An unofficial blog that watches Google's attempts to move your operating system online since 2005. Not affiliated with Google.

Send your tips to

February 1, 2007

Google Works on Unified Search Engine

Google promises to change the presentation of the search results. As most people use web search exclusively, they miss a lot of good results available in specialized searches. Google tried to compensate this using OneBox results, but the solution seems a bit artificial, because there's no correlation between the OneBox and the organic search results.

Marissa Mayer says Google really needs to do something about that:
I think we need to look at results pages that aren't just 10 standard URLs that are laid out in a very linear format. Sometimes the best answer is a video, sometimes the best answer will be a photo, and sometime the best answer will be a set of extracted facts.

Larry Page also thinks the current format, adopted by most search engines, is not the best:
Since users are interested in many types of information, we expanded our search index to include new types of content. We added the ability to search for code in more than 40 different programming languages and indexed more than 7 million U.S. patents. Google Video search now links to YouTube content and with that acquisition, over time, we will integrate the YouTube platform with our search engine.

To make this range of information more useful, we are working on integrating different types of results -- video, images, news, books and so on -- all in one place. We are now blending book results into the main index and we will add more going forward. We are excited about providing a truly seamless user experience in search.

Apparently, a Google Universal Search is already in testing. It will be interesting to see how Google can rank heterogeneous pieces of data and do it well.


  1. We have been doing this for our site search on TechRepublic for a few months. See here for an example

  2. Looks like SearchMash will leave the playground and become official some time soon.

  3. I doubt this search engine will be like SearchMash. A page of search results for "romeo & juliet" could look like this:

    1. Wikipedia page
    2. The full book at Google Book Search
    3. A page that discusses the play.
    4. IMDB page of the movie.
    5. A video of the play.
    6. A Time article from 1940 (Google News Archive)

  4. It should me more than that..
    1. consolidated the data over the internet
    2. best rated content should prioritize in deploy the order
    3. Discarding the duplicate