Film: Making a Location Aware Audio Guide for SFMOMA

How do you create a revolutionary location aware audio guide for a museum?

This presentation was made by Keir Winesmith, Head of Web + Digital Platforms, SFMOMA at MuseumNext Melbourne in February 2017.

in May of 2016, after a three year closure, an expanded SFMOMA reopened. To coincide with the opening the museum launched a gorgeous, immersive, location-aware, audio-first app.

This being San Francisco, where there’s a cultural expectation of risk taking and an embrace of failing forward, they built the app as a partnership with a young start-up, using unproven technology, in a half-finished building site, with voices and stories rarely heard in a museum and, two weeks from opening, it still didn’t work. Every bit of this project, the partnerships, the technology and the content, were risky.

A lot happening in the last two weeks, including the development of a workaround for a previously undiscovered iOS bug, and the museum got the app onto devices the day before opening. Since the launch the museum has fielded a lot ‘how on earth did you do that?’ questions from peers. Here’s how.

Keir Winesmith : Hi everyone. That is what it looked like, actually, inside the museum, in one of the more finished spaces. This is the refurb of one of the older parts of the building, when we started working on something that was supposed to work indoors. I was wondering, could I flip from the presentation timer to the notes? Is that possible [over the back]? They were showing … yeah, great, thank you. So, I’m going to cover a few things today: I’m going to cover the SFMOMA context; like why we did it and where we were; I’m going to talk about what we actually built, and the way we did it … I’m going to try to answer the question ‘Does it work?’ and talk a little bit about what comes next.

So, when we started working on SFMOMA’s location-aware app, we started looking around the city for different ways that we could tell this story. How could we make the SF in SFMOMA stand out? When I arrived at San Francisco from MCA in Sydney, the first frame of that video you just saw was how the building looked; we’d just knocked it down … and I had this idea of what I call just-in-time storytelling, this idea that we would provide relevant content to someone in the moment that they needed it, and I wasn’t sure exactly how to do it, I wasn’t sure how we were going to do, and who we were going to work with to make it happen, but I did inherit something that was really valuable, which is this incredible psychographic study of San Francisco. So I can give you that, because it’s a little bit different to maybe this context. So we looked, as we were closing, at the whole city, and a lot of the tourist population, we learnt that the city was very smart, and it was art-interested but not necessarily art-knowledgeable, and so what does that mean? That means that something like 80% of people who live in San Francisco or visit San Francisco are considered arts-interested. So that means that they’re interested in film or literature or the visual arts, performing arts, music … so that’s a really big chunk, it’s higher than most places. So that really encouraged us, we thought, there’s an audience here, we can triple in size and still be sustainable; SFMOMA, for the five years before we closed, was largely a five and a half to six and a half, 100,000 people visiting every year; very stable.

But we wanted to be three times as big, just about, and we wanted to grow our audience to make us survive in a city that we love and have been in for a long time. We wanted to do the SF in SFMOMA. So we then started looking if those arts-interested people, like who really gets modern and contemporary art? Who really … who’s our audience? So, the big wedge is people who were somewhat … a tiny bit confident about the visual arts, and the thin wedge is people who are highly confident about the visual arts. You know, the stuff that we do. And then the really, really thin wedge is people with a visual arts degree. And so we largely programme for them, which makes it really hard to stay open, and I would guess that in most of the cities that most of you work in, it is really similar, and it may be science knowledge if you’re a science museum, it may be history knowledge if you’re a history museum. Definitely in San Francisco, it’s art knowledge. So we knew that in order to attract that larger cohort, that bigger group, we needed to do something that was relevant to them, we couldn’t speak in the way that we have spoken in the past. We’re pretty good at wall labels, we’re pretty good at intro text, we’re pretty good at publications … if you measure them by their scholarly aspect, we’ve been pretty good at telling that authority voice in a kind of deep, I’m sure masculine baritone, that says, you will understand the power of this work; here is the knowledge that you must have.

And working with our curatorial peers over a number of years, we have been working on that voice, trying to move from telling this story of the artwork to telling a story about the artwork, that gives you, the audience, a way in. We are pitching you a reason to look at that work. And we got good at that, we got better at that, it was a real … it was a serious intent of the opening of the new museum, was to sort of change that voice, make it more welcoming, make it more approachable, make it more understandable. But it was still us speaking, and it was still that authority, it was still that feeling that you get when there’s a joke and you don’t know the punchline, or there’s a language and you don’t know all of the words. And that’s a lot of what we hear from people who are giving these post-visit surveys, is that sometimes they’ll laugh at the art because they don’t have the language or the skills to talk about it. And so, if we’re going to be sustainable and relevant in our city, we need to talk about the art in a different way. We need to do the [Bay Area] for the work that we’re doing. And so this project is one of many projects the museum did that’s a direct response to that need.

So, what did we build? We built something that was location-aware, and we built it for three reasons: navigation. This is a new museum that not even we had been inside; we did it because we wanted a phone-in-pocket experience, we wanted people to look at the art, we wanted people to listen, but look with their eyes, we didn’t want people looking at their screens unless they were taking great Instagram photos and promoting us for free. If they’re having an art experience, we wanted them to be present with the work. And we wanted to enable what we call ‘multi-gallery stories’; so this is this idea that you can take someone on a narrative through the museum that’s going to be relevant to them in their voice. And what does that mean? Relevant to them in their voice? So we decided … because it was San Francisco, and there’s a lot of fun, interesting and smart people around … to not be the only people speaking in our tours. I think I’ve done in the order of 50 museum audio guides as part of this … the research that led into this project … and almost all of them were terrible. I built three before this one … they were all terrible, too. I’m just going to say it out loud. And they’re almost always the dry voice of authority; sometime dressed up with a really energetic [attar], or a local historian or someone, with a really great tone, and really engaging tone, but it’s almost always the same nuggets, pearls of wisdom that we’re dropping from on high that you could scoop up and suck in.

And so we’ve taken a really different approach to that; so we’ve got a diverse set of voices throughout, because we know that people are looking for different things. So, on my right, on the right of the screen, is [unintelligible 00:06:57] who were comedians from the [HPO] series, Silicon Valley, which is incredibly popular in San Francisco. They are sardonic, irritating, hilarious comedians in real-life and in the show. They make fun of each other, primarily, these two, and they’re also confused about modern and contemporary art, which the majority of our audience is also confused about modern and contemporary art. So we paired them with one of our curators, and they walk you through a tour that [we feel, I don’t get it]. It’s 20 minutes, it takes you in 16 galleries, and they’re baffled throughout. Like, there’s a urinal on its side, next to that really beautiful painting? I get the painting, that’s a pretty thing, but why is the urinal there? And along the way, the curator gives them feedback and chats with them, and by the end, they’re kind of having ahah moments, you’re like, oh, wait, so the urinal is in response to this? Oh, that makes sense, hah, that’s kind of cheeky, isn’t it? And so they’re discovering the work in a way that is really attainable for anyone in … for most people in our audience.

Then we spoke to Philip [Petee], who we asked on Facebook; we just messaged him on Facebook, we had a …. hey, we’re doing an audio tour about unbalanced artists; would you like to narrate it for us? And he, within an hour, said yes, and within a week had recorded his content, and it is kick-arse; he can read anything and it sounds great. The person on your right is Errol Morris, documentary filmmaker … one of the problems we have for this sort of immersive storytelling … and think of it like a podcast you walk around inside … is that some of the work we have up is up for a really short period of time. So the first two, they take you through our permanent collection, parts of it … Errol Morris takes you through our photo floor. So the photography collection rotates every three months, so how do you do deep storytelling for something that’s up for three months? Well, in our case, you get Errol Morris to talk about the themes of looking at photographs. And so we know he’s going to tell a story about looking at portraiture. Not which portrait he’s going to look at or talk about, but just the idea of looking at a face coming out of the frame at you. And so we trigger that content when you move into any gallery that has a face in a photograph.

He talks about landscape photography; there’s always a landscape photo somewhere in the museum, and so we can navigate you through any of our shows … photography, and his voice is there with you as he tries to unpack the looking that happens when you look at this sort of work. And we’ve done it in five languages, and there’s about 20 hours of content in there. We also took a chance on what we call nearby audio, so these immersive walks are very subjective, the production value’s a little bit like radio lab … when I first pitched this idea to the director, I said, it’s kind of like the film ‘Her’, [amidst This American Life] and he said ‘Uhuh’, but then after a lot of demos, he really got it, and is now a big fan. So we also took this sort of radio approach to our storytelling, so on your right … sorry, on my right, on your left, is Kiki Smith’s Lilith, so this is a sculpture that’s hanging from the wall, with two piercing blue eyes sticking out at you, and it’s Lilith, you know, the person Adam was with before Eve … and we then grabbed an [aerial] dancer and we hung her from the wall in exactly the same position as Kiki Smith, and we recorded her describing the feeling of fragility and power that happens when you are presented up on the wall like this. And so we have her talking about that, then we have a curator explaining some of the art historical notes, and the artist, who is actually a crazy person, but is wonderful, talking in this really expansive way about this particular work.

And so you’ve got these three voices all in dialogue, and then we [coast back to the end], so that person’s been hanging there for the whole audio [to stop]. Still hanging there, kind of sweating now, describing what it feels like to be hung off the wall like that. So we took that approach. We know that people get more out of museums if they have it as part of a social experience, so we made it social; you can sync groups together and they all hear the same thing at the same time, and we give you a timeline; if you want to, after you leave … and the thing that’s most popular about that is people take photos, and they want their photos; if you don’t give them their photos, they will find you where you sleep. It’s possible because of this partnership that we made with Detour, so Detour is a really smart but very small outdoor touring company, and they made some production tools that for us were the key to this project. It’s an audio word-processor that allows you to edit text that’s editing the audio in real time, and allows you to place that audio in space.

So think of it like there’s a … you’re walking through a new city with two friends; one of them’s playing a guitar track, the other one knows where to go and is telling you a story. So the music track just plays the whole time, so if you want to spend five minutes looking at something, that’s fine, you’re still there. As soon as you move again into the next gallery, the next bit of the story kicks in. And because you’re with a friend, they know where you’re going, they know where you’re going next, and they only tell you the commands on where to turn left and where to turn right when it is the moment to get those commands. And these direction tools allow that to be possible. And does it work? I mean, well, getting a little bit deeper about it, does it work? But when we look at the analytics, it doesn’t always work. Some of the times it works, and some of the times it doesn’t.

So German, to me, is this very subjective, kind of like radio lab in many ways, it’s a story of a German-American radio producer, who interviews her mum and her grandmother, who are both still living in Germany, about living through the war and growing up in post-war Germany, through the lens of work by people like [Polker] and [Richter], who were on the walls. And so she takes photos of Richter and talks about and asks them about their childhood, and what it was like to grow up through the war, when these paintings were being painted, and they reflect on their identity and she reflects on hers, and so it’s really like a self-discovery of modern German identity through familial relations, as told as you walk through space looking at paintings … and that is not something that museums traditionally do, it’s certainly not something that we’d ever done before, and what blows me away is that big bar on the right … that’s all the people who make it to the end. It’s 40 minutes. So a bunch of people at the very beginning say, uh, this is not for me, I’m not interested … and then the drop-off rate is very low throughout, and an enormous number of over 80% of people make it to the end, of forty minutes of very personal audio.

The one next to that is called Play by Play. My boss and I were convinced there’s a massive overlap in the [Venn] diagram of people who love modern and contemporary art and people who are sports-mad, but they don’t actually [hit], it turns out. So as you can see, a huge chunk of people just leave straight away … you can also see the numbers. That says ‘160 people left straight away.’ You can see the other numbers … ‘1,600 people made it to the end.’ So the numbers are even very different, so Play by Play is this sports analogy looking at contemporary art; so taking sports stars and sportscasters to talk about contemporary art. We thought it would be a slam dunk, it’s very funny if you know a lot about both worlds. It turns out people don’t know a lot about both worlds. This isn’t in the app anymore, we’ve taken it out.

So some of the times it works and some of the times it doesn’t, but if you’re going to do something chancy, it does not do you any favours if you don’t go in knowing you’re going to evaluate, go in costing the evaluation, do the evaluation, and then respond to it, make change. And that’s one of the things that we did at the beginning, SFMOMA actually took an early decision to kill two big projects in order to pay for evaluation on all the projects we did do, and leave enough money to tinker with them and improve them over time. So, three months after opening, we started our evaluation phase, and we’ve just finished all of that, before Christmas, and we’re already making changes to all of the digital projects that my team and I worked on as part of the opening. But in the process, we killed some darlings. Because it’s not good enough to take risks. Take risks, measure them, and make change based on the outcome. I think that’s critical.

And is it working? Yeah, I mean, yeah, it’s working. This is the average session length during a particular period last year of how long people [were] actually experiencing the app. That says 120, that’s 120 minutes. The average session length is 116 and a half minutes. So that’s from the first bit of content listened to to the end of the last bit of content. It’s a little bit bloated, some people have lunch in the middle … but largely, you can see it’s pretty consistent through time. We had a show, you can see, in early August we had a new show open, and that had a peak, people were really engaged with that new show, but it’s largely flat. It means that most of the people using the app, and this is 60,000 people who’ve used the app, listened to over two million minutes of audio since we opened in May last year. Seven hundred and eighty thousand individual clips have been listened to. They’re listening to it for a long time, we’ve got a few completionists. Someone has listened to all of the tours, and Tweeted us about them, and cried. So some of that really works, and even in that, we’re also not counting anyone who listens to less than three bits of content, because we think we’ve failed there, and so they don’t fit in our metric of success. So it’s sort of working.

So how did we do it? We partnered with a company called Detour; it being San Francisco, the first thing that we did when I started the museum and a couple of others came from elsewhere is basically step away from the contractors that we currently had, and started doing the thing that you do in San Francisco, which is take smart people to lunch and hope they’ll do something for free for you. And so we took a bunch of different smart people to lunch, and one of the people we took to lunch was Andrew Mason, who founded Groupon, left Groupon, founded Detour, and he came to us wanting us to do outdoor content for Detour, and we were seeking him out trying to get him to work indoors. And so they brought their indoor project that they had in their minds, they brought it forward two years, and then we together approached Apple and tried to convince Apple to help support real-time indoor location using Apple’s indoor-location, core-location tools. We sat down with Apple and we said ‘Where does this work the best?’ Because we’d tried light, we’d tried sound, and we’d tried beacons, and none of them were accurate enough for us to get the sort of storytelling that we wanted to do.

So, we sat down with Apple and the guy … I can really distinctly remember this … he kind of cocked his head on the side and he said ‘Well, we were kind of hoping you …’ And so, when Apple hasn’t done it yet and no-one’s done it yet, it was a little bit risky, and it didn’t always work, especially at the beginning. So now it’s out and it’s got all of these beautiful, superlative, very positive reviews, and for the first four or five months that we were open, we had almost solely four- and five-star reviews on Apple iTunes. And so we decided to do the thing that you really shouldn’t do, which is, instead of waiting for the evaluation and fine-tune the product, we started to keep taking more risks. So we made this kind of non-linear spatial kind of walking tour that wasn’t a linear path from space to space, it was like you walked into any one space, and we used the backing-track feature in Detour to basically trigger these environmental audios that were the story of Bruce Connor, who’s a … at least in San Francisco … a famous American artist who worked across all sorts of mediums pastiched together, and some people call him the founder of the [music] video … and so we wanted to honour him by making something really experimental and kind of crazy, with a re-imagined floorplan, so this is the floor that Bruce Connor was placed in, we changed the walls, we added new walls, we added this little [sinarettes] and [theatre-ettes]. We sent all of the updated floorplans to Apple, and they were supposed to process them and get back to us really quickly. We ended up having to string new wi-fi access points into this part of the museum, because it was one of the old parts of the museum, that didn’t have the same coverage as everywhere else. I had to sneak under a crawlspace and just put wi-fi access points wherever I could, to try and improve the reception.

But … it didn’t work, actually, and we started to get one-star reviews and two-star reviews. We started getting hammered. We went from all four or five to all one or two in a week, because we pushed too hard, and once you’ve committed to location being sublime … and we had committed to location being sublime, to the point where people would feel anxiety … which we never expected … people would feel anxiety, because for the 15 minutes before that, the storytelling was so on-point and so seamless, it was like someone was in their ear and knew exactly what to tell them, that they took a phone call or sent a text or looked out a window for a bit, and forgot what their command was next, forgot what their next bit of navigation was, and expected this thing in their ears to go like, oh, you’ve finished with that phone call, as we were, we were just heading down a corridor to the left. And so people were feeling really anxious, and so we didn’t expect that. And if a location doesn’t work, if you bet everything on it, you actually have a brick in your hand. And so we created a brick for some people some of the time.

So it is dangerous, and we have pulled back, so we’re kind of not doing that anymore, and we’re increasing the rate that we re-survey the museum, and we’re increasing the rate that we have an ordered run on the wi-fi to see if anything has changed, because if something changed, it throws you into the street … you’re walking along, and you’re three-quarters of the way through, ‘I don’t get it’, and you’re laughing with your friends, and all of a sudden, you literally can’t get to the next bit of content. And that’s a problem. So we started this project when the museum was an idea, and we finished the project when it’s this shiny, new reality, and we managed to get some of our more attractive members of staff to pose for it, and thinking about where we go next, a lot of what we’re doing is taking the evaluation that was just finished to improve the product, and looking at the questions that we’re getting and improving our processes around keeping it up to date … because it is … we have created a monster. And doing things like giving away headphones to people who’ve downloaded the app but don’t have headphones with them, so we ordered 10,000 headphones like the ones you see in this picture, to give out, and we’re even thinking about financial incentives and other incentives.

We do this thing whenever we do evaluation of an interpretive or investigative space: we ask you ‘When you arrived, did you feel like SFMOMA was for you?’ and then we ask you ‘When you left, now, as you leave, do you feel like SFMOMA is for you? Is it relevant to you?’ And it’s just a slider, you just use your finger. And then we also ask you ‘Is modern and contemporary art a thing that you get, understand, is it for you?’ And we ask you when you leave the same thing. And we look at is there a delta. You know, have we made a difference, has the visit made a difference? And in general, the visit … unquestionably, the visit makes a difference, but what we’re seeing is, the people who use the app, the visit makes an enormous difference; like of 40% more than the average [unclear 00:22:45], like a crazy big difference. And those people are more likely to come back, more likely to be members, and more likely to bring a friend. But when the app sucks, when it doesn’t work, when you have technical difficulties, it’s evens. We’ve made no difference at all. And so for us, we need to fix the technology, but we need to get it in more people’s hands, because if we want to be relevant, and we want to be sustainable, then we need to use the tools that we’ve created that we have statistically proven are changing people’s connection with the museum, and connection with the work that we put on the walls, [and you get in] more people’s hands. So that’s kind of where we want to go next.

So my name’s Keir, and I’m going to stay here for a bit and answer questions.

Female voice:  Questions, down the front.

Female voice:  Yeah, thank you very much, that was really interesting. I’ve got loads of questions that [I’m interested in] [unintelligible 00:23:39] why, now. I’m interested in the ideation process for the content itself, because you talked a lot about putting the SF in SFMOMA, and all the knowledge you had about that audience. For the concepts, did they come from your extraordinarily creative team alone, or did you go out and source some of those ideas? Aside from the experts who you already knew were great storytellers, did you go to the audience beyond that, and …

Keir Winesmith:   That’s a really good question. So I get asked two questions usually … the first one … and it’s always the same question, which is ‘How did you do it?’ And when it’s nerdy technical people, they’re like, ‘How did you build it?’ or ‘How did you convince someone to build it for you?’

Female voice:  Yeah.

Keir Winesmith:    And when it’s kind of content-focused people, it’s like ‘How did you convince the curators to let you do this?’ And so it wasn’t an active convincing the curators; we brought them along throughout, and so when the museum was closed, we had these massive maquettes … about half the size of this room was filled up with models of the building that we were building, with magnetic walls … they were all about this high, you know, kind of two wingspans … with every painting that was going to be on view when we opened stuck to the wall. And we would meet with the curators for each of the shows, and they would walk us through the narrative of every show, and talk about specific works and specific vantages and specific connections, and so we took that internally, and we came back … we called them ‘hooks meetings’ … and then we came back and presented what we thought were like the great stories for the different shows and the different artworks, and then working with them we kind of picked out the ones that would really sing, and then together, we generated ideas about who would be a great person to speak to that.

And a good example was, there’s this beautiful work that’s basically about a couple in crisis, and the way that it’s been painted is, it’s kind of set up like a film set, or like a stage set, and they’re in different parts of the stage set, and both of the parts represent their thoughts, and so the curator talked a lot about this, the way this relationship is depicted, and said, you know what, you should get a set producer or a set designer to talk about this work. And so we were kind of led by … then we met the set designer, and we talked about it and the ideas of the curator, and they came out with a bunch of their own ideas, and so, led by the curatorial intent, then with our group focusing on a story that would be a hook, we then found someone who’d be a great person to speak to that hook, and then in the end, it went back to the curator to say, did we get it?

And so it’s certainly not a thing we just did and said, hey, it’s all done now. We absolutely involved our peers throughout.

Female voice:  And did the audience … did any audience members contribute ideas around that too, or …

Keir Winesmith:   We’ve done two projects in the past, before my time, that were focused on getting audience voices in front of other audience, and both of them were both success and failure. Both of them were incredibly successful in getting audience voices, and total failures in that audiences have no interest in other audience voices. Then I challenge anyone to find me a project where that’s not the case. But people love leaving a mark, have no interest in other people’s marks. Yeah. I think the trick to that is expertise, I think people are really interested in expertise, not necessarily where it comes from, but expertise. So …

Female voice:  Thanks.

Keir Winesmith:   Just sing out.

Female voice:  Hi, thank you so much for that presentation. I’m from the National Gallery of Singapore, and we also have a location-based app for audio tours, so we’d love to talk to you later about how we’ve done our app.

Keir Winesmith:   Sure.

Female voice:  But my question is about tour retention. You showed a slide earlier about a tour that had a 40% rate of users completing the tour, so how does that compare to the tour completion rate for your other tours?

Keir Winesmith:   It definitely varies from tour to tour. So, we have one about how the building is built, and how it’s designed, and it has a lot of stairs in it, it takes you up through the stairs … it’s narrated by a producer from 99% Invisible, a really great design podcast … [I encourage] you to check out [Avery Truffleman] … and most people make it to the end, well over 50% of people make it to the end. Whereas we have another tour, actually the Philip Petee tour, has this big … it’s like pretty much most people make it through, but there’s a big drop off at the first set of stairs. And so, we’re really curious, is it the content just before the stairs, is it the fact that you have to go up the stairs, and you’re kind of like, oh, it’s been good, but maybe I can try something else now. And so now that we’ve got this really … great analytics, and enough people have used enough of the tours, we really know what is and isn’t working, we’re now diving into the particular moments to understand why there’s these sudden peaks and troughs of use. But overall, the vast majority of people finish the vast majority of tours, which is something that we didn’t expect, we expected more like a drop off of a sensible curve, but it’s more like a U-shape. Most people make it to the end if they don’t leave right at the beginning, for the ones that are working, and the ones that aren’t working have these very pronounced moments where they’re not working, and so we’re thinking, do we actually make that one end there? And can we change it, or do we leave that bit out and bring you to the end earlier? So we’re starting to re-do some of the content based on how it’s being used.

Female voice:  One more question, up in the middle.

Male voice:     Hi, super-quick question, actually, just in relation to what you just said. In regards to stairs, are the tours accessible for people in wheelchairs and things like that, and how does it deal with that if you’re telling someone to go upstairs?

Keir Winesmith:   Luckily, the stairs are all adjacent to elevators, and so we have this sort of joke, because the stairs are really architectural, and quite beautiful, there’s a lot of signage inside the museum that says ‘Take the stairs, they’re beautiful.’ And they’re just about the most photographed thing in the museum at the moment. But every tour is designed so you can do it in a wheelchair, and so [it always] says ‘Take the stairs, or if you don’t or you don’t want to, the wheelchair … it’s on your left’, and they always converge back into the same trigger point, and so the next bit of content will trigger either way. And I know this because five weeks before the museum opened, I tore my Achilles tendon, and so I was working on a walking tour, and I was in a wheelchair at first, and then on crutches, and so I did all of my testing either on crutches or in a wheelchair.

So I advise not doing that when you’re building something that’s an immersive walking experience. OK, the last point I just want to make about accessibility is, SFMOMA really genuinely takes museum accessibility seriously. We’re not as good at it as we should be, and I don’t think anyone’s necessarily that good at it, but in talking with our blind and partially-sighted peers, they [sited] the shared audio-experience, so we have 15 or 20 of the artworks, of the most-viewed artworks, have a little short description, sort of a 90-second description of what the artwork looks like, for blind and partially-sighted, and it plays before the narrative, and so you can do a kind of … essentially blind-and-partially-sighted tour of the museum. But what people have reflected to us is the fact that you can sync means that what is largely an isolating experience, what [unintelligible 00:31:38] calls ‘Visiting a museum is like following around some friends who are having fun’ if you’re a blind person, and so we can sync, and everyone can be hearing the same story at the same time, and that’s been a … I think that’s a huge gift that’s not obvious.

And so we want to do more of that work, we’re starting to make all of our interactives have blind-and-partially-sighted sort of modes, so you can plug in headphone, and you can swipe left and right and hear the content, and hear the work described, and I think too often, visual arts organisations focus on themselves, so they look White, they look educated, and they don’t necessarily look like their audience, and so I think that’s something that as a sector we need to address, and we’ll need to start with every project we do.

Thanks, everyone.