Human Factors Minute will become available to the public on March 1st, 2023. Find out more information in our: Announcement Post!
Dec. 23, 2022

E268 - How ChatGPT exemplifies Chatbot UX

This week on the show, we talk about ChatGPT’s UX and why the conversational UI is powerful. We also answer some questions from the community about transferring to HF/UX as an OT, how to learn about research methods after transitioning to the field, and when to give potential employers the password for your case study.


This week on the show, we talk about ChatGPT’s UX and why the conversational UI is powerful. We also answer some questions from the community about transferring to HF/UX as an OT, how to learn about research methods after transitioning to the field, and when to give potential employers the password for your case study.

Check out the latest from our sister podcast - 1202 The Human Factors Podcast -on The CIEHF - behind the scenes - An interview with Tina Worthy:

News:

It Came From:

Let us know what you want to hear about next week by voting in our latest "Choose the News" poll!

Vote Here

Follow us:

Thank you to our Human Factors Cast Honorary Staff Patreons: 

  • Michelle Tripp

Support us:

Human Factors Cast Socials:

Reference:

Feedback:

  • Have something you would like to share with us? (Feedback or news):

 

Disclaimer: Human Factors Cast may earn an affiliate commission when you buy through the links here.

Transcript

Welcome to Human Factors Cast, your weekly podcast for human Factors psychology and design.

 

 

Oh, yes. Hi. Hello, everybody. Welcome back to another episode of Human Factors Cast. We're actually recording this episode a little bit early on December 20. You're all listening to it on December 22 or later. We know how podcasts work, so you're either listening to it on December 22 or later, but we're recording a little early this episode, 268. My name is Nick Rome. I'm your host, and I'm joined today by Mr. Barry Kirby. Hey there. Hey there. Hello. It's the last time you and I will meet this year, I think. Yeah. Although we will meet next week and we'll get to that. But let me just go over the show first. Let me tailor the tape. We got a great show for you all tonight. We're going to be talking about Chat Gptpt, and we're going to be talking about the UX and why the conversational UI is powerful in some ways. We'll also be answering some questions, asking questions. We always do that. Answering some questions to the community about transferring to Human Factors or user experience from as an occupational therapist, how to learn about research methods after transitioning to the field, and when to give potential employers a password for your case study. But first, we got some programming notes, and boy, do we have some good ones for you today. Next week, like I said, we'll be back well, Barry and I will be back in what we call our holiday tradition, or you can think of it as our Human Factors Cast holiday party, if you will. We all kind of get together. We recap all the stories of 2022. Barry and I will break down every month of the show from all the stories. We're going to do that. We change our clothes a few times. We do, yeah. There's no expense spared in the costume department and hair styling. And also and I hear you do not want to miss the last third of that show because we haven't recorded it yet. We'll see how that goes. Anyway, that's next week, and then we'll be back with another regular show back at the beginning of January on the fifth, so you can look forward to that. I hope you all have a great holiday season. I just want to start the show off in that way. It's a holiday season. Hope you all are well and are just having a great time with your families and the holiday cheers just out there. I don't know, I feel like that's something special that I don't necessarily say all the time, but in the spirit of the holidays, we're giving there's so much stuff, Barry, I don't even know where to start. So we're doing like, a patreon refresh again. We do these every couple of years. We kind of reevaluate internally what's working, what's not working. And so some of the things that are going on here. So we're kind of updating some of the role titles. Our honorary staff are now called All Access Members, and I think there's a couple of reasons for that change. We did clarify a couple of the benefits. No benefits are removed, we're just adding some. And so now is a great time to become a patron, especially as some of this stuff will be rolling out at the beginning of next year. But we're excited to kind of announce Human Factors Cast Academy, which is our way of storing all the resources and sharing all those resources with you, all that we have put together over the years. We're slowly building this repository. We're planning to make our own courses at some point and put them up there, share content that is useful for folks so that's available to our All Access members. We wanted to make that tier a little bit more enticing and also give back for those who are supporting us at that tier. So there's a couple of other things that I want to mention. First off, our Deep Dive series, that is. Now, listenable, you can go listen to those stories to those Deep Dives if you'd like, half of them anyway, but those going forward will have full audio tracks that you can listen to as you read along. And those are awesome. Our Deep Dives are great, and I'm only allowed to say that because I did not write them. So go take a listen or read those. You can find those on our website under the blog. You can find all the Deep Dives right there. And then last but not least, I'm going to tease something really exciting for next year. That's it. That's all I can say. There's something really exciting coming next year that we can't quite share this yet. We're having a meeting about it on Friday and it is going to be an awesome thing. If you've been listening to the show for a while, you know that we've always had stuff kind of in the back hopper working on stuff. This is going to be a good one, so stay tuned. All right, Barry, I talked for like five minutes on this stuff. What's going on over at twelve two? So twelve two. The interview with Tina Worthy is still up. She's given us all the insights into how the CIA CHF the chart of economics and human factors here in the UK, how it works and what goes on behind the scenes. The next episode may or may not drop before the new year, because my workload has been super stacked. But where I've been crunching the numbers on what's been listened to most, what's been watched most, and really, some of the barriers that we've broken this year, some of the targets that we've hit that I didn't even know, we're going to be targets this year, but we've broken them anyway, which I think is fantastic. So, yeah, so keep listening and hopefully if I can get some time, we'll get that episode up before the end of the year breaking. Barry, I love it. I finally listened to that episode with Tina. It's a great listen. Tina is just so great to listen to. You guys have some great rapport with each other. All right, well, why don't we get into the part of the show you're all here for?

 

 

It's called human factors news. This is part of the show all about human Factors news. Barry, what is the story this week? Surprise us. I think I should take over. So the story we're going to talk about this week is why is the user experience of Chat GPT so powerful? Chat GPT, if you haven't heard of it yet, clearly we talked about it last week is a chatbot application developed by OpenAI that is gaining rapid popularity, getting 1 million users in just five days after its release. It's been hailed as a tool that has the potential to disrupt not only search engines, but also as fields such as elearning, writing and editing. The software is really highly versatile, with advanced natural language processing capabilities that allows it to assist users in a wide range of tasks and help them expand their knowledge and skills. One of the elements that's been hailed as a key strength of Chat GPT is its simple and intuitive user interface, which follows a no UI approach, meaning that the interface is designed to be as minimalistic and as unobtrusive as possible. The perceived thinking is that it's better to have a simple and intuitive UI that allows the user to discover the power and functionality of the software on their own. This creates a sense of excitement and adventure as the user can discover new capabilities of the app and be pleasantly surprised by its power. This allows users to focus on the conversation and the task at hand, rather than being overwhelmed by the technology. The simplicity of Chat GPT's user interface also makes it highly learnable, allowing users to start using it quickly and easily without the need for extensive training or instruction. So Nick, what are your thoughts on a Noui approach, a no UI? I think that the chat box is the UI anyway, look, I think generally I'm pretty positive on this technology. There's been a couple of good points brought out by listeners who seem to think that I was really excited about this. I am for the potential. One comment that we got was sort of, yes, this is great and can be transformative, but this is going to be used in the worst ways possible. People are going to use it to collect data and information on people in a million bad ways. And yes, that's true. We've mentioned on this show a million times, policy is not where it needs to be for where AI is right now, and we just keep pushing forward without thinking about the consequences. Yes, that is true. Also, I want to say that even though all that aside, thinking about the UI of chat, GPT and other services that will ultimately come to follow, it is a conversational interface, but prompting is still an art, being able to put in the right words to make it do the right thing. And some guy is mad at us on TikTok because we don't know how to prompt correctly. But I mean, that's not the point, right? The point is that humans shouldn't have to think about it for the system to return what they're asking for. And in a lot of cases it does return what you're asking for. It's just that there's some like edge cases where it needs to understand a little bit more context around what you're asking. So in some ways this is transformative, in other ways this dude on TikTok needs to get with the program. So I don't know. Barry, what are your sort of initial thoughts on this? Well, firstly, I slightly got my high horse about this, I'm Fabulously Honest, where they claim it's a no UI approach. No, it's not. It's got a UI, it's there, it's minimalist, it's very clean, but you've got to put in I think one of the big drives I always have is to have a minimal training or zero training approach to UI because there's all sorts of benefits if you go down that approach. This you have to really exemplify by your guy who's got TikTok. If you have to learn how to use the system in such a way to get the right sort of results, then the UI has failed as far as I see it. But as you point out, if you're just tapping away into it, you can actually put in some really almost bizarre sentences or requests in there and it will come up with some decent stuff. And I think we'll go through some of this about why this whole no UI approach I think is a phrase that I want to consign to some sort of bin and it stays in 2022. I think it is still quite clunky. It's anything that is just purely text driven that demands you to be typing in that way is still chunky. It's not the smoothest UI I've seen, but then that goes down to what is it actually for? Because if it's for document writing and stuff, then actually if it is for you to sit down there and chat with, like we've really stretched it in various ways and I've done all sorts of things with it to see exactly what sort of output it's going to give you, then it's really good for that. If you could upload documents, there's all sorts of bits to it that you could make it better or make it more adaptive or whatever. But what I'm struggling with is what is it actually for? The example, the analogy I'll give is Google Glass, Google Ass, brilliant bitter kit, really, really amazing bit of kit. And what it could do was fantastic. And I spent a lot of time looking at different use cases, everything from police and ambulance services all the way through to the military, through to just general users. But one of the reasons that Google has sort of finished the way it did, and it's not entirely over the idea of more stuff with it, but the way that that whole thing finished the way it did is because nobody could truly find it's great. It's a nice gadget, but what's the actual requirement? And so this is kind of what we need to do with this, is I think there's a lot of utility, but just we've got to nail actually what is it for or what are instances of it going to be for and how is it marketed? But fundamentally, I guess it opens them bigger questions about just how important is a UI going to be for all sorts of AI, lots of two letter acronyms going around tonight. Just how important is that UI going to be to really, truly allow people to get the full power out of this interface, out of this chatbot? Just saying I'm angry, but I love it. I'm wondering, you bring up the example of Google Glass and I'm wondering if this is not sort of a multi tool, right? Google Glass had a lot of applications and AR wasn't mature as it is today, but something like this feels more like a multi tool. And I didn't really talk about it at my initial reaction, but I'm thinking something like this type of interface would make voice virtual assistants actually useful, right? So if you have a UI where you have the in home devices that I've stopped using because I found sort of the limits of their usefulness to be turning on and off the lights and to tell me play a song or something like that, right? But this would make them tremendously more useful if I could have a conversation with them and be like generate a couple of ideas and then have it come back snappy with just a couple of ideas right then and there. Or if I'm working at my desk and I'm writing on something or typing, or give me this so that way I can just ask it and multitask while I'm using it. So I think that in itself and having sort of understanding what the text input is going to feed into those future applications. I think this could also replace something like Google, where it is a true assistant in the sense of if you have something that's chat GPT supposedly doesn't access the Internet, although you can get it to do it there. Are ways in which, if it was attached to the Internet, you could ask for the latest news on something, and it could return something based on the context that you've given it. Give me the news within the Human Factors domain. Boom. Then you don't even need our weekly roundups anymore. See, we're obsolete. Or I still do it, but that's how I get it and to me the simplicity of being able to query in your everyday life is sort of the benefit here. I'm very careful about what information I put in there because it remembers everything. So I don't use it for work yet, although I can see this being very useful, where you put in a list of bullet points and it comes out with a summary for you. And I've been tempted, but again, company data, I don't want to put that in there yet because I'm just not sure it's not been vetted with our It department, all that stuff, but personal uses. I've put in like a bunch of stuff for the podcast because I'm not afraid of anyone else gets in the human factors podcast space. Who cares more than merrier for me? The productivity increase of having this I have almost like a dedicated window up. Funny, I have it up on both of my side monitors right now. But I have a dedicated window up for this thing now because I have now shifted my workspace to have two working windows and one assistant window that I can just throw prompts in there. Just by the way, they've added an update. Definitely since this thing was published and definitely since we talked last, where now it has all your conversations down the left hand side, so it remembers everything and it's just a bunch of new chats for me. And I can't identify which one is which because I've thrown in like, 500 something odd queries to this. So I'm like searching through new chat no, that's not it. New chat no, that's not it. But it's nice as there I can continue those conversations when I want to anyway and then I've also wasted things like AI, toilet paper dispenser if you're in the pre show, you know what that is, chat GTP. So ultimately I think there's some good value to having this thing be easy and ubiquitous to use for users. No UI, that's a little bit of an exaggeration. The UI is a conversational interface and we've talked about conversational interfaces on the show before, but the conversational interface is very low friction because you're talking to it like you would talk to anything else and you're typing to it in a way that you would type to anything else. And sometimes that communication can even be looser than you might talk to somebody else. Like, hey, here's a bullet point, do something with it. And it's like it would do something with it and you're not specific. And so it'll bring back a couple of different things. And having that type of ability with just so little information that you can put out there into the world and have it generate something that is useful to you, ideally is the power here that we're talking about when we look at this UI or the usability of this program. There's a million different directions we can go. Barry, where do you want to go? So I kind of wanted to critique just some what would be the top three changes you would make to the UI to make it useful for you just in using it for me. Because in the article that we've been using and I'm going to hopefully pull that up if I remember where I put it. No, I'm not. They sort of highlight that the other way that this could have gone would have been for the software to clearly display and explain everything to the user about what it is that they could do. Like have loads of stuff up on the side, loads of queues and all that sort of stuff. And they've gone down this clean approach, not have lots of info boxes, not have lots of tool tips and links and training things. And you can argue actually, is there a happy medium between the two? And I think there is, as you've sort of highlighted every time you start a new chat now, you get a new chat save in the left hand column, so you have the opportunity there to change the name of the chat. And so it just says new chat. That's what it's titled as. Wouldn't it be so neat if it just actually bought in the first thing that you'd put in there? Because the chances are you're putting in a prompt about something and because it could be clever, you would actually just put the topic. It does do that. The new chats are for the older chats. But like, for example, the AI toilet paper dispenser that it came up with its own, and it's labeled that chat model. Unable to help is the result of the first query. The last one before that, which was, by the way, this one is. Give me 268 reasons why Human Factors cast might be the best podcast. It's the query it did. There's 268 reasons here and I could go through all of them, but that would take way longer than we have time for. So just know that you're listening to the best podcast, one of the best podcasts. I guess the bit that I wanted to pick up here is really if you're just sitting here conversationally and your topic of it is brilliant, and you can listen to the last episode as well, where we wax their record, just how amazing it is. You show this to people who have not seen it before and literally Jaws hit the floor. It's all, wow, look at the quality of the output you've got. So if you've got the time here to sit and type and do stuff, then it's fine, it works really well. And I think the way that you describe it is kind of the way I'm using it as well. So I've got it open as a half screen side window and I love it for we've been talking about all sorts of things like how do you developing a training course. We're thinking about doing some of our own training courses next year and so I just typed in how would I plan a training course? Well, you'd lay up this and you do this. There's many ways you could do it, but we would suggest doing it. And it's not necessarily telling you how to do it, but it's getting you off that start line. Rather than looking at a blank sheet of paper, it's giving you some really good content that is not copied. It's been developed for you so you can use it, but then you still got to go and tweak it. If I'm honest and open with everybody, the blurb for the news, I didn't write that. I was going to ask I was going to ask with the exception of there's one paragraph in the middle that I put in because I thought it added just a little bit more use and just a little bit more to it. And what I sort of put in saying I give it the entire article and say, can you give me a 200 word overview, 200 word synopsis go? And you come back with like 170 words. And I was like, thank you very much. Whereas normally I'd spend a bit of time copied and you'd synthesize it. It was good enough. Quick note on that, though, to go along with our conversation from last week, where we were talking about sort of understanding what is and isn't AI generated, we talked about the watermarks a little bit. I've become a lot better at detecting the language that has been coming out of this thing. And so looking at your blurb today, I knew exactly that it was the program, at least from the first sentence. I can tell there are certain sentence structures that it's following. There are certain things that it will do that you start to understand over time. So even in the span of a week, I've become even better at detecting this stuff. And I feel like we're as a society going to get better at detecting this as well. So it's going to be a lot harder to plagiarize and to basically use it as a crutch in some circumstances. Like, yes, you can use it as a starting point, but it's very clear to me that it wrote this when the first word is chat GPT. I can tell by the first word. The interesting bit here is take the human factors approach across all of it is what is the point here? Are we trying to get across a message to a third party in a way that is literate and meaningful? Or are we trying to get some text written by an AI? But clearly from a use case perspective, a user goal. I have a message that I want to get across to the audience. I know for a fact that me writing starting from scratch, I find it difficult. I think most people do. If you start with a blank sheet of paper saying, right, I need to write 200 words on Chat AI, you'll be like, Right, okay. By the time you've procrastinated a bit, you've gone and maybe, I don't know, tidy desk because you don't want to get in touch with it, you cleaned up something else and then you'll maybe get down to it. Then you'll stop for a break. And that's how my entire degree pretty much. But at least with this you could turn around and say, I know, I got to get mess across. And you're right, the structure is the way it is because it's following rules and probably more literally following rules than we normally do. When we put in structures together, we will be more dynamic with language. And I can be very dynamic with language because I'm terrible. But the idea behind this is in there, you get a start, you read it through, I changed a few bits like that and actually put a huge chunk in the middle and it saved a whole lot of time. But I've still got that endpoint, which give us a summary of the article for the show and it allowed us to talk about it. So it's achieved its goal. Spot on. I'm very happy whether I've done that or largely what we sometimes also do is maybe chop out the top 200 words of the article and put it straight in, maybe with a nice rounding off. I still haven't written it. We give the credit in the kiodos to the author of the article and we credit them. The end result is still the same. So I think what is quite neat for me here is if you could actually get used this almost as an API, and I think this will be where it will go, you can yeah. So then you have almost different instance of it done, where you can sort of, if you're wanting to, I don't know, summarize documents, then you just have a nice big open text box, cut and paste your stuff in. You're maybe got some cues as to if you want a short, medium, long, or you want an executive summary. This is the phrase to write that. Cut and paste the document in dishwasher. Done. Or if you want to have a virtual assistant, it does that because, like I said, with the connectivity to the Internet, this would be just next level. It stopped learning at the end of 2001.

 

 

I wonder why they did that, really, and why it doesn't keep on learning. But, yeah, I think it's going to be quite cool. Where did I start off with that? You have some sort of assistant to remind you. I know, but this is just so exciting. This is what I think is really cool about this. So what I don't like about the UI is in given what we're doing, it's a bit like you don't really know what you can do. The other analogy I would use is when you get a mobile phone now, you don't read the training manual, it doesn't exist anymore because no deals. You learn by exploring. What does this have to do? What settings have changed? But that's because you can go on an exploring nature, it's very visual or it can be tactic or audio. Audio. Whereas this, if you don't want to type in that box as initial prompts or whatever, if you can't as being pointed out, we can't interrogate it properly. You're not really going to learn because you don't know what you're missing. And so I have a real problem with that. Yes, you can learn ability. You learn by exploring, but you don't know what to explore unless you've got time just to sit there and just go, do you know what blue is a couple of things to this point. So first off, I agree with you the understanding of what this is capable of. I mean, we talked about some extreme cases last week about pushing this thing to be a dungeon master for Dungeons and Dragons, right? That is an extreme case that it can do. And you just prompt it the right way. If you log into this interface, it's fairly simplistic. It is a chat box with a couple of examples, capabilities and limitations, and I think those are great. However, there's more things that exist than just these examples, than just these capabilities, than just these limitations. And what I think would be a tremendous add to something like this is to simply rotate those with some other capabilities, limitations and examples. Say instead of explain quantum computing in simple terms, say, explain to me what human factors is in the form of a limerick. Like there's ways that you can press it or subtly encourage the users to push it in different ways. Heck, you could even make the system itself provide some examples and then have those feed in the waiting screen when it's at capacity. Actually write me a stand up joke about chat GPT. Right, those are examples too, but why are those not included here? Or why don't they at least rotate them to give the user some idea of ways in which you can push it? Because the more we push it as a community, then you and I can sit and talk to I did this, I made it made code for this thing and I think there are communities for it. I follow some of those communities, but I'm not up to date with everything. And some of the top posts all see, oh wow, they pushed it to do that thing. That is incredible. I'm going to try that in my whatever. It inspires me to push it in ways and see if. I can get it to work for certain things. I just would like to see more of that discoverability. Right. When you ask something, what can you do? The voice assistants will say, well, I can do this. That the other thing. If you ask them, what else can you do? They'll continue on. But the answers are mostly the same. They'll not tell you a randomized list of things. They might, but it's going to be finite. And this, it's like, what can you do? It's like, well, what are you trying to do? It can talk back to you and understand a little bit more about your context. It's like, well, I just don't know what I want to do yet. Well, here's some crazy out of the box ideas that you could maybe play with these resonating. Yeah. And I think that's currently almost part of its weakness is that it is so generic. But I think clearly the application of this, given that sort of context. So one of the bits that we've highlighted is a real flavor of the month, and it will be the next flavor of the next two years is human robot teaming and obviously putting that AI in the center of that. So this is a couple of projects I'm working with at the moment. This will be ideal for you to sit there and not only be able to say, I want to know about this, I want to know about that, I want you to be able to go and do this. I want you to be able to do that. But if you're not sure about whether the machine can do something for you to be able to sit there and say, how can you help me? What can you do for me today? Have that conversational piece as well. Will be hugely, hugely powerful and almost it will open up human robot timing in a way that it was meant to be as you watched it on Star Wars, Star Trek, them sort of things, the proper hello, computer type of thing. Right? Well, let me even back up a step because you brought up a couple of things there that I want to make sure we touch on. Right. The API thing first you brought up a little earlier in the show, and then you also brought up sort of this human AI robot teaming aspect. And I think largely to continue the discussion we were just having about its usefulness and being able to communicate what it can do to users. I think we'll eventually see some improvements here when the API is being used in specific applications. Right. So let's take, for example, a couple of different domains, right? Let's take the health care example. You ask it in a healthcare setting. It's in your My chart or whatever system that you're using for tracking your health. And you say, what can you do? Well, I can provide you a list of immunizations that you need or do not have recorded. I can provide you updates on when you should get checked out next. I can provide you some analysis on some of your past records, that type of thing, right? Those are some things I can do for you. What would you like to do so I can give you the context, right? So there's healthcare you can think about in a virtual environment. You could use these as like non player characters in role playing games, right? And they could come back and say, what can you do? And it would be in character, assuming that it could assume that character and say, well, I could send you to the Farm for a quest, or I can have you go kill these monsters for another quest. Or I could have you do that thing and have it respond in a way that's actually in character and makes sense so you can do those types of things. Can you imagine if you had sort of assistance built into, like, public transportation systems where you go up to it and say, I need to get to this location, and they can come back and say, okay, well, you can take a bus from here to here, take a train from here to here and then walk the rest of the five minutes from the station. Or you could take this train and then it's just going to take a little bit longer. It can explain to you the nuance of every option that you have available, including pricing, including accessibility options, all those things, right? Basically, if it has access to a system that you have, imagine these are being used for work someday. They have access to all the code that you have in software development and you can tell it, hey, we need to make this feature, code it up. And then they have basically your existing code and they have access to all the code that's out there in the world open source. And they can then make something that you trial and error and at least most of it is going to be there and you can really facilitate software development. So, I mean, there are ways in which I can see this being incredibly useful based on the context in which you're using it. I don't want to go too much into detail here because there's a million different domains that we can go into, but those are just a few examples of how I think this will be transformative in a lot of ways. When you think about this easily accessible UI, what can you do? It'll be largely based on context. Right now we're looking at Google, a Google type interface where you can search in Google and you know you're coming to Google for search, you're trying to find out some information and I think this is going to operate a very similar way you're coming to this just, it's an omni tool in so many words. And when it doesn't do the thing that you're expecting, it breaks, and that's okay. Then the limits of your understanding of the system have expanded because, you know, it can't do this thing, or it pushes you to push the machine to do a different configuration of that prompt so that way it can actually do the thing, because I've encountered that very scenario. It's not doing the thing, but I know it could do this. How do I adjust the prompt? Anyway, I just wanted to bring that up because I think it's an interesting point of discussion where a lot of that stuff will be, oh, I've activated my voice word by saying one of those things. Nice. Yeah. I think for me, sort of a final thought on this. It is brilliant. I think there is clearly now it's going to go places if you open the Pandora's Box. The interesting bit for me is now around how do we personalize these services? How does your personal data get stored in a way that is useful on the Internet, because it has to be internet based to allow that level of connectivity. How do we maximize the use of your personal data to allow you to have personalized services for your transport, for your whatever it is, for your healthcare, et cetera, et cetera, whilst retaining confidence in it that it's actually still secure as your data in this whatever it is. And that is all how we have faith in the UI. So I'm excited. I think you're excited, and I can't wait to see how it goes now. Yeah, I'll just mention one more thing that's kind of opening up Pandora's Box here. But there are certain populations that might benefit from, I don't know, some qualifiers in the UI, especially around accuracy of information. I'm thinking children who are looking to this thing as like, it can engage with them in a way that's age appropriate, but it might not bring back information that's 100% truthful in some cases. And so how do you provide sort of accuracy measurements of the things that are coming back to say, hey, look, it might be this, but it might not be this. And then also that goes with the aging population too. This is going to be or those with cognitive impairments as well. It's going to be one of those things where you're going to want to let people know that how this model works behind the scenes as they ingest that information, because not everyone is going to listen to this episode of Human Factors cast and understand all the intricacies. Why? No, I don't understand why they wouldn't do that. I don't know. Maybe we're training the model now. Anyway, thanks to our patrons this week for selecting a topic that we heavily influenced last week. And thank you to our friends over at UX Collective for our news story this week. If you want to follow along, we do post the links to the original article on our weekly roundups in our blog. You can also join us on our discord for more discussion of these stories and much more. We're going to take a quick break. We'll be back to see what's going on around the community right after this. Yes, that's right. Patreon rewards are always evolving. We got some fun stuff to talk about here in just a minute. But first, I want to thank all of our human Factors cast all access patrons like Michelle Tripp. You all keep the show running. Obviously, we want to give back to you. So with All Access, there's a new excitement, exciting announcement. Excitement. Is that exciting excitement? That's excitement. Hey, human factors. Cast Academy. We mentioned this at the top, but basically so this is for All Access. We're excited to kind of announce this new benefit. It's human factor's. Cast Academy is what we're calling it. It's a collection of all the learning resources that we put together, put out there ebooks, webinars courses, other resources organized by topic for your convenience. It's currently in development. We're working very hard on this, making sure to bring you a wide variety of high quality educational materials. We have access to a lot of things back here that we want to make sure that you have access as a Patreon supporter to the show as well, and basically making all this stuff available for everyone and for everyone at the All Access tier, I should say, to take advantage of it, log in to Patreon, sign up for that tier. We also have another exciting thing to come later that we'll talk about. I'm hinting at it, but our Human Factors Cast VIPs will be getting something very soon that we're excited about, and it will actually benefit everybody. That's a patron. So if you're interested in learning about what that is, we do have some information on our Patreon site now. So if you wanted to get a little bit of a teaser for what that is, you can go check it out there. Anyway, let's get into this next part of the show we like to call this Came From. It came from. That's right. This is the part of the show it came from. That's it. It's where we look all over the Internet to bring you topics the community is talking about. And for the second week in a row, we got another Discord one, which I'm so thankful for. I love the discord ones because that is truly from our listener community. And so this one here is from Karina, and they write about transferring to Human Factors or UX as an occupational therapist. So this one's an interesting one. There's a lot more context in the Discord that you can go look at and read. If you're interested in the full context, I encourage you all to do so because I've truncated this response to this question to be very short. But the long and short of it is I've really enjoyed being in the healthcare industry but it wasn't until I became an occupational therapist and working with patients from all walks of life that piqued my interest on how tasks, processes, et cetera, can continuously be improved. I'm not sure how I can get into the field in the first place. Are there certain jobs and positions I should look out for? That can be some kind of starting point. So we talked a little bit about starting points into UX generally, but I think this is an interesting situation where you're a little bit further into your career, you're trying to figure out what's the next step, how do you get into human factors or UX? Right, I did do a follow up question. It looks like they're trying to become a UX researcher in the space and so that's another one of my top mentee questions. But Barry, how would you get into the space to sort of look at these tasks, processes, et cetera? Well, it's interesting because actually I would say a good half if not more of the people. If I wind out to the whole human factors domain, I would say half if not more. Not primary human factors professionals because most people have got a primary degree in something or started to career in something, found human factors and went oh, it's amazing, how do I then do that? And so actually transitioning is I would say it's simple but it's difficult. I think it's simple because I think if you've got something that can apply because you've got loads of skills, you've got loads of things that you already do that we are looking for and it's an occupational therapist sitting down and looking at them sort of things and saying what directly transfers. So listen to a few episodes of this podcast, look at YouTube and just do a basic bit of research around what the big building blocks of human factors in UX. And therefore you will be able to go along with that, with what you do already and tick a few boxes there and then use that as a basis to go finding jobs, get your network up and running. You're already doing the right thing by being on our discord because you've got lots of like minded people. The human factors community is possibly one of the kindest, nicest networking communities, I think, of any discipline because I like working with nice people, that's why I've stayed. And so nobody is going to be shouty or off or anything like that for you asking advice and things like that, which you're already finding. And so start building your networks, builder them, contacts and just talk to people. Nick, what about you? How would you tackle this one? Yeah, this is an interesting one and I'm going to suggest maybe a non conventional approach here, which is to look for internships. I know that can be counterintuitive when you're so far in your career. And I think one sort of benefit to looking at internships is the ability to be non committal about something. And I know that sounds bad, but in a way, it's basically doing a trial run for whether or not you enjoy the work. You can then sort of understand what skills you have that are going to transfer well. You understand what skills you have that are not going to transfer well, or gaps in your knowledge even by working in a simple internship. It also gets you work experience, which gets you in the door. And I wouldn't say you're limited by starting at an internship and then going and applying for, like, a senior position, because then you'll know what? Skills transfer. And you can at least use that as leverage to say, look, I've done a job before, and it's been an internship, sure, but it's really taught me which my skills are transferable. And hey, as an OT, actually, turns out a lot of them are talking to people, research. I think there's a lot of really transferable skills there. So that might be one way in, because oftentimes the thresholds for internships are fairly low. I guess the stakes around those internships are lower because a company is more likely to take a chance on you. It's often for a finite amount of time, either a summer, a winter, something like that, a short term, and again, it just gets you exposed to the environment. Now, if this is like something you're absolutely sure of, your gung ho, is there anyone you know that can take a chance on you and put you into a position that matches your skill set? I think that might be another way in. It's all about who you know, right? And if you know somebody through your experiences as an OT or through your experiences, then that might be a second good way to get in. Anything else, Barry? For that one, the only other thing that I was talking to a sort of mentee about this last week is the way that they were heavily involved in the medical profession. And what they started to do was start to apply human factor techniques that they had in their books and things like that, and started applying them to their day job, not necessarily as a real output of their job, but they meant that they did a task analysis. They did a critical design review. They did all the different bits, and they just applied it to what they were doing, and they got somebody to have a look at them to say, I've done this on this task. What do you think? Have I done it right? So with a bit of a mentoring type approach, just having a go at doing some of the stuff in your own time helps. Yeah. Find somebody else who's followed a similar path and ask them questions about it. If they're nice and they're in the human factors field. Let's be clear, there's a couple of real assholes in the field, but they're far and few between. Barry is probably correct. I'm one of them that knows the other. I would say he's probably correct about there being like 99.9% nice people in the field. Like you talk to any of them, approach to any of them can be intimidating in some cases if you're a little earlier on in your career. But I think in most cases, people are more than happy to sit down with you and talk to you. So find someone who's had a similar experience and talk to them about it. All right, let's get into this next one here. This one is by Leon At from the UX research subreddit there's how to learn scientific research methods if I transition from a designer position. So they were wondering if there are courses that teach market research, ethnographic research, quantitative qualitative research, those types of things. I want to get a comprehensive knowledge of tools and techniques. So that's the question, Barry. Where can they go and learn more about research methods? So I'll be then going straight to I don't know whether to do them in the US, but in the UK we have research methods courses. If you're going to go and do a master's or a PhD or something like that, it's effectively the first course you will do it's the first module you'll do is research methods. Then we sort of don't necessarily like them as much, but there are boot camps out there. You could go and waste your money on a boot camp. Some of them, I'm sure one or two of them must be good. And actually, if you just wanted a weekend of it being thrown at you as a starting point to then go and have a taster for other stuff, I think that's probably where the value is. Books. There's some really good books out there. There's two out there that I use like a Bible. Or there's one specific I use as a Bible for it's my first place I go to for all of my research methods. And then finally there are online training courses. We've already had mentioned in the show tonight about a potential new training course resource Nick Works. There a bit more about that one. But YouTube, if you go and do some of that, there is loads of stuff online already that you can do the dry reading on. The problem with doing some of these stuff on books and online is it doesn't bring it to life. And sometimes the methods that we use sometimes can be a little bit dry, a lot of pen and paper stuff. Unless somebody can bring it to life for you, you can sometimes slightly lose the will to live. But once you see the power of what it can do by somebody who knows what they're doing with it, or you can just chat to somebody about yourself. I've just heard about this methodology. What do you do with it then it can actually really change perception. Nick, I'm going to guess where you're going to go, but how would you teach something like some scientific research methods? Can you guess why I picked this one? No. I'm getting some really good ideas for Human Factors cast Academy here. Are we pushing that? I don't know. So, look, here's the thing, though. If you're looking at the entire Internet, there are going to be, like Barry said, resources out there available for you to look at and to understand what each of them do. And he's right, that's about it. Your knowledge will be of the research methods and not necessarily how they apply to your work. And so my recommendation here is to basically learn about when to use each one, when it's appropriate to use each one, depending on your sort of needs or requirements of user research or human Factors research, and understand the basics. And then if you've decided that it is well within your capacity to try something new, to try to get at an outcome that is deemed necessary, then try it. And from there, basically document if it's worked, then great. And use those best practices going forward. Keep up that momentum. If it didn't work or it didn't work the way that you thought it would, document why it didn't return those results that you were expecting, and document maybe why it failed and take those lessons learned forward with you for the next thing. So I think understanding what you've tried and why it did work or why it didn't work is a good post mortem to conduct when you're looking at these methodologies. So I wouldn't necessarily think that you need this comprehensive knowledge, just a starting point to understand what's out there. And then if you have the capacity to try something new in your research, do it and push yourself and try to get it to work. And if it doesn't, then it's your job on the line, not mine. All right, so we got one more here. I know we got one more here tonight. This is by Wiz Califa. Nice. And from the user experience sub. Right? When do you give potential employers the password for your case study? Can I just answer now? So they write, I created a case study for the project that I've been working on, but the product hasn't launched yet. When should I give potential employers the password for this case study? Or should I just write it on my resume and only show it during interview presentations? Barry, what are your thoughts on giving passwords to prospective employers about projects that you've worked on? If it's a project that hasn't launched yet, I'm guessing it's not your project, it's your company's project, therefore it's not yours. I hate to say you're going to be signed by your own contract. You're going to be releasing information out there unless you have an NDA with the person who's doing your interview, which I'm still not entirely convinced in my head would actually work anyway, if the product hasn't launched, I don't think you should be sharing it at all anyway. However,

 

 

I wouldn't do it until the interview itself. If it was launched and I could do it, I would possibly put it into maybe some graphics or something into my application and then walk them through it at the interview. I wouldn't necessarily send it in personally, but it depends, doesn't it? It depends on the type of interview you're going for and what it is personally. The way that this particular one has been described, I would be virgin on Never because I don't think it's your intellectual property. Certainly if it's not launched yet, if you got permission from your current employer, which I think will be if you're going for a job interview, if you got permission from your current employee to do it, that's a different ballgame. I'm going on the assumption that your current employee doesn't know you're going for a job interview. I think I'm going on Never personally or this one specifically. Yeah, good points, good points. I am thinking that in this particular case, you're right in general. In general, if you had something locked behind a password, why do you have it locked? There's probably a good reason that you have it locked. Is it because it's under NDA? Is it because for one reason or another? Anyway, think about those reasons. But if you decide that you're going to share it with them anyway at some point, send it along with your resume. Don't hold them up just from somebody who's looking at resumes. They might already be looking at somebody else. If they can't get access to it. They might just kind of I wouldn't say they write you off, but they'd kind of put you in the awaiting pile where somebody else might have sent it in. And I'm already looking at their portfolio and Barry brought up a really good point about NDAs. Don't break those, that's just a bunch of trouble. And it's especially tricky if you are applying for like a competitor, because then you've just given them a leg up on what's coming out of their competitors face and that is anyway, be professional about it, don't shoot yourself in the foot and have fun. Not my job. All right. Not theirs either. Not theirs either, apparently. All right, let's just get into one more thing. Barry, what's your one more thing this week? It's Christmas. We've been doing so tonight we've been out with the family. There's been a big amazing illuminated walk trail that we've walked around as a family. Eldest daughter's back from university and it's just so nice to have this time together. Tomorrow we go in for a mini expedition up north to go and see my parents. And so a couple of kids have come up with me to do that. But there's just so much around, and I'm telling everybody else and not doing it very well myself that make sure that you take the time. To put down the computer, put down everything else, put work away, which, if you're self employed, I'm fully done with, that being very difficult. But the days between Christmas and New Year, as much as you can put everything away and just enjoy yourself because life is too short. So, yeah, Christmas, I'm really looking forward to it. You know what, my one More thing was something completely different from what I'm about to say, but you have just inspired me with the theme of family. And I am going to spend Christmas with my family this year. In fact, day after Christmas, my wife, my son and I are going to embark on the magical journey of potty training. And so we've got that whole week off, no plans. I'm putting my phone down the entire day after Christmas and we're just going to go and we're going to do the thing because they say, make a whole week for yourself where you have no plans. So we're going to try it and there's going to be a lot of bodily fluids and stuff all over me over the next week. And so if you're feeling sympathetic towards that, then maybe write me in our discord and say, I feel for you. I'm sorry, this is what worked for me. Give me some pointers. Could use them. Can you keep sort of a six hourly diary, audio diary having three now quite older children. Are they potty trained? Just about. Eldest, has just turned 18, so just now she knows drinking eggs, so different for usages, but, yeah, good luck with that. I don't know any of that at all. Thank you. I'm very much looking forward to it and reading the books, getting prepared. Well, I will see what I can do. About a six hourly diary. Although daily, because they say, put your phone away, put your phone away, do not because then you're going to miss it. You're going to miss it when they go to relieve themselves and I don't want to miss that because we got to get them to the place. Anyway, that's it for this week, everyone, if you like. That's why I did toilet paper in the preshow. I see. Anyway, that's it for today, everyone. If you liked this episode, enjoy some of the discussion about if you enjoy some of the discussion about AI, go listen to our last episode. If you enjoyed some of the discussion about chatbots, go listen to episode 253, which is user perspective on categorizing using chat bots and voice assistance. That was a good one. Comment wherever you're listening with what you think of the story this week, for more in depth discussion, you can always join us on our Discord community, visit our official website, sign. Up for our newsletter. Stay up to date with all the latest heat of Factors news. If you like what you hear, you want to support the show, there's a couple of things you can do. One, wherever you're at right now, you can stop. You can leave us a five star review that is free for you to do and really helps other people find the show if they're just poking through those apps. Two, tell your friends about us. That word of mouth really helps us grow, especially a lot for those of you in graduate programs who have other friends who want to introduce us to them, please do, because that really helps. And if you have the financial means to, we are always trying to give back to our Patreon supporters because you literally keep the show afloat and we want to make sure that you are taken care of too. So that is why we're doing things like Human Factors Academy and this other thing that I'm really excited about, that we can't talk about yet, but we'll talk about at the beginning of next year. And so if you have financial means to do so, do that. We'd love to be a continuing resource for you all. As always, links to all of our socials on our website are in the description of this episode. Mr. Barry Kirby, thank you for being on the show today. Where can our listeners find you if they want to talk about potty training 18 year olds? If you want to go and talk about that, then you're going need to go see my daughter's Twitter. But anyway, if you come text me, then find my Twitter. And any my social is actually okay. If you want to come to the interviews with other interesting Human Factors professionals, then come and find me on Twelve or Two, the Human Factors podcast at Twelve or Two Podcast.com. As for me, I've been your host, Nick Rome. You can find me on Discord and across social media at nick underscore Rome. Thanks again for tuning in a Human Factor's cast. Until next time. Until next. Well awake, because then we're going to be anyway. It depends. It depends. Bye.

Barry KirbyProfile Photo

Barry Kirby

Managing Director

A human factors practitioner, based in Wales, UK. MD of K Sharp, Fellow of the CIEHF and a bit of a gadget geek.