Since our beta launch last week, we have been overwhelmed by our users’ enthusiasm. We are very encouraged that some of you said you like the value of our service. We are glad to see that Ambiently is useful in your application fields, such as research and education.
Users also want to know more about discovery engines. Quite often, after seeing the difference between a search engine and a discovery engine, users wonder which engine is a better tool for finding information on the web.
The Berrypicking Model of Information Finding
To answer the question, let’s take a look at the actual process of information finding on the web. Back in 1989, before the web was known to us, Marcia Bates of UCLA proposed a famous model for our information finding process: berrypicking.
Using the analogy of picking berries in a patch, Professor Bates believed that the information-seeking process is iterative and interactive. Finding information on the web is a journey: users start with a search query, check out what they get, modify the query, and continue.
Search and Discovery
Therefore, the best tools should always serve us along this journey. Today’s search engines like Google and Yahoo! play an indispensable role in our daily lives because they provide good starting points. Facing the sea of web information, everyone needs a search engine at hand all the time.
What a discovery engine does is work with a search engine for the rest of the information-seeking journey. It works best when you are already in the “neighborhood” of a webpage of relevant information. When you pick a big ripe blueberry on a bush, very likely there will be many more around! This is something unique about a discovery engine, something it might be doing better than a search engine.
Continuing Improvements
While users are excited about finding unexpected and useful new web sources via Ambiently, we clearly know that this is just the beginning. We have tons of work to do to improve the quality of ambient links on many, many ambient pages.
Sometimes you may notice that ambient links on an ambient page are not relevant. This is caused by a combination of many factors: the data structure of the original webpage, the data Ambiently can get from that site, the current scope/capability of our algorithms and rules, and the limitation of our foreign language support, among others.
Currently, Ambiently is in its public beta trial period. We are working hard to improve our service every day. In this process, we are eager to hear feedback and suggestions from you.
Further Readings
Berrypicking Your Way Through Search
Finding Interesting Information on the Web: It’s Not Just for Search Engines Anymore
Memex: Ambiently
Ambiently: Like a Search Engine, but Cooler.
The image we use for this post was taken by Mike McCune, view it on Flickr.
