By: Matthew Magain

May 21st, 2008

Reader Rating: 8

Andy Budd was in Melbourne as the keynote speaker for the Web Directions UX [1] conference this week. SitePoint's Matthew Magain spent some time chatting to him about blogging, design, CSS frameworks, careers on the web, usability testing, the future of the Web, and much more.

SitePoint: Thanks for taking some time to talk to us Andy! Let me begin by saying that your blog, along with Drew McLelland's All In The Head [2], is one of the first blogs that I began following, many years ago. Can you tell me how your blog has shaped your career on the Web?

Oh wow, cool. It's probably quite difficult to say to what level it's had an effect, because obviously there are a whole bunch of things that have happened.

But I do think it's been fairly important. I started blogging in the UK about web standards quite early on, and at the time I only really knew of two other people who were blogging about the same stuff in the UK, and those turned out to be Jeremy Keith [3] and Richard Rutter [4], with whom I later set up my business, Clearleft [5]. And after a while I discovered a few other blogs, like Drew's blog, and a few other people started doing similar things.

It was a fairly early stage, so my other peers were people like Doug Bowman [6], Dave Shea [7], Cameron Moll [8] and those guys. So I guess I got lumped in with those people, and there's definitely kudos that came with that. Rich, Jeremy and I were the people in the UK doing similar stuff.

But I also think a lot of it was luck -- a case of being in the right place at the right time. I was really interested in web standards, and I really just started blogging as a way of helping sort through my own thought processes. When I discovered something, rather than keeping it to myself, I'd post it up on my blog, because I thought that if it helped me, there might be somebody else out there that it might help.

I guess like most bloggers, it was all driven by fairly altruistic reasons. At least in the early days, I think most people blogged for fairly altruistic reasons. These days you get a lot of people who blog to get fame and fortune, which, frankly, I think if you want to get fame or fortune, you should learn to play the guitar or learn to act. I don't think you're going to get famous by blogging.

But yeah, I guess I got reasonably well-known through my blog. At one stage it was getting an enormous amount of traffic, and I guess in part the ability to go and speak at conferences came out of that. Not directly, but when I approached SXSW, for example, to see if I could speak, I'm guessing the fact that I was a known entity helped that. Although SXSW has always been quite good at breaking new talents; they've always been quite good at getting people who haven't spoken before to speak.

I guess when I wrote a proposal for my book, CSS Mastery, the fact that I had some speaking experience and I had a blog confirmed to my publisher that there was at least some kind of market there. As you would know as a publisher, it's a little bit risky having a completely unknown person write a book when there's no defined market. Whereas at least people who read my blog were probably more likely to buy CSS Mastery.

So I guess it has all kind of fed off each other. The more speaking I do these days, I suppose my blog is really poorly looked after, I have to admit. I'm lucky if I manage to write one or two blog posts a month. And I feel really bad, because I really want to write stuff, and there's loads of stuff that I want to write.

But with running Clearleft, organising dConstruct [9], and doing all this speaking means my available time is quite tight. I think it's quite sad that a lot of my peers who I mentioned earlier -- Dave Shea, Doug Bowman, Dan Cederholm -- have also found themselves in a similar situation where the amount they're blogging has really kind of trailed off in the last 18 months, as we all end up working harder and doing loads of stuff.

SP: Do you think that part of that might be that everything to say about CSS has already been said?

Oh, absolutely, yeah. I haven't really been talking about CSS on my blog for a good couple of years, and my focus is much more in terms of usability and user experience design. I mean CSS was always just one narrow part of what I was interested in anyway, but it was the part that I was learning at the time, so naturally it was the part I was blogging about. And I think you're right -- I think CSS, to a certain extent, has run its course -- at least the people like your Doug Bowmans and your Dave Sheas -- have discovered all of the interesting stuff that we're likely to discover.

So the people who are blogging about CSS now are people who are just learning about it, and discovering it fresh. And there always needs to be that kind of stuff -- there's always going to be people learning, so there's always going to be the need to learn from your peers or from people in similar situations. But for me, the CSS thing is kind of like a done deal, to a certain extent.

SP: You mentioned your business partners before, Jeremy and Richard, who are identities on the Web in their own right. How does the clash of egos work under the Clearleft umbrella?

Pretty well actually. The reason we came together to work on Clearleft together is because we all have fairly similar beliefs about how the Web should be. We all believe in semantic markup and web standards ... we believe in good design and usability and user experience. So we all have those similar overlapping interests. Individually, though, we have things that we are passionate about.

So Rich is amazingly passionate about good typography, and good accessibility, and he's a really awesome interaction designer ... really solid. He gets the work done -- he's like the backbone of Clearleft, really. He's the one who does all the work, and Jeremy and I take all the credit!

I'm really interested and passionate about user experience design and I'm fairly good with the business side of stuff as well, which Jeremy and Rich aren't really interested in. So I like speaking to people, I like going and meeting with clients and explaining what we do, and gettin passionate about all the great stuff that the Web has to offer.

Jeremy is just super smart. I mean, he is really on the ball. He knows his code and his JavaScript, but also he knows his classical references and that kind of stuff. He went to art college, so he's got that kind of background. Plus he's an awesome speaker.

And so it kind of works really well. Obviously we have our own different approaches to certain things, but generally, philosophically, we mesh really well.

SP: So there's not too much butting of heads then?

Oh God no, not at all. I mean there are a few little "catch points" that always spark discussion, and one of them is the "fixed vs. fluid" discussion. And there are always differing opinions around that. But that's because it is one of those "How many camels can you get through the eye of a needle?" type situations. There's no right answer, there's just opinions. Generally, it's those kind of classic discussions that come up over and over again.

SP: Let's talk about usability testing. Why is it so important? And to play devil's advocate, doesn't it take the expertise out of being a good designer?

Hmm. If you believe in the sense of "genius design," that people can innately just create something amazing -- and there are genius designers out there, like Steve Jobs, then you could be seen as saying that usability might stifle that genius design.

But for the 99% of designers out there who aren't these Galileo or Leonardo da Vinci-style genius designers, design is a really useful design tool. And I think one of the problems with the perception of usability testing is that most people's experiences with it is as a summative testing that's done by big agencies at the end of the project. And they tack the word "test" on it, and it makes us designers feel like our work is being tested and graded somehow, and if you fail the test then you're bad designers.

It's terrible, because it makes us designers feel like we're living in this kind of "checkbox" mentality. The kind of usability testing that I favour is called formative testing, and it's using usability testing as a design tool. You know, we can't just sit in our little isolated boxes staring at Photoshop all day, creating experiences in isolation, and believe that the experiences we're creating are going to be usable and perfect. Because experience shows that once you put something out there, people will always use devices in a way that you never expected. And people will find frustration in things in ways that you could never expect.

I think one of the big problems is that clients and employers feel that you've hired a designer, and they should be geniuses, and they should get things right the first time. And frankly, if you hire a company like Clearleft or Adaptive Path, that first iteration is going to be really good -- it's going to be much better than if you had hired somebody who doesn't understand about usability and user experience. But you never, ever get it right the first time. And good designers understand that design is about process, and usability is a really important tool to educate design. So I do think that usability is a design tool. You sit down, as a good designer, you see clients or users use the products that you're designing. When you see that they're struggling with a particular situation, or they don't understand something, you get these "light bulb moments", and it informs your design decisions. So then you can go back, and iterate, and do better and better. And the more time you spend in usability tests, watching users use your products, the better a designer you'll be. And I do think there's a certain amount of arrogance to think that you're a genius designer and not have to worry about how users use your stuff.

So I see usability as a design tool. But then, I see design as Design with a capital D -- I think design means designing the system, the architecture, the user flow, the information, the visual interface, the templates ... it's not just the graphic side of things. You've got to get all of that right, and all of that is meta-design.

SP: The bio on your site mentions Flash. How does your experience working with Flash influence how you design these days with HTML, CSS and JavaScript?

Oh, does it? Crikey! Yes, quite amazingly I used to be quite a hardcore Flash person. Not so much a designer, but I was quite a hardcore ActionScript programmer. And I was out there back in 2000, building object-oriented Flash games in ActionScript 4. But I started to get a little bit disillusioned with the Flash world, because at the time the experiences people were building were not very usable, they were not very accessible, and I discovered this brave new world of web standards, and it just made so much sense to me. So I slowly started moving away from Flash and into the world of web standards, which is kind of a strange transition I guess -- I don't know any of my standards colleagues who have done that.

But on the other hand, weirdly, it's starting to come full circle. I don't do any Flash stuff these days at all, but if you look at Flex, and you look at ActionScript ... basically ActionScript is ECMAScript, which is a standard. Flash uses MXML as a declarative markup language, so friends of mine like Aral Balkan [10], who do Flex development, are writing Flex in a markup language, using CSS, very similarly to HTML.

So it's interesting to see how a standards-based approach is being adopted by the Flash and Flex world.

SP: The reason I mentioned it is because I was wondering whether you missed the freedom that a lot of Flash designers feel like they have over a markup-based solution?

I think that design is all about constraints. Some people love the blank canvas situation, but I don't at all. I think that's really for artists, not designers. I think design is all about constraints -- business constraints, technology constraints. And it's those constraints that actually create the innovation.

And if it wasn't for those constraints, we wouldn't have all of these cool CSS effects, or all of these cool JavaScript libraries. So no, I don't think it's a problem so much. However I am increasingly interested in user experience design, and I do see the limitations of HTML and CSS and JavaScript. And yeah, I am starting to peek at some of these experiences that Flash developers are building now, and think it would be nice to have some of that flexibility -- but with the accessibility and the markup benefits, and the openness of HTML.

So it's not something I will ever do again -- I'm over the Flash thing. But it's interesting, yeah.

SP: You revealed Silverback [11] to the world the other day. Tell us a bit about this application.

Yeah, sure. I've been doing this little world tour of usability workshops, partly because I think that usability testing is the way forward. The whole concept of Silverback was as a way for designers and developers to do really low-cost, guerrilla usability testing.

There's a whole bunch of different software solutions out there which are quite expensive -- in the region of US $1,500, and they're good but they're really feature-heavy and quite difficult to use. So we decided to create Silverback, which essentially captures screen activity, mouse clicks, stuff the user is typing in, audio, and video, as a way of doing really cheap low-cost usability testing.

SP: The marketing of this application was interesting, wasn't it? You put up a sign-up page, and you told me earlier that you didn't plan on promoting it, but it seems to me like the perfect viral campaign.

I guess so, yeah. It really was accidental marketing though. What happened was we bought the domain, (link: silverbackapp.com) and it was just resolving to a domain name company, and we thought people might start hearing about it -- our friends or whatever. So I asked Paul, our designer, to stick up a holding page, and we had the nice gorilla design from Jon Hicks. And Paul, who's an awesome designer, created this nice little Parallax effect that he'd been playing with.

I went to London for the afternoon, not expecting much to happen, I was off to see a client. And he'd just twittered it -- he'd just posted "Guys, I've put this holding page up." And all of a sudden we got inundated with all these twitter messages and emails saying "Wow, that's so cool, what's Silverback?"

So people were getting excited because of the logo, people were loving the holding page, and people were really interested and curious about what the application was. And it was just a holding page -- we didn't want to give too much information away. It just said "Silverback. Coming soon." And that seemed to just generate so much buzz, and within the first day we had about 2,000 people sign up on our announcement list. And by the end of the week we had 5,000 people sign up, and I've got no idea how many people have signed up now.

So for a small little company doing a small little desktop app, that's pretty awesome. But now one of our fears is, because we've been a little bit mysterious about the application, and there are reasons why which I'll come to in a second, people have been imagining the best kind of thing that could possibly happen. I guess, because we're Clearleft and we're fairly well known, and I work with Jeremy and Rich and we've got so many smart people who work with us, suddenly people are starting to think that this is some sort of application that will solve world hunger and will, you know, make the world a peaceful place.

In fact, it's a really simple application that came out of realising that we've got a portable usability lab in our hands these days -- whether it's an iMac or a MacBook Pro or whatever. If it has a screen, it has a microphone, it has an in-built camera, then you don't need these big listening labs. You don't need to set up a tripod pointed at the screen. You can do it all in the one device.

And there used to actually be these portable usability labs -- you can still get them, in fact. They're these boxes with camera equipment and so on. You don't need that -- you've got it all in front of you. But the idea was so simple -- screen capture, audio capture, and video capture -- we were scared that if we gave too much information away, somebody who already had a screencasting app or was doing something similar could add feature in quickly and easily. So we didn't want to start shouting about it until we'd developed it.

Also, we'd never really developed a desktop app before, and we didn't know how long it would take -- and it has taken longer than we thought. It has taken probably a couple months longer than we ideally would have hoped. Getting bugs fixed and tweaked, and tweaking the UI. Plus if you develop an application, you end up obsessing over it, and we obsess about most of our clients' web sites as well. But if it's your baby, you want to make sure it's 100% right. So the fact that we didn't talk about it or saw what it was fueled the speculation even more. And that was part of the mystique -- everyone started chatting about "Oh, what is this Silverback?" People started writing blog posts, and I'd meet people at conferences, who'd say "Can you just tell me?"

So yeah, this kind of mystique did make it quite a popular thing, so it did help us. And I wouldn't say it was purely accidental, because as you saw today we do like building these little things into our applications, and we do have a good network of friends, so if we twitter something, people will find out. But it certainly wasn't a conscious thing to keep it secret so that it would fuel desire... it just kind of happened like that.

SP: What's your opinion of CSS Frameworks, and how do they fit into the approach you guys take in developing a solution for a client?

We have an internal framework that we use for our wireframes -- our prototypes. Wireframes are not things that live by themselves, so it doesn't really matter how semantic or pure they are.

This library, or "framework" if you want to be down with the hip kids, is a mixture of HTML and CSS and JavaScript, which we use time and time again for our wireframing. It makes use of jQuery, and a company we work with called New Bamboo [12] have helped us to develop a jQuery plugin called polyPage, which allows multiple state changing within a wireframe, which is really cool.

So I think these libraries -- frankly, I think "framework" is a bit of a grandiose term for something like Blueprint [13], for example. Essentially it's three or four CSS files and a page worth of HTML. It's not a framework -- a framework, I think, is a much bigger thing. And there's a certain element of trumpet-blowing going on there. By calling it a framework, it makes it sound like it's this big, important thing like Ruby on Rails. Everyone knows Ruby on Rails is hot, everyone knows Django is hot -- let's call it a CSS framework, and it will be big.

I think what you need to do is develop your own framework. You need to develop a series of tools, a library of elements that work for your team, that work for the types of sites you build, that work for the internal belief system structure and knowledge that your team has.

I think generic libraries like Blueprint box you in to certain solutions. They have a certain amount of flexibility, sure. You can choose between one of any number of columns -- I think it's 16 or 24 -- and you can span columns and stuff. But it's still limited to this column behaviour. And you're limited to the size of the gutters and the width of the window, and if that doesn't necessarily fit with your designs, what do you do? Do you spend hours hacking your library to fit with your designs?

Probably not. What I imagine many people do is fit their designs around the library, and you end up getting this templated approach to web design, in the same way that people who are really focussed on a particular CMS -- all of their web sites are designed around the capabilities of that CMS, rather that creating a user experience that is right for that solution.

Ironically I've spoken to some of the people behind some of these libraries.And they have been in real-world situations where they've used a library initially, and it just didn't work and they've had to stop using it and rebuild things from scratch. So libraries aren't always the quickest thing to use, because you may be using the wrong tool for the job.

Jeremy has a saying, which I think is a pretty wise saying. He says, "If you have a hammer, everything looks like a nail." And I think that is the problem with the pre-built tools. If you have a design in place, you try and put everything into that library.

And I actually think Blueprint has some fundamental issues that, as a company, we don't particularly like. One of those is the fact that it is not very semantic at its core. It has presentational elements -- for example, it has classes called "span-1", "span-2", "span-3" ... and what it has essentially recreated is a table-based layout. My feeling is, why add 200 lines of CSS code to recreate a table-based layout, when you can just use a table? If you're not that bothered about semantic markup, and you're willing to use libraries that pollute the semantic nature of your HTML, why not use a table? Because it's already built, and all that code is already there. Why try and reinvent the wheel?

Now, personally I think that tables for layout are bad, but I also think that non-semantic markup is bad.

Then again, people would argue that it isn't really non-semantic. I think that clearly when you've got class names in your HTML structure that are called "span-1", "span-2", "left", "right", whatever, that it's a little bit backwards.

SP: When IE8 comes out some time later this year, it will support CSS tables and will be the last of the big browsers to do so. Do you think there is a light at the end of the tunnel that web developers one day won't have to worry about choosing between using libraries or tables for layout? Or will there always be legacy systems that need accommodating?

This is one of the big issues in the web standards world, with the creation of HTML 5. There's a big discussion about the fact that there are a lot of legacy systems out there. There are a lot of systems built two, three, four, five or more years ago, that the site managers just cannot rebuild. And there are CMS systems that are 10 years old that are using old-style markup.

So when looking at developing the HTML 5 spec, the working group wanted to put the font tag back in. Their argument is, "Well, there are so many people already using it, our browsers need to understand what to do with the font tag, so we need to bake it into the spec." My problem with baking it into the spec for backwards compatibility issues is that you are legitimizing its use, and by legitimizing its use, you're destroying the whole idea of separation of content and presentation which, no matter how impractical it might be in certain situations, I think is a good paradigm to advocate and aim for.

But on the other hand, I do understand the argument for Microsoft and the meta tag that they're introducing. When you update your browser from IE7 to IE8, if a web site breaks, people aren't going to be blaming the web developer, they're going to blame the browser manufacturer -- even if the browser is doing a proper job, they're the ones who get it in the ear.

Unfortunately when they moved from IE6 to IE 7, IE7 was much better and was much more standards-compliant, and the standards world all thought that move was amazing. But they got lots of negative PR and negative phone calls from their sales guys and from their clients, because that update had actually broken existing sites. So backwards compatibility is a big deal. But I do think the idea of bringing the font tag back is crazy and redundant.

So going back to your question, we're always going to be left with redundant pages and redundant markup. And this is a really big problem that the browser manufacturers are always going to have to face. The browser manufacturers might not be supporting IE4, the operating system manufacturers might not be supporting Windows 95, and for a very long time we're going to be left dealing with markup that is very old. And there is important information in those pages -- there are Tripod pages, and GeoCities pages that are 10 years old that have important information, but don't have a web developer or designer or someone still looking after them, but it is still historical data. And talking about digital preservation, at some stage browsers will not be able to backwards-render.

But on the other hand the whole purpose of web standards, partly, is digital preservation -- if everyone created web sites in the proper fashion, we wouldn't have this problem in the first place. Wishful thinking, perhaps.

I honestly think that CSS is basically a dead technology after this next revision. I think that's really sad, because I'm a standards advocate, but that's what I believe. I don't think we'll see a CSS4. I don't think we'll see half of the modules in CSS3 implemented, or even end up in the specification, because CSS3 has been under development for 8 years or something ridiculous, and in 8 years we haven't got a single module complete yet. We've got a few almost finished.

But I do think that in this world where I've got a PS3 at home that performs live texture-mapping rendering in 3D, on the fly, why we can't get a browser that can render rounded corners on a box in 2008. You know, we should all have jet packs here! A rounded corner box or multiple background images shouldn't be rocket science.

I think part of the problem is that there are innate problems with the standards development mechanism, and I think there are innate problems with the browser vendors, and I do think we're going to see other technologies -- sadly, proprietary technologies like Flash and Flex and XAML and XML and MXML and non-native Web technologies -- take over, because we are limited with what we can do with our current technology. The reason we're seeing all of these JavaScript libraries come out is because they're supplementing what the standards and the browsers should be doing. It's kind of a stupid situation to be in, and we've got to lobby the W3C to pull their finger out to get these standards ratified, we've got to lobby the browser vendors to stop arguing among each other over petty little things and implement them, and there needs to be a radical shake-up in the way these things are created.

SP: The Web Standards Awards [14] is a site that you and a bunch of guys put together a while back that regularly showcased good examples of combining function and form in a standards-compliant way. Do you think this site will ever come out of retirement to showcase sites built with a bunch of new technologies?

Wow, you've got a good memory! Yes, it's something that me and Cameron Adams [15], who I didn't know at the time, and a guy named Johan Edlund from Sweden, who I'd also never met. And it was like an antithesis to the CSS Zen Garden [16], which was awesome. But the Zen Garden was a bunch of people showcasing their CSS skills by working on a single web site.

It was awesome, and it changed a lot of people's impressions about CSS being a boxy thing to being a beautiful thing. But there weren't any showcases of actual, live, working web sites. It's all very well sending your boss to look at a "hacker's" site, but they want to see examples of well-known companies.

So we gave awards to sites like the PGA Golf Tour, and big companies that bosses would know. And the idea was that people would send their bosses to these sites with the message that this web standards stuff isn't just a niche thing -- it's something that lots of people are doing. And we ran that for about a year and a half, and the idea was that we would post one or two awards a week -- silver awards -- and then at the end of the month we'd nominate one of those as being the best site of that month.

And it worked for a while, but then we felt we'd reached a tipping point where standards had started to become that prevalent that there were loads and loads of really good sites out there, and awarding good design stopped being unusual and started just being commonplace. And secondly, there was this proliferation of sites that were doing a similar thing. There was Style Gala, CSS Vault, CSS Beauty, CSS Mania, CSS whatever ... I stopped keeping up. I literally have in my RSS reader, a list of about 30 gallery sites, and I see new ones all the time!

Frankly, when there were three CSS gallery sites, we were thinking "Well, who needs a fourth?" I've got no idea how these sites manage to survive, but they obviously seem to. They all share the same designs -- it's shown on one site, then it proliferates to all the others, so there's very little that's new that's being shared around.

So I guess we thought that the site had done its job, and unlike a lot of these new sites, which are there to make Google advertising and to build a brand for the people that are running the site, the Web Standards Awards existed because we were really passionate about highlighting good design. Once that was a done deal, there was no real need to carry it on.

SP: As well as Silverback, what else is in the future for you guys?

We've got loads of stuff on at the moment. We've got Silverback launching, in a couple of weeks' time. We're still doing some last minute tweaks; we're still optimising the outputs, because they're a little bit large at the moment. And we've got the web site's design being built at the moment. We're hacking it into the payment system, all that kind of stuff. So hopefully in two or three weeks' time, Silverback should be launched. There are a whole bunch of demo copies floating around -- I gave some away at our workshop.

We're getting dConstruct 2008 [17] organised, which is a conference we run in Brighton, which has grown so much over the years. The first year we had 120 people in a church hall ... last year it was 750 people! And we capped it -- the venue had a capacity of 2,000 people, and I reckon we could have filled up that venue, because tickets sold out to dConstruct last year in under 24 hours. They go really quickly.

But we didn't want to sell more tickets, because we wanted to keep it to a manageable level, and we didn't want it to turn into SXSW. I mean, SXSW is awesome. But I think that one of the big important things about conferences, apart from the speakers, is the attendees, and allowing these hallway conversations. And if you've got 1,500 - 2,000 people in a room, and there's no room to wander around, it's just difficult to meet up with people. So 800 people is kind of the sweet spot. So this year we'll probably take about 800 people.

So we're really excited about dConstruct coming up. That site's being developed at the moment and will hopefully launch soon. We've got a great line-up of speakers, and we do a different theme each year. This year the theme is "designing for the social web".

We're redeveloping the Clearleft web site as well. Plus we've got loads of personal projects on our cards. Paul, our designer, has also got a personal project that he's working on part on company time, part on his own time, called WalRSS ("walrus") and basically it's an application for iPhone users which lets you turn any web site into an iPhone-capable RSS feed, or an iPhone-style web site. So that's something he's done with Simon Willison.

And of course primarily we're a user experience design agency, so we have loads of clients at the moment, doing some really interesting projects. We've just launched Edenbee [18], which is a really cool social networking site for people who are interested in reducing their carbon footprint. We've just launched a site called Rate My Area [19], which is a local search site. We're doing a couple of sites in the education space, and a whole bunch of other sites that I can't really talk about.

Plus we've just hired another user experience designer, and very soon we'll be looking to hire a new visual designer, and once they're in place then we need to bed down for a bit and get some of our projects done!

SP: Someone who calls themselves a "user experience designer" generally hasn't studied this topic at university. Is it only something that you can learn from doing, and from going to conferences like this one, and taking queues from the folks who are leading the way in the field? Where does someone who has no experience in user-centred design get started?

The problem for people entering the industry now is that all the people who are senior people in the industry started when the industry was in its infancy. So as technology and the way the industry works has progressed, we've progressed with it. So we're up to 8... 9... 10 years experience. When we started, we were crap, but the web was crap. So we could be crap, and the web could be crap, and we just kind of learned.

Now, the level of quality and the level of professionalism is so high, that when people leave university or college, and don't have that experience and aren't experts, as we weren't when we started, there's actually very little room for that. Because if you design crappy little web sites now, people aren't going to be interested because the level of professionalism has gone up. So I don't think there's a bit of a gulf -- a bit of problem -- in how you jump from having the actual desire to being able to actually being able to do the execution. Because things have moved on.

For example, when I built my first web site, it was a horrible little frame-based web site that looked crap. But every other web site looked crap too, so it didn't matter. Now, web sites look good, so when you do your first web site, it will probably look terrible in comparison.

I do think going to university can help. I think if you go to university, you probably want to study something that is quite specific. If you're interested in user experience, then doing an HCI course, or doing a cognitive psychology course or a social psychology course would be a great benefit.

But still, once you've done that, once you leave, you need to do several years of on-the-job training, and that's really difficult. Because you need to, at the same time, also be learning HTML and CSS and learning to do design, and doing freelance work. When you leave, if you've got a good portfolio to show, you need to go and work for an agency for a while that will take you in and nurture you and get you better enough for two or three years. In a small agency, you can learn those skills, then you can go to a bigger agency or do your own thing.

I think doing things like internships can be good. We've just started doing internships at Clearleft, actually having people in over some period, like a gap year over the summer between university courses. To come and actually get real experience in a web company, I think that kind of experience is hugely important for your development. Once the course is over, if you can say "Well, I've worked for three months at Clearleft, or whatever, and here are some real projects that I've worked on" then you're much likely to get employed.

I do think it's really hard though, and I think what you need to do is specialise. I think gone are the days of the generalist, because I think that, sadly, if you're a generalist, it does mean that you're doing everything not to the highest possible quality. Because there are so many specialists now, you've got to compete with those. So starting off and being really good at web standards, for example, if you want to get a job in industry, if you're really good at HTML and CSS then you will get a job, it doesn't take that long to learn HTML and CSS.

But learning the softer usability skills -- you can learn heuristics, you can learn basic principles. But it's not until you've actually sat down and built loads of sites and screwed them up that you can do that properly.
Which is why you tend to find, in my knowledge, very few companies like Clearleft or Adaptive Path or Happy Cog, who do what we do. Because you need that breadth of knowledge. And you need to have been doing it for 10 years and making loads of mistakes. So most web companies are either very generalist or very specialist that only do user testing, or only do user research ... or they do everything -- the programming, the building: a little bit of this and a little bit of that.

So we've definitely focussed on the user experience side of things, and pit ourselves as "web architects". So we do all the kind of stuff that an architect would do, only for a web site.

SP: Andy, thanks for your time! We appreciate it.

No worries Matt, thanks very much!

Image credit: Gary Barber [20]

[1] http://webdirections.org
[2] http://allinthehead.com
[3] http://adactio.com
[4] http://clagnut.com
[5] http://clearleft.com
[6] http://stopdesign.com
[7] http://mezzoblue.com
[8] http://cameronmoll.com
[9] http://dconstruct.org
[10] http://flashant.org
[11] http://silverbackapp.com
[12] http://new-bamboo.co.uk
[13] http://code.google.com/p/blueprintcss/
[14] http://webstandardsawards.com
[15] http://themaninblue.com
[16] http://csszengarden.com
[17] http://dconstruct.org
[18] http://edenbee.com
[19] http://ratemyarea.com
[20] http://www.flickr.com/photos/cannedtuna/2496220544/