Day two of EuroIA had speakers hailing from Italy (actually they were pretty well represented) to South Africa. And amongst the talks we had not one but two different rounds of IA bingo. Who said IA was boring?
People coming in to the second day of talks were greeted with what were called “BS Bingo” cards. While it wasn’t quite a drinking game (yes, there’s an app for that), it did have a prize — a free ticket to EuroIA 2012, won by Jeroen Grit
Extending the Storytelling, Boon Sheridan
It takes a great talk to drag people out of bed and been in a conference room at 9am on the second day of a conference, but Boon Sheridan did that, with a packed room to hear his talk on blending IA and content strategy.
He’s noticed that the word ‘deliverables’ have become a dirty word in IA and UX (“don’t worry about deliverables, just do the work”). However, he feels that there are many good elements of deliverables that are useful for a project. Therefore, he proposes using blended deliverables.
Their benefits are:
- Strategic approach: they’re not throwaway documents as they’re meant to encapsulate the big picture
- Tactical focus: they help you get sign off!
- Perfect brainstorming The documents are open for deliberation and easy to make.
- Ideal for collaboration
He suggests implementing them through the following forms
- Audience personas: What comes before personas: Who are we speaking to? Key messages? What content do they need? Where? This is were audience personas come in. They’re your widest audience that you want to reach, and help you keep perspective
- Content flows: Where is your content going to come from? How are you going to host it? What functionality is needed? These are a great way to identify problems up front/periodically
- Building on it: the key concept behind blended deliverables is that they’re finished but changeable. You should be able to sign off your documents as done, but then be able to review them at a later point and amend them if the system has changed. They give the clarity and aligning nature of deliverables without the pressure of them to be ‘finished’.
The other key concept that Sheridan brought up was the idea of designing for disagreement. (Apparently it was an aside in a speech by Kevin Cheng about Twitter’s design process, even though he doesn’t even remember saying it). The idea behind it is that many of the problems in a design process happen because stakeholders think they’re all agreeing to the same thing when in fact they all have different ideas. Creating deliverables that actively cause people to disagree can help bring up any ambiguity between stakeholders and get them all on the same page.
Pervasive IA for the Sentient City, Andre Resmini & Luca Rosati
Resmini and Rosati gave a talk on the IA of cities based on their recent book Pervasive Information Architecture. Above all, they suggested that we need to consider an information layer in the physical environment, and as a living, resiliant ecosystem.
Rosati (one of many at this conference) referenced Marcia Bates’ methods for information seeking as an important way of understanding a structure for a city:
One thing we need to do is change our thinking from top-down strategies to bottom-up ones.
- Top-down is ‘traditional IA’. Examples of this in cities are map and wayfinding systems,
- Bottom-up is a basic system that is adapted. Twitter is one such example: a ‘stupid’ technology that has been adapted over time. In the physical world, these can be found as desire paths: a result of least effort and time.
Places are used as wax. Places are the site of a mnemonic palimpset
Resmini believes that a lot of the architecture/town planning research was done in the 60s/70s and hasn’t advanced much since then. (He’s also not that much of a fan of Christopher Alexander’s design patterns).
In terms of resiliance and the physical world, this is explored well in Stuart Brand’s How Buildings Learn. Here, it’s shown how change in building happens at different speeds (the outsides slowly decay, which you may redo your kitchen once a decade, your decor every few years and move around your furniture every few months).
Given that neither Resmini and Rosati are town planners (they are an architect and linguist respectively) someone from the audience did ask the inevitable question: isn’t that their job? The work of the IA isn’t seen as being one that takes over from a town planner, but instead collaborates with them to ensure that various offerings such as signage and communication systems are appropriate.
Designing Interactions that Help Customers in Decision Making, Stefano Bussolo
Bussolo’s talk was a breakneck (if beautifully cadenced) tour of the world of neuroscience and its relation to decision-making. His key point was that we should be thinking about chooseability rather than findability, and that different types of users deal with choice in different ways.
When choosing a product, consumers fall into one of three categories:
- they know exactly what they want,
- know their preferences for a product,
- or only know the attributes.
More interestingly, while those two types who don’t know exactly what they want don’t like too many options (decision paralysis or fatigue), those who do like seeing all the others.
So, how do we understand and deal with this? Bussolo explained that it all comes down to heuristics. While we would like to always make decisions logically as it’s highly accurate, it also takes a lot of cognitive effort. Heuristics gives us more bang for our buck by being relatively accurate as well for far less effort.
The heuristic strategies he suggested:
- Elimination by Aspects (EBA): cut out what you don’t want.
- Majority of Confirming Dimensions (MCD) — e.g. opening up browser tabs of all the options for a car and comparing them
- Satisfying Heuristics (SAT): take the first satisfactory alternative e.g. finding a carpark or searching on Google
- Lexicographical Heuristics (LEX): sorting via terms.
- Equal Weight Mean: aggregating set of scores into a whole e.g. Trip Advisor ratings for cleanliness etc and the total score.
- Faceted Information: letting people drill down. Suggests looking at Peter Boersma’s EuroIA 2010 talk on this topic.
- Facilitate both rational and heuristic decision strategies
- Divide the processes of decision making
- Design for different users (those who are decided, and those thinking of preferences and attributes)
- Give users some external aid (external cognition, suggestions)
Understanding the Nature of Resistance, Alla Zollers
Welcome to therapy. That was how Zollers introduced her session, and it was all about feelings: namely now to identify and deal with resistance from clients. As she paraphrased from Star Trek, resistance is not futile, but natural and a result of emotional processes that we can’t ignore.
- Acknowledge it: if you have a feeling somethings wrong, you’re probably right (and it’s probably something far bigger than you know about).
- Identify it: you need to talk this through with your client, in a non-threatening (i.e. therapy-talk) kind of way e.g. “you seem… I feel…” (One person in the audience pointed out that “you seem…” could still be considered aggressive, but as Zollers pointed out, if it helps bring to light the underlying problem).
- Wait: silence is golden for bringing out the truth if you’re prepared to not fill up the space with chat.
Teaching Design Thinking, Jason Hobbs & Terrence Fenn
Based of their research (a project that was funded by the IA Institute) Fenn and Hobbs talked about how design education needs to change to accommodate the changes in design, with a specifically South African perspective.
Fenn asked: as design educators, are we applying the rules of UCD to design education? If Don Norman says that if someone can’t use a product, it’s most likely the product’s fault, then that suggests that failures in design education is the design education system’s fault.
Their project is specifically focused around the idea of indeterminacy. These days, designers are more than ever expected to be able to do a range of things. But if you send your interaction design students out to investigate transport, and they find that the problem is signage, do you let them do graphic design or force them to do a website? How do you train — and encourage — them to be able to work in areas where they don’t have core skills?
Fen pointed out that while design thinking is promoted by IDEO etc, it’s locked behind copyright and thus difficult to use in education.
So how do we teach these design wicked problems? As it turns out, IA could be a useful model.
- IAs deal with wicked problems (complex problems, multiple users, huge amount of data) every day.
- It’s very normal for us as IAs to start unpacking all of these aspects “scuplting with the data”, and use qualitative research
- IA problems are usually informational and especially digital, but our exploring problems usually leads to solutions beyond this context.
What was particularly fascinating about this talk was also it’s unique cultural perspective. South Africa’s liberation from apartheid in the 90s had some obvious repercussions, but also others that outsiders might not think about. Fenn and Hobbs highlighted how it has affected the local transport system : a formerly highly structured and thus easy to manage system (different races travelled on different buses and at designated times) has struggled to cope with the change in the overarching system around it. Similarly, when they used examples about designing out crime, it was easy to realise the complexities that they have to deal with ranging from the police to communication.
Fill in the IA Gap, Mags Hanley
Industry veteran (and inspiration to many, including fellow speakers) Mags Hanley finished the day both ruminating on the mood of the conference, and the changes to the industry she’d noted since recently coming back.
I’m proud to call myself an information architect. Not an interaction designer, not a user experience designer … an IA.
Hanley still feels a lot of pride for and in the industry, but feels that IA had both narrowed and forgotten to teach a lot of its fundamentals. She told us that Louis Rosenfeld had confessed that he had workshop slides that were over a decade old … but she realised he needed them as people didn’t know what many core concepts of IA were.
And to test us all on whether we did actually know all our fundamentals, she got us to play IA Bingo (not related to the earlier BS Bingo from earlier today):
(Eric Reiss may or may not have won).
She suggests all IAs should be able to do the following four things:
- List organisation structures & be able to consciously choose “ya can’t defend it unless ya can choose”
- Create models of the structures — navigation model, app model, data model — without information in it and be able to show content moving back/forth. One of the key phrases from this conference was around domain models. It’s clear that you need to be able to understand these completely when proposing a solution.
- Understand deep IA — content objects, CVs and semantic web — at least enough to hire the right person. Deep IA may be a strange and specialised area of IA (like typography?) it’s one you can’t afford to not understand at least a little.
- Understand how people seek out info (this is different from usability/UX). There is a whole field of research devoted to information seeking in the real world (for example, how women look for information in doctor’s offices). Look for it and draw out skills.
She urged us to tell our stories — junior IAs know methods, but IA provides value around methods — and shared her own from the BBC. She admits that the IA team back in 2002 “lived in their own little bubble”, and she learnt the hard way that your users may not always use your meticulously designed prototypes (BBC music reporters shunned the complicated music content type and just hacked the general one as its output was more pretty) unless there’s a reason (provincial rugby reporters started inputting their game results into the until-then neglected sporting post types as it automatically promoted their page on the website).
We also need to keep doing IA user research (which is more than just card sorting). Other ways of doing research include information seeking (look at search log etc.), hierarchy of infomation, facets, and task flows.
Conversely, be broad in our profession. We need to understand domains such as UX and cognitive psychology to do our job properly.
Be the glue with editors, business, stakeholders, and designers. We’re the people that understand what it’s supposed to be.
Much like the night before, the talks finished up with a call to action:
- Get IA equal standing as a UX field of practice. There’s too much talk about IA disappearing or being a part of UX. It is different.
- Data visualisation: we need to know how to do data visualisations for interaction (and getting decisions made).
- Make IA cool again. As Hanley admitted: “I was going to be a medical librarian … then I found out I could work with computers.”
- Find our voice — blog, talk, tweet, and take away — read, review projects, practice IA, and above all speak about it.
Image CC by Pedros