London Online Conference

Citation

Halliday, C.A. (2002), "London Online Conference", Library Hi Tech News, Vol. 19 No. 3. https://doi.org/10.1108/lhtn.2002.23919cac.002

London Online Conference

Carol Ann Halliday

What follows is a representative sample of presentations illustrating the themes of the 25th Annual Online Conference held at the Kensington Olympia in London, 4-7 December 2001. The report reflects comprehensive summaries of the individual presentations from relevant library- and technology-oriented strands, via notes made during the presentation and the use of the supporting Conference papers included in the proceedings.

Web search engines

"Web search engines: changes and current condition" - Keynote, presented by Greg Notess, SearchEngineShowdown.com, Montana State University-Bozeman, MT, USA

The search engine market has undergone significant growth in terms of the number of competitors and sophistication of technologies developed and used, since 1994. Although market consolidation has been suggested to be imminent, the number of search engines continues to increase as fast as existing search engines go out of business or are acquired.

Search engines have struggled to find a reliable income stream (with advertisements being on the decline), and as a consequence a number of them have disappeared, as noted by the following scenarios:

• •

  Infoseek changed to Go, was bought by Inktomi and then dropped altogether;

• •

  Inference Find came and went;

• •

  Lycos no longer uses its own technology but that of "Open Directory and Fast Search";

• •

  WebCrawler and Magellan uses and is owned by Excite;

• •

  Dejanews, Remarq, Altavista Usenet and Contentville no longer wholly exist, and so on.

Teoma.com is owned by AskJeeves and is a new search engine which has its own database as well as Spiders crawling the open Web, thus retrieving different results from those engines that only search the open Web. WiseNut also operates in this way and both "automatically group search results by topic", using elements of Artificial Intelligence. Teoma has "Internet resource guides" and uses a technology called "link analysis" to sort the "metasites" from the main search results. WiseNut offers customisation and has done so since its inception.

The following search engines are still in operation and are building their own databases:

• •

  AltaVista;

• •

  Direct Hit;

• •

  Excite (question mark over this continued existence);

• •

  Fast (at AllTheWeb and Lycos);

• •

  Google;

• •

  Inktomi (at Hotbot, MSN etc.);

• •

  Northern Light;

• •

  Teoma;

• •

  WiseNut.

Whilst the search engines appear separate, they do partner to provide a more complete technology offering and reach, for example: NBCi moved from Inktomi and now pays for high ranking by Overture, as does iWon. Lycos uses Fasts' database instead of its own; and Yahoo! now uses Google after having changed from AltaVista and Inktomi, previously. Results from the different search engines are taken from different types of databases, e.g. advertising, own content, directories, news etc.

Google retrieves results from a number of databases, their own database, telephone number listings and PDF files.

Advertisements can take different forms; graphic banner, text banner, skyscraper, side ads etc. Google shows text banners and side ads related to the "search terms used". "Pay for placement" search engines have become prevalent with Overture, FindWhat, Kanoodle and E-spotting operating in this way. Payments are also made to the search engines by sites who pay "per click"; this payment is related to the amount on "specific keywords" and produces the Web site ranking as a result. Altavista, America Online, Hotbot, Infospace, iWon "also will display some Overture results". Sometimes these results are highlighted as "featured listings", as at Hotbot, for example. Overture is the only large search engine that will change the order of listings on the basis of payment.

Other search engines request payment for a review of other sites, without a guarantee that they will include in the Search engine directory, e.g. Yahoo!

Web sites can pay to have specific pages to be crawled more frequently. Spiders specifically focus on breadth rather than depth and search engines have traditionally only indexed HTML pages, but now other file types can be included.

The search engines do claim that non fee-paying sites are still represented in the results.

The search engine industry is buoyant but dynamic in its participants. Innovation and differentiation are still happening and, whilst there may be concerns regarding the ranking practices, there is still no charge to the end-user.

Super searching

"Developments in Web searching", presented by David Green, EMEIA Head, Public Web Strategy and Operations, Andersen, UK

This presentation sought to contrast the development of searching and to help provide a picture of future search functionality development.

Searching today allows for:

• •

  Search for terms (but keyword may be specific to publisher).

• •

  Historical indexes (pages may be "timed-out").

• •

  HTML addresses aesthetics and machines present the data, but do not understand them.

• •

  Results listed as published on the page, then the user must scroll through the page to look for their answer.

• •

  Flat, non-linked concepts, links between data set with no visual representation of the data relationship.

Tomorrow's searching will allow for:

• •

  Search for a concept.

• •

  Real time environment.

• •

  Automated process; state high level goal via PC/Web.

• •

  XML equates to content with meaning (machine processed).

• •

  Results distilled from many sources of data from different pages.

• •

  Visualisation of the concept of space/relationship between data.

• •

  Search on PC and get results sent to a mobile phone.

Search Genera. Searching will change as publishing changes, e.g. XML tagging as a standard, etc. Large sites compiled by people and advertising has neglected search technology and so allowed for a new approach in the forms of Natural Language Searching and Link Technology. This creates the opportunity for machine manipulation of information for our needs.

Search\Meta Search\Search\ Intelligent\Symbiosis Index/NLS /Utilities/Agents/

The semantic Web provides meaning to structure of XML documents leading to the potential for connectivity between Web sites, televisions, mobiles and various other appliances. Currently meaning is expressed in triplets: things, properties, and values. Humans endow meaning via: subject, verb, and object.

XML will be the successor to HTML that will open technology across different platforms and allows the publisher to address meaning, but in an arbitrary structure. Resource Description Framework (RDF) is linked by URLs and provides meaning to XML. The result of this will be a much more powerful structured query resulting in an integrated and semantic web (the user will be unaware of this technology and does not need to be able to see it, as searching is an unnatural event for people, so only the answers required will be presented).

Ontologies technology. This relates to files that formerly define relations among terms, e.g. taxonomies and a set of inference rules.

Open ontologies provide equivalence relations, i.e. the same concept, e.g. Zip code and postal code, and so does not look for literal words but for inferences/concept, regardless of the term/word that was used to describe it originally.

Intelligent agent. The Intelligent Agent adapts searching to your preferred results, e.g. a search on music, where the user only requires opera; for the next search the agent will get rid of pop and country music, etc.; thus these agents are proactive and personalised.

Agents will "talk to one another" and the transition from autonomy to accountability will be evident. What type and how much information can be passed from agent to agent? The future of crime is knowledge-based - the theft of personal data, with data signatures for security and verification. The agents will work together, not as one independently, as the information value chain.

Agents will delegate other agents already identified as appropriate through an "Agent Service Directory" (not unlike a series of advertisements as in the UK Yellow Pages). The process of delegation from agent to agent will be the "sub-assembly" of information, each agent adding value to construct the user's final answer and successfully distil the high value multi-page cross-referencing of XML categories. The Intelligent Agent will also adapt links dynamically to other concepts with the increased use of AI.

Global brain. "A Web-Global Super Organism", populated with adaptive (regenerative pages), with limited reasoning agents. It will "know" the concepts between the human-related terms to allow for cross-domain subject searching. It is a dynamic, self-organising Web.

Is the "Global Brain" a "Digital Dictator?" Could the individual assume a secondary role in society? Could the adaptive intelligence supersede humans?

A more positive perspective would be that the combination of people and computers would equate to a better understanding of the future.

"The concept of identification will establish increased relevance, as the term (subjective) is divorced from the concept (objective)."

Digital libraries and collections

"The DNER: weaving rich information and learning resources in the fabric of the Web", presented by Philip Pothen, Joint Information Systems Committee (JISC), UK

This presentation sought to explore the meaning of its own title in the context of "our education sector today." The Distributed National Electronic Learning Resource (DNER) is intended to be the point of access for the education sector to "quality" online information. The DNER is a "work in progress" initiative or set of initiatives.

The DNER have adopted, as a means to achieve their vision, ways of providing a set of isolated resources that are presented in the way the user has come to expect, via the open Web. With the Internet explosion comes the sophisticated user, who brings his/her functionality and navigational expectations, cross-searching, search recognition and retrieval of the various file types; PDF etc.

In order to answer the needs of the end user, the DNER presents itself as "an online space within which resources are accessed and which is managed to allow a direct user experience".

The reasons for the DNER development are:

• •

  Collection growth and lack of fully articulated and cohesive strategy for the functionality and co-ordination of the development of the online collection.

• •

  Lack of brand cohesion, in that the user is not presented with one strong overall brand, but a proliferation of brands resulting in each branded resource and service within JISC competing with the others.

• •

  A need to provide a "system of provision" taking into account "commercial, organisational and technical" requirements, i.e. the delivery, different purchasing models, subject-specific portals with cross-searching capabilities across the JISC resources. Involvement in the JOIN-UP project; thus JISC standards will need to be compatible with potential industry delivery and search partners.

• •

  JISC need to be recognised as providing both learning and teaching resources.

JISC services both the higher education (HE) and further education (FE) communities, with the latter being more advanced in the adoption of "convergence". FE initially brought about the integration of "libraries, media centres and IT [...] into shared spaces and shared organisational structures, [...] user-focused way of supporting and facilitating student learning". JISC has working relationships with the FE community via various associations.

The DNER provides services to support resources becoming a "wider community resource"; service initiatives include the archive hub and COPAC. These services provide data management and/or assist in locating community-specific information.

The UK National Union Catalogue report proposes a "serials union catalogue" JISC is jointly working on finding this. It is these activities coupled with the content provision that explains the title of this presentation more fully "... weave resources more strongly into the DNER by supporting the descriptions of collections in a consistent way and the disclosure of these descriptions through collaborative services designed for this purpose".

The DNER face challenges in fully providing the HE and FE communities with a full range of content delivered in a way that satisfies the end user and is simplified in terms of brand presence which will allow for a more "seamless experience". The DNER aims to provide the services in a way that does not require multiple brands. Thus the DNER plans to rebrand itself. JISC has identified that there is a need for "a consistent thread running through local, regional and national e-info activities and services," so JISC have opted for an identifiable term, rather than a brand. They are to use the concept of "Portfolio", as it is adaptable and customisable, has meaning across all subjects, will provide less confusion than a set of acronyms and brands. The examples provided by JISC are "My Porfolio, Oxford Portfolio, Physics Portfolio" etc.

The DNER are working on overcoming brand proliferation, providing integrated services across communities and focusing on the supportive "end user experience". Their future challenge is to become "... a world class resource for the twenty-first century".

"Hawks' claws and hickory nuts: managing information dilettantes in the corporate environment", presented by Alma Christine Vathis Howell, Intel Library Research Group, Intel, USA

The presenter was very clear in her view that the themes identified in this paper were also applicable to public and academic sectors.

This presentation addresses the difficulties for information professionals, as each customer (user) faces the same factors that influence performance as the information professional and the served organisation:

• •

  Internet economy;

• •

  globalisation of information;

• •

  competitive advantage.

The customer values immediate access to the information they need (as and when they require it); no "intermediary" and no "delivery boundaries" are required.

The customer may now begin research on the Internet or intranet, (this is reinforced by the industry Super I-AIM 2001 study conducted by Outsell), instead of beginning at the library.

Information professionals must "align" their services with the company objectives and apply measurements that support and prove this to be the case and that the library is helping to achieve those objectives.

Attention must be paid to the above three factors at: the customer level (how does this improve the user's ability to do their job?), the library level (how can the information professional take account of these and, still better, support the organisation and the customer via measurable benefits?) and the organisational level, (what difference do the library services make to the bottom line?). Information professionals should also include "more than the corporate performance-management approach that is tied to compensation ..." by "understanding the information-gathering behaviours and personal values that drive employees and the dynamics of successful teams in the global marketplace".

The Intel library has addressed some of these challenges through management problems:

• •

  limited budget and resources;

• •

  occasionally "overlooked" as a quality resource;

• •

  image;

• •

  respect;

• •

  expanding remits to include non-core services to ensure "political survival".

On the positive side work has been done to "strengthen the library image" and show the value of the library to the organisation, as "the management has addressed the challenges of the Internet economy, globalisation of information and competitive advantage".

In order to provide a better service to the customer the information professional sought to understand the environmental motivations for information use:

• •

  Triggers for information use were found to be an initial technical complexity at the beginning of projects.

• •

  Customers all search the Internet prior to contacting the library.

• •

  Customers only ask the library what they believe the library can deliver (which "illustrates preconception").

Intel conducted a loyalty study in spring 2000, the principal findings being:

• •

  Intel employees spend 16 hours per week searching/using information - this is higher than Outsell's identified industry average of seven to nine hours per week.

• •

  Only 40 percent of users would be willing to share information with the library.

• •

  Other sources are used for information. Some claims that even previous users were unaware of library services.

• •

  A total of 80 percent of library users experienced a time saving.

• •

  A total of 60 percent of users saved money in using the library (using a mean calculation).

As a consequence of these, and other industry survey findings, strategies in planning activities included "better navigation and organisation of the intranet site, reference interview refreshers for librarians, and targeted marketing campaigns".

Intel's resources are held on a global intranet, with a call centre to manage help requests. There are plans to integrate internal and external content with electronic versions of desk references etc.

The library strategy is in part to accommodate the different cultures of the subsidiary organisations in order to align it with "corporate initiatives". Continuous performance feedback on the information professionals having successfully supported the corporate and individual value, thus accommodating diversity.

Competitive advantage is addressed by providing the information professionals with the skills required to support new demands and change in remits, e.g. business analysis research and research support for technologists and Web management, etc. A number of the corporate objectives are highlighted in the libraries' strategic plan and benchmarked against a set of achievement criteria in order to demonstrate success. The library uses the same methodology as the business plan. Individual professional development plans have business benefit and are owned by the individual and his/her manager.

Thus the library must understand its users and have business measurements in place to demonstrate achievements and continuity.

Carol Ann Halliday(cahalliday@emeraldinsight.com) is a member of staff at Emerald/MCB University Press.