What's happening

Download the proceedings of the 2007 summit.
Download the conference timetable.

Ricardo Baeza-Yates,
Yahoo! Research, Barcelona, Spain & Santiago, Chile

Download the presentation (pdf)

Stanislaw Skorka , University of Crakow, PL - Download presentation (PDF - 1.6 Mb)

Among the IA’s background disciplines, library and information science are probably the most important. The research area and methods from these disciplines are borrowed by information architect, user experience specialists, usability engineers etc. It is said that IA evolved from information and library sciences. However, librarians in Poland do not seem to be interested in making relation with this new field. For significant number of them IA remains unknown discipline. There is need to translate IA to librarians, because they create, organize, search, and manage information in various forms. Moreover, they teach information literacy. Unfortunately, they also forget about their user’s needs and preferences.

There are at least areas in which librarians can benefit from applying IA methodology: 1. Library webpage designing; 2. Organization of physical library space (reading room, book collections etc); 3. Evaluation of web resources. Web users are very often educated by librarians or information servicemen. Therefore, librarians ought to be capable of designing information for concrete users (both high school and elementary school students), and teaching children and youth the methods of searching the Web. The presentation focused on the triple role of librarians listed above.

Joanna Wiebe, Orbitz, US
Scott Confer, Sears Holdings Corporation, US

How can a group of people from various software development disciplines rapidly develop high quality requirements for complex, evolving web applications? The criteria for “high quality” include that requirements will be organized, coherent, complete and “buildable”. Two analytical methodologies have been integrated to achieve this goal. Cognitive Organization for Requirements Elicitation (CORE) integrates an empirically-validated conceptual graphing method from the field of cognitive task analysis into a soft systems inquiry framework. CORE enables information architects (IAs) to lead a process to build, clarify, and iterate requirements, in large or small projects, in any domain, in a way that reveals inconsistencies or problems with existing solutions, and encouraging innovation.

This collaborative methodology uses visual display of information to make it accessible to all disciplines. The methodology was developed at Orbitz Worldwide, where travel booking web sites are enhanced with dozens of new releases every year. A case study approach illustrates how when project requirements are vague or missing, information architects (IAs) lead project teams through a seven-step process, to build, clarify, and iterate high quality requirements. Attendees will learn how to draw a Rich Picture, how to use Visio templates (available to participants) to construct Conceptual Graph Structures for goal hierarchies, taxonomies, causal networks and spatial relationships, how to use these diagrams to build preliminary requirements, and how to reach consensus with their team on final software requirements. The CORE methodology will offer actionable information that will help attendees translate information architecture concepts and processes to other software development disciplines, to collaboratively build high quality project requirements.

Are Halland, NetLife Research, NO

True findability needs to be designed from the inside and out - starting with the findable object itself, and focusing on the Core content and functionality that solves user needs as well as business problems.

This presentation outlines a practical design framework for thinking about websites and webapps in terms of findable objects (Cores) and Inward and Outward Paths to and from these.

Luca Rosati, University for Foreigners in Perugia, IT
Andrea Resmini, University of Bologna, IT - Download presentation (ODP - 5 Mb)

This paper aims to figure out some cross-boundaries principles of information architecture, from digital to physical environments and vice versa. Complexity and the ubiquitous computing phenomenon imply a new challenge: designing bridge-experiences -- i.e. consistent cross-contextual user experience patterns which apply to everyday objects as well as architectural spaces and digital environments so that users do not have to radically modify their behavioural patterns every time they move from one context to another. We will show that information architecture may play a key role in designing bridge-experiences and we will try to define what we call the “bridge-laws” of findability by comparing physical and digital environments.

Peter van Dijck, BE

Let’s say you are localizing a website. There is no alphabetical ordering in Chinese – and your widgets are organized alphabetically! Can I really use users’ tags in all languages? “Public schools” mean almost the opposite in the US versus the UK – now what? And my international car selling site has different car categories for every country?

With lots of examples and a practical approach, this will be a “meaty” presentation, bringing together existing theory about translating thesauri with new insights about global IA for the web. First, we look at translating taxonomies. Issues like semantic overlap, differences in hierarchy and in granularity between different classification systems are discussed, with various approaches to fixing them. We’ll also look at other IA systems that need to be localized, like ordering systems and localized search. Finally we discuss the more “web2.0-y” aspects of global IA: how to localize tagging systems, social networks and more.

Susan Webber, LBi, UK

The Map of Medicine (www.mapofmedicine.com) is a tool for clinicians that provides medical best practice information in the form of visual care journeys. Currently the Map of Medicine exists as a web application used widely in the NHS. LBi have recently been developing a Mobile Map of Medicine for use on PDAs and other mobile devices in order to reach out to clinicians on the move.

The Mobile Map was developed for the following reasons:

• A pilot carried out in Kenya with a paper based version of the Map of Medicine proved that the product is useful to clinicians in the developing world. The research recommended presenting the Map of Medicine on PDAs and other cheaper devices as users in the developing world don’t always have access to desktop PCs, or mains electricity
• There is an ever increasing use of PDA devices in the medical arena, especially among junior doctors
• The National Health Service in the United Kingdom is looking into providing a healthcare specific PDA for clinical staff.

This case study shows the journey and challenges of the translation of the Map of Medicine into a mobile form. This journey will focus on the user-centred design and Agile development approaches we took to take the product from conception through to launch.

James Kalbach, LexisNexis, DE - Download presentation (PPT - 3 Mb)

Principles of Long Tail economics point to the increasing growth of niche markets, or small bounded domains. With these comes an abundance of specialized, online information. And as Chris Anderson, author of The Long Tail, says, “You can find everything out there in the Long Tail.” But with so much specialized content, actually finding the right information will only get more difficult. Trying to organize everything from the top-down is futile, but as IAs our goal is rarely to classify everything. Instead, we work in bounded domains--for our companies or for our clients.

It stands to reason, then, that as more and more niche information spaces emerge, traditional navigation, taxonomies, and controlled vocabularies for organizing and filtering will not only continue to be exist, but will become increasingly important. Even detractors of taxonomies and controlled vocabularies agree that these traditional means of organization make sense within bound domains.

The navigation layer of long tail information spaces will rely on several modes of organization and access:

• Categories of niche markets will help us find the right domain
• Taxonomies will help us navigate within a niche market
• Classifications and groupings filter information quickly and efficiently
• Groups of tags will help us deal with emergent structures better
• Automation and category searches will do some of the work for us

This talk presents some fundamental considerations for navigating Long Tail information spaces and how they relate to both traditional and modern means of classification. Examples from the web and from the presenter’s own work demonstrate principles of navigating large amounts of information in complex bounded domains, including tagging and automatic indexing.

Frances Forman, Nomensa, UK

The complexity, ambiguity and fuzziness inherent in our language and categorisation is enough in its own right to frustrate an Information Architect trying to find suitable terminology or classification, but predictable menus are even harder to design because our interpretation of language is so heavily reliant on context and related information that we attend to whether visual, textual or auditory. People with disabilities who use different access technologies or information foraging strategies interact with Web content in a variety of different contexts.

Screen reader users have access to written language as spoken language whilst users watching captioned video have access to spoken language as text. On the Web, users are also able to translate information into new structural forms or summaries. This can either happen through browser and technology features or through toggles and switches that are designed into Web content. Different forms, formats and modalities of the same linguistic information suit different people and will have different usability implications.

Understanding how information access is compromised for different people with disabilities will help us to design more inclusive structures and support navigation.

Inclusive information architecture is about ensuring that the importance of context is not lost in translation and joining a cross disciplinary debate about where aspects of Web accessibility should take place.

Olly Wright, Media Catalyst, NL - Download presentation (PDF 9Mb)
Joe Lamantia, US
Thomas J. Froehlich, Kent State University, US

Ethical concerns are becoming increasingly important to information architects. This shift is being fuelled by many factors: including the increasing ubiquity of technology, the pressing need for sustainability, and the continuing rise of corporate transparency and accountability in reaction to blogging and other long-tail publishing. More and more research is telling us that effects of information technology on individuals and our culture are powerful and profound. As information architects we are in the front line when it comes to designing systems that manifest these effects.

However in responding to these concerns, information architects have only a paucity of of tools, resources and best practice to fall back on. The desire to work ethically and design ethical products is there, but the means to do so are scarcer.

This panel will contrast the perspectives of three speakers on the complex subject of ethics. Each will bring their experience to bear; describing both theory and practical approaches to ethics that should help any IA get to grips with this challenging topic.

Ariel Guersenzvaig, Multiplica, ES

This is a presentation of a comparative study of 30 onlines stores from the Spain, the USA and Chile. The stores have been analyzed according to a set of 100 factors that influence online persuasion. The factors apply to: Positioning, value proposition, differentiators, Customer Experience, Credibility and Trust, Pressure Selling, Brand Consistency and Customer Service.

The central objectives of the study are:

• To reflect on the main variables which affect conversion and the purchase process in an e-commerce store.
• To identify best practices in the multiple factors which affect conversion in an online presence.
• To describe the ideal store.
• To identify the most relevant lines of work and trends in the conceptualisation and design of online stores.
• To see the actual gap in the way of approaching Internet presences between the United States, Chile and Spain.

The surprising findings are – well – see you in Barcelona!

Kars Alfrink, NL/DK  - Download presentation (PDF - 14.5 Mb)

In many cases, playful architectures – dynamic spaces whose state can be manipulated in meaningful ways – are an effective way of delivering information. Traditionally, there’s been a divide between the internet and games industries. There has been little exchange of knowledge between the creators of games and web sites. This presentation hopes to offer a small step towards improving that situation.

Game designers talk of game play, possibility spaces and flow. IAs concern themselves with user experience, information spaces and attention; the parallels are obvious. The games industry has recently been abuzz with talk of serious games – using games for education. With the advent of RIAs, IAs have found themselves thinking more about more dynamic ways of transferring knowledge. Here again, we see important overlap and exciting opportunities for cooperation.

We’ll look at the state of the art in game design at the structural and behavioural level. When outsiders look at games they typically focus on the visual surface, but the defining aspects of a game can only be discovered through actual play. Similar to how IA concerns itself with the ‘invisible’ aspects of a site’s design, an important part of game design involves designing a ‘rules set’. This rules set is made accessible to the player through the game’s visual layer. Finally, every game is played in a social context. Rules set, visual layer and social context are the three components of a basic model of games that allows us to see how we can bridge the gap to the practice of IA.

Peter Boersma, info.nl, NL

1. Translate your favourite design solution into a design pattern.
2. Translate your favourite way of working into a process pattern.
3. Combine the two and the result is: Best practices on steroids!

Design patterns are an accepted way to document best practices in a structured format that allows for quick learning, reuse and inspiration for designers of interactive systems. Patterns for typical IA problems such as "breadcrumbs", "advanced search" and "booking process" are well-documented. A designer tasked with designing a system that might contain these elements can browse pattern collections and choose to apply the knowledge documented there.

Process patterns have been applied in the development of object-oriented software for at least 10 years. Scott Ambler describes them as "a collection of general techniques, actions, and/or tasks (activities) for developing object-oriented software" and distinguishes 3 levels of patterns: task, stage and phase. Collectively, process patterns help organizations and teams document knowledge about which part of a process works best in what situation and how to execute the part.

In the past I have proposed to add process attributes to design patterns expanding the focus of patterns to include the context of design. In this presentation, I'd like to expand those ideas and describe what a combined process-and-pattern library could look like. My aim is to encourage attendees to think about combining design patterns with design processes and profit from their synergy, which includes the creation of a common language for clients, designers and developers to talk about design.

Claire Rowland, Seren Partners, UK - Download presentation (PDF - 7.5 Mb)

A user’s experience with an organisation is determined not just by a single product, the web site, call centre or other channel, but by all of the interactions they have with the organisation. Many an IA has found themselves in a situation where they have wished they could have an influence over factors over and beyond the web site which they know are affecting the user experience, but how to go about this? The emerging discipline of service design applies a holistic design approach to craft the total user experience across all the channels through which the service is delivered. Service design is essentially interaction design, applied to a larger canvas and timeframe: and much of the innovation is happening in Europe.

This presentation will introduce service design, discuss why it’s important in the service economy, provide an overview of service design process and techniques (including ways in which IAs can translate skills they already have to the broader service context), and talk about why it can be hard to sell but help you to try anyway! It will be illustrated with examples of the speaker’s work from projects covering the UK, France and Germany.

Svein Ølnes, Western Norway Research Institute, NO
Nils Arne Hove, Western Norway Research Institute, NO - Download presentation (PDF - 730 Kb)

Users far too often encounter deficient search functionalities on web sites. This is due to the fact that the search module often does not have any semantic information of the content of the site at their disposal. In this presentation we will sketch how the introduction of a (semantic middle) layer in the web site architecture will provide such information. We also argue that information architects often underestimate the importance of such layer, representing a conceptual model of the site.

A middle layer is put in between the database and the web interface. In this layer, a semantic structure is represented by an ontology. There exist two kinds of standardised technologies to support such implementations:

• Topic Maps (ISO 13250)
• RDF/OWL (W3C recommendation)

There are several advantages in using such an architecture, it:

• gives a more flexible and scalable solution
• enables exchange of information
• permits reuse of both information and structure
• enables more efficient search modules

Margaret Hanley

As we develop or re-develop web sites, many of us are pulling together an increasingly diverse group of applications into one site; some of which are controlled and managed by our companies like core product information; others that are provided by third parties like job boards. Our challenge is to make this experience as seamless as possible to our users, ensuring that it reflects one brand; while enhancing functionality.

In these cases, we need to be able to move between the finer details of classification for our products to understanding how we enhance our companies' core offerings to create more value for the users and the bottom line.

In this talk, Mags Hanley will discuss how she has worked with companies using the principles of "the web of data" to enhance product and business offerings. Examples include using web services to reduce risk for credit checks, the creation of product information for the sole purpose of being used as a feed and enhancing product offerings by finding and re-skinning applications to create "one-experience".

Bogo Vatovec, bovacon, DE

An implementation of an e-commerce website to be delivered in 22 languages, 15 countries and sell 30000 products sounds like a nightmare. And the nightmare continues. A company with no experience in e-commerce, business units that never worked together, companies acquired but never integrated, new technology and no tools and methods available to handle a project like this. This case study will demonstrate the challenges, approach and solutions and failures in this challenging project, includine issues such as:

• Organizational challenges in setting up and running an information architecture and implementation project of this size.
• Content assessment and content modelling challenges
• Taxonomy and meta data
• Localization of individual e-business websites for specific markets and handling of language issues

Alexander Jongman, Hogeschool Utrecht, NL - Download presentation (ODP: 12 Mb)

UiaML is the result my “basement research” to improve the models that information architects create during a project. Until now information architects model their projects deliverables using storyboards, wireframes and sitemaps. Those models are a more or less abstract representation of the user interface to be build. They are created to discuss them with the various related discipline groups, e.g. customers, graphical designers, developers, etc. so that they together can translate the model into reality. Creating a model using those tools (storyboard, wireframes & sitemaps) worked well in the past, but today it’s getting harder by the day to use them to create a good reference model. Today the “sitemaps models are dead” [Eric Reiss, EuroIA 2006], wireframes of web 2.0 sites requires lots of additional unstructured documentation and agile software development represses the latter by a prototyping approach. So it’s time to learn some improved modelling techniques get a glimpse of future development of UiaML itself.

Giorgio Venturi, UCC Ltd., UK

Heathrow Airport is soon going to inaugurate one of the world's largest and most luxurious terminals. The Heathrow website redesign aims both to improve global user experience and to seduce users and induce them into spending more time in the airport. One particular challenge was to optimise user experience for a site that is at the same time an information hub for departing and connecting passengers and an appealing display for airport shops and restaurants. We tackled issues related to using personas, communicating them to a wide range of stakeholders, having different IA deliverables tailored to each audience, balancing user and business needs, designing personalised pages ('My Journey') and generally enhancing user experience with search and help

Tjeerd De Boer     NL
Wolf  H. Noeding   DE
Søren Muus           DK
Tomy Lorch           ES
David Caruthers    UK
Filip Borloo           BE (Moderator) - Download presentation (PDF - 1 Mb)

What is the future for European Information Architecture?  A number of European networkers and thought-leaders who earned their spurs in setting up local communities try to answer the question.