5 Findings: Conversational Practice

interviewer: …she was doing something in Willoughby. #00:00:02-9# 

respondent: Ahh… she was planning to, run a study in Willoughby… #00:00:08-5# 

interviewer: Yeah. #00:00:09-2# 

respondent: Um, that didn’t happen in the end, so we had the X-Communicate symposium here. Did you come to that? #00:00:14-3# 

interviewer: No, I didn’t. #00:00:15-1# 

respondent: But, um, she ended up doing that and having a study at the Seymour Centre as well.  #00:00:20-3# 

interviewer: Okay.  #00:00:20-8# 

respondent: Yeah.  #00:00:21-1# 

interviewer: Yep. Um, all right. So, how I’ve been doing this previously is it’s roughly… well I aim for about 20 minutes.  #00:00:31-8# 

respondent: Do you need to turn this on, or…? #00:00:32-5# 

interviewer: It should already be recording.  #00:00:33-8# 

respondent: Yeah? #00:00:34-2# 

interviewer: Say a few words for me.  #00:00:37-4# 

respondent: Just testing. Test.  #00:00:38-6# 

interviewer: Yeah. Perfect. Um, the idea is that um, I’ll kinda just lay it out. This work is showing how my process unfolded, um, so I kinda take you through that, and then I wanted to talk to you about your process. Perhaps how that compares to mine, if it does, in any way. It’s sort of around 20 minutes that I’m after, but usually it goes a bit longer. We just get chatting, which is fine. Um, do you know anything about my work? #00:01:08-6# 

respondent: Ahh, no.  #00:01:09-7# 

interviewer: Okay. So I have spent the last few years designing interactive space for autistic kids. Um, the age group is about three to five. Um, I’ve done a few studies with um, autistic children and then a group of neurotypical children of the same age. Um, this is the space that they’ve come into for the studies. Um, I don’t consider any of this kind of finished work, it’s all prototyping and exploration of some of those ideas. So it’s probably easier if we go in here first so you can get a sense of what the kids did.  #00:01:49-8# 

respondent: Mmm hmm. #00:01:50-6# 

interviewer: Um, so they would come in here, one child with their parent, and it was non-directed. They could stay for as long as they wanted. It tended to be sort of between 10 and 20 minutes. Um, and they could play in whatever way they chose. Um, so this is the third and final interface that I made. Actually I’ll just let you have a play with it, first of all. #00:02:23-8# 

respondent: (long pause while pressing buttons) #00:02:42-8# 

interviewer: So the interaction with the buttons and coloured light, um, the idea with this interface is to have a very immediate connection to the response of the space, so obviously any coloured button, first press turns into that space, that colour. Um, if, I was thinking a little bit about issues with repetitive behaviours with autistic children, so just doing the same thing over and over again. Um, and so, it’s this program to, over time, if they stay within the same panel, the mapping becomes more complex, so same panel on the tent and interface. If they continue in the same space for even longer, eventually the mapping is totally gone, so you get different colours all over the space and different sounds. Um, I’ve been working with that idea as something from um, psychology called sabotage. So um, what I’m trying to do is I set up the sense of control really quickly and then eventually, that is taken away. So that’s the sabotage and what that, the idea with sabotage is by doing that, it kind of motivates the children to be more explicit about what they’re thinking or trying to um, get out of the experience. So the best example I had was an autistic boy who came in and immediately figured out that this was a scale of notes and so he went around and around and around and around. And the sabotage is programmed in, but I’ve also got like a sabotage button that I could trigger from outside the space. And when I did that, he immediately sat down and spoke to his mum and was like, “Can you help me? Can you try pressing that button at the same time?” So, it went from what was just a repetitive, kind of, I guess internal behaviour to being outwardly expressive. Even when I switched the sabotage back off, he continued to interact in that way rather than going back to the repetitive behaviour. Um, that’s really it. #00:05:06-3# 

respondent: Okay. Cool. #00:05:07-5# 

interviewer: Um, I’ve got a few cameras around the bottom here which is how I observe and record the um, the interactions, which I code later on. Um, so I’ll show you a couple of the steps that took me to there. (leaving dome space) This was a work I did for Honours a few years ago now, when I worked with a two-year old girl with autism. And the idea was to, um, kind of through her interaction, figure out which sensory preferences she had. So this first device was like a RFID data reader for sound, pressure sensors for vibration, and proximity sensor for like a visual feedback. She was clearly attracted to vibration and so the second device that one in the middle were these joysticks, which um, they all vibrated different speeds depending on which direction you pushed them. Um, but they had different material, so I was trying to separate whether it was really vibration she was interested in or just the materiality of the object. It seemed to be vibration, and so the um, last object I designed was a change table. So her mother, who I worked with as well, said that changing her was really difficult. She was active and hard to um, just settle down. But the vibration interaction we found really sort of focused her, so we thought maybe a vibrating change table would be helpful to kind of get her settled in that process. The trick with it was the two opposing corners were the triggers, and they were too far apart for the girl to um, trigger on her own, so she needed to actually ask her mother to help her make that happen. So I was trying to get some cooperative interaction happening.  #00:07:05-1# 

respondent: What do you mean with vibration change table? #00:07:09-5# 

interviewer: Ahh, so this last one here. #00:07:12-0# 

respondent: So as in vibration? What do you mean with, what was changing? #00:07:17-5# 

interviewer: Oh change as in ‘change her nappy’. She was a two-year old girl.  #00:07:21-5# 

respondent: Oh, yeah. Okay. Got it. Yes. Okay. #00:07:23-0# 

interviewer: So it’s kind of a funny angle there, but it’s kind of, you know, this size um, and underneath the surface, there was rows of vibration motors. Um, to trigger them, both of those red corners needed to be pressed at the same time. Um, the second device, when we introduced it to the girl, ahh, and she started using it in real life and realised that she was able to make that vibration happen, she immediately looked at her mother and showed what… there was a therapist there as well, and the therapist identified that as joint attention, which was quite important for the development of a child with autism. So that’s really where all of these ideas started. How does, how can this kind of technology facilitate better social communication between the child and their parent? Um, but I wanted to look at that on a larger scale, so they were very much designed for the one girl, um, whereas this – #00:08:25-8# 

respondent: For the two-year old? This is all for a two-year old girl? #00:08:28-5# 

interviewer: Yep.  #00:08:28-8# 

respondent: Okay.  #00:08:29-2# 

interviewer: Um, and so what I’ve done in this space is I’ve mapped out on the floor how this idea kind of led to some material explorations. I was looking at um, different materials, the way that they could… I wanted to create a space that was quite contained, um, that felt safe but not claustrophobic. Um, also, I’m using a lot of um, I guess sensory input that could be overwhelming for a child, so lighting and sound. I wanted to make it all quite subdued, so um, I worked with a material that’s got a nice, kinda softening effect. Um, and then I came into this prototyping stage where I thought through the more technical side of things, like how the lights work, um, the system involved. At the time, I was thinking that lighting and sound could be kind of a narrative way of exploring. But (cough) excuse me, um, it felt far too complex for the age group I was working with, and so this became more of an idea to think through what it might be like to have a sensory processing problem, which is quite common in autism, um, and how that changes the way you interact. So I made these headphones that have microphones on them, um, that pick up quite different sounds. So if there’s a lot of people around, it feels, um, it’s sort of a strange sensation. It’s a bit like binaural headphones where you feel a little bit… Ahh, what’s the word? It’s hyperreal, the way the sound comes in. Um, so I just wanted to get a sense of that change in sensation and how that would change how you experience something like that. Um, and then I went into the interface design. The first one was… You might be familiar with the Reactable.  #00:10:42-1# 

respondent: Yep.  #00:10:42-7# 

interviewer: Yep. It’s a very similar set-up.  #00:10:44-5# 

respondent: Yep. #00:10:45-1# 

interviewer: Um, these surfaces just replace the dome on top of that table, so that tabletop is removable. Um, I found this was just too abstract for the age group I was working with. Cause there’s no mapping of position on that surface. Um, I went into like a peg-and-hole kind of game where each coloured peg corresponded to one of the lights on the tent.  #00:11:15-3# 

respondent: Right. Okay.  #00:11:16-7# 

interviewer: It was more successful, but um, I had some issues with the autistic children particularly. Um, I had lighting underneath, which was these lights. And the quality of light was… Um, somebody gave me the feedback, it felt very sharp and so there was one boy who couldn’t look directly at the table. It was his way of kind of filtering out um, that intense light. Um, and so we ended up with that scale model interface, which I think is probably the most successful, particularly because it’s so immediate with the buttons. The camera tracking the with both of these had a little bit of lag and that was just enough to throw the kids a bit.  #00:12:03-0# 

respondent: Yeah. Right. Okay. #00:12:05-3# 

interviewer: Um, and I guess this kind of iterative process went back and forth with um, interviewing um, therapists, which I brought into this space, as well as um, doing post-study interviews with the parent and child after they’ve come into the space. Um, and that’s really it.  #00:12:29-6# 

respondent: Cool.  #00:12:30-6# 

interviewer: Yeah. Sorry, that was kind of like an unloading very quickly. (laughing)  #00:12:36-7# 

respondent: Yeah. Good, good. #00:12:37-1# 

interviewer: (pause) (cough) (long pause) Um, this was a really early interface design I was playing with. Um, perhaps not using an object at all, just the body to… there’s a camera underneath and blocking out lights.  #00:13:09-4# 

respondent: (pause) Okay. Cool. This is interactive? (referring to dome model) #00:13:22-4# 

interviewer: Yeah. It just triggers different scenes, I guess.  #00:13:26-3# 

respondent: Yep.  (pause) Did you evaluate as well or was it just like a short model? #00:13:33-0# 

interviewer: This was a pretty short model. It was set up in an exhibition. Um, the feedback I got was mostly that this was really kind of off-putting for a lot of people, that I didn’t evaluate at any kind of… Um, I didn’t quantify it in any way.  #00:13:53-2# 

respondent: Yeah. (pause) Good. #00:14:05-0# 

interviewer: Yeah. This is also the model that I’m um, trying to think through at the moment. It brings together a cybernetic system, which is kind of the way that I designed the interactive system and I’m trying to join it to… I don’t know if you know of activity theory. It comes in HCI a little bit. #00:14:27-4# 

respondent: Yeah. I’ve heard about it. Yes.  #00:14:28-9# 

interviewer: Um, so particularly, using this model for analysis. So thinking about how the person wants to change the object or objective, and kind of what happens in that area when they are sabotaged. So um, there’s the breakdown that I think Susan Bødker talks about it a little bit. So when subconscious actions become conscious, when there’s a breakdown occurring. So I’m kind of trying to make that breakdown happen so that the child sort of reassesses what they’re trying to achieve, and that just gives me something to observe. Um, so yeah. That’s where I’m introducing sabotage. As an external effect on the environment.  #00:15:21-1# 

respondent: (long pause) So this is sort of retrospectively describing your process or the model you implemented? Or like is it going to be used as a model for designing similar installations? #00:15:52-9# 

interviewer: It kind of, this is always something I’ve been working with in cybernetic um, systems, particularly Gordon Pask’s conversation theory. Um, that idea of the perceived boredom comes from his Musicolour system. I don’t know if you’re familiar with that one? #00:16:11-9# 

respondent: No. I don’t think so. No.  #00:16:13-1# 

interviewer: He created this lighting system which responded to the way musicians played music, so it would pick up a melody or a tempo and reflect that back. But, the best, well the most interesting part of that system was at some point, it stopped responding in kind and went off on its own thing, so it was like the perceived boredom. And then it was up to the participant to try and re-engage the system. Um, so that was an adaptive system which is beyond what I’ve got here, but that idea that um, it stops responding, and then so it’s up to the participant to you know, drive whatever happens next. And that’s the conversation. Um, so that’s not quite where I got to, but it’s the ideas I’m working with. The activity theory is very much the analysis of the video um, the video observations in the space and sort of thinking about what is it that I’m looking for? What kind of behaviours? And I think sort of joining them is my way of um, situating the sabotage within my practice. It’s, as far as I can tell, not something that is in HCI or interaction design. It’s a term used from psychology, but I can’t find work of anyone that has deliberately sabotaged interaction, for that kind of purpose anyway. I mean they talk about the breakdown as something that happens more when someone doesn’t understand how an interface is supposed to work rather than um, doing, you know stopping something from working on purpose.  #00:18:11-3# 

respondent: Well it’s most likely you can find it in the area of Critical Design. Have you looked at that at all? #00:18:17-3# 

interviewer: A little bit. Yeah. #00:18:18-5# 

respondent: I guess maybe some of Bill Gaver’s work.  #00:18:20-5# 

interviewer: Yep. #00:18:21-5# 

respondent: You’re familiar with Bill Gaver? #00:18:23-1# 

interviewer: Yeah. Yeah I’ve looked at creative probe… #00:18:26-0# 

respondent: Cultural probes.  #00:18:27-5# 

interviewer: Cultural probes. Yep, and then technology probes. #00:18:31-4# 

respondent: Yeah. Yeah. I’m not sure if it’s sabotage. It’s more, it’s sort of, I don’t know, I guess it’s taking away some of your control that you have. But um, or abstracting that control. Like um, have you seen the drift table, for example? #00:18:51-1# 

interviewer: Yeah. Yep. #00:18:52-0# 

respondent: So right where you have one button which brings you back to your home base, but you don’t have buttons to interact with um, that map in front of you. And it’s um, you have to work out that it’s weight-based and you put weights on the table to go in a particular direction. It’s sort of taking away some of your control that you otherwise have or used to have in a HCI system. But it’s not so much sabotaging the control as in changing the parameters. Um, and I guess in critical design I guess, it’s sometimes used to um, provoke a response in the participants. Like an unusual response or to have unusual discoveries. And it’s also like adding an element of randomness. Like there was this a really nice study called um… What was it called Photobox? Something like Photobox. So they built a little box that had a lid and a photo printer inside. And um, it was connected to your Flickr account and it would um, randomly print, at random times, print five photos a month from your Flickr account. So you’d open a box and there might be photo in there or not, but you don’t know of when it’s happening across the month, so it’s adding this randomness to where people then develop a very um, well an interesting relationship with an object. Because like some of the participants got very frustrated because they would keep checking it, and there was no photo in there. Then they didn’t like the fact that they didn’t have that control of that process, but then there was also this element of surprise, which made them actually build a relationship with it. Because of that element of surprise, that it sort of had almost this life of its own. But I don’t if you can – I’m not familiar with the theory of sabotage, so I’m not sure if that’s similar.  #00:20:46-2# 

interviewer: There’s some similar ideas. I think where this is different is um, I’m trying to establish control, very clearly straightaway, and then take that away. So it’s not… #00:20:58-6# 

respondent: It’s different.  #00:20:59-7# 

interviewer: Yeah. And then, what happens at that moment when they’re like, “Oh, this is not working the way that it just did or I expected it to.” Um, but it was interesting. None of the children… I kind of expected, especially the autistic kids, to be upset at that point, but I didn’t see any of that. They kind of worked harder to figure out what, how they could re-engage it. Unfortunately, I didn’t program it to the point where it could be re-engaged. It was more of an external thing where I could just turn sabotage on and off. Um, but I guess it had the same effect anyway because I was observing them from outside. Um, but yeah, I found that really interesting. I wasn’t working with low-functioning autistic children, um, which I would expect to have quite a different response. Um, but that’s probably where I’d like to go next with it if I was to turn it into a bigger study.  #00:22:01-9# 

respondent: Mmm hmm. Are you still doing that or is it the end of your PhD?  #00:22:06-8# 

interviewer: I’m hoping to submit for March.  #00:22:09-1# 

respondent: Okay, so you’ve wrapped up your studies? #00:22:11-2# 

interviewer: When I pack this up, I just write.  #00:22:12-6# 

respondent: Okay.  #00:22:13-5# 

interviewer: Write for the next few months.  #00:22:15-1# 

respondent: So it would be like in your next position or whatever you do next to become a new experiment. Okay.  #00:22:22-1# 

interviewer: Um, I’ve spoken to quite a lot of therapists, particularly occupational therapists, and they’re really interested in how this space would work in their own practice. Um, so there might potential work with those kinds of people, which would be great. It’s what I wanted to do this time, but um, I found it really difficult to get those people on board coming from Art and Design. Um, I guess the way I structured this, it wouldn’t give them the kind of hard data that is valuable to them. So that’s probably something I look at next.  #00:23:04-2# 

respondent: Because you’re already doing this as a practice-based PhD I imagine? So exhibitions are part of that.  #00:23:08-8# 

interviewer: Yeah. Yeah. And I didn’t start with a hypothesis like they would, I’d be like, “I just want to see what happens.” (laughing)  #00:23:15-9# 

respondent: Yep.  #00:23:16-1# 

interviewer: (cough) Um, but yeah. Their feedback has been great, so we’ll see.  #00:23:25-1# 

respondent: Mmm hmm. What questions do you have for me? #00:23:32-3# 

interviewer: Well, I would… In your own work, um, how do you involve participants? Do you involve participants at all? #00:23:43-0# 

respondent: Yes.  #00:23:43-2# 

interviewer: Is it just after you’ve produced the artefact, or… #00:23:48-8# 

respondent: Ideally we start with the participants. Yeah. So it depends a little bit um, as always it depends a little bit on timing, time, and budget also. But um, usually we try to start the participants and um, to what is known as user research, as you probably know and heard, and collect data from our participants before we actually do an installation or an intervention. I would call probably design intervention, rather than installation. I’m not coming from an artistic background, more from a design background. And ahh, again, depending on the particular project, it would take the data whether it’s interviews, surveys, or observations, or a combination and then analyse that in order to gather the requirements. Um, we often develop things like personas for example, and then develop artist storyboards for example. Both in terms of using storyboarding, in terms of representing the current situation and then using the storyboarding to describe the future scenario. So where we sort inject intervention into the same scenario and say this is how it’s going to look when, once that has been installed and the situation change and how the behaviour changes of people through installation. And then we built um, prototypes, and that can be paper-based or mockups and then ideally, we can go back and test it with participants, so similar to what maybe you have done here. With some of the prototypes are in particular I guess this is sort of a series of iterations. Um, and uh, so depending on the scale and the type of project that um, takes place or does not take place and some it has to take place with constraints. Like um, for example, the installation piece (incomprehensible) at the AMP building where right, which was massive media-facade projection, so that was very hard to test. So, what we did in that case, for example, was… Look, my PhD student at the time was doing a lot of simulations on this desktop computer and that was sort of a starting point, sort of like setting up with two floor projections. Setting it up as two displays, and just looking at that with live interaction input. And first of all, evaluating it ourselves and then we also made an installation in the Seymour Theatre, so we booked a theatre for two days or so, and used their stage. Cause they have like (incomprehensible) a stage mechanism, we used that to mount the projectors and actually have a projection of one of the surfaces and then send participants through this space and get to them the first experience from them and it was observation actually only, very informal. Just send them through and watch them, how they respond and um, engage with the work. Um, and then we also did a series of controlled studies where we tested different parameters in the same theatre space as well. Change parameters. Like for example, sound feedback, so we tried different ways of using or implementing sound feedback, or sound as a feedback mechanism and then adding surveys after that, so it was very structured research to ask people about their experience and look as that was happening. Um, before installing that, then actually at Circular Quay, where like it’s much harder to do any adaptations once it’s installed. I think we were still doing software adaptations in terms of the responsiveness and obvious things, but once it’s out there, it’s harder to make changes. But even then we sort of observed people in the space, how they interacted and also recorded the things on the computer, so we recorded their movement patterns, which is very similar to what you do with your cameras here. And then sort of wrote a reflective piece on that, on how people would interact with it in actual space.  #00:27:49-1# 

interviewer: Okay. Did you take any, I guess feedback data from people once you had the final installation? #00:27:57-4# 

respondent: With that work, no. Um, it was partly because of ethics.  #00:28:02-0# 

interviewer: Okay. #00:28:02-2# 

respondent: I think we, I can’t remember the time we didn’t have ethics or it whether was too complicated, like in um, some cases, we have ethics approval to distribute flyers for example for interviews. But um, it’s very hard, as you may know, it’s very hard in a public space to get people to interact with you and pause and to talk to you. So in that case, we didn’t do that. Another work we had last year with a student of mine in Vivid which was this augmented bin that you could play Tetris with, so we put something in to play Tetris. And there was an LED bowl around the bin so the bin was a display essentially, like a game world. Um, and that was also installed at Vivid, but without doing any, um, direct, seeking direct feedback from participants, but then we installed it again on our campus later. Where we had more control and we could um, have it over a longer time period and turn on and off different things and then both collect data in terms of like how much interaction is happening with the bin as well as doing interviews with people in the space. So often, we often try to collect a more structured feedback in a more controlled environment. #00:29:18-3# 

interviewer: Yep. #00:29:18-8# 

respondent: Um, which is probably partly because of a lot of our work being in the urban space where it’s very hard to collect information. For example, the project we did with Ian, um, we didn’t do any user research in the beginning because um, that’s not right, we did indirect user research, so we spoke to the Council and other stakeholders about the program and that they were trying to solve and so we’re trying to get information from them rather than going straight to the participants directly and ask them. Which is probably a bit of a time as well as permission and ethics reasons, for those reasons, we didn’t have the chance to actually go out and collect information. Um, so we sort of ended up doing more like an action research model where we’d install something, like we built something, installed it, observed, do interviews, and then take that back and make the next iteration. So, it’s a slightly different process. #00:30:14-7# 

interviewer: So in those projects where you don’t get direct participant feedback in the final stage, do you, how do you measure success of a project? Do you try? #00:30:26-9# 

respondent: Um, it’s not about measuring success of the project for us. Which, I don’t know if that’s maybe different to your work here, but um, in many cases, it’s more about identifying patterns or strategies or a combination of those. So, um, like, I mean with the Tetris bin for example, so the Tetris bin installation, one of the goals was that um, people would stop littering right, and put rubbish in the bin rather than on the street. Um, and we tried to measure that I guess in a campus study. But there was so many variables you can’t control. It’s actually hard to quantify that. And because most of the research I’m doing is qualitative rather than quantitative, I’m also not usually working with hypothesis and hypothesis driven my research. It’s more explorative. It’s usually not about saying whether you’re successful or not. Right, if you’re going to have a hypothesis, then you can’t really say anything was more successful or not. It’s more about um, identifying new knowledge in a particular domain. So, um, and with like the AMP installation, for example, we identified different ways of how people would interact in that urban space. Augmented urban space and the social interactions that were occurring out of that as well. And how their listening to the time of the day or the time of the night and the number of people in the space and all those things.  #00:32:03-7# 

interviewer: Yep. How do you work through that kind of data?  #00:32:07-4# 

respondent: The observational data? #00:32:10-5# 

interviewer: Yeah.  #00:32:10-7# 

respondent: Um, so, different – in an another project where we had a public interactive screen that evaluated. We have video recordings similar to this. And so we have, because we’re using a Kinect camera in that case. We are recording depth image, depth-camera image um, which shows a number of people and what they’re doing. And then we also had a recording of the screen that they’re interacting with and then we were superimposing those so we could look at, okay, obviously behaviour and what you see on the screen and then we would go though those videos and coding them manually. And in that case, we actually just used, we didn’t use any software, we just went through all the videos manually and had an Excel spreadsheet to code.  #00:32:56-7# 

interviewer: Okay. #00:32:57-6# 

respondent: Um, so it was, we did, oh I keep forgetting, but it’s inductive or deductive. I think in this case it was deductive because we had a um, either way we had a criteria beforehand so we were looking for these three or four different patterns and we go through the videos and just identify when those patterns occur and make notes along the way. And then also get, depending on the research we drew, we’d get at least two coders, so that we can look at an agreement score at the end and then calculate that or we would get someone to look at the random subset of the whole amount.  #00:33:33-8# 

interviewer: Yep. #00:33:34-3# 

respondent: So if it’s, especially if it’s a more straightforward coding system, we’d get a random subset and then give it to the second person and then get the second person to code it and then see whether there, how much overlap there is to the first person. So that’s how you sort of like, you know justify that your coding, um, there was some rigour to the coding you’ve done. Um, with interview data, we usually either using software or just pen and paper or like, just paper like print out of transcribed interviews. Cut them into chunks of information and then just sort of shuffle them around into groups. And um, often (incomprehensible) the inductive content analysis where you sort of create categories as occurring through the contents. You’re trying to look for patterns in the content you collected. Um, like in one research, we looked at for example, it was a visualisation projection on the ceiling of your um, social media interactions, for like on Facebook, um and like if we both be talking loud, we’d each have a halo essentially projected above us and we would show our interactions as well on Facebook. #00:34:43-5# 

interviewer: Okay. #00:34:44-3# 

respondent: And so we validated that in focus groups and then we found patterns like um, I don’t know… privacy, um, I think interaction in terms of how to interact with it, usability problems, social um, interactions sort of occurring, these sort of things. So then we had those categories and could say within those categories and made those observations and this is what we think. Therefore it needs to be considered with these sorts of interventions. That’s the generalisation that we do (incomprehensible). #00:35:15-7# 

interviewer: Good. #00:35:19-7# 

respondent: Does that make sense? #00:35:20-7# 

interviewer: Yeah. Yeah. It’s kind of similar to where I’m headed. I haven’t analysed all the data yet, but um, it feels like- #00:35:29-6# 

respondent: I mean there’s, for video analysing, there’s software as you probably know. I keep forgetting the name of the package. Envideo? #00:35:37-5# 

interviewer: Yeah. I’ve heard of that. #00:35:38-5# 

respondent: And most universities provide that for free. And that then allows you to I think annotate video segments and like, you know make marks on the video at all, so it’s… um, if you have a lot of video to go through, it’s probably a good thing to do. Um, in our case, we had to do the analysis quickly, because there was a paper deadline and we were faster doing it that way manually rather than importing all the video (incomprehensible) direct form (incomprehensible) and office installing software, etc., etc.  #00:36:09-2# 

interviewer: Cool. Interesting. Um, what else was I going to ask you? I think that’s all.  #00:36:24-0# 

respondent: Really? #00:36:24-9# 

interviewer: Yeah.  #00:36:25-2# 

respondent: Okay.  #00:36:26-1# 

interviewer: I mean, really, I just wanted to compare practices. Um, so yeah, that coding information’s really interesting, the way that you work through… Um, kinda, you bring in participants at several stages in the process. That’s really interesting.  #00:36:49-2# 

respondent: And ideally, you want to bring them in as much as possible right. And there’s of course it’s the idea of co-design or collaborative design or participatory design where you bring them in as designers, considering that they’re not designers. Um, but that takes just a lot of time and commitment.  #00:37:08-3# 

interviewer: So when you say participatory design, would you ask sort of specific design questions of the user or kind of take their more general feedback as something that directs you to the next iteration? #00:37:22-4# 

respondent: Um, it’s a combination. So you actually get them into design solutions for you, but you don’t use them as solutions. You use them as user feedback.  #00:37:33-4# 

interviewer: Yep.  #00:37:34-3# 

respondent: If that makes sense, because they’re not designers. But um, it requires, it’s a long process and depending on which methodology at, it has four stages where you first of all um, immerse them into the problem space so they’re familiar with that and you get them to talk about their current experiences, which is also easy for us to do, right? It’s easy for us to talk about existing things as iterations. Then you get them to design solutions for the problem that you’re working on. And that can be using paper, cardboard, or low-fi materials, and then you get them to enact that. So you actually get them to take the prototype that you just built and pretend to use it. And then all that sort of feeds into your user research and you analyse it the same way you’d analyse interview data, for example. You don’t say, “This is the solution that they wanted so I’m going to build this now”. Which is often a misconception.  #00:38:31-9# 

interviewer: Have you ever worked with a population who couldn’t give you that kind of feedback? So, like, one of the reasons that I dropped autistic kids straight into here is because they couldn’t give me that feedback. And I think there’s the argument with technology probes, that you have to present kind of a functioning prototype to start getting feedback.  #00:38:51-9# 

respondent: That’s the idea of technology probes right, where you don’t start with the user research. You sort of build something and then you test it. And um, I think Bill Gaver does that in his research too, in that he doesn’t necessarily go out and collect user research. Um, I’m trying to think of this project. Some of his projects, is sort of immersed himself into this space and it was the cultural probes as was a way of collecting information and knowledge about the users. But there’s still then, I mean with the cultural probes start to take one bit of information that someone wrote on a postcard and use that as a design inspiration so the cultural probe is more of a way of getting inspiration as a designer rather than doing like a user research or an interview, where you know you interview a number of people from each of your sample populations and then you try to extract and generalise all that information. You know, just one piece of information is enough to come up with an idea, like the Drift Table idea underlying for that was someone wrote on a postcard, um, I think the question is “what do you do when you sit alone and you’re really bored at home and someone said, um, imagine my living room as a spaceship.” That’s what they wrote on a postcard, right? And so they went, “Oh, this is a cool idea. Let’s build this.” And then everything else was their designer, what does it call it, intuition in a way. They said, “Okay it’s going to be a table and you know, buttons.” It’s weight-based and all that stuff. It’s sort of the designer input and then they built that as a technology probe and put it to people and then see how they use it and often they use it in a different way, like that example that I had, which was the, what was it called? Um, I think that, the cupboard where you put your keys on when you come in. I don’t know if you’ve seen this project. When you come into your home and you usually have a cupboard and area where you throw your keys. And it was linked to a picture and depending on how you throw your keys on the thing, the picture would tilt or (incomprehensible) it would visualise your emotions when you come home and you slam the key. And the picture gets distorted. But then people used it in a very different way and started developing a personal relationship with the picture and trying to use the key to control the way that the picture was responding. So there was, they were using it different ways.  #00:41:08-8# 

interviewer: Yeah. Yeah. #00:41:09-8# 

respondent: Yeah. And that’s just a different approach to design. And that’s not one of them is better or worse than the other. It’s just a different way, and um, in many ways, if you do, if you’re using technology probes, you might discover um, solutions to what you otherwise might not have discovered, right. If you start by going to the user and collecting, you get sort of lost in the process of collecting information about the current problems, then you get stuck on that bit and maybe just sort of make a very small incremental improvement to an existing problem or an existing application you have. Whereas if you come in with your, um, using your design skills and knowledge as inspiration, you might develop something that they, that would’ve never come up as the data otherwise, and then see how they use it and refine it that way. Yeah. It’s the idea of critical design too, right. Critical design is not based on what users want or need. It’s sort of throwing something crazy at them and then seeing how they use it.  #00:42:13-0# 

interviewer: Yeah.  #00:42:13-1# 

respondent: Yeah. Yeah.  #00:42:15-1# 

interviewer: Great. All right.  #00:42:19-3# 

respondent: Good? #00:42:20-8# 

interviewer: I might stop it there.  #00:42:21-9# 

respondent: Yep.  #00:42:22-6#

Back to top