WP3KnowNet PrototypeTool Documentation

Know-Net

Project number:EP28928

Project Title:KnowNet

Title of deliverable: Tool Documentation Prototype (1st Version)

Availability:Confidential

Number:D3.1 & D 3.2

Version: Final Version

Date of delivery:30th June 1999

Workpackage:WP 03

Nature of document:Report

Table of Contents

1. approach for prototype Tool Specification...... 1

1.1 Background & Protocycling...... 1

1.2. Developers Meeting in Cambridge...... 2

1.3. Broad Architecture for the Know-Net Tool...... 3

1.4. Integrating the Method with the Tool...... 4

1.5. KM Application Modules………………………………………………………………….

2. Implementation & testing strategy for user sites...... 5

2.1 Different perspectives of Business Value……………………………………………….

3. developing the prototype tool......

3.1 Evaluating Third Party Tools......

3.2. Demonstration of the Prototype…………………………………………………………..

4. protoype tool specification (V1)......

4.1. Introduction......

4.2Distribution, Publishing, Filtering......

4.3Collaboration Services......

4.4Search & Retrieval Services......

4.5Indexing Services......

4.6Storage Services......

4.7 Integration Services……………………………………………………………………….

4.8 Administration Support…………………………………………………………………….

4.9 E-Mail Explorer……………………………………………………………………………..

Know-Net Tool Documentation (Prototype)

KNOWNET/D3.11

 PLANET, KNOWLEDGE ASSOCIATES, INSEAD, DFKI, ICCS, FHBB, UBS, GW

WP3KnowNet PrototypeTool Documentation

1.Approach for Know-Net Tool Specification

1.1 Background & Protocycling

(WP2) examined the requirements engineering for the Know-Net Tool and described

  • The User required services and functionalities (with a summary of User Ratings of importance)

KNOWNET/D3.11

 PLANET, KNOWLEDGE ASSOCIATES, INSEAD, DFKI, ICCS, FHBB, UBS, GW

WP3KnowNet PrototypeTool Documentation

  • Process specific requirements

KNOWNET/D3.11

 PLANET, KNOWLEDGE ASSOCIATES, INSEAD, DFKI, ICCS, FHBB, UBS, GW

WP3KnowNet PrototypeTool Documentation

  • Product specific requirements
  • Description of the Individual Knowledge Portfolio.
  • A first ‘Suggested Integration approach’

The overall development approach for the Know-Net project has been agreed to be of an ‘iterative prototyping’ nature, as opposed to a ‘fixed specification’ and development. The primary benefit of this approach is that the Know-Net Partners recognise that ‘technology enabled learning’ will occur amongst both the Users and Developers as prototypes are developed and tested through protocycles. That is to say, the requirements specification will and must change as new concepts, new possibilities and new thinking emerge. A fixed specification approach does not allow for, nor capitalise on, further innovative thinking or experiential learning during the project.

Of necessity, therefore, no detailed specification will be written prior to development. Instead a broad specification will be written, prototypes will then be produced, and through the prototyping sessions the detailed documentation will finally emerge as Deliverable 4.1 Tool Specification.

1.2Developers Meeting in Cambridge

On 1st & 2nd June 1999, a Know-Net Developers Meeting was held at St Johns Innovation Centre, Cambridge. The purpose of the meeting was as follows:

  • Meet and get to know all the development team members & present their work to date
  • Discuss and explore optimal ways to better work together as a virtual team
  • Better understand the Know-Net Tool Vision & Objectives
  • Discuss technical integration issues
  • Allocate work, responsibilities, deadlines to developer teams
  • Discuss first proposal of Know-Net Tool for Users
  • Discuss User site pilot implementation and prototype testing strategy
  • Discuss work to be presented at Copenhagen meeting for WP3 at end June 1999.

1.3Broad Architecture for the Know-Net Tool

The Know-Net Tool will consist of three main elements

  1. A 'Knowledge Server' that User sites will be able to install as either:

a)their main Knowledge Server in a KM system

b) add-on Knowledge Server to complement their existing KM system

2. A library of specific 'KM Application modules '

3. A Know-Net Navigator

Our User Requirements Analysis, undertaken in WP2, has underlined the importance of recognising that the market for Know-Net will consist of a great variety of user types, sizes, stages of maturity in implementing KM systems and implementing IS/IT generally. To expect all users to consider Know-Net as the only KM tool to consider and use, with no regard to their current stages of implementation of KM and IS/IT would be severely limiting and almost impractical. Equally, to expect all users to implement Know-Net in exactly the same way, and precise order of implementation, would be equally unrealistic.

The approach for the Know-Net Tool, therefore, is to provide all users with, at least, the basic Know-Net 'Knowledge Server', to either add on to their current KM system, as a KM Tool to enhance their existing KM services, or to use as the central KM server for their organisations KM system.

Optionally, users can add specific 'KM Application modules' as required.

The Knowledge Server will have two main components

a)Knowledge Services

b)Knowledge Objects Store

The seven key functions provided by the Knowledge Services component of the Knowledge Server will be

a)Distribution, Publishing and Filtering services

b)Collaboration services

c)Search and Retrieval services

d)Indexing services

e)Storage services (Knowledge Object store)

f)Integration services

g)Administration and Support services

1.4Integrating the Method with the Tool

WP6, Method for Knowledge Diagnosis, will link directly to the Knowledge Objects Store.

WP8, Method for Performance Evaluation, will link directly to Knowledge Objects Store and KM Applications metrics.

WP1, Holistic Framework, will link directly to the Know-Net Navigator

WP7, tested through WP11, provides a Transformation Method that will enable the user (with assistance from KM consultants) to adapt and integrate the Know-Net Tool to the business processes and technical infrastructure of the organisation. One part of this Transformation Method will be a simple Knowledge Server checklist/tool to assist the user with prioritising which Knowledge Server services to implement, in which sequence, to be of most value to the business in the first instance.

1.5KM Application Modules

All Users could optionally add-on specific KM Application modules to the Know-Net Knowledge Server.

The KM Application modules will primarily focus on facilitating and supporting the key KM processes, e.g.,

  • Client & Market Knowledge
  • Sharing Best Practices
  • Virtual Project Management
  • Learnings & Competence Management
  • Innovation Management

The Transformation Method (WP7) will enable the user (with assistance from KM consultants) to adapt and integrate the Know-Net KM Application modules to the business processes and technical infrastructure of the company. A simple KM Application modules checklist/tool to assist the user with prioritising which KM Applications to implement, in which sequence, to be of most value to the business in the first instance.

Knowledge Associates will initially offer a series of specific KM Application modules to augment the KM Knowledge Server from its Knowledger 2.0 KM product suite.

FHBB will initially offer an Individual Knowledge Portfolio, as detailed later in this document.

DFKI will initially offer an indexing and retrieval support based on metadata and ontology-based document-content modeling. The metadata handling and retrieval machinery will be complementary to the KM Knowledge Server. It will enable comfortable access to multiple information sources and will allow for flexible and powerful content-oriented document organization.

2.0 Implementation & Testing Strategy for User Sites

2.1 Different perspectives of business value

Considering our three Know-Net Users in turn, we can see that they may have different perspectives of ‘best value to the business’ and different expectations from the testing. For example,

  • UBS may decide that an implementation of aspects of the 'Collaboration services - threaded discussions' of the Know-Net Knowledge Server may be of most initial value, followed by the ability to test feeding a section of the existing static one-way intranet 'Bank Web' into an interactive Intranet, supported by the Knowledge Server, with full 'comment' feedback, opinions etc. In this case, the Know-Net Knowledge Server could be considered as an enhanced add-on to a current established system.
  • Gooch Webster, for example, may decide that an implementation of aspects of the 'Distribution, Publishing & Filtering services', using agents to better push/pull information to people may be of most initial value. In this case, GW have an established Notes/Domino environment with aspects of KM in practice on a daily basis. The Know-Net Knowledge Server could be initially an enhanced add-on to a current system, but possibly with a strategy to continually and gradually integrate existing organisational services around the Knowledge Server.
  • Planet, for example, may decide to evaluate all aspects/services of the Know-Net Server, as most initial value, with a view to establishing a new KM system for the organisation, based around the Know-Net Knowledge Server.
  • The Know-Net Community, for better general communication, collaboration and knowledge sharing must also be considered to be a valuable user test site. The Community may decide, for example, to evaluate 'Distribution & Publishing services' in the Knowledge Server to ensure mandatory readings and changes/modifications are notified easily, and then, perhaps, to evaluate the 'Same time technologies' for shared applications, electronic whiteboards, real time chat etc, as most initial value to the Community.

Within the Know-Net Community, we should recognise the value of testing the Know-Net Knowledge Server for virtual team working, perhaps using all the Know-Net Knowledge Server capabilities.

3.0 Developing the Prototype Tool

3.1Evaluating Third Party Tools

Throughout the duration of WP3, the developers researched and evaluated a variety of possible tools and technologies as candidates for possible integration into the Know-Net Tool. In particular, the following tools were evaluated in some depth by Knowledge Associates (KNAS)

  • Autonomy Knowledge Server - This KM suite of software is produced by Autonomy plc from Cambridge UK. The suite uses neural networking technologies and extensive user profiling. Information is read and analysed into ‘concepts’ and stored in the proprietary Knowledger Server. Information can be read from all types of internal and external sources, eg Websites, e-mail, RDBMS, on-line feeds, WP sources etc. At the same time, users are profiled and their areas of interest are stored. The Knowledge Server then matches user interests to stored knowledge ‘concepts’ and advises the user of the match. Furthermore, the Autonomy server compares similar user profiles and matches ‘user to user’. The intelligent agentware and searching capabilities of Autonomy looked very promising for the Know-Net Tool.

The technology, as a concept, has much to offer and KNAS tried to install the Autonomy Knowledge Server and link it to searching and profiling the KNAS applications, selected websites, and profile selected users. Much difficulty was encountered with installation, performance and stability of software. Secondly, the user price was, at the time, estimated to be in the order of £50,000 plus per installation. KNAS abandoned the Autonomy evaluation to seek other alternatives.

  • Lotus Domino Version 5 – This groupware and web centric technology is produced by Lotus Development Corporation. KNAS had much prior experience of developing KM applications, including Knowledger 2.0, in Lotus Notes Version 3 and Lotus Notes/Domino Version 4. In the early stages of the evaluation and installation, immediately after Lotus had released the software to the Lotus Business Partner community, (KNAS is a LBP) there was instability of server performance. At a later stage, Lotus issued bug fixes and the software became acceptably stable. The increased and improved functionality and new concepts introduced into Lotus Domino Version 5 confirmed that this would be the primary platform for the Know-Net Tool.
  • Sametime Server – This technology was released by Lotus Development Corporation for Lotus Domino Version 4.6. Sametime uses its own server in a separate box to Lotus Domino Version 5.0.

The Sametime discussions, meetings centre, chat, sharing of applications and whiteboarding capabilities, live buddy lists on the Web seemed to work well, during the evaluation, but the videoconferencing capabilities did not work well.

The overall performance and pricing of Sametime, compared to competitor offerings (KNAS could not find any real competitors to this product) were key factors in adopting the Sametime Server as the optional add-on server for sametime communication and collaboration work for the Know-Net Tool.

Until the Lotus Sametime videoconferencing is working (Lotus advised late 1999) the Know-Net Tool will use Microsoft Net Meeting for videoconferencing functionality.

  • Muscat Software – This technology is a competitor to Autonomy KM suite. KNAS installed and evaluated Muscat. The installation was much easier and quicker than Autonomy (days as opposed to weeks). However, the performance whilst spidering and indexing external websites was much slower. However, the Muscat suite pricing starts from £4,000, as opposed to £50,000, but it does not claim the sophisticated neural networking technology and user profiling of Autonomy. The search capability and semantic analysis functions looked promising.

Although Muscat could be a promising candidate for the Know-Net Tool, the company has yet to develop the Lotus Domino Version 5 release (it is promised, but currently only runs on Lotus Domino Version 4.6).

Accordingly, we must wait and see if Muscat produce a version that Know-Net can incorporate within the project time deadlines.

  • WebTrends Software – This technology effectively monitors website usage and provides a variety of management and statistical reports, including broken link analysis (one of the requirements for the Know-Net Tool). KNAS found this tool to be performing satisfactory. It is included as the Know-Net Knowledge Server Maintenance Tool.
  • IBM DB2 Universal Relational Database – this technology wasevaluated as a potential candidate for managing the relationships of the Know-Net Server Knowledge Object Store.

Because of its universal acceptance in the marketplace, and its capability to be used via a web browser (the only one encountered) we found this to be the satisfactory platform for the Know-Net Tool.

3.2Copenhagen Demonstration

At the Know-Net Meeting in Copenhagen in late June, the Know-Net Prototype Tool was demonstrated to the group. In particular KNAS demonstrated the following:

  • A live connection over the Internet to the KNAS Cambridge, UK, Know-Net Prototype Server, based on Lotus Domino Version 5
  • A demonstration of the ‚Whats New‘ agentware from Lotus Domino
  • A live connection to the Sametime server and a realtime discussion/chat with KNAS developers in Cambridge

In particular FHBB presented the following:

M. Bettoni presented in Copenhagen the MailTack design from a developer's and from a user's point of view. The developer's view focused on the MailTack Architecture (elements and their connections) with a special emphasis on the importance of a relational DBMS as a basis for efficient and effective MailTack data handling. The user's view, on the contrary, focused on the suggested uses of MailTack and on the underlying concept of an E-Mail Explorer and its E-Mails knowledge base (MailKB) that allows a user to exploit the hidden potential of E-Mail communication: doing knowledge intensive, different-time conversations (discussions, negotiations, argumentations, decisions, consulting, ask-answer sequences, complex know-how interactions, etc.) with many partners spread over different locations.

In particular, DFKI discussed the following :

The basic „look and feel“ of the intended search and indexing interfaces were shown which combine a browsing-like navigation through metadata structures with graphical concept maps for indexing ontologies. It was shown how the external metadata handling is integrated with the usual Know-Net applications‘ input forms. The advantages of powerful, independent indexing mechanisms compared to the conventional Notes categories were discussed. Several extensions to the basic serach and indexing approach were proposed and discussed: defining active delivery services as automatically started „query scripts“ posting their retrieval results actively to the user; search over different sources independent from Notes and the Know-Net knowledge server; automatic or semi-automatic indexing by integrating a learning text categorization component.

4. Prototype Tool – Broad Specification (V1)

The Know-Net Knowledge Server will be based on a Lotus Domino version 5.0 Server and, optionally, for real-time collaboration services, a Lotus Sametime version 1.0 Server. Users of the Know-Net Knowledge Server will need to purchase the appropriate Lotus server licenses. At the time of publication, the Lotus Domino V 5.0 Server retails in the UK for approximately £1,600 and the Lotus Sametime Server V 1.0 retails at approximately £4,000 plus approximately £16 per sametime user client (£160 per pack of 10 user licenses). The Know-Net search and indexing mechanisms will run on the server side as a separate Java application communication with Lotus Notes via the Notes Application Programming Interface. At the client side, the search and indexing mechanisms will be supported by Java applets started by the browser.

The seven key functions and services provided by the Knowledge Server listed below, are described in this section following in more detail.

a)Distribution, Publishing and Filtering services

b)Collaboration services

c)Search and Retrieval services

d)Indexing services

e)Storage services

f)Integration services

g)Administration Support services

Distribution; Publishing; Filtering

Publishing support – (DFKI)

The Know-Net tool is based on the Lotus Domino version 5.0 technology for storing documents. All users with proper access to shared databases can submit their documents for discussion.

Thus the Lotus-Notes basis does support publishing. Moreover, threaded discussions about the document itself can be achieved in the same manner. Users are able to submit their comments as Notes documents using the response form . Indexing is integrated with the Notes input forms via a Java applet.

Document Libraries - KNAS

The ability for users to instantly create and manage document libraries will be provided. The default access control (ACL) will be open to all. If the creator of a discussion database wishes to restrict the access control rights, he/she must directly contact the Knowledge Server Administrator who will be able to enforce the necessary rights.