April 17, 2007

Computers in Libraries: Day 1 Morning sessions

Library 2.0: Building Communities, Connections and Strategies by Ken Roberts, Hamilton Public Library.

It was a little odd to come so far to hear such an inspirational story from my own back yard. Hamilton is just a little west of Toronto. When the city was amalgamated by the provincial government a few years ago, it presented problems and opportunities for the library. The highlighted project, a community portal for the city, had many community partners including other libraries in the area and mostly community organizations. The goal was to make the library the information destination for community information, to generate traffic and interest for those organizations. Kind of a Google for Hamilton.

The idea was to integrate with municipal services, using a CMS, rss and an events database. There were risks, for example getting uptake from community organizations to list their events. It is a success -- 70-80% of all library visits are online. The pathfinders and other subject guides are also online, there are also online bookclubs. One of the most important partner was the CFL team Hamilton Tiger Cats, they gave away 22K football tickets, they also used billboard advertisements.

The most important aspect is that this project represented partnership at it's best; if your library is a good partner in these things, it's reputation will precede it. You have to recognize different partner organizational cultures, to commit to common goals; you also have to find a sustainable model for these partnerships, a way to go forward on tight resources and differing levels of commitment. There are also challenges: some partners have more clout than other, like the city government; a changing, evolving environment, sustaining what's built and find a way to move forward.

What you need to succeed: build trust and respect, have shared values, engage the right people, start small and thing big, integrate with your core mission, test test test, celebrate success and excuse mistakes. And what's next: get into Second Life, wifi, changing roles for librarians, IM reference.


Social Search Engines had two speakers. Gary Price presented on Issues to Consider in Social Search. One important issue is when you build social search systems you have to make sure that a significant number of people participate, to create the social connections worth searching. The have to contribute consistently over a long period of time. The dark side of social search is the danger of spam, commercial organizations gaming the system and the quality of tags that users create. And how new is the idea of social search anyways, for example librarians have long used tools like the Librarians Index to the Internet. Other examples of social search engines are custom search engines, questions answering services and services like Intute and Globaledge. The rest of Price's presentation was mostly a laundry list of various social search tools, which are easy to check from the link above. A good presentation, covering a lot of ground in a short period of time.

Next up was Steve Mansfield on Humanizing Search: Internet Searching...Evolved. Mansfield is the CEO of social search company Prefound, so you have to take his comments in that context.

So, what is social search: it's answer-bound, social bookmarking, people search, directory search, personal page search. A key point is that from PreFound's perspective, the definition of social search is still open because there is no dominant, big player to get in the public consciousness first, to define social search the way Google defined web search. He hopes that social search will be more than just social bookmarking sites, that it will encompass ranking of links, relating links together in types of bundles, that is will reduce search chaff by introducing an element of human judgment. But how to get that human element into the search process? For example, humans can relate text, image, audio and video together in way automated crawlers cannot.

Real social search is ranked, links to a group of human created related links. Issues include spamming & gaming of tags and we have to develop technology to overcome this. The demo of the PreFound product was pretty interesting, with a lot of potential both as a individual product and as a representative of a product class. The main challenge, to me, seems to be to find a way to get enough people to convert their own personal leisure and procrastination time to building the human connections to make social search work, or to find a business model that will allow them to pay people to do it.

No comments: