Alan Smeaton
Online Information Seeking: Preparing for the Big Shake-Up
Online information seeking is an intrinsic part of what we do in our work and leisure pursuits. So much of our day-to-day living depends on us being able to find information online, or online information finding us. The success of this critical path between us and the information we desire is determined by the techniques, the algorithms and implementations of the search and recommendation systems we use to find information and for information to find us.
Over the intervening 25 years the development of (web and enterprise) search has matured only slowly. We do have better snippet previews within those ranked lists, colored links to show if we have already visited a page, the knowledge graph with factoids as candidate answers, we have search across different media, mobile search, and more. On the engineering side we have some of the most complex and impressive systems ever built. Web crawling is great, machine learning and learning to rank replaced our guesses at keyword selection when formulating queries but we still don't have personal search, unified search across personal sources, unified search and browse and recommendation in support of our tasks, and we juggle multiple tasks (how many browser tabs have you open ?), so functionally we are a long way short of what information science wanted us to have all those years ago.
In this talk I will offer my perspectives on search services and search functionality and I will present my horizon-scanning on where I believe we are headed in the short term and what we as customers and users of web search, are likely to see. Whatever happens we are likely to experience some major shifts in the ways we find information online. I will indicate my own personal laundry list of features I believe could have been incorporated into search and which are now credible features for inclusion and delivery as we experience the greatest shakeup in the 25-year history of online information seeking.
Over the intervening 25 years the development of (web and enterprise) search has matured only slowly. We do have better snippet previews within those ranked lists, colored links to show if we have already visited a page, the knowledge graph with factoids as candidate answers, we have search across different media, mobile search, and more. On the engineering side we have some of the most complex and impressive systems ever built. Web crawling is great, machine learning and learning to rank replaced our guesses at keyword selection when formulating queries but we still don't have personal search, unified search across personal sources, unified search and browse and recommendation in support of our tasks, and we juggle multiple tasks (how many browser tabs have you open ?), so functionally we are a long way short of what information science wanted us to have all those years ago.
In this talk I will offer my perspectives on search services and search functionality and I will present my horizon-scanning on where I believe we are headed in the short term and what we as customers and users of web search, are likely to see. Whatever happens we are likely to experience some major shifts in the ways we find information online. I will indicate my own personal laundry list of features I believe could have been incorporated into search and which are now credible features for inclusion and delivery as we experience the greatest shakeup in the 25-year history of online information seeking.
back to overview
Watch Recording
