Getting UX Done

It’s here! bit.ly/GettingUXDone.

I’m currently at UXLibs 7, so this is good timing. My colleague Paul Harding and I presented at the last User Experience in Libraries conference, in 2022, and then wrote the paper up as a chapter in the 2023 UXLibs Yearbook, which I’d highly recommend you buy for your library… The link at the top of the page is to an Open Access copy of our chapter in the White Rose repository, and you can see lots of other OA stuff (including previous UXLibs year book pieces) on my Publications page.

Ask not what your organisation can do for UX; ask what UX can do for your organisation.

The title of the chapter refers to our central thesis, that we need to spend less time talking about the concept of UX in our organisations, and more time focusing on the results. Sell by doing. Save the library money and give UX the credit afterwards…

We have lots of tips in there about how to truly embed UX, and a case study about the YorSearch Improvement Project, a UX-led piece of work to usher in the new UI for our catalogue. That particular piece of work had a fairly profound ongoing effect on the way we approach User Experience at York, and the ‘three rounds of five’ approach to the fieldwork (detailed in the article) is one we use all the time, including in the recent Website Improvement Project.

Anyway, have a read of the article and let me know if you have questions or would like to discuss any of it in more detail.

---

A structured introduction to UX in libraries is a whole page of this website, recently revamped, and it displays posts about UX work we’ve done at York. It also features a new quote from Andy Priestner (UXLibs founder and chair) about what UX really is:

“At their core, UX methods are all about making sure that the user is at the centre of what you do. They are about accepting that library services should not be built around staff agendas, convenience, assumptions or gut feelings, but instead around what your users really need and really do. This naturally requires that you regularly connect with, empathise with, involve, and understand your users.

A myriad of UX research techniques are available to explore user activities and needs: attitudinal techniques such as semi-structured interviews, cognitive mapping and photo studies; and behavioural techniques such as observation, behavioural mapping and usability testing. These approaches should always be conducted in balance so you are researching both what people say they do and think they need (attitudinal), and what they actually do and really need (behavioural).

The UX Process begins with user research, continues with data analysis and concludes with the ideation and testing of prototypes. The end result should be investment in more valuable and relevant user-centred services and initiatives, which have been generated from, and tested with, your users.”

— Andy Priestner, May 2023

UXLibs II: This Time It's Political

I'll never forget #UXLibs, for reasons beyond the excellent content and camaraderie pic.twitter.com/u6KQfo10os

— Emma Thompson (@LibraryEmma) June 26, 2016

At 9am on Day 2 of the UXLibs II conference, 154 information professionals sat in a large room feeling collectively desolate. I don’t want to be glib or melodramatic but the feeling of communal sadness at what had happened in the EU Referendum overnight felt to me akin to grief, like someone close to the conference had actually died the night before.

Was there anyone present who voted Leave? Possibly. But it seemed everyone was devastated. There were tears. UXLibs is, as Library Conferences go, relatively diverse (although it's still something we need to work on), not least because well over a third of the delegates - 60 this time around - are from outside England. Our North American and Singaporean friends felt our pain, our European friends were sad our country had chosen to leave them, and for the Brits it was already clear what an omnishambles the vote had caused.

The committee had met for an early breakfast to process how we should proceed. We agreed on two things: first that however we all felt, organisers and delegates had to deliver the best possible conference experience in the circumstances; and second that this was not time for neutrality. (In fact I was talking to Lawrie Phipps from JISC a little later that morning and we agreed that perhaps if so many libraries and educational institutions generally weren’t so neutral by habit, people might have a better idea of when they were being systematically lied to by politicians.) Conference Chair Andy Priestner was due to open the conference: say what you want to say, don’t hold back, we agreed. There had been a lot of jokes the day before - humour is an important part of the UXLibs conference as it leads to informality, which in turn most often leads to better and deeper communication, proper relationships – but there would be no attempt at making light of this. Don’t gloss over it. Don’t be glib. Don’t be neutral. But do be political.

So he was. You can read Andy’s reflections on his opening address here, and this is what he said:

“Today is not a good day.

I’ve worried for several months about this moment in case unthinkably it might go the way it has gone. I am devastated. Everyone I speak to is devastated. This is a victory for fear, hate and stupidity.

But as Donna said yesterday when describing her experiences in Northern Ireland – ethnographers have to get on with it. WE have to get on with it. Perhaps it’s a good thing that we will all have less time to dwell on what has just happened. Perhaps it’s good that we’ll be busy.

What I do know for a fact is that we have to be kind to each other today however we might feel. Let there be hugs. Let there be understanding.

For me one of the most precious things about UXLibs is the networking and sharing we enjoy from beyond the UK. The collaboration across countries, the realisation that despite the different languages, cultures and traditions that we are all the same and can learn so much from each other.

But it’s too soon to be cheerful. It’s too soon for silver linings.

Today is not a good day.’”

I was proud of him.

And then Day 2 happened, and I was proud of EVERYONE. What an amazing group of people. Shelley Gullikson put it like this:

“Last year I said that UXLibs was the best conference I’d ever been to. UXLibs II feels like it might be the best community I’ve ever belonged to.”

Everyone found a way to help each other, support each other, make each other laugh, and work together – after Lawrie’s keynote the first thing on the agenda was the Team Challenge so no one could spend any time sitting in dejected silence, there was too much to do… Collectively everyone not just got through the day but made it brilliant. It wasn’t a good day overall – a good conference doesn’t transcend political and socio-economic catastrophe. But it was the best day it could possibly be.

---

I attended the first UXLibs conference in 2015 and I was blown away by it. It felt like the organising committee had started from scratch, as if there were no legacy of how a conference should be, and designed it from the ground up. They kept some elements, the ones that work most, and replaced others with new and more engaging things, especially the Team Challenge. It was the best conference I’ve ever attended.

The follow up, UXLibs II, had something of an advocacy theme – as I put it in the conference, if UXLibs I was ‘how do we do UX?’ then UXLibs II was ‘how do we actually make it happen?’. As communication and marketing is something I do a lot of work around and, as Andy so kindly put it, he wanted to see if we’d actually get on and not hate each other if we worked together, I was invited to join he and Matt Borg as the main organising committee (although we had a huge amount of input from several other people in planning the event). This was in September last year; Andy and Matt had already been planning for a while and by October we had our first provisional programme in place.

Andy and Matt...

Matt and me...

I find organising events approximately three trillion times more stressful than speaking at them, and hadn’t got fully involved in putting on a conference since 2011 when I swore ‘never again’. But I couldn’t resist the chance to work with Andy and Matt because we are pretty much on the same page about a lot of things, but disagree on a lot of the details, which makes for an interesting and productive working arrangement. So, around 400 emails later, a couple of face-to-face meetings later, many online meetings using Google Hangouts later, we were in thestudio, Manchester for the event itself. At the end of the two days, despite the dark cloud of Brexit hanging over us, everyone seemed exhausted but fulfilled. We’d built the event around the community and what that community said it needed, and I think it worked. It’s a great community and I felt excited to be part of it – challenged, stimulated, and I’d echo the delegate who came up to me at the end and said she’d never laughed so much at a conference: it was FUN.

Several things made this conference different, for me, apart from just the content. There's the fact that all the delegates have to be active participants (they were 'doing doing' as I put it, somewhat to my own surprise and certainly to my own mortification, when introducing the team challenge), there's the mixture of keynotes, workshops, delegate talks and team challenge, there's the informality and fun but with the Code of Conduct to ensure people can work together appropriately, there's the fact we individually emailed 100 delegates from UXLibs I to find out what their challenges were so we could help shape the conference, there's the fact that 150 people got to choose which workshops and papers they attended, there's the blind reviewing process for accepting papers, there's the scoring system for the best paper prize that was far more complicated than 'highest number of votes' because different papers were seen by different sizes of audience... There's the fact there's less fracture and division than in most conferences: I truly feel we're moving forward together as a UX in Libraries community. There's the fact that the venue was not only excellent but had a trainset running around near its ceiling that you could stop and start by tweeting at it!

Turns out it's quite easy to avoid All Male Panel. What you do, conference organisers, is you don't put all males on the panel. (Pic by @GeekEmilia)

There's the fact that Matt made completely bespoke badges with individual timetables for all 154 attendees! I can't tell you what mine said (let's say Matt was experiencing some remorse at saying he'd do the badges by the time he got to 'P') but so many people commented on the good-luck messages he put in to all presenters for their slots...

For all who think THE UXLibs Conference is a jolly :) #UXLibs pic.twitter.com/KaThDgLQXQ

— Sonia Wilson (@sonia_wilson1) June 23, 2016

This is very cute :-) #uxlibs pic.twitter.com/CrpH3Zyxcb

— Rachel Claire Walker (@phys_librarian) June 23, 2016

So it was pretty great, overall, despite everything. Thank you everyone invovled.

UXLibs III planning has already begun.

#UxLibs 4: Ethnography You Can Try At Home

Tweet

Post 4 about the UXLibs Conference. This one is about the actual ethnography we went there to learn about. If you're interested, there's a post about the conference organisation and format here, plus a post about the Usability keynote here, and a post about the Ethnography and Design keynotes here.

But first, the Conference badge. This genuinely was the best user experience I've ever had from a badge. The attention to detail was symptomatic of the whole approach to the conference, I felt - it was designed to be a good experience. Timetable, maps, where your accommodation is, which parallel sessions you were in, name, institution, Twitter name, the conference hashtag AND it was reversible so it could face either way and still be useful. All little things but together they made a real difference.

Badge magique!

Before attending this conference my knowledge of ethnography was limited to having read about it. I'd not done any. I knew it was useful, and I knew why I felt we needed to make use of it - but if someone had caught me by surprise with the question 'So what IS ethnography in libraries?' I would have crumbled... But not anymore!

Here are some of the techniques we learned about ethnography - in the final post next time I'll cover the process of designing a service or product off the back of what you learn.

Cognitive Mapping

I've put this one first because it's my favourite. I didn't actually get to do this one (on Team Space Grey we all went to different workshops, and mine was the Observation one), but I really loved the results of it. If you're reading this thinking "I did do this and you've misunderstood it" then correct me in a comment!

Cognitive Mapping is in essence asking your subject (student, staff or whoever) to draw a map of the library - or, ideally, of their wider learning landscape - in order to understand how they perceive the space, what they actually use, what they value and see as most important and so on. Often the subject is asked to change colour of pens every 2 minutes, for a total of 3 colours over 6 minutes, so you can later see which order the items were drawn in, an indicator of their importance.

Here's a map which Team Space Grey got one of our students to draw (thanks for sending me this Kristin!):

We got two maps overall - what was particularly interesting for us is that one of them mentioned the basement area of the library as being dark and scary, and another didn't put the basement on the map at all! This was a big part of the idea we later pitched (which I'll cover in the next post).

There are some examples from Donna Lanclos of congitive maps, here.

What we didn't really have a chance to do at the conference itself was code the findings. To quote library anthropologist Andrew Asher - one of only two people to hold that position in libraries, the other being Donna Lanclos - here's how you might go about it:

“Coding these images basically involves counting the elements drawn in order to construct two indexes: a identification index, which is the number of times that an element is drawn divided by the total number of individuals participating (i.e. the percentage of the time the element occurs), and representativeness index, which is the number of times an element is drawn divided by the number of times that category of element is drawn (e.g. the number of times a study room on the first floor is drawn divided by the number of times all study rooms are drawn) (See Colette Cauvin’s “Cognitive and cartographic representations : towards a comprehensive approach” for additional discussion). I also constructed a temporal index for each element by coding the three colors in order (1 = Blue, 2 = Green, 3 = Red) and calculating the mean value for each element (you could do more complicated things by combining the indexes if you are mathematically inclined, however, I’ve found that these three get at most questions).”

— Andrew Asher | www.andrewasher.net

Observation

Observation is exactly what it sounds like - you occupy a space in the library or wider campus and you note exactly what's going on and how people use the space. In our workshop we were encouraged to focus on the location itself, the pathways users took through it, the interactions they had (both with other people and the objects and machinery), and the tools they used.

I drew a heat-map, or rather a behavioural map, using the Paper app on my iPad - tracing the lines of every single user who came through the library over a set period of time. You quickly get a picture of how people move through the space. (Not a single person went down to the basement...)

I can't draw at the best of times, and this was done using my finger - so my handwriting which is usually terrible is now completely unusable, and the weird mass of colours near the bottom left is me repeatedly trying to get the action right to bring up the colour selection tool - but I really enjoyed making this!

The idea of course is to build an understanding of how the space is used, and then adapt the space to better suit the reality for the users. For example, at Judge Business School they moved the digital display screens so they were easily viewable from the 'desire line' - the path most students took through the library. There's more on this in slides 13 - 19 of Andy Priestner's deck here. (That presentation is also a great introduction to ethnography generally.)

Interviews

The key thing about interviews in this context (as opposed to the more traditional focus group methods libraries often use) is asking completely open ended questions. I mentioned this in my post about Donna's keynote: if you ask how a student writes an essay, you get a potentially more illuminating answer than if you ask 'what library resources do you use for an essay' or 'how do you use the library'. All of this makes the data you get messy and harder to process, but ultimately in my view more worthwhile.

Touchstone Tours

Just as the Cognitive Mapping gets the user to draw the map rather than view the library's own, Touchstone Tours ask the user to take the librarian on a tour of the library, rather than the other way around. By the user telling you how the tools, systems, building and spaces work, you get to truly understand how it feels to be a user without the in-built knowledge and understanding we have as library staff. (Our tour subjects didn't take us down to the basement because, for them, it really didn't matter.)

Love-letter / break-up letter

I'm reluctant to try and describe this one in too much detail as I wasn't in the relevant session and we didn't end up using it in my team. But it involves writing a love-letter to the library if you like it, or a break-up letter if you don't. When it works well it gets to the heart of the user experience, and helps understand the emotions Matt Reidsma talked about in his Keynote on usability.

There is more info on interview techniques, touchstone tours, and the love letters, in Georgina's workshop slides.

Anything I've forgotten, let me know!

#UXLibs 3 | The Art of the Keynote: Donna Lanclos and Paul-Jervis Heath

Tweet

After yesterday's post about Matthew Reidsma's Keynote on Day 3 of the User Experience in Libraries Conference, here's an account of Donna and Paul's keynotes from Days 1 and 2 respectively.

Donna Lanclos | Once Upon a Time: a story about ethnography and possibility in libraries

Donna is completely passionate about what she does, it's infectious, and she's a very powerful communicator. There were times listening to this when I felt like I was hearing a better informed and more articulate version of my own interior monologue about certain things I get frustrated about... I agreed the heck out of it.

But it challenged me too, which is part of the art of the keynote. Contrast is what makes a presentation engaging. We like the feeling of hearing what we know and agree with, and are made uncomfortable by hearing things which challenge us or that we don't agree with. This push and pull is what moves a great presentation forward. Too much of one and not enough of the other diminishes the potential impact of a presentation. If Donna had just come out and said a bunch of stuff I already knew and agreed with, I would have enjoyed it a lot but it wouldn't have stayed with me. A keynote must, on some level, provoke (but not for provocation's sake).

One of the particular things I was aware of Donna doing and aware that I don't do enough (or basically at all) is allowing us the audience to connect up the dots for ourselves, so we were learning rather than just being told. The use of stories, analogies, and flat-out asking questions, helped us to provide our own answers, which is much more powerful and long-lasting than just hearing people say the answer off the bat.

She also used something I mention in presentation skills sessions but have never done myself - The Black Screen! The idea of this is it focuses the attention on the speaker by, in contrasts to the normal picture-heavy slides around it, being completely blank. It's a device which says, okay, let's have a serious conversation now, for which I want your full attention. One person didn't understand this and actually called out 'Can we have the slides back please!' and when Donna said 'what?' they changed it to 'Is there anything on the slides..?' Donna said 'no'. She then stopped the poor guy feeling too awkward by saying 'you talkin' to me?' in mock indignance / aggression, which made the moment pass less painfully and more humorously than it might have done...

There were a few really key themes that emerged for me.

• Simply asking people what they think, and what they want, is a pretty unreliable process. This has always troubled me. We KNOW it doesn't work (people often don't know what they want, and they often simply take the path of least resistance through surveys and focus groups) yet we do it anyway and base decisions and even policies off the back of it. This method of gathering feedback is not enough on its own. We need find out what our users understand, rather than what they say they want. 

• So ask open ended questions. As soon as you say 'what resources do you use when you're writing your essay?' you're limiting what you can learn from the interaction. It's like asking someone to take you to a destination and then putting them on rails which only go one way. Ask them 'what does it look like when you write an essay? Talk me through the whole thing'. Then you learn where the library fits in with a larger process or life-cycle. You learn what the true user experience is. You (in a totally non-creepy way!) follow the user home.

• Ethnography is messy and uncertain. You need institutional support for that uncertainty. You need to try things out, try and provide a better experience but it's ITERATIVE. Don't wait for all the answers, just go for it. Fail. Then do it again, more effectively.

• If we have to show people how to use the resources - in Donna's words, spend time telling people how to click a link on our websites - then we're doing it all wrong. The User Experience has failed. I am painfully aware of how much time I spend doing this, in one form or another. I once made a VIDEO about how to use the catalogue. Aaaargh.

• Donna asked us to 'dispense with the idea that all important things in education are measurable'. I don't think, sadly, we can do this and still be as accountable for our impact as we need to be in this day and age. My hope is that if we ethnographise the heck out of everything, we will actually, possibly for the first time, truly understand the student experience, our role in it, and how to make it better. As a result of what we do about it, all the countable measures (like the NSS scores and the LibQual Surveys etc) will go up anyhow. At the moment I feel like I spend a lot of time trying very specifically to address certain issues which will mean the things we measure are more favourably assessed. That's all wrong. The tail is wagging the dog. The results of the stuff we count should make our superiors happy as a BY-PRODUCT of us truly understanding our users.

• You need to know enough to ask good questions, but be happy to admit you don't know much more than that. It's really the opposite of 'We're thinking of introducing X, Y and Z. Which would you prefer?'

Donna asks a lot of us. I hope we can take what we learned from her and start to deliver.

Paul-Jervis Heath | Transforming insights into services

I really, really enjoyed Paul's talk because it took something which I was aware of but which was ambiguous to me, and broke it down into smaller concepts I could understand. We all know roughly what 'design' is, but what are the structures that facilitate its success? Paul told us, so now we know.

So here are the rules of design, as I took them from Paul's talk:

• Brainstorming is but one method of generating ideas, and not a particularly good one at that. In fact, people working alone come with twice as many ideas as groups doing brainstorming!

• Using a 'yes and' method to build on group ideas - making things better rather than shooting them down - is the way to work together. 20% more ideas are generated when people constructively critique and discuss ideas

• You need both divergent and convergent thinking. It's a process of widening out the scope of ideas, then focusing in on where the common ground lies and developing something from there

• Implement by making. There's no eureka moment, so don't sit around waiting for inspiration to strike. Go out there and DO stuff

• You can get fixated too early on one idea, and if you put all your energies into pursuing it, you might be missing out on better ideas that never get a chance to come to fruition

• People's goals are key. Mostly they will stay the same, even if the tasks they need to complete to accomplish them change

• Students have a triangle between spaces they habitually use. Anything outside that triangle feels inconvenient and far away, even if technically it isn't...

• If you can't draw your idea, it's not worth having (I personally didn't buy in to this one, but the rest I found really useful)

---

Header pic by Jamie Tilley. See the original on jtilleyphotographic on Flickr, here.

#UXLibs 2 | The Art of the Keynote: Matthew Reidsma

Tweet

Don't think I've been to a conference before where all speakers not only have really great things to say but know how to say it #UXLibs

— JamesA-E (@JamesAE) March 19, 2015

My second and third posts about the User Experience in Libraries conference are about the Keynote speeches. (Email subscribers, there's a load of embedded content in this one - here's the link if you want to read it online.)

There were three, and they were all brilliant. I found them not only very informative (as someone new to UX) but also fascinating and educational in their delivery (as someone who teaches presentation skills workshops). I want to cover both aspects here.

A keynote is very different to a regular conference talk. It has a different role. You can't just report back on a project you've done, in a keynote - it has to address wider or more fundamental issues, and be transcendent somehow. But for me the best keynotes still give people something to DO with the information they've been given. It's a real challenge. It's an art.

Matthew Riedsma | More than usable: library services for humans

All of the keynotes were fantastic. Matt's in particular, I thought, taught me a lot about the art of the keynote. He had so much TIME. He left long pauses.

He let things sink in.

He allowed us to digest what he was saying.

And of course, allowing pauses is all about having confidence in your material. When I'm doing an infolit talk a Department has asked me to do, with them selecting the content, I tend to keep the momentum up and move quickly through it, because I'm not always convinced the audience is enjoying it. It needs urgency to work. But if I'm doing a talk at a conference, a talk of my own choosing, I believe in it a lot more so I try and leave more gaps. But not as longer gaps as Matt did. This was a masterclass of pause-leaving. It was awesome. No one was shuffling their feet or wondering what was going on. We were all captivated by the talk.

Matt's slides were the perfect example of how to create a visual theme without using a template. They did what a template is supposed to do: keep the audience aware they were part of a certain landscape, that all the information they were seeing is connected and of a piece. But without all the bad stuff templates also do: because all the slides didn't actually look the SAME, they were able to express their contents uniquely, and be formatted for the best method of communicating that particular set of information.

Matt used blue, black and white as the colour scheme. This meant blue and white text, black backgrounds, pictures with a lot of black in (the night sky, for example), and, a really nice idea this, black and white pictures of the people he was quoting, behind their quotes.

A selection of Matt's slides to illustrate his visual theme. All of a piece but no template in sight.

The visual theme was cemented by the use of fonts:

@thomaskistell @ned_potter Fonts: Bemio for quotes, Montserrat and RaleWay for titles.

— Matthew Reidsma (@mreidsma) March 19, 2015

Three fonts, which is the maximum you should use in one presentation, generally. Two of which worked together and complemented each other nicely: in effect, Raleway is for the set up and Montserrat for the punchlines, so we as the audience are being guided to what is most important about each slide. The other font, Berio, contrasted to denote that something different was happening (quoting others' words rather than creating his own).

The slides were pretty. But that was a by-product of them being EFFECTIVE. Big fonts that everyone could read, and which guided us as to the key messages. Pictures which helped us learn and told a story. One point per slide so we never had to decide if we should continue reading or continue listening. The presentation supported and enriched, and possibly prompted (although Matt seemed to know exactly what he was going to say without using the slides) - but not duplicated. Slides and presenter working together, not competing.

Here's Matt's full presentation:

If you have three-quarters of an hour to spare, there's a recording (slides and audio) on Vimeo - it's well worth watching!

Matt's talk was ostensibly about usability. It covered a lot more - philosophy, immigration videos, chair awareness, urine in space.

I've created a Storify of the tweets people wrote during the Matthew Reidsma keynote, which goes some way to capturing what is was all about, and which I won't embed here as this post is already taking up a lot of space, but I'd recommend you follow the link and have a look.

To try and boil it down, the phrase that summed up the essence of the talk for me is 'Libraries are PEOPLE, all the way down'. Usability is all about putting people and their experiences first. Experience is messy and complicated. How someone experiences something can be affected by the kind of day they're having (the kind of LIFE they're having), rather than just, as we might assume, the tools they're using in our libraries. Our users are not coming from an emotionally neutral place. A task based approach to usability assumes they are; an experience based approach better allows for an emotional experience. The Andromeda Yelton quote in the slides above is key: often people talk about libraries as being about information, and access to information, and more frequently these days we talk about people in that equation too. But not just librarians. Other users. Libraries let people transform themselves through access not just to information but to each other.  

And, absolutely crucially, we need to rethink usability from being an attempt to produce a perfect experience, to instead an attempt to design for breakdown. Accept things will do wrong. The key is the user's ability to self-right them.

Design our services (online and in person) so when they breakdown, it is intuitive to rescue them and carry on working. Matt used the analogy of using a mouse on a small table: at some point you may move the mouse too far and it falls off the table. At that point, no one goes 'oh it's broken' and seeks help - they just put the mouse back on the table and go back to work, very quickly forgetting the 'task' of using the mouse and getting back to the 'experience' of the technology being an extension to themselves. This is what we need to aim for.

This is far more revelatory to me than it probably should be. I feel I should have been more aware of this before. But I wasn't. I frequently try to get everything to just WORK - but when I think about what works for me outside the library, it's the procedures or technology which I can correct, on my own, when they break down. I've been designing everything as task based, forever. Now I can focus on transitioning to designing for experience and usability.

It was a brilliant talk. It had stories (that were relevant), humour (but no jokes), philosophy (which underpinned the practical stuff), and calls to action. Ace.

(As an aside, we also challenged Matt to get the words 'sac magique' into his presentation - a reference to 90s classic Tots TV - which he managed to do so brilliantly as to satisfy our juvenile need to get him to do something silly, but in such a way that no one else saw them so it didn't detract from the professionalism of his presentation... They were in the search box of an American Citizenship website he showed us a print-screen of. I mention this because although I was trying to spot the reference, I kept forgetting about doing so and getting engrossed in Matt's talk - which is actually a neat metaphor for a lot of what he was talking about. I had a task - spot the sac magique - but actually the experience was so good I forgot all about it...)

---

Header Photo by Jamie Tilley - the original, on jtilleyphotographic's Flickr account, is here.