- Seb Chan is the Chief Experience Officer at the Australian Centre for the Moving Image, so that's a fantastic title that I think he can be proud of. He's had a long stint also at the Cooper-Hewitt in New York and now with the events yesterday and overnight, he's promised to slightly tweak his talk because one of his patrons of the Cooper-Hewitt is Michelle Obama, so a very personal connection to the events, but he'll tell us why museums are awesome and why he thinks we should be thinking more about them and making more out of them.
Thank you, Seb.
(applauds) - So yeah, so I was working on these slides for the last week and then threw them all away and went back to my American slides, 'cause I figured I hadn't done these slides for this sort of crowd in Sydney.
So this will be very fast because it's between me and lunch. Last time I spoke here, it's a nice font actually, it's the Cooper-Hewitt's font, see the nice kerning and all that, ones are good.
I worked at the Powerhouse Museum, which of course as you know is being moved out to Parramatta, possibly. Last time when I was at Web Directions, in 2011, I spoke about the changes museums were going through that I'd witnessed for the previous decade at the Powerhouse, this shift from 2001's view when Wikipedia launches, that museums were trying to be this encyclopaedic resource, and when I left Powerhouse that the mobile explosion was working across museums and we were starting to think about the museum without walls and this data provider role.
But at the heart of all of this is really that museums are democratic spaces, they're supposed to be democratic spaces, they're also meant to be curiosity machines. They're not supposed to be places you go on excursions to learn a bunch of facts, they're supposed to inspire you to learn about new things you didn't know you wanted to know about.
And the V&A commissioned a really beautiful work that sat in their foyer a couple years ago, which was this. The V&A's a big design sort of museum in the UK, and this was really saying to all of the people that came into this beautiful public museum, that this is actually all your stuff.
Museums collect all of these things, they collect for you. So then I got on a plane in 2011 and ended up here at the Smithsonian, so I was working for the Smithsonian for the Cooper-Hewitt, which was the design museum in New York, it had just closed when I started, and it's located a little bit north of the Guggenheim, which is where all the tourists go, Cooper-Hewitt's there, and just as I left, they got really good coffee in between.
Just do that again, 'cause you know...
Which was a shame, 'cause I really missed coffee until, yeah, anyway.
But it was in an old house and it was in this house that belonged to one of the wealthiest guys of the early 20th century in America, who also set up a lot of libraries, provided a lot of money towards the public libraries.
He also did a lot of not-so-nice things as well as he ran a steel empire.
So the museum was pulled apart and then rebuilt and the museum had a whole bunch of crazy things in it. Bird cages, buttons, an amazing collection of wallpapers, ceramic animals, more ceramic animals that animate well into gifs, posters, product design. And when I started, a month before I arrived, Bill Moggridge was gonna be the director and Bill was featured in Fast Company and they said to Bill, "Bill, you've come from a design firm, you're now "running the National Design Museum, it's in this old house, "you've got this weird collection, what are you gonna do? "Because this stuff you have, the stuff the National "Design Museum of America has, is kind of old and weird "and you're in this old, weird house." So one of the things that we started to do was started to collect new forms of design, so I acquired for the Smithsonian's collection, this app called Planetary, but we didn't just acquire this as a piece of software, we actually acquired the source code and then released the source code, publicly available, and it wasn't just the final source code but the versioned code, so the GitHub Repo now sits in the Smithsonian's collection for you to take, look at, and modify, remake, and commercially use. So even though this was occurring, I'd been hired to help build this building that was about, think the purpose of a museum isn't just putting all of this stuff up on the web, it's now about creating a space where this physical space of the museum is reformatted for a digital age.
Because in that small house, a very small amount of the collection actually fitted inside it, right, there was 210,000 pieces in the collection and perhaps 1000 at a time could fit on show. So the museum became this interface, this physical opportunity to access things that should already be on the web, but the interface looked like this, in a restored way, and when Michelle Obama was talking about museums and public spaces, when she launched the Whitney in New York, she said this. And these, the physical design of these spaces, particularly in America, is designed to exclude people. I think that's so important now with Trump and this is quite different to the design of public buildings here. So really what the job and the renovation of the museum was about was about this, it was building a new system to change the people who felt welcome within that physical space, and also change the way the collection was experienced.
So Josh mentioned magic wands this morning, so I changed this too, so we made a thing which is really a magic wand, so imagine you're walking around the museum with a magic wand, so local projects, Madia Design Firm in New York worked with us and they pitched us this idea very early on, they said, "What if you give "people who come into the National Design Museum a pen. "Not a museum app, not an app on their phones, but a pen, "and it wouldn't it be great if it just looked like this." Because the pen would do a bunch of things, it would be functional and you could draw things but it would also communicate to visitors, the design was for doing, it wasn't, the Museum of Design wasn't just for looking at, it was about designing a series of processes.
And the pen then could connect to this original purpose of the museum's collection, this is the museum's collection being used in the 1920s, a tool for people to copy, learn, remix, and transform.
So we eventually made this thing with a whole bunch of other firms, with a whole bunch of other support, Bloomberg Philanthropies gave us a lot of money to make that possible, but it started off like this, my team did some very early video storyboarding to look at the way you might use a pen, a stylus, within the museum. And we had to think about the way people would wear the pen, and what would you do with this pen before we started designing the system that ran it. So you can see the person moving through the space, connecting to big interactive tables where you could add notes to the things you've collected, then you put in your email address, 'cause that's what we thought was good at the time, you can keep on going and you wouldn't need your phone. But as it turns out that there are a lot of other challenges, so out of a design sprint we did for two days with GE, we came up with a whole bunch of other things that made it, that were really the things we knew we needed to design for it, durability, cleanliness, New Yorkers are obsessed with things being clean, right, so if I give you a pen that maybe someone else has used beforehand, how do I know it's been cleaned between the last person and me? All that stuff.
Batteries, batteries are hell, so when we started this process, we used 3D printing to come up with these things. We had this manufactured in Taiwan, which is an amazing piece of work in itself, I had a little video of that but I took it out for time reasons, 'cause the main thing was, you get 3000 of these arriving in the museum, and then it's actually about also designing for the back of house logistics. But we found that using three triple-A batteries gave us 30 days life in a way that a rechargeable battery didn't. And you can imagine this going on each couple of weeks at the museum, when you have to switch the batteries in all those pens.
So that's kinda what it's like, it's an NFC reader at one end, it's a stylus at the other, the backstory's all online with the gruesome details. But this is it working in the museum now.
So as you walk around the museum, every visitor gets one, you touch a pen to any museum label, you can bring it back to big interactive tables and explore the things that you've picked up throughout the museum, you can design and draw your own things on these tables as well, makes recommendations based on what's in the collection, all of this other stuff, and because you have the pen connected to your ticket, you just hit save, you touch your pen to it, know our email addresses, and you're there. Also that wallpaper collection suddenly becomes alive and you can start drawing wallpapers in a large projected space, and it makes that museum experience social in a way that using an app wouldn't have done, and people of course use their phones and cameras to picture themselves in the museum alongside their own creations. And people made crazy stuff, I mean, wallpapers are not the most exciting thing until you can see them projected, and then it's like, "Oh my god, "wallpapers are kinda cool," and it worked with people of all ages as well, this is one of my favourite ones, the bacon and egg wallpaper, people loved their stuff. And they shared them and all of that.
But also important in this was the ability to use this device without looking at it, because the idea of using an app in a museum, is you're always staring at your phone instead of the stuff you've come to see. So the pen also vibrates, so it's this magic vibrating wand, it's kinda cool.
It's kind of also like pre-shopping design, so imagine a museum was like a pre-shopping experience. And as it turns out, it got used a lot, so in the first year, the Cooper-Hewitt released the data set anonymized of the first year of use so you can look at the patterns of what gets collected when and all of that, four million things collected in the first year of use. This was all built on the architecture, but importantly, the philosophies of the web. So an API at the middle of this, that runs all of the things, it connects what's in the collection with the ticketing database and then publishes that out through web content protocols that then the pen interacts with.
The API as well is of course public too so other people can now and have been making things on top of that. The API methods are all published.
When you go home, you get this, so you get your visit to the museum, you can download all of your things, you can delete from the system for privacy reasons, the things you've collected and made yourself, you can download as vector shapes.
If you've designed a three-dimensional thing, you can download a 3D model file that you can now print on your 3D printer at school or at the office works near you. And all those videos that you might have walked past, you've got all those too, and you've got all the stuff you've seen, you've got all the people, the designers, you've picked up works by, you can navigate by colour and see all of the things that are green, you can see things that our museum acquired in 1987, and you can also see all the things that the Cooper-Hewitt has that are made by people the Powerhouse Museum has too. 'Cause that was kinda cool.
You can do this anonymously.
And I think this was super important, we moved away from the email account process to one that tied it to an anonymized ticket, because we were very conscious of the need to provide value to visitors without them needing to provide us anything back. Libraries do not release borrower records for a reason and it's important that public museums do similar things. So then 2015 comes around, I leave the U.S. I love the "cancelled without prejudice," so awesome, and I arrive here in Melbourne which is kinda awesome, now I don't know if people know this museum. This museum is actually one of the most visited museums in Australia.
It's also the national museum of film, TV, video games, digital culture, so it's a museum for you people. And it's in the middle of Melbourne surrounded by craploads of great coffee, 'cause it's Melbourne. And that's awesome, you may know it also from the David Bowie exhibition we had from V&A, also make exhibitions, the Game Masters Exhibit that was toured in Sydney a while ago, that's now in Europe, this one's in Taipei that we also made in house, it gets a lot of visitors, this is more visitors than any museum in Sydney gets, 22% of them are international, well that's Melbourne right, anyway the problem with this museum that I'm at now is all of the things we have are better experienced at home on your couch, right, 'cause they're TV, maybe bed, video games, these are the things, and so what happens when we do VR exhibitions? We have Lynette Wallworth's Collisions work on at the moment, the museum experience looks like this. So that's kind of a design problem, right? We need to solve for that, or this, this is one of our commissioned works, this is a commissioned work where people go into a VR experience with the Sydney Dance Company, which is quite an interesting experience, but it really turns the visitors into a spectacle that other people watch, it's kind of awesome. But anyway, what is different about seeing this in a museum to seeing it at home or somewhere else. Different challenges, too, with the sort of stuff we collect.
The stuff we have isn't digitised but is often already digital, the stuff we show isn't unique, us putting a video game in an exhibit, you can play that video game at home.
The things that we show are not things you can walk past and spend maybe 20 seconds with, I think the average time spent in front of a painting in a gallery is about 20 seconds at best.
Viewing a film is a 90-minute experience, or more if it's a Peter Jackson film, and what's the balance of social versus not-social? We don't like it when people talk in the cinema, so what does it mean if your museum has cinemas in it? And if we go out to the web and we wanna show all this stuff on the web, we're in Australia, so we often get this, right? And there's this problem as well, and then there are other people doing better stuff, so the archive is putting all the video games we remember from our youth like this one and making them playable in Javascript, JSMESS powers this, Malware Museum also on the internet archive if you want to experience what a DOS virus was like circa 1993 with emulation in Javascript, you can. Which is that challenge of also what is the atomic particle of a moving image work, what is the smallest piece of it that makes sense, is it a frame, is it a moment in a film, is it a scene in a film, how do we talk about those bits? If I had this chair, that chair is static, but that chair is not a film, a film has time and has a series of moment, scenes in it.
But there are other people doing great work, this is from 2007, this is a video search, text-based video search from 2007, this allowed you to do a full-tech search of a whole bunch of stuff, this is the word museum appearing in one episode of Mad Men, I think it's series five, episode 12.
They did this by creating a full-tech search based on the subtitle files shared on bit torrents. That same tool has now been turned into a tool that runs the Bollywood film archive, which allows you to search about 30,000 Bollywood films dating back to the 1920s, by scene, which is incredible. So I'm interested in the ways machine learning can help us catalogue and extract moments out of films. But with machine learning of course comes biases. What is it identifying as important and how do we make those biases visible to other people, visitors who are looking at the things the machine learning has identified.
How do we work with rights holders so that people can take away the things that they have seen? So if you come to the museum and you play this video game, how does that transport to your computers at home or at school? How do we work with public broadcasters around this, but also Steam, what does it mean to be able to go the museum, play a game that is available on Steam, and then go home and get it in your Steam queue. Because really what we're trying to do is build interfaces that reveal the connections between media types, so this film has a reference to that film that has a scene reference in this moment in this video game. And museums are very good at revealing their connections. And the interfaces that museums of film have are very bad at doing that.
But more importantly, what is it that you want from a democratic museum of film and video games? So I'd love to converse around that for the next couple of years, we've already released our collection data and some other source code for a bunch of products on our GitHub Repo, and thanks a lot.
(applause) - Thank you Seb, for a very inspiring talk, you touched on so many interesting things, maybe before we head out to lunch we could just sort of sit down for two minutes, I know you're all hungry so we'll let you get to it very quickly.
You said something about the museum as a democratic place and as an open API to the world around it.
Do you see any of the sponsors that are actually involved, sort of thinking about their own companies in the same way, do you see any sort of transference of that idea? - To a degree, I mean I think it's that sense of using the museum as a testing lab for ideas and I think a lot of companies are very interested in using the physical space that a museum provides to test out products and services and to connect those products and services to products that have existed in the past.
And museums have some ethical issues to negotiate there but I think it's kind of an interesting space. Where I am now because of all the material we exhibit is generally circulating in a commercial marketplace. Those ethical issues are slightly different, I mean we have a cinema, we show video games, these are not outside the marketplace and I think it's about us, if you play Donkey Kong when you come and visit us in Melbourne, understanding that Donkey Kong is part of a whole series of designs of platform games and designs of the interfaces for Donkey Kong have changed, or Donkey Kong's variants have changed through time as well, and that's what's interesting and is different to you playing Donkey Kong at home on an emulator. - Yeah, and if I can draw an analogy maybe between the museum that still deserves a visit but what does it mean if everything is online, what about shops now that you can basically buy everything very conveniently from at home, what's the point of a shop, right? - Well like that pre-shopping thing, it's sort of that pre-shopping notion but I think you have seen the success at Readings in Melbourne, a bookseller, as with physical shops, the rise of Amazon setting up its own storefronts now, that the purpose of the shop as a place to just buy products has begun to shift, and it's now about experiencing products that you might buy through delivery later on. - Yeah, speaking of experience, the photo that you had from Collisions, from the Collisions exhibition, everyone was sitting sort of two metres apart, no collisions whatsoever, it was quite funny. - Yeah, exactly, but when you step out of that VR experience, you're actually put into a space where you're meant to discuss what you've just seen and so that's a sort of staging piece, but I think VR is particularly challenging for museums, we just, I think there's a lot of work to be done about how to make VR experiences actually social without them becoming second life again.
- Right, second life, yeah. - I live through that, right?
- We've all been there once or twice, yeah. And I think Mike Peche will re-visit that topic in great detail, so that's awesome.
One last thing, you said something really interesting about revealing bias via machine learning and I think everyone who codes saw these fascinating examples of how you deconstruct videos and games with it. Do you think that's possible, that machine can detect bias without us feeding that into it and explaining what bias is? - We feed the biases in, those data sets that are publicly available to run your images against, for example, museums have been cataloguing things for centuries, at least a century.
They classified the world, and what's interesting about museum taxonomies, ways of classifying, is that there's an inherent realisation that these are, we call this a particular type of chair, you may not, but we understand that that's in, when we get computers to say what type of chair this is, we obfuscate the choices people have made to, say, identify this as an arm kind of chair, versus, some other type of chair, I'm sorry, I'm bad today, but you know what I mean, it's that sort of thing, so revealing those hidden biases and making those data sets that we are running machine learning against visible to the public is one of the roles that museums have. - Okay, thanks again for a very inspiring talk, please do make good use of Seb's time when you're out at lunch and enjoy his brain and his thoughts and his theories, so yeah, thanks again, Seb, cheers. - Thank you, cheers.
