Skip to main content Accessibility help
×
  1. Home
  2. Publications
  3. Publishing partners
  4. Facet Publishing
  5. Content Listing

Refine listing

Refine listing

Access:
Content type:
Publication date:
Apply   From and To filters
Subject: Show more
Journals Show more
Publishers: Show more
Societies Show more
Series: Show more
Collections: Show more

Actions for selected content:

Select all | Deselect all
  • View selected items
  • Export citations
  • Download PDF (zip)
  • Save to Kindle
  • Save to Dropbox
  • Save to Google Drive

Save Search

You can save your searches here and later view and run them again in "My saved searches".

Please provide a title, maximum of 40 characters.
Cancel
×

6023 results in Facet Publishing

Page 274 of 302

  • 7 - Search usability

  • Martin White
  • Book:
    Making Search Work
    Published by:
    Facet
    Published online:
    08 June 2018
    Print publication:
    18 March 2007, pp 75-88

  • Summary

    In this chapter:

    ■ The importance of search usability

    ■ Options for presenting search results

    ■ Three case studies of excellence in search usability

    Supporting information discovery

    Usability is a measure of the ease with which someone can make use of a system to undertake specific tasks. Over the last few years there has been a very significant increase in the level of interest in usability as organizations strive to get the best returns out of their investment in websites.

    Among the elements of usability are:

  • ■ Functionally correct – the system allows the user to undertake the task

  • ■ Efficient to use – the time and/or number of clicks that are taken to perform the task

  • ■ Easy to learn – the task requires only a few clicks and advice is provided (or is available) to assist the process

  • ■ Easy to remember – the system does not tax the human memory

  • ■ Error tolerant – the user is never left at a dead end and there are routines that track problems as they arise and enable designers to take remedial action

  • ■ Subjectively pleasing – users want to use the system because of the overall ‘feel’ of the site and the way in which it reacts to their requirements.

    • Although there is a vast literature on usability testing there is very little advice on usability issues regarding search. Even the normally invaluable Useit website from Jakob Nielsen has only a few comments on search among the several hundred other entries. Probably the best general resource on the design of search pages is the much-expanded chapter on search in the third edition of Morville and Rosenfeld's Information Architecture for the World Wide Web, and Jenifer Tidwell's book on designing interfaces is also a valuable resource.

    Search in particular has to be a dialogue, because of the need for the search engine to assist the visitor/user to determine what they already know and then how they wish to proceed with the next stage of the search. The page navigation on a site is more about presenting terms that the visitor/user will be familiar with as a starting point, and then supporting working down through some form of hierarchy to find the information required. This is, of course, a simplistic view of navigation versus search, and elsewhere in this book the point has been made that search and navigation must be complementary.

  • 4 - Making a business case for search

  • Martin White
  • Book:
    Making Search Work
    Published by:
    Facet
    Published online:
    08 June 2018
    Print publication:
    18 March 2007, pp 41-48

  • Summary

    In this chapter:

    ■ What are the hardware, software and staff requirements?

    ■ Understanding total cost of implementation

    ■ Using risk management to make a business case

    Introduction

    It is not easy to build a business case for investing in search applications. The benefits of a search engine are very difficult to measure, even though there is good evidence from surveys that searching for information is timeconsuming and often frustrating. In the case of content management software applications often relatively standard processes are being undertaken on a regular basis. Search is not like that. Implementing a search engine may reduce the time spent on searching for information, but in most organizations that is not a measurable quantity. As with most productivity tools, is the time freed up then used for other tasks, or is it still used for searching but now over a wider range of documents and with increasing care being taken to find the most relevant?

    When implementing content management software, many companies have found that the costs of professional support from the vendor or systems integrator are considerably more than the base licence cost. However, once the implementation has been accomplished the staff costs associated with maintaining the software are quite low. This is not the case with search software, where there will be a continual requirement to optimize its performance.

    This is also the hidden cost of using open source software. Although the use of open source software is still very low in enterprise applications, there are many open source products that can be used for website searching, where in many respects the requirements are less demanding.

    One of the standard approaches to justifying investment in software is a return on investment (ROI) analysis. Generally speaking, only in the professional services sector is time logged and costed on a sufficiently normalized basis for a case to be made about the impact of productivity tools. Even then, trying to build a business case that will satisfy an accountant is a major challenge.

    Chapters 9 and 10 cover some of the issues that need to be considered when making a business case for investment in search for a website and for an intranet, but this chapter sets out some generic issues about the development of a business case.

  • 12 - Multilingual search

  • Martin White
  • Book:
    Making Search Work
    Published by:
    Facet
    Published online:
    08 June 2018
    Print publication:
    18 March 2007, pp 127-134

  • Summary

    In this chapter:

    ■ Why multilingual search is so difficult

    ■ The value of Unicode

    ■ The problems of transliteration

    Searching the Tower of Babel

    The issues concerning searching in multiple languages are often poorly understood, and yet the need to be able to do so is going to be increasingly important. Research from Byte Level Research (http://bytelevel.com) indicates that the majority of internet users are not native-English speakers.

    From the perspective of website search, this means that there may be a considerable number of visitors searching the site who may have only a limited range of synonyms and linguistic awareness. This is not just on a cross-national basis. It is estimated that over 300 languages are spoken in London alone, though this is probably the most linguistically diverse city in the world. Clearly, in the period up to the 2008 Olympic Games, the growth in Chinese users is considerable.

    The management of multiple languages also needs to be carefully considered in the enterprise environment. Just because an organization has English as its global corporate language does not mean to say that all documents will be in English. Documents relating to staff contracts and policies, and contracts with local suppliers, will invariably be in local languages. Patents and other legal documents will also be in more than one language, and if any one individual user is to have global access to the resources of the organization, the problems of how to search in a language-independent way as regards both the language skills of the searcher and the languages of the documents need to be addressed.

    Searching multiple languages

    Many search engines claim to be able to search in multiple languages, but care must be taken over just what this means. It usually means that the search engine can parse a document written in a wide range of languages, create an index, and then run a query against that index to present a number of relevant documents. Although not easy to undertake, this is now quite well developed technology and uses Unicode to convert a language to a standardized (or rather normalized) format. This enables a search to be carried out using a query in the destination language.

  • 5 - Specifying and selecting a search engine

  • Martin White
  • Book:
    Making Search Work
    Published by:
    Facet
    Published online:
    08 June 2018
    Print publication:
    18 March 2007, pp 49-64

  • Summary

    In this chapter:

    ■ Using personas to develop a specification

    ■ The critical role that a document audit plays in search selection

    ■ IT systems architecture issues

    ■ The stages of the selection process

    Introduction

    The basic principles of specifying and selecting search software are the same as for any other category of software, including a clear identification of the business need, and defining user requirements. However, with search software the process of defining user requirements is more complex, because:

  • ■ There is a range of tasks that the search software will be supporting.

  • ■ There is unlikely to be any previous experience of specifying search software.

  • ■ Specifying the performance of search software involves both hardwarerelated elements (which are relatively easy to specify) and user-related elements that are much more difficult to specify.

  • ■ The performance will be highly specific to a specific collection, or collections, of documents.

    • Indeed, where there is currently no search feature, or at best very limited search functionality, there can be some substantial issues in developing a specification against which to evaluate the search products. For example, at least one search vendor requires information about the number of searches to be conducted in order to scale the hardware for the system.

    Using personas to specify user requirements

    The use of personas to specify user requirements for websites and intranets is now well established, and there is a considerable literature on how to undertake persona development.

    A persona is a real virtual user. The ‘person’ described does not actually exist, but is created through research to typify some of the characteristics of a group of users. Biographical details are developed, even down to a photograph, so that the persona is so ‘real’ that the web team or intranet team start to identify with them as individual members of staff or as visitors to the website.

    The main characteristics of personas are as follows:

  • ■ Personas are hypothetical archetypes, or ‘stand-ins’ for actual users, that assist in defining and prioritizing user requirements.

  • ■ Personas are not real people, but represent real people throughout the specification process, and are defined in sufficient detail to create the illusion that they are real people.

  • ■ Personas are defined by the tasks they undertake, not by the organizational structure.

  • 11 - Enterprise search

  • Martin White
  • Book:
    Making Search Work
    Published by:
    Facet
    Published online:
    08 June 2018
    Print publication:
    18 March 2007, pp 113-126

  • Summary

    In this chapter:

    ■ An introduction to enterprise search

    ■ Technology and implementation issues

    ■ A ten-step procedure that will ensure effective enterprise search implementation

    Introduction

    The term ‘enterprise’ tends to be widely used in the IT industry to signify a high-performance software application for use by Fortune 500-type companies. Over the last few years there has been much discussion about what the difference is between a web content management application and an enterprise content management (ECM) application. The same applies to the search business, although, in common with ECM applications, there is the sense that these solutions are designed primarily to work behind the corporate firewall.

    The concept of enterprise search is to be able to locate information from any specified server/application, which may include not only internal servers but also external information services. This information might be held in:

  • ■ unstructured text files (e.g. web and intranet sites, record and document management systems, e-mail servers)

  • ■ structured relational databases (e.g. finance systems, enterprise resource planning, customer relationship management)

  • ■ video and image collections (e.g. digital asset management, video conference recordings, video training)

  • ■ specialized collections (e.g. project management applications, engineering and architectural drawings, maps, geospatial data, genomic databases).

    • This list is very incomplete, and of course the complexity is increased quite substantially by the extent of language diversity. In many enterpriselevel applications the two languages are UK English and American English, and although there are no issues of transliteration and translation, the problems of synonym management (e.g. gas rather than petrol) are very substantial. It does not help that few organizations would have an accurate list of all the database applications and servers that might contain information broadly relevant to the totality of the enterprise.

    Another issue with the term ‘enterprise’ is that it has a connotation of a for-profit organization. The reality is that all organizations are an enterprise when it comes to specifying a search solution.

    In many organizations the requirement for enterprise search starts with the need to provide an effective search solution for an intranet. It is often when implementing an intranet search solution that the organization realizes that employees would benefit from access to a wider range of information applications, and then finds that the intranet solution is not adequately scalable in power or extensible in functionality to provide an enterprise search solution.

  • 3 - The search business

  • Martin White
  • Book:
    Making Search Work
    Published by:
    Facet
    Published online:
    08 June 2018
    Print publication:
    18 March 2007, pp 31-40

  • Summary

    In this chapter:

    How big is the search market?

    ■ The impact of Google, Microsoft and IBM

    ■ What is a search appliance?

    ■ The benefits of a hosted search application

    ■ Profiles of over 50 search vendors

    The market for search

    The challenge for the software industry is to develop robust and effective software products based on the principles set out in Chapter 2. Until the launch of the CMS Watch Enterprise Search Report the search business was very poorly documented and understood, unless the organization subscribed to services from companies such as Gartner, IDC and Forrester. The Enterprise Search Report was written by Steve Arnold and contains detailed evaluations of around 30 search products. The level of detail in the report, both on the basic technology of search and in the features of the main search engine applications, is substantial and is essential reading.

    The search business is very fragmented, and compared to other software markets not very large in sales of standalone search products. In March 2006 the Gartner Group estimated that sales of new search software licences would total just under $370 million. The total revenues of the search software vendors is probably around twice this amount thanks to the support, maintenance and upgrade fees charged to existing customers, so that the total market value is perhaps around $700 million. Autonomy (including revenues from Verity) probably accounts for around 45% of this revenue, and Fast Search and Google for a further 20%. These figures can be no more than estimates, as many of the companies profiled below are privately owned and disclose little or no financial information. For example, one major search vendor announced in November 2006 that it had a 109% year-on-year revenue growth in the third quarter. This is certainly impressive, but no mention was made about the profitability of the company.

    Over the last couple of years there have been some acquisitions in the industry, but only the Autonomy/Verity deal involved two quoted companies. Triple Hop was acquired by Oracle in September 2005, and a couple of months later IBM bought iPhrase, in both cases for undisclosed amounts.

  • Foreword

  • Martin White
  • Book:
    Making Search Work
    Published by:
    Facet
    Published online:
    08 June 2018
    Print publication:
    18 March 2007, pp xi-xii

  • Summary

    Why do you need a book about search? Isn't search easy? A searcher enters some text into a box, clicks a button, and hey presto! – all the pertinent information appears in a nice handy list. Google made it simple; why can't your company do the same?

    It turns out that effective search is complicated – and hard. Ask any enterprise knowledge manager, website owner, e-commerce director, or, really, any fellow employee, ‘Are you satisfied with your search engine?’ The responses will catalogue a litany of woes: ‘It gives the wrong answers’; ‘We don't understand how it works’; ‘It doesn't find all of our information’; ‘It's slow’; ‘The results are cluttered and irrelevant.’

    This is often followed by the almost inevitable kicker: ‘We're investigating new search technology to fix it.’

    Yet technology is rarely the problem. Although people have been searching electronic information for more than four decades, search as a discipline remains quite young. You have significant choices about what technologies to apply and how to deploy them. Security, usability, scalability, and information architecture all present complicated, but important, challenges. Enterprises that succeed at search address these challenges as an ongoing process rather than a discrete technology project.

    That's why you need this book.

    I can think of no one better than Martin White to create a businessperson's guide to good search practices. As a search expert of global repute (and outlook!), Martin has taken the time and care to condense a complicated topic into an easy read. There are a variety of different technical approaches to search in the marketplace, but this book is essential to any search manager.

  • 6 - Optimizing search performance

  • Martin White
  • Book:
    Making Search Work
    Published by:
    Facet
    Published online:
    08 June 2018
    Print publication:
    18 March 2007, pp 65-74

  • Summary

    In this chapter:

    ■ The parameters of search engine performance

    ■ Search engine tuning

    ■ Using web analytics software

    ■ Data privacy issues

    Introduction

    The previous chapter outlined the elements that go to make up a search application, whether for a website or an enterprise. A major challenge for the organization is then to make sure that the entire system works as effectively as possible. For better or worse, Google has set some standards for search performance, but these are mainly in terms of systems performance and the time taken to run a query against an index of the spidered web. Users of websites in particular, but also users of enterprise search applications, use Google as a benchmark when judging search performance.

    Optimizing search performance requires attention to be paid to three different but related areas:

  • ■ the performance of the search system

  • ■ the usability of the search application

  • ■ the extent to which the search provides the information being sought.

    • Search system performance

    Some search vendors display the time taken to undertake a search as a metric of search performance. Obviously system response speed is important, but it is doubtful whether the user actually cares how long it took so long as it took no longer than they expected, even though they would not be able to specify in tenths of a second how long that is. Among the factors that need to be tracked on a regular basis are:

  • ■ average response time for different types of query (which requires a definition of query types)

  • ■ average time taken to retrieve and then present individual documents or groups of documents from each server

  • ■ time taken to update the indexes (measured in megabytes per hour).

    • The initial indexing can be quite a prolonged exercise. A representative time might be of the order of 10 Gb/hour. Subsequent updates to this will be much quicker.

    In an enterprise application the system response times for queries and document presentation are determined by the organization's server/network architecture. This may vary with the time of day, the day of the week and many other factors, because certainly as far as the network is concerned the latency will be affected by other activities taking place over it.

  • 8 - Desktop search

  • Martin White
  • Book:
    Making Search Work
    Published by:
    Facet
    Published online:
    08 June 2018
    Print publication:
    18 March 2007, pp 89-94

  • Summary

    In this chapter:

    ■ What are the benefits of desktop search?

    ■ What to look for when selecting a desktop search product

    ■ A list of 22 desktop search products

    Introduction

    The rate of growth in the storage capacity of desktop PCs has been quite extraordinary, with the standard hard drive in a PC being 160 Gb; even in a laptop the standard capacity is 40 Gb. At the same time the number and size of documents being created and stored on a local drive has been increasing at a similar rate – a sort of electronic Parkinson's Law that states that the volume of documents increases to match the memory capacity available. Marshalling all this information into a folder structure is well beyond the capabilities of most of us, as the sheer rate of change in business requires new folders and files to be created without leaving any time to reorganize the existing array. The end result is that it is impossible to find all relevant documents (including e-mails) when responding to a request for information or commencing work on a new document.

    Windows XP does have a search feature, but this string-searches files and is a very slow process. For several years there have been indications that Microsoft was planning to launch a more sophisticated search feature in its Longhorn desktop product. This is now called Vista, and by the time this book is published more information should be available on this feature.

    Ahead of this a number of the public web search companies launched desktop search software that would build an index of the documents on a hard drive, with Yahoo! and Google being among the early leaders. Over the last two years many other companies have joined the market, and there is now a wide range of products available; these are listed in Table 8.1. These all enable users to index files on their own PCs and laptops, and also on shared public drives.

  • Preface

  • Martin White
  • Book:
    Making Search Work
    Published by:
    Facet
    Published online:
    08 June 2018
    Print publication:
    18 March 2007, pp xiii-xvi

  • Summary

    Over the last few years search has come centre-screen, thanks largely to Google, a company less than ten years old but with a market capitalization that is already 30% of that of Microsoft. Entering just one or two words into the search box is the entry point to relevant websites containing billions of pages of information. It all seems so easy. Look behind the scenes and you begin to realize the scale of the operation and the degree of innovation. The applied mathematics of Google takes a book to begin to comprehend, and then there are probably a million or more servers located in secure bunkers. There is a similar story behind Yahoo! and Microsoft, and the other public web search services. The result is that the public web search services do a remarkably good job in sifting through millions of websites to highlight those that could be of some relevance.

    By comparison, searching for information in the digital repositories inside organizations is usually very difficult, with employees having lost faith in their intranet because of ineffective search, and visitors to the organization's website are hindered rather than helped by the search functionality. A search for the term ‘intranet’ on a UK central government website returned files with titles that included File30958, Microsoft Word – errs32.doc, file 22013, Business Case and Sarah in the top ten relevant documents.

    It is an often-quoted suggestion that 80% of the information created by an organization is unstructured. Whatever the real answer is, there is now a general concern that the rate of growth is so great that the problems of finding specific information are now among the most serious challenges faced by an organization of any size. Many organizations have adopted content management software to manage the addition of content into websites and intranets, but have found that the distributed authorship that this software encourages makes information discovery even more difficult.

  • 10 - Implementing search for an intranet

  • Martin White
  • Book:
    Making Search Work
    Published by:
    Facet
    Published online:
    08 June 2018
    Print publication:
    18 March 2007, pp 105-112

  • Summary

    In this chapter:

    ■ A ten-step procedure that will ensure effective intranet search implementation

    The ten-step process to implementing intranet search

    Although the basic principles for selecting an intranet search engine are the same as for a web search engine, there are some important differences and hence the steps are somewhat different:

    Step 1 What will be the benefits to employees?

    Step 2 Consider the technical options

    Step 3 Develop and obtain approval for a business case

    Step 4 Write a project plan

    Step 5 Define a test document set

    Step 6 Scope the user interface

    Step 7 Select a vendor

    Step 8 Install and test for usability

    Step 9 Optimize search performance

    Step 10 Have a development strategy.

    Step 1 What will be the benefits to employees?

    Arguably the expectations of employees are much higher than those of a visitor to a website. There is often an alternative website, but in most organizations that have an intranet this is the default information platform, and contains (at least in theory) all the core information an employee needs to meet organizational and personal objectives. The primary role of an intranet is to enable better decisions to be made. Certainly there is a subsidiary role in being a communications channel, but the business case for investing resources in an intranet is increasingly about supporting tasks. Whereas with a website hyperlinks are an important way of supporting information discovery, the scale of an intranet, the diversity of the content and the number of contributors means that hyperlinks are not as rigorously implemented and managed as they are in good websites. There should be no excuse for poorly implemented search in an intranet, as the search function has to make up for the inevitable weaknesses in hyperlinks.

    Compared to website statistics, similar information for an intranet, even when it is available, is much less helpful, though of course if the site already has a search engine then the search logs should be carefully reviewed. In the case of an intranet, its future development within the overall information management strategy of the organization must be considered from the outset.

  • 1 - Search must work

  • Martin White
  • Book:
    Making Search Work
    Published by:
    Facet
    Published online:
    08 June 2018
    Print publication:
    18 March 2007, pp 1-12

  • Summary

    In this chapter:

    The evidence that search can have a significant impact on business performance

    ■ A brief history of the development of search technology

    ■ Understanding how people go about searching

    ■ The potential for the role of information discovery manager

    Searching but not finding

    In July 2005 the Quarterly Survey from McKinsey Consulting1 reported on the Global Executive Survey that the company had conducted among 7800 executives in 132 countries, a fifth of them at Chief Executive Officer (CEO) or Chief Information Officer (CIO) level. Overall, 29% of CEO/CIO-level respondents and 40% of other senior managers reported that it was difficult to find information on which to make company-wide decisions. This is a very worrying finding. Companies are flying blind, and making highly risky decisions without being able to find information that they have already created and stored.

    Among many others was a survey carried out in 2004 by Vascom Bourne on behalf of Inxight Software among IT Directors in the UK financial services sector. According to the survey:

  • ■ 73% of respondents said that the main barrier knowledge workers face in sharing corporate information was not being able to use one information retrieval tool to capture data across several repositories.

  • ■ 58% of respondents also said that their company's search tools were ineffective at sourcing information quickly and efficiently.

  • ■ 66% of the companies interviewed said employees were regenerating information simply because they were unaware that the documents already existed.

    • For a number of years Susan Feldman and her colleagues in the Content Management and Retrieval Solutions research group of IDC (International Data Corporation)2 have been surveying the time taken to undertake a range of office tasks. Their latest analysis indicates that, on average, an office worker spends 9.5 hours a week searching for information, but, of this time, 3.5 hours is wasted in not being able to find the information required. As a result, either this information has to be recreated or a decision must be made on the basis of inadequate information.

  • 9 - Implementing web search

  • Martin White
  • Book:
    Making Search Work
    Published by:
    Facet
    Published online:
    08 June 2018
    Print publication:
    18 March 2007, pp 95-104

  • Summary

    In this chapter:

    ■ A ten-step procedure that will ensure effective web search implementation

    Introduction

    Although good web design will assist site visitors to find the information they want through navigation aids and hyperlinks, visitors also expect to be able to search the site. The extent to which there is value in implementing search does depend on the content of the site and the way in which visitors will wish to locate information on it. Adding a search function because of concerns that visitors are not able to use the navigation, and hyperlinking to locate content without an overall consideration of the information architecture of the site, is not the optimal approach. It could well be that search enables the navigation itself to be revised and optimized, removing layers of hierarchy, because search is clearly positioned as a way to find information in document libraries that may well improve the usability of the site to other users.

    Search is especially important in e-commerce sites, where there is product- or service-related information. One of the issues in such sites is not just the ability of the search feature to locate products, but also to be able to provide a range of ranking opportunities, such as price, size, colour, and other attributes. Any failure in metadata tagging could result in the product being found in the initial search but not when a ranked list is requested.

    As a result faceted search is especially useful in e-commerce applications, and one of the best examples is that of the wine retailer www.wine.com. The Librarian's Internet Index (www.lii.org/), The Gateway (www.thegateway.org/) and the Oracle Press Navigator site (http://pressroom.oracle.com/) are all examples of the Siderean's Seamark Navigator product. There are also a number of search vendors that specialize in e-commerce applications, and these are listed in the Appendix.

    The ten-step process to implementing web search

    The ten steps are:

    Step 1 What will be the benefits to site visitors?

    Step 2 Consider the technical options

    Step 3 Develop and obtain approval for a business case

    Step 4 Write a project plan

    Step 5 Agree the outline search parameters

    Step 6 Design review

    Step 7 Revise the content

    Step 8 Select a vendor

    Step 9 Install and test for usability

    Step 10 Review and revise.

  • Appendix - Search software vendors

  • Martin White
  • Book:
    Making Search Work
    Published by:
    Facet
    Published online:
    08 June 2018
    Print publication:
    18 March 2007, pp 143-158

  • Summary

    In this section:

    Brief profiles have been provided of nearly 40 search vendors. This is not intended to be a comprehensive list of all vendors, nor does it imply any endorsement by the author or Intranet Focus Ltd. The companies with the notation ESR are those listed in the 2006 edition of the CMS Watch Enterprise Search Report

    Active Navigation

    www.activenavigation.com

    HQ: Southampton, UK

    Initially founded under the name Multicosm Ltd, Active Navigation has its roots in the research carried out at the University of Southampton into the structure of navigation, innovative models of hypertext, and the thematic analysis of text using linguistic and statistical techniques. The company is privately owned.

    Arikus

    www.arikus.com

    HQ: Toronto, Canada

    Arikus was set up in 1997 as an information management consultancy. The core of the Arikus products is the Arikus Information Refinery Engine, AIRE. The software is priced very competitively, with an enterprise licence for up to 3 million documents being priced at $19,995. There is also a developer version for incorporation into specific enterprise applications.

    Atomz (ESR)

    www.websidestory.com

    HQ: San Diego, California, USA

    Atomz was the trading name of Avivo corporation, which was founded in 1996 with the objective of offering outsourced content management solutions. It moved into search solutions and in February 2005. was acquired by WebSideStory. The revenues of WebSideStory are around $70m.

    Autonomy Corporation (ESR)

    www.autonomy.com

    HQ: Cambridge, UK

    Autonomy was founded by Dr Michael Lynch OBE and Richard Gaunt in 1996 as a means of commercializing the work that they had carried out on the use of Bayesian statistics to underpin pattern recognition as a search technology. In 2005 the company acquired Verity Inc. for $502m, and in so doing became the largest search-specific software vendor. Autonomy had ready acquired etalk Corp. Autonomy is continuing to support all current versions of K2, as well as developing the latest and future generations of K2, to be known as IDOL K2. The company has stated that it intends to continue support of K2, and that the first release (K2 IDOL 7) will be usable by customers at their discretion, either as a standalone product or in a new mode running as a part of the IDOL infrastructure.

Page 274 of 302