Episode 8: Animation and Accessibility
Animation impacts accessibility in more ways than you might think. In this episode Cennydd and Val discuss the pros and cons of interface animation as it applies to accessibility and animating responsibly. Animation can lessen cognitive load and help prevent change bias, but it can also make using your interface more difficult, or even impossible, for some. Thinking strategically about your animation and implementing it well does more than just cater to people with special requirements, it makes your designs better for everyone to use.
Welcome to Episode 8 of Motion & Meaning; a little podcast about motion for digital designers. I'm Cennydd Bowles.
And I'm Val Head. In our last episode, we talked about physics for motion design and got super-science-y and all that; it was really fun. And today it's still going to be a little bit science-y. We're going to be talking about animation and accessibility which, as it turns out, is something we both have a lot of strong opinions about. Who would have thought? We want to talk about the various ways that animation can affect accessibility, both the good ways and the bad ways but I think we should get one thing about of the way first which is our take on accessibility. Accessibility, especially in web design circles, often gets framed as this totally super-annoying thing that just holds back all your design ideas and is just this bad thing you have to deal with because you have to and it's terrible and it sucks. And that's not how I think of it and I don't think it's how you think about it either?
No, no, it's not. I think we have a tendency in the web community to sometimes view it as a downstream function; it's an engineering thing, it's development, right?
It's like, on the end of the list! It's the last thing you do!
Right, yeah; make sure you've got your alt tags in and all this sort of patronising…well, so long as you do those, then problem solved. But I really don't think that's what accessibility is to me. It's not just about catering for people with special requirements; it's a way of approaching design that can make design better for everyone. When I tend to talk about this, I actually like using the label "universal design".
Oh, that's a good one.
Because that talks about the all-encompassing nature of accessibility. It's not just about tailoring for people who have particular needs but it's making sure that everyone can benefit from your service and so it's a design challenge as much as it is an engineering technique. I also think there's an ethical angle to this as well; it's important, I think, for us to make services that the whole world can use, not just a select lucky few. We should be contributing to the empowerment of everyone rather than just people with perfect eyesight or without any motor problems or anything like that so for me there's also kind of an ethical angle as well, so that's why I don't like seeing it relegated as just a code thing.
Yeah, and along the same lines, I really like that idea of calling it "universal design" versus…kind of framing it that way. But for web stuff as well, when it comes to web animation, we have all this cool stuff we can do and the idea of considering accessibility isn't actually going to hold you back: we can make amazing new things that also are accessible or universally designed that everyone can use and benefit from: it's not an either/or situation. They can exist together; we can make really great things that are also easy to use. So, with that in mind, let's get to some specifics. I guess we'll start with things animations can do to make UIs more accessible. I guess the positive part.
There's a couple of main things that I guess various studies and stuff have studied and how animation and adding transitions can actually improve just how easy something is to use. We see it a lot; it often gets talked about with performance where this idea of, instead of just waiting for things to jump on screen once they've loaded, you add these transitions to show that content is coming and it helps people understand what's happening that way; makes it seem like things are happening faster, which is always a nice bonus, the perceived performance. Other things that have been studied a lot in academic circles is the idea of animation actually reducing cognitive load; the idea on the most basic sense is the idea that if something is moving around on your interface, if you animate that movement, it's all on screen showing it going from point A to point B and that means that you don't have to do that in your head, you don't have to track: where did that button go? Where did that item go? Because you just saw it and that gives more room essentially in your memory or thought processes. There's more scientific terms than that, but whatever, there's more space in your head for you to keep track of more important things like maybe the content you're reading or the task at hand.
That's really nice because that reminds me of, I think it was Bruce Tognazzini, I did a course of his; he's a fairly old-school interaction designer, one of the pioneers of the field at Apple and he talks about knowledge in the world; I think it was Bruce, knowledge in the world versus knowledge in the head and how it's essentially always better to have the former than the latter because then you have less stuff bouncing around and this is a nice example of it. You reduce that cognitive load by getting things out actually into the system so the user doesn't have to remember state and so on. I think this is a really powerful thing for interaction design people as well, because they love talking about cognitive load.
I remember someone did a…like a UX designer or an interaction designer cheat-sheet or guessing game or something like that and cognitive load was the answer to pretty much everything; kind of in a jokey way, just throw cognitive load in half your sentences and you'll sound like an interaction designer. So, any interaction design folks listening: hey, here's your opportunity to talk about animation and cognitive load.
That might be actually the same reason why so many of these academic studies actually have cognitive load in the title or in the summary; it's just like…cognitive load; cognitive load; I'm like, oh my goodness, OK, we get it! You've studied cognitive load and animation a lot! There's some other things they've studied; there's a couple, I've been going through some of these studies and picking out the best ones and there's some good ones from the University of Rotterdam that suggest that well designed animation can actually help people solve problems more easily; essentially the whole cognitive load thing again, because they don't have to think about where things have gone on screen, they can use that to solve problems but also queuing techniques and complex animations can enhance learning performance basically meaning, if you actually guide people with motion as to what's important, they absorb or retain that information better but you're like, oh, that kind of makes sense!
When you're talking about queuing techniques there, is that highlighting certain things, that kind of motion to draw the eye to particularly memorable phrases or actions or whatever it is, that kind of thing is it?
Yeah, that study in particular, I think it was some part of how the heart works, they had these really complex animations people were supposed to watch and learn from and the main thing they were studying was if they actually put in little hints or animated cues of which parts were most important, where they should be focusing, people did better in the test at the end, which…hey, that's kind of…
That makes sense.
It does; it's the movement catches your attention and you can use that for good things.
I think advertisers have known that for a while, haven't they? Since the early days of banner ads on the web, they've learned that you can draw attention to certain things in hopefully memorable ways, so this is a more positive application hopefully of that.
Exactly. The banner ads aren't trying to teach you anything but the studies show that actually does work, which is why they do it and you should totally try to use that technique for much better things than banner ads; there's better uses for it. There's a lot of studies on cognitive load and less amounts for that interaction design bingo for anyone. But also other studies have shown that animation can improve decision-making and even help people better learn spatial relationships and I really like the idea of how animation can help people better retail spatial relationships because we tend to do that a lot: we have these layered interfaces where things are behind other things and we're dealing with this spatial idea a lot more in the things we're designing so the idea that, having transitions and animations and thinking about showing the motion of how these different pieces and states are related can actually help people better track where things are and even just make a better mental model of what it is they're looking at, whether it's your UI or specific information. It's kinda powerful.
The spatial angle I think is really interesting as well because in the really early days of the web, we took a lot of spatial metaphor: you had all these houses and office blocks and so on; this room was for this and we used those kind of metaphors that you move between those rooms and we quickly found that approach wasn't terribly good, so we moved away from spatial metaphor and spatial relationships for describing information space. But I kind of think that we're coming back a little bit toward that.
It seems like it.
Yeah, I think partly because the devices are smaller so you can fit less on screen so you have fewer of those navigational cues: you have to build up more of a spatial mental model of where things are in relation to each other because they can't fit on screen, particularly on a smart phone or a wearable so that suggests that kind of thing. And then also the bringing together of the physical and the digital world; I've studied way-finding quite a lot; way-finding systems and augmented reality and things like this; I talked about the VR piece I read a couple of episodes ago. I wonder if we're going to see more of that sort of spatial information space, building of spatial awareness in that, so maybe that becomes more important again; we tried and found a way that didn't work right the start and we're just using houses and all these kind of tacky illustrations. Maybe there's something deeper there that we're starting to realise where it is beneficial.
Yeah, I think so; I think it's Version 2 of that; we kind of force you, like you said, in these smaller spaces. Maybe we aren't drawing pictures of houses but we kind of use that, maybe one step removed or higher level or something. One of the studies that talked about the spatial relationships was actually dealing a lot with family trees and most of these studies are very learning-oriented and I feel like that's a result of them being in academic circles; they often study in e-learning contexts so we can take stuff from this for what we do, but it's not one-to-one, unless you're making e-learning courses which is probably not all that many of us, when it comes down to it. They're actually having people try to re-create this family tree that they'd seen and were showing different ways of presenting it and found that having animations to help show them that initial information helped them retain it better. And I think that applies to interfaces too; you might not be showing people a family tree but you're showing them how all the elements of my app are related and that can take on a similar structure. So it's really interesting stuff, I think, anyways.
It's not quite the same; there's not quite the learning outcome but there's still a learning process at the heart of a user coming to terms and understanding the system writing an application, figuring out how it works and so on is a learning activity, even if it's not, there's not going to be a test at the end of it. So, some of that will be applicable.
It's not like a course that you're like, oh, I need to learn this, but when you open a new app or a new site, you have to learn where things are; that's why we have all this on-boarding and craziness like that. There is a learning process there. On-boarding's a whole other situation though!
Oh, isn't it just? Yes!
We won't get into that! So, that's some of the benefits of animation that have even been studied by academics. There's some science and research behind it which really helps, especially if you're trying to make this case to your team-mates or your boss, that kind of thing, but there's also, when we're dealing with animation and putting UIs in motion, there are some accessibility factors or things that animation can affect, audiences animation can affect, that maybe other bits of our design don't as much. Some of them would be…well, I guess the main one would be folks with vestibular disorders which is an umbrella term for any sort of disorder or whether it's temporary or actually some sort of degenerative disease type thing; anything that has to do with your inner ear and balancing systems. A friend of mine described it as kind of like if you think that your inner gyroscope is totally off and just can't recalibrate itself; that kind of stuff. A lot of those vestibular disorders, not all of them, but many of them are triggered by visual things and those visual things can be movement on screen: that can actually trigger a reaction for people with vestibular disorders. Also, similar things can trigger people who suffer from really bad migraines or epilepsy; anything that falls in that area as well. A lot of those are visually triggered as well. It's that whole group of people or potential audience that can be triggered by visual movement and for people with vestibular disorders they could get really dizzy; they could have a headache and be nauseous for the rest of the day and that kinda sucks: you don't really want to do that on purpose!
No, that's right. This is all slightly scary because obviously the last thing I want to do is be a trigger for that kind of illness; that sounds like something pretty terrible I want to avoid. So, what sort of strategies do designers and developers have to try and reduce the risk of that happening?
Yeah, I've talked to a lot of people with vestibular disorders; I've been trying to research this a little more because, especially when it comes to web animation, holy crap, web designers, we've never had to deal with this before, right? Not never, but very little because the web is so static, we never really had the potential to make someone dizzy for the rest of the day and now that could happen. Oh my God, what do we do? And the really interesting thing about talking to people with vestibular disorders is, none of the people I've talked to were like, all animation and interfaces should go away and it should never happen. There are people who say that, but it's not the same people; there are just some people that are angry about everything. But that's not the folks with vestibular disorders; their main thing is that they want to know if they're about to walk into some huge amount of animation. This New York Times article they are about to open: if it's going to be this huge parallax, what was it, snowball skyfall…
It was a snowfall, yeah.
Snowfall! Skyfall is not a thing. That's a movie! They want to know if that's what they're about to open versus whether it's just your basic article of text. If they see something, if you tweet something that's like hey, check out this super-crazy web GL experiment, they can be relatively sure that that's going to be super-animated and they can decide whether right now is a good time to open that and check it out. But if you're like, hey, here's this New York Times article, they don't know: is it going to be crazy animated? Is it just going to be text? So, having some kind of warning and good context is helpful. A lot of the stuff actually we've been talking about in previous episodes of just having purpose for your motion; having it be there for a reason. All that stuff helps make this make more sense and lets people know if they're about to get into this or not. So a lot of that is super-helpful; the context. Things like not having stuff auto-play. If you have a big animation that's more along the lines of a video or a big demonstration; if that can be something that's not moving until you tell it to move: all the better. People can make that call whether they want to see it animated or not. And on a larger scale, this is something that I'm really interested in seeing how we can solve on the web, but there's also the idea of actually offering some kind of switch or toggle to reduce animation; we're doing a lot of, seeing a lot of super-parallax sites and a lot of places where large amounts of motion and the main triggering factor is really large amounts of motion. If you have a button that flips into place, it's probably not going to be a big deal, but if you have this huge full screen transition, that's probably going to be a problem, so this idea of having a reduced animation toggle it's been suggested by some people like Greg Tarnoff who's actually a web developer who has a vestibular disorder, so he's pretty well-versed in all this and that's basically some way for people who are very sensitive to this to tone it down or…
To opt out.
Turn it off so they can…because these very highly animated sites, it's not something that you can necessarily have pause at the beginning; if you have a very parallax motion or something, how do you pause that? It's part of your interface, it's part of your experience, so then this idea of a toggle to reduce the animation becomes maybe a better solution. We've seen it in iOS; they have a reduce motion when iOS7 came out, they added that.
Yeah, that's right. And that actually kind of brings the question for me then is: is that something that's best handled at the individual site or app level, or is that something that the OS or the browser should be turning off and saying, well actually, we're going to block these CSS properties or transitions or are we going to turn off this type of…I'm not too literate in native code but there will be some methods that are called and say well actually, we won't show those. I suppose from the individual site or app basis, if you're building in a progressively enhanced manner anyway, that should be relatively straightforward; you can put in some toggle somewhere, whether it's front page or whether it's in Settings, if it's a frequently visited app or something like that and because hopefully, build this the right way, it will essentially say, well these transitions will just essentially block the, suppress them. But it would be nice if we could find a way to standardise that. Do you know, are there any kind of conversations taking place along those lines?
Mmm, that makes sense. On iOS, I know the reduce motion setting is honoured across apps, if it's coded correctly or if you follow certain rules. For instance, I know the opening twitter animation which…I have reduced motion turned on, just because I find it excessively animated, but on the Twitter app then you've got the bird and you kind of zoom in and so on; that doesn't appear if you have reduce motion turned on in iOS, so obviously again, a Twitter engineer would know much more about how this was done, if it was coded in such a way that that is respected. Now, I'm almost certain it would be possible to code it in a way that it isn't respected…
That the reduce motion flag doesn't come into effect. I suppose that's where it does become an important part of engineering things the right way; choosing the right code solution, so that if someone has overridden that preference to say, look, I really don't want this stuff across any of my apps, then you'd better respect that because you may actually be putting them at risk by routing around their preference.
Yeah. Plus if that preference is there in something that's exposed, why wouldn't you, if you wanted to?
Well, never mind! That's a silly question. But it would be so awesome if we could have something like that in browsers too; almost like a moderniser for, do you want animation? Can you handle animation? Not just can your browser do it, but do you…can you handle it? That would be amazing, so I would love to see the web do a better job maybe than even some of the stuff OS has done. I'm hoping it's coming around and I know I'm talking about it a lot so hopefully other people are too.
That's great. Sounds exciting.
So that's kind of the main focus of these visually triggered vestibular disorders and that kind of thing. Things to think about and look out for and possible solutions. Of course, there's the rest…or not the rest, but there's other conditions animation can affect as well; things like reduced motor control, people using assistive pointing devices and that kind of thing, so I've definitely noticed on the web that not all animated navigations are keyboard accessible, so that's a thing that we should maybe try to do better. Some of them are though; there's some really good ones that I've been surprised you could tab through and use your arrow keys and everything worked and I was like, oh, you did such a good job, you want to hug them!
It's kind of a shame that it's come to that though, we want to praise someone for doing something that should just work; we should have higher expectations here.
There should be a basic consideration that your navigation is keyboard accessible.
You would think.
Let's hope we can improve there.
Well, for now, it's still an amazing thing when you're like, oh wow, that has 3D transitions and I can use my keyboard; you guys are the best! One other one for that, that I've seen for folks with motor control issues are using assistive pointing devices is, be really careful about when you have automated animations and the thing that you need to click moves. I know carousels aren't cool any more, but this happens a lot in carousels where the read more link, every time it transitions, it's in a different place. I was at an accessibility meet-up last year where there's this guy who was a college student, he was showing us how he uses his college's new site and when he used one of those Dynavox systems to use the computer and it was this really dramatic part where he wanted to read this article that was in the carousel and he was trying to get his cursor there but it was really slow because this Dynavox thing is just, that's how it works, and before he got his cursor to the link, it changed to the next thing and there's this whole room of us watching it and it was just so suspense-full and every single person in that room is never going to design a carousel like that ever again! It was just awful; every time he would almost get there it would change and there was no way to pause it, there was no way to stop it from transitioning; it wasn't one of those ones where if you put your cursor in it, it stopped animating or anything. So those are some big ones that I think are easy to forget.
I think this kind of reminds me of what I was saying at the top as well, in that if you get this right, it improves the design for everyone, because that's also just a basic usability mistake, to have touch targets, click targets that move. It's not just that someone will take a long time to acquire them; it might be just they're a long way away, that they only just noticed it at the last second; Fitts's Law, the time taken to acquire a target is proportional to the distance you are from it, so even if it's ten seconds, a user who can acquire a target quickly may still be trying to get there right at the last second and it still goes, so if you solve well for the cases that you may think, well, these are edge cases; they turn out not to be because they affect everyone. It's that universal design angle again.
Yeah, and that's totally, when you see some of that you're like, wait: I would like this better, even though I'm not using one of the systems. I have a mouse and I would like it if that didn't change; I could have been about to click on that too and sure, I would have gotten to the next link faster but still, that would be really annoying and I wouldn't like it. So it's one of those things that would improve things for everyone and it's really just, even just thinking about in more detail of how that thing would be used as opposed to just being like, animate it: yay, done! Which is never a good system. One other thing I've noticed that people have been talking about when it comes to low vision users or blind users is this idea of using aria roles, at least for the web, to actually hide things that are purely visual and that happens sometimes a lot in animations where we've created all these extra elements that maybe only visually have any…they're just there to be seen, they don't have any semantic value otherwise. And they actually make an effort to go in and set an aria role of I think, hidden. I'm not very good with aria roles; usually other people do that stuff. But basically just saying, this is purely visual; don't even try to read the alt tags on this or don't even try to read this content because it's purely visual. Which, I think it depends on the circumstance whether that's the perfect solution or maybe not even perfect, but the best solution, but I can definitely see how you get to these situations where maybe instead of having a video, you've got this very detailed animation and if everything in it is purely visual and that content is also in other places, maybe it makes sense to just…don't worry about this, just hide it, so it doesn't cause confusion or make a mess of things. So that's what I want to look into more because that's one area I still need to research now is how screen readers behave with animation because it seems like on one hand you want to be like, oh it doesn't matter at all, they can't see it; but I don't think that's actually the case!
I don't think it's the kind of thing you want to make assumptions about; it's one of those areas that a bit of research and then hopefully some thoughtfulness when you're actually building the thing and designing the thing could save quite a lot of users quite a lot of hassle so it sounds like something important.
Definitely, definitely, but that's one place that I don't have a lot of answers for yet, but I'm looking to get some because it's a really interesting thing to look into and I like that; the things we find out are just things that we can do to make things better for everyone and it's not like, oh, never animate anything because it ruins it for all these people; that hasn't been what I've found in any of this research so far. So that's always very encouraging.
Yes, I suppose that's good to hear.
I was a little worried at first, looking into all this stuff I'm like: oh no, what do I do if they all say that animation is horrible?
Right! We need to just announce the podcast actually ignore Episodes 1 through 7; exactly!
Delete it all! Anyways, I think that's…I guess some other helpful things we can do, things to look at, at least for the web, Web Content Accessibility Guidelines; suggestions are super-basic, they were written in 2008; kinda shows, but they're still a really good place to start. They've got some really basic things. Basic in the sense of like, this should be the first thing you do and then other things go from there based on what you've designed and how it might be used. They have the stuff about non-autoplaying animations having pause and play controls; not flashing the screen more than two times a second, I think; just basic stuff like that. So that's another place to check out. Also, there is a site, The Accessibility Project. It talks a lot about how to handle this for the web and another one that's similar is webaccessibility.com and we'll link both of those in the show notes. They talk a lot about best practices for accessibility, at least on the web; I don't know of any of these for native apps or any similar resources for native apps though; I don't know if you do?
I don't, to be honest. Maybe that's one thing we can throw open to our listeners; if you know something that we ought to about native apps and building those in accessible ways, then please get in touch and we'll happily pass on any information that we get on that. But yes, that's beyond my level of expertise, to be honest.
Yeah, I imagine it's got to come up in some places, right?
Oh, I'm sure, absolutely
I just have no idea what resources to look for there. So if you know some, let us know, because I would love to check them out too.
Yeah, absolutely. Shall we move on? Let's talk about what we've been reading, let's go with our typical reading list just to round things off.
So, for me, I keep doing this; I've been watching a lot of videos again! So I'm not going to link them up but I've been taking a motion design class, a traditional motion graphics class and so I've been watching a lot of videos for that. It's been super-interesting; it's very challenging, but I've learned so much as far as talking to people who have been motion designers for years and I'm like, hey I do animation on the web and it's kind of two different worlds. But I'm learning a lot. So I've been watching a lot of videos for those but a little bit of reading stuff; I read the Design Machines article which I know has been out for a little while but it's a long read, it's probably something you want to read on a Sunday morning or something, but it really changes the way you look at web design and designing for the web and I think even for apps and stuff, just the idea that we do so much of the same thing, we've basically turned design into something machines can do and we should maybe stop that in some sense. I think changing this…this call for change and how we approach this, I feel like motion has a big part of that in one of the things we can change. Just an interesting read overall.
I agree and I think that piece, I read as well, and it's been controversial; I think it's largely on point to be honest, I think there are a lot of really strong points in there. It reminded me a lot of Eli Schiff's Fall of the Designer series which I don't know if you've read, there's five articles in that. Not specifically about motion design; it's really about the almost local minimum we're finding ourselves with in design that everything's starting to look very similar and Eli mentions flat design and deprecation of visual design and so on, so I think Eli's five part essay and then this article as well, on louderthanten.com, I think they work quite nicely to pinpoint some of the problems with where we're at and again, as you say, hopefully motion offers us a bit of a route to clamber out the other end of this local minimum.
I think some of it is, we just need some good spark to get people out of this; we're in a rut of sorts just as an industry. We need to get out of it and we totally can, so it's super-interesting stuff; I'm going to read those essays as well. And one other thing I read recently, which it's kind of high level light sort of thing, it's The Importance of Time in Design up on creativebloq.com which isn't really a place you'd normally see super-detailed articles so this isn't a very detailed one but I just like the premise of this idea of, we do need to think of how time happens in design and how we design for the fact that time passes and include that in how we're thinking about design. So it's a good one to read and unlike the Design Machines, it'll probably take you about two minutes!
Right. OK, a couple of things from me and they're both on a cartoon slant because we've talked about cartooning a little bit through this series as you'd expect, I think, from a motion podcast. The first one is just a stupid little throwaway Twitter account; I had an Instagram account recently, didn't I? Well, this one's a Twitter account and it's @toongif or toonjif, if you're soft-g inclined; if you have certain opinions that are regrettable! So, toongif, it just has little gifs of cartoons, and that's it; just little loops; two, three second loops and it's…Wile E Coyote falling down a hole and all this sort of stuff…
And some Jetsons.
It's all that kind of stuff and it's just really nicely done and again, you're not going to look at it and go, I suddenly have the answer to my motion design challenge, but I really like it just as a way to keep things in mind and to see again how the masters tackled some of these challenges with all the principles we talked about right at the start of the series, so it's a nice little account, I recommend following it.
I just followed them.
And the other thing which is maybe a little bit related as well is, I was recently in Chicago at an event called Prototypes, Process and Play which was a great event and one of the talks there was just a short talk from Jen Myers called Cartoon Creativity: What I learned from Chuck Jones, Chuck Jones being, you'll probably know, one of the master animators going back to forties and fifties or whenever it was and it wasn't really a talk about motion design or animation per se, it was really taking more of a kind of creativity angle, but it was a very nice little talk and it highlighted some of the techniques and some of the…at least the new approaches that those animators brought to that emerging field and so kind of made me pause and think and recognise again; it's always helpful to have that reminder that there are people who've been doing this for a while and maybe you can look at some of the techniques that they've had and some of the tips that they've used, so it's a lovely talk and hopefully, I don't know, I think Russ's plan is that he's going to put all those talks up from that event, or at least the transcripts or something like that, so if you get a chance, do check that one out.
I'd definitely like to check that one out. I've seen Jen talk about similar things before; not this particular one, but it sounds super-interesting.
It was great. OK, so that pretty much wraps it up for today. In the next episode we're going to be talking about getting this stuff actually into your workflow; some of the process on how you deliver motion design as part of your project, how you overcome resistance, how you convince people to get this done and any shortcuts that are useful along the way so that hopefully will be quite a useful practical one, so I hope you can join us for that.
You've been listening to Episode 8 of Motion & Meaning with Val Head and Cennydd Bowles. You can find out more about this show atmotionandmeaning.io and we'd love to hear your feedback on Twitter where we are @MotionMeaning. We are also on iTunes which you may be listening through right now. If you're enjoying the show, give us a rating or write a review on iTunes so more people can find out about Motion & Meaning. We would like that really a whole lot. So thanks for listening, and we'll see you again soon.
Transcribed by Alison