Adding Value To On-line Collections For Different Audiences

Stephen Brown, David Gerrard, De Montfort University, and Helen Ward, East Midlands Museums Libraries and Archives Council, United Kingdom

Abstract

This paper is about models for effective on-line museum, library and archive design for diverse audiences. Successful Web site design depends inter alia on understanding and accommodating users' needs, and the more heterogeneous they are, the more difficult this is to achieve. In this paper we examine two models for handling diversity in on-line audiences, one developed specifically for museum Web sites, the other developed for curriculum design and media selection.  Both models consider the need to go beyond mere publishing of information. We argue a case for synthesizing the two models and use a hybrid version to deconstruct the design of two different heritage Web sites.

Keywords: Diversity, user centered design, design models, layered content, multiple points of entry, connecting storylines, learning experiences.

Introduction

While early commentators viewed the World Wide Web simply as a dissemination medium; e.g. McAndrew (1995), it is widely accepted today that "Organizations such as museums have a more vital role to play in the online world than just publishing information" (Peacock et al 2004). Although ultimately what visitors come to a Web site for is information of some kind "... there's more to it than meets the eye. Information needs can vary widely, and each type...causes users to exhibit specific information seeking behaviours." (Rosenfeld and Morville 2002).

Designing for diverse audiences is a major challenge for Web designers because of the tension created between the need to provide relevant, important, information quickly and easily for each specific segment of the audience and maintaining wide appeal and overall coherence. The more heterogeneous the audience is, the more difficult it is to achieve a Web site that is usable and relevant for the majority of its users (Krug 2000). "With so many different audiences, online exhibit organizers often struggle with the shape and approach of their online project." (Marable 2004).

Diversity

Museums, libraries and archives traditionally serve three core constituencies (Bowen et al 2001): researchers, the general public and schools. "Researchers" need not be accredited scholars and curators. They may simply be enthusiastic amateurs such as collectors or historical enactors (Marable 2004). The key characteristic is that they are experts. According to Preece et al (2002), experts have more knowledge and experience of their domain and can thus select "optimal strategies" for reaching information goals and can better predict the consequences of their actions. Thus researchers tend to search and seek to impose or derive connections between the disparate pieces of information that they find.

At the other end of the spectrum lies the general public, a large and heterogeneous grouping with widely differing levels of domain knowledge, interest and needs. Marable (2004) suggests "general audiences may not have enough knowledge of the topic to navigate a powerful search tool, and would need help in putting objects into context." Donovan (1997) similarly commented on the need "to wrap layers of interpretation around the bare fact of an object before the public can begin to grasp its significance." Thus the members of the general public tend to browse and need to be presented with information that makes the connections for them to help them appreciate the links between and significance of the disparate pieces of information presented to them.

School students and teachers can be regarded as a hybrid of the other two groups with students closer to the general public and teachers closer to researchers in terms of their knowledge and skills:

Figure 1. Three core constituencies.

School students need to be able to develop the characteristics of experts, i.e. to be able to discover, analyze and synthesize information that addresses identified topics, but their domain knowledge and skill levels are such that, like the general public, they still need a degree of scaffolding to engage their interest and provide the conceptual frameworks required to assimilate new knowledge. Teachers are likely to be more domain knowledgeable and have better research skills but may still need help to find specific information and resources to support specific projects quickly, and to address particular curriculum goals.

The idea of progression from novice to expert is supported by a qualitative study by Cunningham and Connaway (1996), one which focussed on the information seeking strategies of academics and research professionals. Interviews were conducted among 19 staff members from computing departments at two universities in New Zealand. It was discovered that the academics interviewed preferred the use of keyword searches to browsing for information and that they: "...rarely browse to come up with new research ideas, although several had done so in the past, in earlier stages of their careers." [our italics].

Models

In response to the problem of audience diversity, Marable (2004) proposed a user centered "integrated on-line exhibit model" designed to present common content to different user groups in different ways best suited to their particular needs. The model comprises three elements:

1. Multiple points of entry
2. Connecting storylines
3. Layered content (3 layers defined)

Multiple points of entry allow different groups to access or select a presentation mode that best meets their needs, e.g. Storylines for visitors who would like some context and inspiration, powerful search engines for those who require access to specific information rapidly.

Connecting storylines provide links between presentation modes, encouraging and assisting users to vary their modes of enquiry and broaden their searches.

Layered content means superimposing on the content different presentation modes to accommodate the requirements of different user groups. Marable specified three distinct modes or layers as "research", "exhibit" and "experience". The research level is defined as the historic artifact and secondary sources that comprise the basic archive or collection. This level, he proposes, is most relevant to expert users.  The exhibit level "provides a curatorial context to the primary sources held in the research level. Aimed at a general audience...visitors can choose among topics and themes selected by the exhibition organizers".  Level three is described as "an emotionally engaging (and entertaining) experience that brings to life the story of the exhibition for novice and general audiences.  This tier …usually takes the form of a story."  He suggests that "cinema and theatre" are appropriate expressions for this layer.

Marable (2004) demonstrated how the model can be applied successfully to the design of heritage Web sites, building exciting and informative sites that are based on searchable collections or archives. It is noticeable, however, that although these sites refer to "learning" as an activity, they do not explicitly set out to address the needs of teachers and students in a formal learning context. In other words, they address only the first two of our three core constituencies. Re-examining Marable's model, we can see that although he proposes three layers, layers 2 and 3 are in fact targeted at the general public, thus giving us only two user groups: experts and generalists. Conceivably another layer is needed to manage the more formal structured learning activities required by teachers and students.

One of the most influential scholarly researchers in this field is Diana Laurillard: her seminal book Rethinking University Teaching (Laurillard 2002) contains a useful conceptual framework for describing different kinds of learning activities or aspects of learning and for mapping them on to media forms, methods and technologies, as shown in table 1.

Table 1. Laurillard's taxonomy of educational media (Laurillard 2002, 90).

Laurillard's taxonomy has previously been used to guide the conversion of face to face teaching to on-line delivery and support. (Brown and Cruickshank 2003). We suggest that it may be similarly applied to the on-line presentation of museum, library and archive objects by matching the required learning experiences of the main target user groups to appropriate modes of presentation. To understand how, we need to consider Laurillard's model in more detail.

Laurillard (2002: 82-90) disaggregates learning into five different kinds of experience,

• Attending or apprehending a lesson as a largely passive recipient of information.
• Investigating or exploring some bounded resource in a more active way where decisions about what to attend to, in what sequence and for how long, are managed by the learner.
• Discussing and debating ideas with others.
• Experimenting with and practising skills.
• Articulating and expressing ideas through the synthesis of some new product.

She suggests that these five different kinds of learning experiences are best supported by different media types which she characterizes as Narrative, Interactive, Communicative, Adaptive and Productive respectively.

Narrative media are essentially linear, highly structured and non-interactive. They are a vehicle for transmission of information and ideas but not, on their own, appropriate for supporting the iterative dialogue that is central to knowledge construction. Videos, animations, exhibition information panels and storylines are examples of narrative media employed by museums. Notice that, unlike Marable, Laurillard makes no distinction between exhibitions and other linear media such as film and performance.

Interactive media offer resources for learners to explore in a nonlinear way. Users can decide for themselves what to look at and in which order. It is important to understand, however, that in interactive media the given text, in its widest sense, remains unchanged by the user. Catalogues, databases, search engines and physical layouts of galleries, bays and shelves offer opportunities for self-directed exploration, but their contents remain unchanged by the viewer.

Adaptive media are similar to interactive forms but with the crucial addition of "direct intrinsic feedback" on learners' actions (Laurillard 2002, 126). That is to say, actions result in consequences that are inherent to the task/system under consideration. A tennis analogy would be that serving a ball so that it clips the top of the net provides intrinsic direct feedback to the player about the need to raise the serve. Additional, extrinsic, commentary from a coach is unnecessary in such a situation. Simulations and hands-on exhibits that can be used to experiment with phenomena such as light, electricity, mechanics and sound are commonly used by museums.

Communicative media are simply those that support feedback and discussion; e.g. email, discussion groups, video conferencing, etc. Laurillard argues that feedback and discussion are fundamental to knowledge construction, enabling an iterative dialogue between tutor and learner through which theories and ideas are conceived, shared and transformed into knowledge and understanding. (Laurillard 2002, 127).

Productive media are defined as tools that allow learners to express themselves and to demonstrate their understanding. Creative and communication tools are less commonly found in museums, libraries and archives Web sites, no doubt because of the problems of ensuring appropriate quality and network security. A good example is the "Tell Your Story" section of the Moving Here migration Web site (http://www.movinghere.org.uk/stories/yourstory.asp?c=1) which allows visitors to publish their own stories about migration memories, illustrated with their own pictures or pictures taken from the Web site itself.

Applying Laurillard's definitions of different kinds of learning experience to the characteristics of the three main audience types, we can create a matrix in which the black dots indicate a mode of interaction likely to be favored particularly by that audience.

Table 2. An ideal theoretical model showing required interactions by audience type.

This analysis suggests three different, but overlapping, approaches would be necessary to accommodate the needs of all three audience types. If we combine this idea with Marable's suggestion of multiple points of entry, then the three layers suggested by Marable can be redefined as the five activity types described by Laurillard. The synthesised model would then comprise:

1. Multiple points of entry
2. Connecting storylines
3. Layered content (up to 5 layers)

In the examples that follow, we deconstruct the designs to show how this hybrid model can be employed to guide the design of museum Web sites by ensuring that the identified needs of different user groups are mapped to appropriate media forms and learning activities.

Examples

Recent years have seen major advances in creating and providing access to digital resources in the public sector at both local and national levels. The UK has three major examples of this. The People's Network project (http://www.peoplesnetwork.gov.uk) has connected all public libraries to the Internet. The NOF-Digitise initiative has funded the generation of an enormous range of material that is being made available, free of charge, to users of the People's Network and the National Grid for Learning through www.enrichUK.net. The DfES Curriculum Online initiative (http://www.curriculumonline.gov.uk) is working to make much of this and other publicly funded content available to teachers seeking digital resources to support their teaching programs, thus bringing together the cultural sector with local and national government and education.

The Transport Archive

The first of two examples described here was funded through the NOF-Digitise initiative.  The transport archive (www.transportarchive.org.uk), is an Internet based multi-media learning resource focusing on three key transport achievements in the sphere of railways, waterways and aviation in the UK. It was developed by a consortium of 4 UK local government authorities in partnership with De Montfort University.

The purpose of the site is to provide integrated access to a number of disparate collections and archives relating to the Great Central Railway, the Bridgewater and Manchester Ship canals, and the development of British aviation at Filton, Bristol. The target audience was initially defined as just about anyone. Through a Delphi style consultation exercise (Weaver 1971), we reduced this at first to 10 categories including transport enthusiasts, transport workers, historians, school students, teachers and undergraduates. This was subsequently refined into 3 core constituencies in the following order of priority:

1. General Public (including parents and young children).
2. Experts (including scholars, curators, amateur enthusiasts, high level students).
3. Schools (including students in the range 11-18 years and teachers).

From the preceding discussion we can see that this implies a need to incorporate all five different types of learning activities as defined by Laurillard. However a condition of the NOF-Digitise funding explicitly excluded development of formal learning activities, so the site is targeted primarily at audiences 1 and 2, although in a separately funded development some attempt has been made explicitly to address audience 3.

The transport archive comprises five main sections: an overarching gateway, three different transport sites dealing with canals, railways and aviation, and an integrated, pan-transportation learning zone. The gateway provides brief information on the other components and access to a site-wide search engine. Within each of the three different transport sites there are stories, multimedia, timelines, maps and links to other relevant sites, plus the search engine, which categorizes the results by transport mode and by theme (see below). These three transport sites are the main content providers.

The Learning Zone is quite different. It is designed for teachers and students to use actively to build upon the resources provided in the rest of the Transport Archive. Registered users can upload lesson plans and completed class work based on the archive's content. Browsing of uploaded content is unrestricted. To stimulate ideas, the Learning Zone offers four major themes: Feats of Engineering", "Changing Landscapes", "Community History",  and "Socio-economic background" linked to stories, exhibits and keywords related to these themes.

Experts wishing to investigate the 8,500 plus digital objects in the archive can use the keyword search facility to generate thumbnail images and titles of more detailed records held in a MySQL database, marked up with Dublin Core Metadata. Figure 2 shows one of the full object records.

Figure 2. Letting in the Water at Ellesmere Port - with Engineer on the Bank T2171. http://www.transportarchive.org.uk/getobject.php?rnum=T2171&searchitem=gap&mtv=T1&pnum=1

Unsuccessful searchers are offered prompts via a set of controlled vocabularies reflecting the most commonly used terms among the first 4000 or so records.  Figure 3 shows a list of suggested terms for which there are matching objects in the database.

Figure 3. Sorry no results for this search. http://www.transportarchive.org.uk/search.php?searchitem=hippo

Alternatively, visitors may navigate the site via a range of stories designed for general public use. They are intended to provide visitors with some context for and interpretation of the underlying objects and to present the content in an inspirational and interesting way. The stories are introduced by short summaries and illustrated by images drawn from the underlying database. Stories are further subdivided into shorter freestanding sections. These stories can be accessed in any sequence but individually provide a linear narrative. However, visitors are encouraged to broaden their investigation by following links from each story to related stories and within the stories to the images and their underlying metadata. Further opportunities for contextualized investigation are provided by timelines and interactive maps.   Figure 4 shows a section of one of the timelines providing an overview of the historical period and situating specific objects within it.

Figure 4. Railway archive timeline. http://www.transportarchive.org.uk/timeline.php?l=l&c=1900&d=1900#year

So, in terms of our hybrid model, the site offers:

Multiple modes of entry: The main sections offer free text search engines; prompted searches via controlled vocabularies; and browsing via stories, timelines and maps. In addition, the learning zone offers opportunities for structured learning activities designed to encourage investigation, discussion, experimentation and articulation.

These different entry modes provide direct access to the different layers of content required by the different target audiences. Thus the search engine provides direct access to the lowest level of archival records, facilitating the process of exploration and investigation.  At the next level, stories, timelines and maps within the main transport sections provide access to the contextualizing layer corresponding to tasks of apprehending and comprehending meaning. The separate Learning Zone section provides a framework for experiment and practice and articulation and expression of ideas, although this section is less well developed because of funding restrictions. For the same reason the site does not provide facilities for discussing and debating ideas.

Table 3 summarizes the transport archive content layers and audiences, using black dots to indicate needs that were met and hollowed dots to show unmet needs.

Table 3. A summary of the transport archive content layers and audiences.

The different layers are linked by connecting storylines (i.e. the themes in the learning zone), just as the search engine results are returned categorized by transport mode and by direct links between different content layers such as links to images from stories.

emsource

Our second example is drawn from a regional archive education initiative, the emsource project, that has been operating in the East Midlands, UK, since January 2003. This is funded by East Midlands Regional Archive Council (EMRAC) and East Midlands Museums, Libraries and Archives Council (EMMLAC) and is coordinated by a project education officer who works across the region. The key focus of the emsource project is the ongoing development of a Web site for teachers (www.emsource.org.uk) that provides archive based teaching and learning packages that support the UK National Curriculum. This work is complemented by advocacy and training events targeted at teachers and education providers, as well as ongoing consultation with the education community.

The emsource elearning project (http://www.emsource.org.uk/learning) was developed in response to teacher demand for on-line pupil activities. Piloting of this project will inform the long-term expansion and development of the emsource Web site. The elearning activities are aimed at secondary pupils studying Key Stage 3 History and aim to promote the use of archive collections for educational use. They are a good example of the kind of content anticipated by the transport archive learning zone. The project was developed jointly by the East Midlands Museums, Library and Archives Council and De Montfort University, with assistance from the University of Nottingham School of Education and the East Midlands Broadband Consortium.

The elearning activities make use of two distinct Web sites. The Archives for Learning Web site describes the learning activity for students and provides links to historical resources drawn from The Record Office for Leicester, Leicestershire and Rutland, Nottinghamshire Archives and Leicester City Museums Service. The site includes support pages that help provide context for the events with which the archive items are concerned.

Stages 2 and 3 of the activities require the use of a discussion board, a service available to schools through East Midlands Broadband Consortium's regional learning platform and used by all schools which participated in the Autumn pilot. A Teacher Support Booklet provides an overview and an explanation of the aims of the activities, guidance on how to run them, and hard copies of photocopyable resources.

Learners role-play different characters on-line to develop their understanding of historical events during the 19th century, based on access to original archival material. The exercise is structured around conditions in the framework knitting industry during the 1840s - a key area of employment in the East Midlands and not then converted to the factory system - and conditions more generally relating to working and living during the mid nineteenth century. In addition to historical knowledge and understanding, the on-line exercises are designed to develop skills in research and interpretation of historical sources, presentation of arguments, and on-line collaborative working.

The activity is in four sections.  In stage 1, students are assigned an historical identity, based on a real character belonging to one of three groups: framework knitters, the establishment or political activists. Taking on the role of a character is an important aspect of the exercise. The sense of character identity is strengthened by giving each student a character name and password  to log on to the site.

Their task is to study a range of on-line archival material. Students use an on-line notepad function to collect evidence relating to their characters as they progress through the exercise. Once the students reach the end of the initial evidence collection phase, they are given an opportunity to review and change their notes.  Then the application generates an email message containing the notes: this is sent to the student's school email address.

Figure 5. This screen shows the notepad for a Political Activist character.  http://www.emsource.org.uk/learning/Activist/charPage.asp

In stage 2, students log on to an on-line discussion board to discuss with other members of their role group three possible courses of action. Based on the evidence they have appraised, they reach a decision regarding the best course of action.

In stage 3, all three role groups are brought together in an on-line meeting chaired by the Commissioner (teacher). The task is to discuss ways forward that would be acceptable to all three groups; but in practice, a compromise is rarely reached.

In stage 4, students log back in to the Web site to find out what happened post 1845, involving their own character's fate. Pupils use the on-line notepad function to write an autobiographical account for the character that they have played.

A teacher-led debrief explores issues relating to the availability and interpretation of historical evidence, the decision-making process and its outcome, and the practice of studying history.

In terms of our framework this site offers:

Single point of entry. Notwithstanding the different role assignments, each student is required to work though the activity in the same sequence, progressing from one stage to the next in tandem.  The universe of archival objects is closed, small and cannot be searched freely.  It can only be browsed in a predetermined sequence, and members of a role group do not have access to the archives assigned to other role groups. (It's also designed so that the archive items become progressively more dense/problematical, thus supporting different learning speeds).  Table 4 summarizes the emsource content layers and audience.  The hollowed-dots sections indicate that in this example the site is available only to registered school users.

Table 4. A summary of the emsource content layers and audience.

Different layers are explicitly embedded in the different stages of the exercise. Students initially have to attend to and understand their individual roles and character briefs from the linear narrative materials presented on-line. Then they are required to investigate the archive to collect evidence. The task is interactive in the sense that they can select which objects to read and in which order, but they cannot search outside or change the contents of their allocated archive. The next two stages entail presenting and discussing their views through the communication medium of on-line discussion. Stage 4 provides the opportunity to articulate and consolidate their learning through the writing of an autobiographical account. The whole role-play activity constitutes an adaptive experiment in Laurillard's terms. The processes are authentic (reading newsletters and handbills, discussing with neighbours and arguing with rival role groups), and open ended, i.e. there is no predetermined or correct outcome. The fact that there is rarely a broadly acceptable outcome at all is immaterial since the real learning outcomes of the activity relate to understanding historical and historiographical issues rather than resolving a mid-nineteenth century dispute.

Connecting storylines are constructed by the students themselves through the on-line discussions. In the discussions they compare different viewpoints derived from different selections and interpretations of evidence and debate issues from standpoints derived from partial evidence bases.

At the time of writing, evaluation data from the pilot was still being collated and analyzed. However, early indications are that students have responded enthusiastically to the role-play element of the activities:

Because they're all real (the characters) you learn... These people are real people and they actually went through all this, so you learning about them is fun because you can see how lucky you are that you're alive at this time and that you're not alive then, because if you're a framework knitter you haven't got a good life. Work and that's it! (Year 8 pupil, Brunts School) 

Being a political activist, you had to listen to both sides of the story and work out a way to make them (the framework knitters and the establishment) compromise. (Year 9 pupil, Meden School)

Conclusions

Museum Web sites can accommodate the different requirements of different target audiences by presenting the same content in different ways. Key to this idea is the concept of layered content proposed by Marable where presentation modes are matched to the known characteristics of specified audiences, but Marable's model is incomplete because it does not explicitly address the needs of teachers and learners in schools or interaction modes which entail discussion and creative expression. Laurillard's taxonomy of educational media offers a framework for comprehensively identifying the needs of different user groups. We have shown that it can be applied to the selection of presentation modes to ensure appropriate user-content interaction takes place, but Laurillard's sequence of activities needs to be reordered to fit with the idea of layers of content, starting with the most fundamental: the object collection or document archive.

References

Bowen, J., R. Brigden,  M. Dyson, and K. Moran (2001). On-Line Collections Access at the Museum of English Rural Life. In D. Bearman and J. Trant (eds.). Museums and the Web 2001: Proceedings. Seattle: Archives & Museum Informatics, 2001. consulted September 9, 2004. http://www.archimuse.com/mw2001/papers/bowen/bowen.html

Brown, S., I. Cruickshank 2003. Pedagogical re-engineering: Using Laurillard's taxonomy of educational media to guide curriculum design. In J. Cook & D. McConnell (Eds.) Communities of Practice. Research Proceedings of ALT-C 2003. 10th International conference of the Association for Learning Technology. Sheffield: Association for Learning Technology. 56-69.

Cunningham S. J, and L.S.Connaway (1996). Information Searching Preferences and Practices of Computer Science Researchers. consulted December 15 2004. http://ieeexplore.ieee.org/iel3/4125/12214/00560024.pdf?isNumber=12214&arnumber=560024&
prod=CNF&arSt=294&ared=299&arAuthor=Cunningham%2C+S.J.%3B +Connaway%2C+L.S.

Donovan, K. (1997). The Best of Intentions: Public Access, the Web and the Evolution of Museum Information. In D. Bearman and J. Trant (eds.). Museums and the Web 1997: Proceedings. Los Angeles: Archives & Museum Informatics, 1997. consulted September 9, 2004. http://www.archimuse.com/mw97/speak/donovan.htm.

Krug, S. (2000). Don't Make Me Think. Indianapolis: New Riders Publishing.

Laurillard, D. (2002). (2nd Edition) Rethinking University Teaching: A conversational framework for the effective use of learning technologies. London and New York: Routledge Falmer.

Marable, B. (2004). Experience, Learning, And Research: Coordinating the Multiple Roles of On-Line Exhibitions. In D. Bearman and J. Trant (eds.). Museums and the Web 2004: Proceedings. Arlington: Archives & Museum Informatics, 2004. consulted September 9, 2004. http://www.archimuse.com/mw2004/papers/marable/marable.html

McAndrew, P. (1995). Using the World Wide Web to support the dissemination of courseware. ACCIS 2, 2, 12-17.

Peacock, D., D. Ellis, J. Doolan  (2004). Searching for Meaning: Not just Records. In D. Bearman and J. Trant (eds.). Museums and the Web 2004: Proceedings. Arlington: Archives & Museum Informatics, 2004. consulted September 9, 2004. http://www.archimuse.com/mw2004/papers/peacock/peacock.html

Preece L., Y. Rogers, and H Sharp  (2002). Interaction Design: Beyond Human Computer Interaction. New York, Chichester: John Wiley and Sons, Inc.

Rosenfeld L., P. Morville  (2002). (2nd Edition) Information Architecture for the World Wide Web. Farnham, Sebastopol, CA: O'Reilly & Associates.

Weaver, W.T. (1971). The Delphi forecasting method.  Bloomington, IN: Phi Delta Kappan, 52, 5, 267-273.

Cite as:

Brown, S., D. Gerrard and H. Ward, Adding Value To On-line Collections For Different Audiences, in J. Trant and D. Bearman (eds.). Museums and the Web 2005: Proceedings, Toronto: Archives & Museum Informatics, published March 31, 2005 at http://www.archimuse.com/mw2005/papers/brown/brown.html