abstract

The core value of search lies in providing effective access to the right piece of information in a timely manner. Traditionally, information retrieval systems have focused on retrieving the correct piece of information. In an effective search system, the following user-centric items play a key role in determining what right information and effective access mean:

1. User context,
2. Device used to access search,
3. User interface being used.

Information retrieval systems such as Google Search perform extremely well when addressing results in the long tail. Addressing accessibility is about recognizing that user needs and abilities vary over time — said differently, accessibility is about serving users in the long tail. As we move toward an increasingly mobile world with users accessing the Web from a variety of devices and usage contexts, overall effectiveness of search systems is determined by the user's ability to complete a given task in a timely manner. In reaching this goal, information retrieval needs to match the result set with the user along a variety of axies. For example, consider the query:

 Lufthansa 455

a query for a specific flight.

1. On a desktop, one might serve up a detailed Web page showing flight tracking,status and , available future travel.
2. On a mobile device,one might show a light-weight version of the above.
3. When using a voice-only interface, one might only speak the current flight status.

In the age of information overload, the band-width between man and machine gets increasingly overloaded. Thus, the importance of search goes up directly as:

1. user's attention span goes down.
2. User's display size goes down.
3. User's network band-width goes down.

Accessible Search is about building information retrieval systems that take all of the above into account. I'd like us as a field to formally define the various axies along which we determine the right result to serve users appropriately.