This episode of the Boagworld show is sponsored by Express VPN, Docsketch and Easy Data Transform.
Related Links
- Inspired: How to Create Tech Products Customers Love
- Creativity code
- Can Machine Learning Write the New Yorker
Transcript
Paul: This week on the Boagworld Show, we look at the UX challenges surrounding security, artificial intelligence, and data visualization. This episode of the Boagworld Show is sponsored by ExpressVPN, Docsketch, and Easy Data Transform.
Speaker 2: Boagworld.com.
Paul: Hello and welcome to the Boagworld Show, the show about all aspects of user experience design, digital strategy and working with digital. This season, we are doing virtual meetups where we come together and it’s pretty much an unknown as to what we’re going to cover in this show because it will depend on people in the chatroom willing to jump on the mic and say hello. We’ve got some questions and other things as a backup, but most importantly, we have Marcus Lillington joining us. Hello, Marcus.
Marcus: Cheers, Paul.
Paul: Yes-
Marcus: Evening drinking beer, lovely.
Paul: To recording a show in the evening when you can afford to drink alcohol.
Marcus: Exactly.
Paul: We need to do that more often.
Marcus: Yes, quite.
Paul: So what’s been going on-
Marcus: Hi [crosstalk 00:01:21]-
Paul: Oh, [inaudible 00:01:21]. It’s all right. Perhaps, we could persuade her to come on the mic because she’s a regular-
Marcus: You never know.
Paul: In the chatroom. So we’ll see what happens. Seriously guys, we need some of you to step up and get on the mic. You don’t need to have anything to say. Just to say hello and introduce yourself. Excellent, Andrew will be up for it so we’ll get him on in a minute, but first of all, Marcus, what’s been going on in your exciting life this week?
Marcus: Well I said… what I put in the show notes was I’ve been working on a landing page and I thought, well, that’s really dull, which is true. I have been doing that, but after our little chat this morning, Paul’s been helping us out, dear listener, with a bit of a marketing push at Headscape for basically to do with the work we do in law, and Paul suggested that we write a blog post and support the survey that we’re going to be putting out, which I’ve been working today. It’s a sort of pet subject of mine that I’ve been finally start put into some words. Can’t talk now. I shouldn’t drink beer really and try and talk on a podcast, but anyway.
Marcus: So that, and I thought well, what things have come up over the many law projects that we’ve worked on over the years and one of the things that often happens within a law based project is… Well to give you a bit of background. The main reason why a law firm has a website is to display the people within the firm and to show their bios and to show their expertise and that kind of thing. So basically, long lists of people and one of my sort of little bug bears over the years and this is quite similar to my bug bear about icons, I’ve always been a fan of well why don’t you just use words. Why do you have to use an icon? You’ve got these sets of little sort of shapes called letters that if you put them together they-
Paul: That have meaning associated with them.
Marcus: Yeah, exactly. People can read them and they know exactly what it is. So similarly, I’ve always been a fan of listing things alphabetically. I think that people like things listed alphabetically. Well, except for those people who have a Latin-based alphabet mainly.
Paul: Yeah. It depends on what you’re listing, Marcus.
Marcus: Yeah, I’ve been trying to think today of examples where that’s not the case and you-
Paul: Countries.
Marcus: Yes.
Paul: If you primarily sell to the United Kingdom or to America, then having… United Kingdom’s especially bad because you don’t know whether you’re going to be United Kingdom, Great Britain or England and you spend-
Marcus: However-
Paul: Your whole life scrolling through a frigging long list.
Marcus: But if you wanted to find a particular country in a list of all the countries, Norway, for example, if they weren’t listed alphabetically it would be a nightmare trying to find Norway.
Paul: Yes, okay.
Marcus: For example, okay.
Paul: Yes. Ooh ooh, I like Andrew. Sorry, I’m just now going to disagree with [crosstalk 00:04:22]
Marcus: That’s fine, whatever.
Paul: With everything you said. Andrew said you should list by last used. I like that.
Marcus: All right, yes. It depends, but I’m going towards the fact that I’m talking about lists of people here, all right?
Paul: Okay.
Marcus: The problem is-
Paul: What about by department?
Marcus: Ugh. [crosstalk 00:04:45]
Paul: That’s perfectly valid.
Marcus: You can certainly filter by things like department or-
Paul: Or position. What if you don’t know the name of the person you’re looking for?
Marcus: Then you can filter by other criteria, Paul.
Paul: By salary. Andrew says you should filter by salary.
Marcus: I’m not talking anymore. I’m just going to drink beer.
Paul: No. No, sorry, Marcus. Sorry. Carry on. No, I [crosstalk 00:05:11]-
Marcus: There is a point to this.
Paul: Okay, yes.
Marcus: The point is, and it relates to the law firm thing and there are probably other lists of people where the same thing applies, but let’s say you’ve got 1000 people working in a firm, right?
Paul: Yes.
Marcus: So that’s a lot, big long list of people. If you click on their people link in the navigation of that particular firm’s website, you’ve got to ask yourself well what would you expect to see when you first click on that link, and I would argue that the first thing you would expect to see would be the startings of a list of people. You can then filter by things like department and salary and position, all those kind of things. So how would you list those people? I would argue alphabetically. So you start off with Jane Abbot, Michael Adams, whatever, ba ba ba ba ba ba. The problem is, with law firms is it might be that all of those first half a dozen people out of 1000 are all super junior associates and they’re not the big stars of the firm.
Marcus: So we then get told… Well, we can’t have it like that. We can’t be showing these nobodies as a first impression of a list of people. So the post I’m writing basically shows the different things that we’ve done and it’s different every time because it’s an issue. Well, what do we do about this, and sometimes you just don’t show anything. You can sort of say well, okay, that’s fair enough, but the user then has to take an action to see anything. So you’re not helping them in. If you show a list of people, then you’ve given them a prompt just to what they’re going to get when they start filtering, but if it’s nothing, then there’s no prompt. So it’s not helping them from a [crosstalk 00:06:59]-
Paul: Nothing is not good.
Marcus: Exactly.
Paul: I agree with that.
Marcus: Another one is oh we can show the stars then. Another example we’ve done is it shows all the search tools, your A to Z and your search and all that kind of thing. Then underneath it’s got sort of the leadership team, but that gives you the prompt of what it’s going to like when you filter on list, but I would argue that it’s also slightly confusing because the list isn’t alphabetical. It’s just showing a few people. Then I’ve looked at another example, the most recent one we’ve done where they do start with a list of just all the As and then the Bs and Cs and so on. I think that it doesn’t matter. You can show that. No one is going to arrive at that page thinking “Wow, this is a really low end firm because they’ve got just their associates showing.” The point is if you’ve designed it well, people don’t think anything.
Marcus: Then going back to don’t make me think, the Steve Krug thing, if you’re showing them an alphabetical list because alphabetizing stuff has been drummed into us since we were small people, then you’re giving people a natural search and they don’t have to think about it all and they will just instantly see a list of people and they’ll think right, that’s not what I want. I’m going to click on D because the person I’m looking for their name starts with D or I want to see all the partners so they’ll filter on that. Point is I don’t think you need to worry about it is the thrust of the article.
Paul: I’m not convinced I agree with you. I normally agree with you over stuff Marcus, but on this particular one I’m not sure I do. I mean, I’ve got no more evidence to support my approach because you’ve worked at a lot more law firms than me. I think maybe you’re making an assumption.
Marcus: I am.
Paul: Maybe you’re making the assumption that as people go in, they’re going in because they want to find someone specific, and in which case, I agree with you. If that is the use case. If they want to find Roberta or whatever, then going in and having an alphabetical list makes perfect sense, but I suspect there are a significant number of people that are just going in to see the kind of people and the caliber of people that they have in an organization, in which case the assumption that showing the junior clerk is number one is a problem on that use case. You do want to show the big flyers. So I actually think you’ve got almost two different use cases there which need two different solutions.
Paul: So when you first go through to a our team or whatever, I think it’s perfectly valid to show the top people. Then you have the option, if you wish to, to then see a list where you can find a particular person or search on a particular person or filter by department or whatever else. It’s an interesting one.
Marcus: I think you do need to do that, but I think the homepage is the place to show the stars and not on the kind of people search of the site.
Paul: For a law firm, you’re probably right because it is so much about their superstars isn’t it?
Marcus: Mm-hmm (affirmative). Yeah. Well it depends. Even that depends. Some firms we’ve worked with are they’re more about volume and we can deal with your problems, there’s so many of us, rather than we’ve got a handful of stars. It depends.
Paul: Yeah. It does depend. But it’s more-
Marcus: It’s like anything, yeah.
Paul: It’s just this idea of-
Marcus: Don’t hide it I think is the one that really bothers me. It’s like oh well, we can’t think of an answer so we’ll just show nothing.
Paul: Yeah, no that’s just insanity.
Marcus: You’ll see lots and lots of law firms do that because of this issue.
Paul: Because it’s politically easier to do that than it is to make a decision. Yeah.
Marcus: Much. Much the case, yes.
Paul: It’s ridiculous isn’t it, really? Oh, that’s an interesting-
Marcus: So anyway. That’s what I’ve been working on.
Paul: Well I’ve got quite an interesting one as well actually. Something I’ve never done before. I’ve got someone I’m mentoring at the moment who works for a logistics company. So let’s say for example you order something from ASOS or, I don’t know, Next or a retailer. You order online and it comes through to you. What I didn’t realize is very few of these companies actually dispatched the goods themselves. They have a logistics company that does that kind of stuff for them. So they pack all the items and they send them out and they deal also with the returns that come through.
Paul: So I’ve got someone I’m working with at the moment who is trying to improve the productivity of items being returned, right?
Marcus: Okay. That’s a big deal I would have thought.
Paul: Yeah, it makes a massive difference. If you can shave off even a second from each return, it makes millions in a year and it is such a fascinating project to work on because what makes it a really good one is you’ve got this combination of obviously what goes on on the screen and the desk itself that they’re working at and, for example, they’ll get some returns that’ll be too big to fit on the desk properly and they’re kind of leaning around items to scan the barcodes that they’ve got stuck to the front of the monitor and all kinds of stuff. We were getting really excited about the idea of using touch screens, and then we started to think yeah, but is that going to be awkward? Is that going to cause repetitive strain injury if all day every day you’re standing up and leaning over and touching a screen that is a vertical screen?
Paul: So it’s been absolutely fascinating because I’ve never done usability stuff that’s got an ergonomic component, a physical component to it as well. So it’s a really interesting project. I’m absolutely loving it and I’ve got all this GoPro footage where they stuck a GoPro above people’s desks in order to record what they’re doing, but also they break these videos down frame by frame, okay. So they’ll record, say, an hour of someone doing the job and they’ll go through and say well how long is it taking them to print the label or to stick the label on or to move the package from the table across onto the conveyor belt or to, you know… Oh, it’s this whole nother world that is utterly new to me and it’s so exciting to be involved in. I’m loving every minute of it.
Paul: So yeah, it’s been a good time. Good project that one-
Marcus: Excellent.
Paul: I know, right? Something different. After doing this for 23 years, it’s unusual isn’t it to get something that’s totally different come out of the woodwork so there you go.
Marcus: Yeah, sort of health and safety related aspects of things like that I would be scared to go anywhere near, if I’m honest.
Paul: Yeah. Well I mean, to be honest, yeah, exactly. I don’t have to deal with those elements of it, you know? All I’m looking at is I mean most of it is still the user interface elements because interesting, it’s like all of these internal tools isn’t it whether you’re talking about an intranet, whether you’re talking about a system like that, whatever it is. No designer has ever looked at it, you know?
Marcus: No.
Paul: It’s all been done with developers just kind of putting together what works and no disrespect to developers, it’s not their job is it?
Marcus: Yeah.
Paul: And of course, then getting the IT department to change this bit of software is a major nightmare. So at the moment, we’re looking at potentially installing an extension for Chrome which the application runs in on every single device and essentially overriding the CSS and java script of the application rather than getting the IT department to redesign their app which is going to override it. It’s a real bodge it solution but it kind of works, yeah. So I love projects like that. It’s made my day.
Speaker 2: Boagworld.com, Boagworld.com.
Paul: Anyway.
Marcus: Excellent.
Paul: Talking of really weird scenarios, I do want to do our first sponsor. There is a link there which is it’s Balsamiq and their weird “Oh yeah, we’re going to sponsor the show but we don’t want to be the sponsor. We want to get other people on the show and we want to promote what they’re doing.” Fair enough. Cool, sounds interesting and cool to me.
Marcus: Interesting. It is interesting [crosstalk 00:15:54]-
Paul: And actually it’s quite interesting because it’s floated a few apps that I was totally unaware of, didn’t know existed and they’re actually really quite good ones. So for example, the first one that Balsamiq have asked us to promote for them is an app called Easy Data Transform, right. So we have this huge amount… as digital professionals we have huge amounts of data, don’t we, spread across Excel spreadsheets and in all these different apps and just it’s all over the place and getting it from one system to another and one format to another is just this endless nightmare of pain and misery isn’t it? This is where you need people like Chris Scott who have got an unnatural affinity with data; data wranglers, data whisperers. It’s like horse whisperer.
Marcus: He is a data whisperer. That’s going to stick.
Paul: Yeah. Yeah, exactly. But fortunately, we don’t need him anymore which is wonderful. He can now retire. You can let him know that.
Marcus: Okay. I won’t, but yeah.
Paul: Yeah. He’s been replaced by Easy Data Transform. So basically you can take your Excel file or your CSV file and there’s like a Wiziwig environment. So the kind of thing that I understand where you can go through and reformat the data and repurpose it without any programming whatsoever. So you can build up really quite complex data transformation step by step using kind of, I think they’ve got something like 36 simple built in transformations you can do, and it runs locally on your Mac or PC. So the data never leaves your computer, which is obviously, these days, a really important thing, and one time fee. No subscription model. I just want to kiss them on that basis alone. Everything has to be a subscription model.
Paul: So if that sounds like it might be interesting to you, check out Easydatatransform.com, Easydatatransform.com. Thank you Balsamiq recommending those guys and pointing us in the direction of them. Sounds very awesome and I want them involved. It’s good that they’re involved, should I say. I don’t want them because they’re already here. Anyway, we have had our first volunteer to come and join us on the show today, which is the very brave Andrew Davis. I’m just sending him an invite at the moment. So Andrew, keep an eye out. There should be a popup invite that you can select and get you on. We’re going to want someone else on the show at some point as well. So if you’ve arrived a little bit later and you’re up for coming on and having a little chat with us, just say so in the chat and we’ll get you on.
Paul: Hey, hey! It’s Mr. Andrew Davis. How are you, sir?
Andrew: Evening gents.
Paul: Hey.
Marcus: Hello, Andrew. Hey. It’s like a-
Paul: [crosstalk 00:18:52] He’s just waved a bottle of beer at the screen for those of you that are listening to the podcast. So Andrew, nice to speak to you. I feel like have we met?
Andrew: We haven’t met, Paul.
Paul: Oh.
Andrew: I’ve been following the podcast for years, but I haven’t met you.
Marcus: You have now.
Paul: See now it doesn’t matter what I say, if I say-
Marcus: You get it wrong.
Paul: To people, “We’ve not met have we,” they go, “Oh yeah, several times.” Then you who for some reason I’m looking at and going I’ve met this guy, I haven’t. So, shows how useless I am.
Andrew: No, I think we spoke on LinkedIn a few times, but that’s it.
Paul: Yeah. Perhaps I just recognize your avatar. It might be just that.
Andrew: Possibly.
Paul: So Andrew, I’m going to guess what you do, right? I’m going to guess purely on the basis of your appearance, although you have been in the Slack channel but I can’t remember what it is that you do. So let’s see if I’m right. I’m going to go you’re a service side developer.
Andrew: You are wrong.
Paul: Oh. [crosstalk 00:19:54]
Andrew: Aye.
Marcus: That’s two out of 10 shall we say?
Paul: I am doing very badly. Right, all right then, Marcus, you’ve got to guess now.
Marcus: UX designer.
Andrew: Close.
Paul: UX… Okay, go on then Andrew. What do you do?
Andrew: Product designer.
Paul: Ooh.
Andrew: It’s one of those weird scenarios where you’ve done user interface design, data visualization and I’ve done some development as well. So [inaudible 00:20:25] as product designer. Catchall.
Paul: These days, don’t they call you a unicorn or something because you can do lots of different things [crosstalk 00:20:35] I think that’s the new trendy phrase.
Andrew: I heard omnicorn today.
Paul: Omnicorn? That’s awesome. That’s it. Marcus, from now on you have to refer to me as your omnicorn.
Andrew: No, I’m definitely not a unicorn.
Paul: Okay. [crosstalk 00:20:50]
Andrew: I’ve just had lots of experiences.
Paul: It comes from getting old, I think.
Andrew: That’s correct.
Paul: The older you get. Yeah, exactly.
Marcus: So where do you work then where terms like omnicorn are thrown around.
Andrew: Omnicorn was not thrown around here.
Paul: Ah, okay.
Marcus: Okay.
Andrew: I work for a data analytics company who are owned by McKinsey.
Marcus: Okay.
Paul: Ooh. So grownup proper stuff then.
Andrew: Yeah, proper grownup.
Paul: Please tell me you don’t have to wear a suit or anything like that.
Andrew: I’m at work.
Paul: Aye. He’s wearing a red hoodie, and I’ve got to say he looks like he’s been dragged through a hedge backwards.
Andrew: Thank you.
Paul: So [crosstalk 00:21:32]. Yeah, I know. I always do. So as a product owner, what do you do-
Andrew: Designer. Designer.
Paul: Designer, sorry. Product designer. What does your normal day consist of?
Andrew: It varies and it’s a great variation. Some of it’s interviewing users, which we’re lucky because we have users. Some of it’s conducting research, but most of it is my kind of forte which is UI design. So taking all of the insights that we’ve got from the UX designers and the users and turning that into some kind of UI.
Paul: Okay. How long have you been doing that kind of role for?
Andrew: Here I’ve been doing it for about four years.
Paul: Right, but you’ve been elsewhere beforehand. Have you always worked on the client side or have you worked on the agency side as well.
Andrew: I’ve done client agency, consultancy, contract and freelancing. Sorry, and [inaudible 00:22:33].
Paul: Okay.
Marcus: Full Monty.
Andrew: Yeah.
Paul: So as somebody that has done the full Monty, which do you enjoy the most?
Andrew: That depends on what the actual day-to-day job is. I did eCommerce for a long time and I got totally sick of designing things for people that didn’t need them. I wanted to do something more useful in society. So what I’m doing right now kind of ticks all those boxes.
Paul: Oh, okay. So in what way? So you’re doing data analysis type stuff. Can you tell us more about it or is it all hush hush.
Andrew: A lot of it is hush hush.
Paul: Okay.
Andrew: I can tell you the product that I’m working on takes data hypothesis and features and kind of manages all of them. So what you were referring to earlier about data wrangling, all of that happens within a product that we designed and the output of that is to hopefully successfully match the metric to whatever the client needs using those hypothesis and functions.
Paul: Ooh. That sounds far too complicated for the likes of me.
Andrew: It’s too complicated for me.
Paul: Right. It doesn’t sound like an easy interface design job because I always think that data visualization is quite a tricky area to get right, isn’t it?
Andrew: It is and it takes a lot of testing. One of the things that we’ve done is to make sure that what we’re doing is not only useful for the people here that use it, but also for the clients. So you’ve got a trade-off and you have to get both right which is really hard.
Paul: So I’m guessing that people in-house that use it are your Power-users that want all of this extra functionality while your clients maybe need things simpler. Is that a fair assessment?
Andrew: Correct. So one of the things that we’ve done is a data manager. So you’ve almost got like a Trello board, but obviously from a client’s point of view, the beauty of that is lost especially when you have something that goes off the screen with about 10 to 15 columns. so you have to design a data vis which they get in one screen which is also downloadable from PowerPoint because McKinsey love a PowerPoint.
Paul: Okay, I get it. So essentially, have you looked at areas like… Oh, there’s a fancy word isn’t there for it that’s gone out of my head. This idea of progressively showing more and more features over time the more someone returns to the application. Is the application sophisticated enough to evolve based on how much you use it or is it more of a static thing?
Andrew: It’s fully interactive. We are looking into ways of doing that. There’s certain amount of things that are timeline based. So as a client and you’re spending millions of pounds, you obviously want to see the benefit and the return on that money. Over time you can see how the data’s changed, whether or not things have gone through ingestion or dropped out. The same with hypotheses because you may have a load of hypotheses to start with. The client may have a load of hypotheses and you’ve got to marry those up. So yes, we’re doing things like that but it is hard and it’s a case by case basis.
Paul: So if you get a new client, someone using the application for the first time, do they have training upfront because it sounds like quite a heavy duty enterprise thing or is it have you got some kind of onboarding process or is there a mixture of both?
Andrew: We have both and fortunately a magnificent product owner.
Marcus: Cool.
Andrew: Yeah, the product owner goes through with the clients. Obviously the users of this stuff day-to-day are Power-users. They know it all, but the clients have no idea whatsoever. So it really takes someone to go through and instill all that information into them.
Paul: Yeah, which it’s quite tricky isn’t it because you always get that… Training is something that I think about a lot, right? You do a training course for people and then they move on to another company or another job or whatever else. So you’ve got this constant cycle of retraining. So increasingly I’m exploring things like onboarding processes, self-learning training, documentation and how to make documentation better rather than relying all of the time on just one-off training courses. Ah, oh, it sounds like… Well, what’s the big new thing? What’s the thing that you’re looking at next that you want to start working on?
Andrew: I’m working on something at the moment which is clients tend to have projects and in the projects they have use cases. We need a way to manage the amount of projects and the amount of use cases. So we’re building a project manager which will also be able to inherit some of the earlier learnings that we’ve done through the product. So if you’ve done a, oh let’s think of an example, an automotive project, then maybe some hypothesis that we know are kind of 80% across all automotive projects where you can almost say, “Do you want to use this stuff again?” So we’re trying to build in some artificial intelligence and machine learning, which is tricky.
Paul: Ooh. Yeah. Yeah, always good to throw in simple stuff like machine learning. I mean go for those quick wins [crosstalk 00:28:44].
Andrew: But it’s also who owns the data. So some of the hypothesis and some of the features that you may have built, sort of models that you’ve run will be specific to a certain customer and obviously we can’t reuse that data.
Paul: Ah. Ooh yeah.
Marcus: So you must have some very, very clever people behind you, Andrew, developing algorithms and that kind of thing. Someone who actually has [crosstalk 00:29:07]-
Andrew: Everyone here has a PhD pretty much.
Marcus: Right.
Andrew: I don’t.
Paul: Ooh. So one question-
Marcus: Yeah, but you’ve got to make it so that people can use it. That’s the main thing.
Andrew: Correct.
Marcus: Sorry, Paul, go on.
Paul: Sorry, one question that’s just been asked in chatroom which is for people that are interested in product design and product management and that kind of thing, are there any books that you’ve seen or read personally that have made a big difference to you?
Andrew: Ooh, I’ve got it on my desk and I can’t remember what it’s called. Yes, I’ll answer that in a minute when I can remember it.
Paul: Okay. Well why don’t you, you go and check that out and then pop it in the chatroom and I’ll mention it in a minute, but for now, thank you very much for joining us Andrew. This is great to chat.
Marcus: Yeah, much appreciated. Absolutely [crosstalk 00:29:56]-
Paul: Yeah, it’s a good one. All right then, bye-bye.
Marcus: Brilliant.
Andrew: Thank you.
Paul: So there you go. That’s Andrew, product designer. Data visualization stuff really just blows my mind. It’s not a easy area to work in because often you’ve got so much data that you’re trying to communicate and it can just be somewhat overwhelming doing that. So yeah, excellent stuff.
Marcus: I’m reading at the moment about basically a AI and the designing of algorithms so that they learn from the bottom up instead of top down and all this kind of stuff.
Paul: Ooh.
Marcus: It’s very good actually.
Paul: That’s very unlike you to read a book like that, Marcus, isn’t it?
Marcus: Yeah, it’s normally just fairy stories for me.
Paul: No, I wasn’t being rude.
Marcus: Joke books, you know.
Paul: Yeah. No, no, no. I honestly wasn’t being rude. I’m just saying you don’t read a lot of work related books normally.
Marcus: I suppose not necessarily. This isn’t really work related.
Paul: Oh, okay.
Marcus: It’s sort of sciencey, I guess. I’m interested in the history of where machine learning came from and where it’s likely to go.
Paul: Ah, okay.
Marcus: It’s by a guy called Marcus du Sautoy, and he’s kind of like the science laureate and he wrote a great book all basically explaining all the major aspects of science like-
Paul: Ah.
Marcus: Going back to Newton and bringing all the discoveries of scientists over the last 500 years and explaining it in ways that even I can understand.
Paul: Ooh.
Marcus: This book is similar but related to AI and machine learning. It’s brilliant.
Paul: So the book that Andrew was talking about was a book called Inspired: How to Create Tech Projects Customers Love by Marty… How would you say that, Marcus?
Marcus: Cagen.
Paul: Cagen. I’m having a moment, glass of wine and all that. So yeah, check that out because it looks quite good. I need a new book. So I need to definitely check that one out. Okay. I won’t read Marcus’s book. It’s probably rubbish. I don’t know. Just-
Marcus: I’m going to go and find it. I’m going to put my headphones down and just go and remind myself of what it’s called. Actually no, it’s on my phone.
Paul: No, you go and do that.
Marcus: As you were.
Paul: That’s fine.
Marcus: As you were.
Paul: Okay, all right.
Speaker 2: Boagworld.com. Boagworld.com.
Paul: Well I’ll read our second sponsor which is ExpressVPN. So we all know that VPNs protect our privacy and our security, but they’re also really good for watching TV and taking your games to the next level because VPNs can unlock movies and shows that aren’t normally available in your country. So, for example, I’ve used it to watch Dr. Who when I’m abroad. I’ve used it when I’m in the UK to watch certain episodes of various TV shows that I really want to see but don’t have access to the UK because we still live in this world where people will release programs related on geographical region which just seems nuts to me.
Paul: So anyway, ExpressVPN does this by hiding your IP address and let you control where a site thinks you’re located. So you can choose from like 100 companies or countries even so that you can be located in whichever country you so wish, which gives you access to streaming services like Hulu, BBC iPlayer, whatever else that are region specific. Unlike my current connection which isn’t connected to ExpressVPN at this exact moment, it’s not laggy which is really good. I don’t know why I’ve been having lag problems recently, but it’s also compatible across all your devices, your phones, media consoles, smart TVs, you name it. You can send it all through ExpressVPN.
Paul: So if you visit my special link which is expressvpn.com/boagworld, you can get an extra three months of ExpressVPN for free. So you can support this show, watch whatever you want to watch and protect your security by going to expressvpn.com/boagworld. So there we go. Thank you very much Andrew for joining us. Is there anybody else in the room at the moment that fancy jumping on a mic and telling us a little bit about what they do and where they work because we’re always interested in talking to new people?
Marcus: Yeah.
Paul: In the meantime, while people are putting that in the chatroom, I thought it might be quite nice to share an app that I came across recently. Now this was recommended by somebody in the Slack channel and I’ve not actually personally used it yet, but it looks really, really good. If you’re a freelancer or a small agency or something like this, it’s an app where you can send proposals, you can invoice people, you can get paid, you can track your time, you can track your tasks. So it’s all of those kinds of products all bundled into one. This sounds like another sponsor slot but it isn’t. It’s just a really cool app that I love the look of, and it’s called AND.CO.
Paul: So I thought I’d mention that because… yeah. I don’t know about you but I always like cool new apps and playing the new stuff. Have we not got anybody in the chatroom that wants to come on?
Marcus: Oh. It doesn’t look like it.
Paul: Very disappointing, miserable lot.
Marcus: My book is called The Creativity Code by Marcus du Sautoy, and it’s subtitled, if I can read it, How AI is learning to write, paint, and think.
Paul: Some of the stuff you see is absolutely fascinating with which AI can do. Have you seen where they run across… where an AI has to work out how to get across an obstacle course and it kind of does it by itself?
Marcus: No, I haven’t seen that.
Paul: All right, it’s a Google thing, and it actually will… It’s given the impetus to get to the other side of this virtual landscape and it has to learn to walk and do everything itself. It’s just mind blowing.
Marcus: Oh, Rachel says she’ll join us.
Paul: Yes, I know. I’ve just invited her on.
Marcus: Lovely.
Paul: And here she comes.
Marcus: Yay.
Paul: Hello, Rachel.
Rachel: I didn’t expect that to happen.
Paul: Well, this is what happens. If you show a moment of weakness and offer to join us. So Rachel is a long-term regular member of the Boagworld Slack channel. Rachel, even though I speak to you probably every single day, I know very little about what you do or where you work? So tell us a little bit about what you get up to.
Rachel: Well I’m technically titled as a senior UX engineer at the moment.
Paul: Ooh.
Rachel: But it’s a very, very large playing field and I’ve not been with the company very long. I joined last August and I’ve been doing lots of little bits rather than getting into anything too deep at the moment. So the two things that I’ve been focusing on mostly are I’ve been doing a lot of process flow with a client, looking at some wireframes, really just trying to understand. We’ve spent about six or seven weeks looking at a registration process which is really exciting, trying to understand the security implications because it’s a very high risk client for security attacks and anything that people may want to break into versus usability and really trying to sort of balance those two aspects and the third aspects of what’s technically possible. So what can the developers actually do.
Rachel: So the order of the wireframes, for example on this one, have changed from having an email login one side of the process to having it at the other side of the process to having security questions to having two-factor authentication, and it’s been really interesting absolutely. For every reason there’s a security reason to do something, there’s a counteraction that the user will find it difficult and development wise. So I feel a little bit like trying to keep three sets of people happy and wire framing and sort of trying to get people to tell us if they like that process and going back over that.
Rachel: And then often flipping that on the heads and doing some front end CSS is sort of where I’m at at the moment just to make things actually look how they’re supposed to work. So we’ve got sort of angular developers in the back end working on the core of the functions. So once I’ve designed it I’m asking them how are we going and trying to get in and actually making it look how it’s supposed to work. So more UI than UX I think really, at the moment. So yeah. It depends.
Paul: It sounds, not that I want to put words in your mouth, but what could be quite a frustrating experience because I’ve been in situations like that before where there’s been this kind of battle between security and usability and I found it quite a frustrating experience. So I guess it depends on [crosstalk 00:39:37] your working relationships.
Rachel: Yeah. I mean everybody is frustrated for the same reasons. So-
Paul: Oh, okay.
Rachel: We’re frustrated because we have a common goal that it needs to be really secure, but we have a common goal that we also want people to be able to use it because we could have a really secure product that nobody wants to use because it’s really difficult. So I believe the team that we’ve got talking to each other are brilliant because there’s nobody sort of waving the security baton at users saying we don’t care about you. So we do have everybody caring that we get the right outcome, but the frustration comes from I don’t know that there is a perfect solution.
Paul: No. Mm-hmm (affirmative).
Rachel: I think we’ve got to kind of trade-off something and what do you trade-off, usability or security? I think we can probably say development wise that’s the last of the three things that needs to be satisfied in this area. Technically I think we’d have to find solutions rather than not make it secure or not make it usable. [inaudible 00:40:54] anything’s possible. But yeah, that’s my fun and obviously, that ties in with registration of the user with usability of like forgetting your password and-
Paul: Yeah, yeah.
Rachel: Losing your mobile phone if you’ve got two-factor authentication. So security and registration are the fun I’m having at the moment.
Paul: Mm-hmm (affirmative). It’s an area that is all about risk, isn’t it, in terms of business risk and the management of that because it’s like to what extent do you mitigate the risk for the company and pass the hassle then on to the user, and everything has an associated cost with it as well. So for example, you could make the web application less secure and take the risk of potentially breaches and the financial implications of that or you could pour huge amounts of money into the technical development in order to enable something that is both secure and usable or you can take a hit in terms of usability, but that comes with its own costs as well. So it’s this constant balancing act, isn’t it?
Rachel: Yeah, you don’t win.
Paul: No.
Rachel: I don’t feel like you can win. I feel like you just don’t lose, if that makes sense.
Paul: Yes. I do, I know what you mean.
Rachel: There’s never a time when you’d say “Okay, let’s make it really unusable but secure.” I think it’s just a trade-off. Yeah, it’s just a [crosstalk 00:42:37].
Paul: What I think is nice about your scenario is that everybody’s on the same page about what you’re trying to do. I had a really interesting one where I worked with a company that basically I was working with them on digital transformation, and we decided to set up an internal blog, right. So just for people inside of the company to educate them about digital transformation and what it was involving. We decided we’ll just throw this thing up in Squarespace because it’s quick and it’s simple and it’s easy. So we decided to do that.
Paul: Couple weeks later, after making the decision to do it, it still hadn’t happened and I couldn’t understand why the client hadn’t implemented this. Creating a Squarespace blog is like a job of five minutes. So I went back to the client and I was told that IT blocked it because of security concerns, right. That they felt that it potentially could be hacked because it was outside of their control.
Paul: So I told this story at an internal conference to demonstrate how internal roadblocks can be a problem. I came off the stage and this woman came bee-lining for me looking really angry and very annoyed at me and said, “I was the one that blocked it from IT.” I said to her, “Well, why did you do that?” She said, “Well, for the reasons you said. ISIS could hack that site and cover it with their propaganda.” I asked her what the likelihood of that happening was, and she said a very telling thing, which she said, “Well admittedly, it’s not a very likely, but I would be fired if it happened,” right, and that can often be the problem. That if company priorities are not agreed that you don’t know that she’s going to be assessed on security and another person’s going to be assessed on usability and another person’s going to be assessed on something else, then it causes conflict, but it sounds like you’re not in that kind of position which is good.
Rachel: Yes. I think potentially it may have to get passed further along the security chain still to get somebody to greenlight it, which could then cause problems if it comes back with problems. I think that sort of highlights as well that this sort of job when you’re looking at usability it doesn’t just mean it’s the button green or red. You actually have to think can I safely a log a user into this site without potentially having their data compromised. So there’s so many scenarios that I wasn’t even aware of that security doesn’t just mean it’s a bot. It can mean that somebody’s maliciously trying to take somebody’s account due to the nature of the application that we’re building. It’s sort of people logging in to find out information and somebody might want to go in and find out information about someone else.
Rachel: So I don’t know a great deal of security stuff, but you kind of have to figure out as you go along things that you wouldn’t want to happen if you put something in the wrong place. So you wouldn’t want to put an error message, for example, saying ‘this email address is already in use’, so that aha, I know that my friend’s got an account now because I’ve just typed their email address in. Things like that that I’m not a security expert, but as a usability consideration you need to start learning all these things if you’re doing something.
Paul: Absolutely. That I really admire you for because I think you get a lot of usability designers or user experience designers that just bang the usability drum all the time. Oh, you’re making it difficult for the user and they don’t take the time to understand that broader context, but a good user experience designer, and this is the difference between a UI designer and a UX designer. So although you’re saying you’re doing a lot of UI stuff, and yes you are, that’s how it’s being delivered as UI, but actually it’s UX thinking you’re doing behind that because you’re considering the other components and that I think is amazingly valuable. Sounds a really interesting challenge although I imagine frustrating.
Rachel: It is interesting. Yeah, it is fun but often frustrating when you just think how hard can it be just to log someone into a website.
Paul: Yeah. Nothing’s ever straightforward is it unfortunately. All right, well thank you so much for coming on. You said you had nothing to talk about but actually I think that was absolutely fascinating. So thank you very much.
Rachel: Thank you.
Marcus: All right, bye.
Paul: Talk to you soon, Rachel.
Rachel: Cheers.
Paul: So there you go. That was Rachel talking about the challenges of security and balancing that with user interface design. She didn’t even know that she was doing something amazing, but she is. That’s what I’m really enjoying about this season of the podcast, Marcus, is all these people that every single one of them says, “Well I’ll come on the mic”-
Marcus: I’ve got nothing to say, yeah.
Paul: Yeah.
Marcus: It’s like-
Paul: And they’ve all got really interesting and valuable stories and the kind of thing that we all kind of struggle with and I love that. I love the fact that people constantly underestimate the value of what it is that they do and how actually some of the challenges that they’re facing are not unique challenges, but are quite unusual and quite intense. I mean the kind of stuff that Rachel’s dealing with is that usability security thing taken to 10. I get frustrated when someone has to reenter their email address in a form because of security, let alone that kind of thing. So it’s brilliant.
Speaker 2: Boagworld.com. Boagworld.com.
Paul: Let’s do our final sponsor for the day which is another Balsamiq sponsor and another really interesting product actually. It’s a product called Docsketch which Balsamiq want you to check out. So Docsketch is a really useful tool if like me you’re a freelancer or you’ve got to get people to sign documents and contracts and stuff like that. If you are running your own agency and not getting people to sign contracts, then you should. That’s one of the things that I was being rude to Chris earlier and saying that he was redundant, but that is one of the things that he really put in place from the very beginning of Headscape is that we need proper contractual arrangements with our clients and that I think is a trait that I’ve picked up from him and has been absolutely invaluable.
Paul: Then it’s not even just contracts is it? We’ve got to sign NDAs and we’ve got to sign all kinds of bits and pieces. So this is where Docsketch comes in. It’s one of those electronic signature tools, but what’s really good about their tool is it’s focused on speeding up that turnaround of getting contracts, proposals, sales documents all signed. So every document signed in Docsketch receives an audit report that meets or exceeds both the US and international e-signature laws. So it’s a legally binding document, and the reason we like them the most, because there are other tools like this out there, but the reason that we’re featuring these guys is because you can get three documents signed for free every month, which for most freelancers is probably enough, right?
Marcus: Done. Yeah.
Paul: Exactly. Yeah. So there are paid unlimited plans as well and unlimited documents and branding options and all the advanced preferences you’d expect, but if you are just a freelancer or even someone like me, I suppose I’m just a freelancer really. I don’t know why I’m describing myself like I’m something special, then service like this is just perfect. So Docsketch.com you can check that out and give that a go.
Paul: Okay, Marcus, do we have a joke?
Marcus: Yes, we do. Now I think you need to be of an age to get this one.
Paul: Okay. So in other words, I’m going to get it.
Marcus: Well, even you might not. I don’t know. Anyway, right. So I write songs about sewing machines. I’m a singer songwriter.
Paul: Oh no, I get that. Yeah.
Marcus: That’s pretty good.
Paul: Yeah.
Marcus: Thank you Don Livingstone for that one.
Paul: So just before we wanted to go, I wanted to mention one other article, which is [Anseem 00:51:10], [Ansleem 00:51:10]. I’m useless with names.
Marcus: Anselm.
Paul: Thank you. Gave, which is he’s pointed out there’s a good one about AI writing bots. Can a machine write good copy, which should be quite interesting. I haven’t read it yet. So it’s a New Yorker article and I’ll put a link in the show notes because that looks depressing. As I’ve just finished my book… Oh it failed miserably. Oh good. I feel much better then. So talk spoiler. So now I didn’t need to talk about that article at all because the answer is it was rubbish. Right. But it can complete a sentence well.
Marcus: Well okay.
Paul: It’s the beginning of the end.
Marcus: But according to the book I was talking about earlier, AI’s been very successful at basically recreating paintings in the style of great masters and things like that.
Paul: Oh they do that with music as well don’t they?
Marcus: And I was going to say onto music as well. Like recreating Bach, fugues and things like that. There was a test at a music school I think in Germany where, basically, they would play… it wasn’t a fugue. It was a type of Bach piece anyway that he wrote 400 of and they’re pieces of music that you always study as a music student and basically the students, they would hear two pieces and they were asked to say whether it was by Bach or by the AI, and the AI got 50% right, but they said that one in four of the Bach pieces weren’t Bach as well.
Paul: Right.
Marcus: Basically it did extremely well-
Paul: Yeah, that is very well.
Marcus: But what it can’t do yet is just make up stuff that’s it’s own as it were. It needs to have a framework to work around.
Paul: Yeah. So Marcus, what was the name of that book you said again? The [crosstalk 00:53:14]-
Marcus: The Creativity Code.
Paul: Creativity Code. Okay, I’ll put a link in the show notes for that as well. All right, another really interesting show. We’ve covered AI. We’ve covered dealing with data visualization. We’ve looked at security. My word, that sounded almost like a developer podcast. Whatever happened? It was. We did it from a user interface element. So that makes it all right. But there you go. That’s it for this week. If you fancy joining us on the show we would absolutely love you. As you can the tell, we really enjoy talking to people and look, you don’t need to worry about whether you’ve got anything to say or not. That is our problem. We’ll find interesting stuff because I guarantee you there will be something you do which the rest of us find interesting and we’d love to have you on the show.
Paul: So to make sure you don’t miss the next show when it’s recorded live, go to subscribe.boagworld.com/community. Sign up then and we’ll get you on because I don’t care who you are, you’ve got something interesting to say. But till next time, thanks for listening and goodbye.
Speaker 2: Boagworld.com.
Speaker 2: Boagworld.com
Speaker 2: Boagworld.com
