try Metahint on your own blog, before we release it to everyone. Sign up for our beta program if you are interested. Taking part in the beta test is free and without any obligation. Note that the current demo shows only a preliminary version, giving a hint of the features we are planning, but it is still rough around the edges.
How is Metahint different ?
We know that many websites have some built-in search functionality and there exist already other widgets for searching within sites. However, these do not offer the suggest/autocomplete feature that has become indispensable on generic search engines such as Google or Bing. Google and Bing show you a list of queries that have been typed by other users and this works well when searching the whole internet. However, for searching within a single website this approach is not suitable: it is unlikely that someone has searched before for the exact same thing that you are looking for. Therefore, instead of showing what others have searched, Metahint shows you what you can find on the website itself.
Very well put this time, the key notice at the top rather than buried way down the page. I would, however, like to know more detail about the indexing/ keyword weighting principles “according to Metahint” (as per “These phrases are extracted automatically and ordered by how well they describe a given page, according to Metahint.“).
Me too! As in I’d also like to know what “according to Metahint” means. This is pretty key for a “customer” of your product id suggest.