Episode #18: Design roundtable (Part 3 of 3)

Matt Linderman: Hello, welcome to the 37signals podcast. I'm Matt Linderman. This is the third and final part of our design roundtable discussion. You'll hear from first Jamie, then Ryan, then Jason, on this episode. What we do on this episode is we take some reader questions from Signal Versus Noise.

Matt: So Shubham asks, how important is the designer to the success of a product? Do you think a developer could be a good designer or vice versa?

Jamie Dihiansan: I think definitely. Design is not something that you're born with, it's something you learn. I think a developer can be a great designer, just in the same way a designer can be a great developer. It's just whether you take the time to learn what's right and wrong, or be influenced by whatever. Sure.

Ryan Singer: I'd say that both of them have to do design. When we really understand design and we know it's not just colors, or which curtains do we match with which pillows or whatever. The software person who doesn't think like a designer is just connecting things together until they run out of things to connect, and they hope it all stands up. The software maker who really knows design, is thinking about... How to separate the different concerns in their software? How to make things where they can easily change them, without breaking other things? How to disconnect certain parts of the software, and connect other parts of the software so that all the joints bend in the right places, and other things stay together where they need to stay together. They're doing design, but the object of their design is code, it's functions, it's data. The UI designer is looking at other things. They're looking at what does a human being see when they look at the screen, and what is important to people, and how is the aesthetic? I think that the question that people don't ask enough, people get fixed on designer versus developer. Instead of designer and developer, and looking at... How should these two be working together? Who starts the process? How do we take turns? How do we give each other feedback? That's, I think, the more interesting thing to look at.

Matt: It's almost more of a loop than just an arrow going one way.

Ryan: Yeah. Not only a loop, but it matters where you start. It is chicken and egg a little bit, but it matters if you start with the chicken or you start with an egg.

Matt: Which one should you start with?

Ryan: For sure, you should start with the interface because you're starting with the customer experience there. Even what that means is something you can talk quite a bit about. Does starting with the interface mean you do a rough sketch? A very convincing rough sketch, enough that you know enough about the data model that the programmers can do some really rough database hooking up while you're getting more details together? Or do you need to get a completely clickable HTML prototype before you have enough confidence in the design to talk to the programmer? Then do you hand it to them and they start building it or do you sit down and say hey what do you think? That can depend on your relationship, it can depend on your goals, but there's plenty there. But I would definitely say starting with the interface design.

Matt: The next question is from Seth. He says it seems your designs have always been quite minimalist and you avoid framing in the sites in solid boarders and everything is laid out in blocks going down the page. Do you feel that the simpler you make the designs for the various product sites the better they convert?

Jamie: Well, I'm not really sure what that means about blocks and stuff like that, but I think that goes back to what we said earlier. We're just trying to communicate, especially for the marketing sites... What does this application do? Not necessarily oh it has an address book, it has this and that, but what problems do our applications try to solve. We're trying to communicate in the clearest way.

Matt: How much discussion is there of conversion rates and eye tracking and this design is working at this percentage success rate versus this design?

Jamie: I think initially we were really interested in AB testing. We have Google Analytics hooked up and we're keeping an eye on that, but I think the true testament is... Are people signing up for the products? The answer is yes. Could it be better? Sure. It could always be better. That's actually what's interesting about design is you know that the design is working if someone's buying your product because if it isn't working then people are bailing and your conversions are down.

Jason Fried: We do look at some of the stuff occasionally. We did some AB testing for headlines and whatnot. We don't do it as often as we probably should, But we have done it and I'd like to do more of it. It's one of those time consuming things that you just tend not to get around to. As far as the word minimalism, though, I tend to not like the word because I don't really even know what it means. To some people it means white space and to other people it means just a shorter page. I'm not even sure what that is. For us, the goal is always clarity. If that means it's short that's fine, if it means it's long that's fine too. As long as it's clear. That's kind of the goal. Certainly we could always be better. If we could test nonstop I'm sure we could refine and refine and refine. I think there's also a point where you want to make some educated guesses and decisions and move on to other things because we've got other stuff to do, we have a really small team. We don't have an analytics department. We have five designers really and that's all we have. We have six products and a bunch of other stuff going on, so we have to make some choices that may not be the best choices, but we're pretty comfortable with them most of the time.

Jamie: The other thing I want to add about usability testing, sometimes there is definitely a clear choice. If you don't really know what's going to work then usability test it, but a lot of times, especially at previous jobs, we would usability test stuff and whatever was the outcome it didn't really matter because we were just going to do it anyway. Sometimes it just doesn't matter. This whole industry about usability testing...

Ryan: You have to be open to the results.

Jaime: You have to be open to the results and a lot of developers and project managers and designers, when they're up against a deadline, they don't really want to look at the results in a clear way. Because they're also kind of controlling the tests, which is kind of bizarre, too.

Ryan: On the topic of eye tracking, something I wanted to say there is when I open up an interface to look at it critically, the first thing I look for is actually where am I looking. I think that's more interesting than seeing where ten other people look. I can try moving something, and then reload the page again and say where am I looking. The different elements on the screen are moving your eye around, they're attracting your eye based on their intensity or contrast and stuff like that. That has a lot do with what you read and what you don't read and that, in turn, has something to do with what you know and, what you don't know about what's on the page. I think that's an interesting thing to pay attention to is where does your own eye go when you open up a screen.

Matt: Let's move on to Colin, he actually has three questions, so we can breeze through these. Have you ever felt limited to the possibility of HTML compared to native apps? How do you approach that? Simplify until it reaches something "easy" to do with actual technology or trying to tweak and distort and use coder's genius to make great things?

Jason: I personally like HTML. I've done design in Filemaker Pro a long, long time ago. It was more you had to do something somewhere else and copy and paste it somewhere else and see if it works. I like the directness of HTML where you just make a change and you reload your browser and where the design shows up is where it lives. I like that element to it.

Jamie: I think what's interesting, just to bring it back, is the fact that we have developers and these tools is different than design. Design is not HTML. Photoshop is not design, it's a tool. What I think is interesting is design will always exist, whether it's hard or not. The thing is communicating, clarity or proportions, or whatever. Whatever tools are coming up, so if it's native app or HTML, whatever's appropriate to communicate what you want to do.

Ryan: Colin's next question: are you designers or designers/HTML/CSS coders and when does the actual coding of a layout start to emerge? Really early, or do you try to avoid this in the first place to be focused on pure ideas?

Ryan: We don't have anybody here who is unable to do HTML and CSS. I don't think we would have anybody like that because we don't really have time for that. If you need to have somebody else translate all of your work into something that's usable, then that's a lot of lost time.

Jamie: I think that the analogy could go to even print design. To some extent, you have to know what colors, with overprinting or whatever -- when it gets screwed up on the printing press. You might not know how to run a press, but you have to know how the thing will be produced. I think the same goes for web design, or app design, or whatever. You have to know what it's capable of, and...

Ryan: Yeah, it's your material. It's what you make your design out of. The idea that you could make a website and not know HTML blows my mind, because it's like building a building and not knowing what a brick is. I think there should be... There's definitely different levels of how deep you go. You can be a CSS wizard, or you can know basics, but you should at least have some foundation knowledge of what your building material is.

Jamie: And I think, also... Emphasize this. We're sitting in this room with Coudal's Field Notes in there. I know Brian, who does the design for field notes...they go through a lot of paper samples. They see how the ink is printing on paper. If the black is breaking up. Those are details, as much as HTML and CSS are for web design.

Matt: Colin's final question. HTML 5 seems promising. What will be the trigger for 37signals to start coding sites using the new functionalities it provides. Do you think HTML five will make your life at 37signals easier?

Jason: We're using it in a few spots already, I think. Some of the new stuff. But, I don't know. It's just another stack. There's new, good stuff, but until all the browsers support all of it... It's not going to... I don't think it's going to change the way we approach design. It just makes certain things easier.

Ryan: It's another tool.

Jamie: It's better in a lot of ways, but it's not going to change the way our stuff looks, for example.

Matt: Duncan asks, "If you had a month and as much money as you needed to complete a study of some quirky area of design or a nagging question that always bothered you, what would you do the study on?"

Jamie: I'd keep the money. [laughter] Let it collect interest.

Ryan: How to get more money.

Jamie: I wouldn't spend it on design.

Jason: I don't know. That's kind of a neat question, but... I kind of like... There's a lot of shit going around when Google was revealed to have tested 20 shades of blue or something. I heard you talk about this. That's not really design, it's research and science, almost. We don't do any of that, and I think it's kind of interesting. So, I would like to learn a bit more about that kind of stuff.

Ryan: I'd like to spend it on...

Jason: Behavioral psychology, almost, of colors and design, or something like that.

Ryan: I would like to spend it on... In person with customers, somehow, or having some kind of experience that we don't already have. We spend all day, every day, looking at stuff that we already have and thinking about how it could be better. So, sitting down for a month on one thing isn't going to change anything. But if we took it and said, "We're going to actually break away from our normal pattern of doing things and we're going to spend some time with our customers, or we're going to meet people that we haven't met who use our things and see what they think." Then this would be something new.

Jamie: I think it would also be discovering what new problems people are encountering, because as the Internet evolves, or as business evolves, people have new problems all the time. That's what it's all about, is solving problems. There's always going to be new problems coming up.

Ryan: Alright. Last question, from Hendrick. He says, "What's your take on the future of design for web-based software?"

Jason: I don't know. I don't really know. I don't think about stuff like that so much. I think we're going to see... I guess, if I had to guess, we'd probably see more iPad like metaphors coming to the web, or something. Maybe more fixed designs. I don't know. Maybe that's totally wrong. I don't know. I just feel that the iPad is going to influence a lot of how people do... Like the iPad UI controls, because over the past 10 years, people have been trying to imitate the desktop. Pulldown menus, and all of that kind of stuff. That came from the desktop, and people were trying to... For years, people have been trying to emulate that in HTML and JavaScript to make it work. So now, there's a whole new thing. There's no such thing as pull down menus on iPads, for example. It's like this pop-up, hovering menu sort of thing. So maybe there's more things like that. There's also going to be stuff like... Actually, hovering is something that we've been thinking about. A lot of our designs require hover. You hover over something to have something pop up. That doesn't work on touch based devices. We're talking about the iPad now, but touch is where it's going. Hover doesn't work on touch, so how can you reveal things that people don't know are there in a new way, and let them know that they can reveal them? Things like that. So, some interesting things are along those lines, but I don't know. There aren't going to be any deep predictions.

Ryan: I think an area with a lot of question marks hovering around it right now is just the file system. If you use Google Docs and you try to share a Google Doc, it's not easy, actually. It's not easy. If you try to sync files to your iPad, it makes no sense. You go through iTunes, and if you use a lot of different web apps, your information is happy inside of the app, but going between apps is not so easy. This seems to be where a lot of pain points start to come together. I don't know what the solution is going to be, but there definitely seems to be an area where a lot of the different things that are changing with how we use web apps, what's happening with the iPad. They all point to the fact that the file system is a little bit like... We don't know what's going to happen with it, or what's going to take it's place. It's a real fuzzy area right now.

Jamie: I guess for me, I guess... The future, I'm not really sure if it's going to be a browser, or an app, or what, which I guess I'm a little nervous about. I guess I should learn Cocoa or something. We'll see. Who knows what's going to happen?

Matt: I just have one final question, based on that a little bit. When you were talking about touch is where it's going, or we hear discussions about HTML 5. How much of being a designer do you feel is seeing where the future is going? It's like a fashion designer or something being able to say OK, here's the way things are moving, and I need to be ahead of it, or skate to where the puck's going to be, as Gretzky said. And how much of it is just waiting, and being like, "We don't need to be at the vanguard and what's new and what's shiny. Let's see what happens, and when the time's right, we'll figure it out." How do you balance that, or is that something that you even think about? Do you just sort of go with it intuitively?

Jason: I think that... Me personally, I want to be at the absolute cutting edge of what we're good at, but we aren't producing a whole technology stack that we rely on. So, we're really good at thinking through things, and using the technology that we have to make things clear. When we have the opportunity, and we see some new technology coming we can take advantage of it to make even better products that are even clearer, and faster, and easier, and fit with people's lives better. We do that. We only have control over a certain part of the world, which has to do with the tools that we know, and the tools that are arriving, and the tools that are nearby. We do our best to follow it, but at the same time, we don't control the whole stack. I think that if we continue to focus on things like copywriting, and clarity, and knowing what matters... That isn't going to change, regardless of what the technology is, and then we just do our best to follow what the possibilities are with changing technology.

Jamie: I think, for me, cutting edge on the web is always going to be something that's art based, which is fine. If you look at how the world worked, the art world is always ahead. And then it somehow trickles into the consumer design world. I think that's the same for the web. I've seen a lot of beautiful sites that do crazy things with new tools like HTML 5 and CSS 3. Sure, we're trying to do some of that stuff now, but it's not like that's our main purpose of existence, is to push the limits of this spec. We're just trying to solve... Or make it easier for people to use our tools. [music]

Matt: Alright, that will wrap it up for this episode. You can go to 37Signals.com/podcast for related links, transcripts, and a list of previous episodes there also. Thank you for listening. Bye.

More episodes of the 37signals Podcast.