EuroIA 2011: Day One


EuroIA is the primary European Information Architecture conference. In 2011 it took place in Prague, The Czech Republic.

See all posts

Nestled between ornate medieval and stark modernist architecture, EuroIA opened to a sold out crowd from twenty-seven different countries ranging from Japan to New Zealand. And of course, a lot of Europeans.

We’ve also done our first experiment in using Storify for curating conference streams. Love it? Hate it? Prefer it to these reports? Check it out and let us know in the comments. (Also check out the conference Lanyrd or Martin Belums “all your EuroIA slides are belong to us“).

Luke Wroblewski, Today’s Web

Wroblewski (who, organiser Eric Reiss was quick to point out, is in fact a Polish-born European, despite the American accent) kicked off the conference.

In short, tomorrow’s web is social, and mobile.

Social is big

Wroblewski showed how Britekite has changed from using the dreaded webform (after 20 years, you’d think that it would remember my name), to Facebook Connect, which both eases the transition of logging into a new app, but also encourages activity through people you know.

Wroblewski also decribed the 0-1-2 model: you’re twice as more likely to engage in something if two friends are doing it already than if only one is).

Facebook Connect boasts not only an 800m userbase from Facebook, but also that 500m of them will be logged in at any one time.

It’s also had huge pickup in use in 3rd party apps: ( 60m/17% in 2008, 100m/22% in 2009, 250m/30% in 2010).

all software will become social, because everything humans do is social.

The nature of social is also changing how people behave on the web. Mark Zuckerberg is quoted as saying “the best check on bad behaviour is identity”, and has been shown in Quora, which has only had to ban one person in their 250 million userbase.


The mobile field is also increasing exponentially. For example,  Amazon has done >$1bn on it in the last 12 months and Best Buy doubles sales through it each year (now at 30m).

Wroblewski pointed out that while we hear about mobile first in developing countries (50% of primary access in Africa/Asia, and 45% in India), the developed world is not far behind: A… but developed countries too: 22% of  people in the UK are predominantly mobile, and the US is predicted to reach 50% by 2015.

This brings up interesting challenges: using a travel website on a desktop not only involves a different screen size, but usually a different context (the mobile may be checking in at the airport).

80% of the crap you design for desktop has to be killed off for mobile

Mobile first philosophies:
Growth = opportunity
Constraint = focus
Capabilities = innovation

The Future:

It’s going to be a zombie apocolypse. Seriously, it won’t be about a few devices, but a plethora of them. Because of this, Woblewski is part of a movement known as Future Friendly investigating how we think about devices in the future.

Today’s web is exciting and scary.

Beyond The Polar Bear, Michael Atherton

Who knew domain modelling could be so interesting? Perhaps when you have arguably one of the richest, and thus most complicated, datasets around to deal with. Atherton explained the process the BBC has gone through to clean up and standardise their vast web offerings, while still allowing for the customisation formerly done with microsites.

He used the concept of Disneyland’s domain system (everything, from theme parks to hotel food, is linked together in rich and non-hierarchical ways) as a great analogy, and one that can be reflected in some of hteir projects.

A Web of Connected Things

A Web of Connected Things

  • The BBC Programming System paid a lot of attention to URIs. While Tim Berners-Lee tells us that they should be hackable, permanent, and persistent , the impermanence of the Beeb’s media (a series may change numbering when it moves overseas, have a mini-series extended to a full one, and may jump channel or even medium), means that they have have had to sacrifice hackability for the other two (for example, BBC1 programme Sherlock has the URI ).
  • BBC Food challenged what audience and medium you design for. The site wasn’t doing as well in Google as might be expected, because there was a lot of churn of content: chefs retain copyright of their recipes, so they tended to come and go, thus confusing Google (or as Atherton amusingly calls it, “splitting the Google juice”). However, they realised that people are more interested in finding a type of dish (entering via Google) rather than ‘the dish’, so came up with the idea of dish as canonical work — while recipes may come and go, the page stays. This, in combination to paying a lot of attention to mobile display (as might be expected, most pageviews were on a mobile device, presumably as people were in the kitchen) led to traffic doubling from 650 thousand to 1.3m, and much higher ratings on Google.
  • BBC Nature is about unlocking and exploring, but also about finding way to managed vast arrays of content. Rather than have to create thousands of pages that might never be seen, the BBC team pulled information on animals from Wikipedia — and had their wildlife experts edit the Wikipedia articles if they weren’t up to par, thus improving the quality of the BBC site and the general information available on the web.

The overall takeaway was the importance of domain modelling (he called Domain Driven Design by Eric Evans “his new bible”):

You need to be able to define the thing to be able to point at it!

and that a shared model + shared language + shared understanding = consistent UX. In other words, the model should be consistent enough that anyone in your team can draw it.

And the web is changing:

design for a world where Google is your homepage, Wikipedia is your CMS, and robots are your users.

Users, Experience, and Beyond, Eric Reiss

Eric Reiss led the audience through a behaviour-centred framework that his team at FatDUX use.

The need matrix is a way to consider the different attributes to any experience (Attitude, Expectation, Schedule, Environment, Origin). Your behaviour when booking a trip is very different from calling the tax office!

Reiss stepped us through how to use the framework:

  1. Do customer research
  2. Create mental models (ala Indy Young)
  3. Write scenarios
  4. Tag the interactions throughout the process
  5. Create snapshots
  6. Do quantitative analysis. Reiss suggests weighting using 1-3 for primary, secondary, and passive interactions, and then coding responses from -3 to +3. The negative answers are important as they can be used to easily show problems.

He urged us to:

  • Understand the ergonomics of need for key scenarios
  • Consider user experience as the sum of a series of interactions
  • Write and chart a scenario to identify, quantify, and prioritse key interactions (snapshots)
  • Go out and make the world a better place.

The Information Architecture of Culture, Martin Belam

EuroIA veteran (we reported on his previous years’ talk) and conference reporter Martin Belam gave a refreshingly frank discussions of the bumpy road to implementing APIs at the Guardian.

One of the key ares the Guardian is looking at is how to move discussions beyond a small hallowed circle of critics and reporters. They’re keen to help “mutualise” the relationship between newspaper and their arts audience (as in with mutual funds, find a way for both audience and paper to be supported, a bit concern these days in the eras of paper closures and paywalls)

When you have a bigger audience, where do you hang these conversations?

They had some success with pulling in external content with a music project that pulled information from MusicBrains, and so decided to create a larger system with books. However, it wasn’t quite so simple.


  • Didn’t get API right first time up. The domain model for books is difficult, as ISBNs can change for editions, and are added to CDs, calendars, and even card displays.
  • Ignored previous experience Person with library experience said that tried tagging with ISBNs in the past and found it difficult as they’re physical. It turned out that that was still true.
  • Too few devs in too big a team. “Fifteen people can change their minds far more quickly than three people can build”
  • Got obsessed with design details (45 minute discussions about start rating details!)
  • Went for ‘big bang launch’. As there were still a few bugs to be ironed out, this damped a lot of interest in the product.

However, they did have some successes:

  • Used an Objects/Properties/Actions Map. This also helped with later mobile first strategy.
  • Giving the developers a chance to be creative again. The team was sent to SXSW11, and the developers made an app that scraped information on band members and made a site.  However, the big issue this brought up was the content’s quality and uniqueness (or lack of in both cases). Guardian readers balked at the content often not being up to the site’s usual standards … and Google penalised the entire site as it had a huge number of pages that was scraped rather than original content. They had to deindex the pages, emphasise the site wasn’t usual Guardian content, and create site specific ways to search it.)

He have the following tips:

  1. Know what is important. What is the goal you’re trying to attain? And is new technology the answer?
  2. ISBNs are evil. [“F**king evil . Worse than mini-bars, which are evil as they put terribly overpriced alcohol in your hotel room”].
  3. Trust good developers. Engaged developers can be the most valuable asset on a project. “Coding is actually really creative. Don’t ruin your developers with boring and uninspired briefs.”
  4. Listen to all of the team. Job titles and age don’t matter if they have the right answers or knowledge.
  5. Get the model right. Lists (rather than ‘pages’ or ‘fronts’) were the key to success. The Guardian got everyone together to create a really strong framework. A good model makes the rest easy.

And a bit of fun: the Guardian was established in 1821. The developers used the API to serve up the news as it would have looked on the original broadsheet. “The developers were particularly proud of ‘Entweet this'”

Out of the Echo Chamber, into the Fire Jason Mesut

The day came to an impassioned end as Jason Mesut played truth or dare with the UX industry. Having been in the field for over a decade, he’s worried with a lot of the precedents and dogmas around.

The Dogmas:

  • Mobile first: is confused as an always use strategy. He poined out that Luke’s view is balanced in calling it ‘a way’, but that others (such as clients) quote it as gospel without the nuance.
  • The open web: open source and open web are not the only way. While our developer friends care about it, to be honest those in business don’t, so we need to maintain a critical distance. Sometimes proprietary is better!
  • Agile: There is more to Agile UX than Sprint OS and Sprints ahead, this is only one aspect, doesn’t always work
  • Service Design: most service design and design is a lot of talk and corporate entertainment. “All fart and no shit”.
  • Responsive design: an old argument in new clothes (fixed vs fluid, separate access etc). Technology changes rapidly.
  • Breaking down silos: this is naive. Organisations are complex, people better in small  groups, change takes too long. “You can’t reorganise people because the website is crap”

One of his most powerful statements is that UX is eating itself with its insular, rockstar centred culture. The picture (for those who’ve seen or at least know of the movie) is priceless:

The Human Centipede of UX Dogma

That said, it wasn’t all fire and brimstone. He challenged IA and UX people to map what they are and where they want to go, and provided a very useful way to show it:

Map Your Own Adventure: What Type of UXers Are You/Want to Be?

Map Your Own Adventure: What Type of UXers Are You/Want to Be?

His some of his key truths and dares:


  • There is no universal truth in UX. Anyone who says otherwise is a liar
  • Sometimes the quietest people have the best things to say. As someone commented on a post by Andy Budd “the people doing the best work are people we’ve never heard of”. They’re not on the conference circuit or hawking a book, they’re just doing their job, and doing it well.
  • There are no silver bullets in UX. Repeating them can weakens us. We need multiple weapons, and to know when and how to use them.
  • Most UX people don”t articulate what they do and how they are different from others. If we don’t know, how will others? Already we have business and marketing taking on design thinking since no one is saying otherwise.
  • Our bubble will burst unless we stamp out the greedy pretenders. There are too many freelancers with scant experience getting too much money and not hanging around. How about paying permanent staff more, and calling out the people who don’t know what they’re talking about?
  • There are more non-UCD success stories than UCD success stories. Many business people are successful without UCD, so don’t push it.
  • Most UX people are not built for strategy. UX people are nice. Do you really want to be like Jobs, Trump, or Sugar?


  • Don’t tweet soundbites. Or in other words, don’t take comments out of context.
  • Critique conference talks.
  • Call bullshit on celebrity UX rockstars. Just because they’re charismatic and entertaining doesn’t mean they know what they’re talking about.
  • Map your UX shape and focus your future. Designer, know thyself.
  • Get into the heads of others
  • Try more designing, less researching. UX can get obsessed with research to the detriment of the actual product (just as marketing will focus on the marketing of it). The devil is in the design details, or in other words, execution.
  • Commit to strategy, or focus on UX. You can’t do both.
  • Share opinions & be prepared to change. Be passionate, but flexible. Don Norman has changed his opinion several times, but at least he has one to change

Image CC by George M. Groutas

Vicky Teinaki

An England-based Kiwi, Vicky is doing a PhD at Northumbria University into how designers can better talk about touch and products. When not researching or keeping Johnny Holland running, she does the odd bit of web development, pretends her TV licence money goes only to Steven Moffatt shows, and tweets prolifically about all of the above as @vickytnz.

2 comments on this article

  1. Pingback: Euro IA 2011 » Zeix-Blog

  2. Pingback: Truth or Dare: My Response to Jason Mesut's EuroIA 2011 talk | Redux'd