Online marketing information can change quickly This article is 10 years and 3 days old, and the facts and opinions contained in it may be out of date.
Todd’s notes: Federated search is essentially the information industry’s terminology for meta-search. Federated search is used for querying multiple premium content databases. The panel did a nice job in discussing where the value propositions are in aggregating multiple data sources. Paul Levy’s viral video “The Content Butcher”, was definitely one of the highlights of the conference.
Moderated by: Randy Marcinko, CEO, MEI
Jill Konieczko, Library director, US News & World Report
At US News, our end-users have access to over 70 databases, from LexisNexis and ProQuest to the community of science the whole suite of yellow book directories. End-user access provides them with authoritative content from reliable sources, but it can be daunting, and time consuming for end Ã¢â‚¬“users to search in tand around 70 databases, especially since each requires a different syntax and offers different interfaces; consequently, we don’t always see he usage we need to ensure appropriate return on our (significant) investments. Employing a federated search helps us to achieve other objectives. (I didn’t quite get all of this)
Why Employ Federated search?
Paul Levy, CEO Deep Vertical
The boundaries of federated search, vertical search, news search and standard web search are merging. The industry needs to focus on improving the users front-end toolset to better manage the vast depth of information that the search industry has created.
Yield management mentality must be brought to the content industry. Matching up marketing with pricing models and yield-management.
Peter Noerr CTO, Muse Global
Federated searching is an exciting growth area in searching, and it is moving front and centre in almost every search tool around, offering immense coverage, strong results refinement.
Jerome Presenti, Chief Scientist, Vivismo
Paul Levy, CEO Deep Vertical
Paul created an amazing viral video on the “content butcherÃ¢â‚¬Â that got some great laughs from the crowd.
Jerome Pesenti, Chief Scientist, Vivisimo
Question and Answers:
Discussions of Google premium content.
Paul Levy discusses the importance of “long tailÃ¢â‚¬Â clients and equates it to the gambling industry where a $40 gambler in a casino is viewed as very important in a much different way than “whalesÃ¢â‚¬Â are, but still very important. Content publishers need to recognize the value of smaller long tail clients.
Peter Noerr discusses how social search and tagging may actually dilute importance. His comments remind me a bit of Steven Colbert’s idea of “wikialityÃ¢â‚¬Â.
Peter Noerr talks about the difference between screen scrapings origins and the difference between it and html parsing.
The problem is not a lack of standards Ã¢â‚¬“ the problem is the lack of USE of standards.
Q: Why would scale be the source of error with social networking data?
A: Peter: Expertise is the important point. In context is the importance. The noise level is the concern, and must be filtered somehow.
Paul comments on the methodology for blogs being indexed adds to the noise level of search results (words from multiple posts being indexed and creating fairly irrelevant results).
Jill discusses methods for identifying the most credible content and why social tagging is often less credible as a tagging methodology.
Q: Why can’t a control vocabulary work WITH social tagging?
A: Jill jokes Ã¢â‚¬“ because there is no spell check to big laughs. Jill would like a controlled taxonomy for social tagging for better control over results.