Expert mode

Sunday 28th Feb 2010

The ‘one size fits all’ approach to operating systems is an outdated idea born of economic limitations and design inexperience. It needs to go. People have wildly different capabilities and interests, and to try and design a user interface that will work well for everyone is an idea destined to failure.

Computer interfaces have come an awfully long way in the past 30 years. In fact, most of the time when we use a computer we don’t even realise we’re doing so — it’s become a cliché gotcha moment to ask a member of the public how many computers they use in a day. Often they’d mention the one at work or maybe the one they use at home to check their email, at which point our gleeful smug-faced twat of a presenter would trounce them with the fact that computers are everywhere, even in that triple-decaf skinny sub mocha dry venti soy latte they’re holding! Aha! Ahaa! Gotcha!

Tedious all round. But I’ll save my ranty tangent for another post. Computers are all over the place (though not yet in coffee) and we use them often without thinking. That’s because most of them have simple task-based user interfaces; A car may have a computer to run fuel management, aircon, ABS, amazing oh-god-is-that-true-AI traction control systems and more, but since our UI of the thing is the same as we’re used to — wheel, pedals, a couple of switches maybe — that even though it’s one of the most amazing computers we’ll ever use we just don’t notice it or realise it’s there.


So what about the traditional idea of a computer, the thing with the screen and the keyboard that lets you get to Facebook, online shopping and porn? That’s most definitely a computer, but the UI is anything but natural or transparent. Windows, Mac OS, Linux, whatever you choose to use, they all have roughly the same kind of interface. There is (supposedly) a desktop metaphor, a bizarre storage method — lumps of data called files and a hierarchical, often recursive, taxonomic system that apparently lives only for depth-first searches and goes by the folksy 1950s-office name of folders. Add to that the disconnected method of interacting with and changing these systems, a mouse, which translates forward and backward motion to an up-down axis on the screen, and a keyboard which, rather than for just typing words, also allows for a cryptic range of modal commands requiring the user to essentially play chords of keys to get the computer to do something with, usually, no visual feedback that it’s done anything at all. These chords are also not discoverable, meaning that there’s nothing in the interface that suggests they exist, and frankly, how would anyone guess? Command-shift-R? What? I press three keys?

Still, there are a fair few people who’ve been using these computers a while, and all these systems seem, if not natural, then at least familiar. I’m one of these people. I’ve used computers of various sorts since I was about 14 and despite all the screeching bores going on about how one system is so much better than the other, there’s not much to choose between them. In fact, most people find them equally difficult, unpleasant and confusing; they’re made to feel pretty fucking stupid when trying to use a traditional computer to do even the most basic things. This is not a good thing.

Non-expert users aren’t always stupid

I’ve always worked for companies that make websites or online tools that the ‘general population’ are going to be using — people who aren’t trained in the ways of traditional computers, and frequently have little interest in learning more about them. These are people who have other things going on in their lives and (one assumes) have a high degree of competency in whatever it is they do. Indeed, for some of our projects, we knew for damn sure that the target audience were highly intelligent and capable people, leaders in their fields and often internationally reknowned. Of course, we also knew that a hell of a lot of our users were stupid as hell and probably shouldn’t be trusted to tie their own shoelaces, let alone operate a cooker or drive a car. But still.

Clever or not, almost none of these people were very good when it came to computers. Designing an online application for something is a big enough task on its own, but when you’ve got the computer itself making things confusing and perhaps even preventing the user getting anywhere near your work you’ve got frustration enough to make you spit feathers. We often talked of an expert mode for computers. This was an idealised system by which the computer would start with a super basic and simplified UI: no confusing functionality, distracting icons, technical jargon or anything else, something perfect for task-orientated (or technically disinterested) users, which could open up like some glittering flower to reveal all the technical gubbins when expert mode was enabled.

It’s never happened. We’ve never had a computer like this. Thinking about it, it’s not surprising why. It’s hard enough designing one UI for a computer, but to design two? Two interfaces that have to interrelate and give access to the same systems and controls? Not only that, but one where errors, feedback and responses from the system and the broader networks have to be handled intelligently and appropriately. Also, what happens if your hapless user accidentally activates expert mode? Suddenly their world would become hideously complicated. It’s not impossible, but there’s not much of an economic incentive to make such a thing.

Security should be the easiest part

Besides, if you’re going to design a computer, you need to consider some important technical issues first, a lot of which seem to have been missed over the years. The internet is infested with virus laden PCs, connected up to and controlled by botnets organised and run by criminal gangs. These PCs don’t belong to criminals, they belong to ordinary people who don’t know there’s something wrong. If the computer is actually behaving strangely, they have no idea that it is, because, to them, computers always behave strangely — in fact, their computer could have been infested with viruses from the first moment they hooked it up to the internet with that modem PC World mis-sold them. The one with the default username and password and a stupidly opaque UI for getting into the damn thing to change it, assuming you even know there’s such a thing as a username and password and that such a thing is even important. Yes, that one. My parents have one of those. It’s shit.

An analogy

If you put an average person in a kitchen stocked with food, jars of seasoning, cooking appliances and all that, but nothing was labelled and some of the jars contained poison and a couple of the appliances could explode if you didn’t turn them on just right, would you expect a good result? No. You’d expect something bad to happen. Yet this is what we’re doing with computers. You’re giving someone access to an information device that they’re going to want to use for online banking, shopping, communication, socialising and all that, and yet quite easily it could expose all their personal information to complete strangers and make them unwitting accomplices to massive international criminal networks.

No user-serviceable parts inside

So where am I going with all this? Well, mainly I’m just ranting that computers-are-a-bit-shit-really, but a lot of these issues have come to the fore because of the iPad. OK, it’s still early days, but I think that the closed, all-in-one device with curated apps is exactly the kind of computer many people need. There was an awful lot of screeching and howling about how Apple controls what apps go onto the device, and how it’s all DRM, and this will bring about the end of technical innovation and that the youth of today will somehow be prevented from learning programming. Well that’s only so much bullshit. If it gets people online in a safe, technically mediated manner, this is a good thing. The saga with ReadWriteWeb and all the Facebook users turning up and typing in their Facebook credentials because it was the first result on Google for a search for “Facebook login” just shows that the open, normal, technically-literate assumptions of computers and the internet don’t work for most people. Just give them a Facebook App they download from the App Store — it is Facebook, it’s not going to be another site they’ve found by accident, it is safe, because it’s been vetted, and while it won’t get them Farmville (yet) it’ll get them poking-that and liking-this or whatever it is they do with a minimum of fuss, because that’s what they want to do, and they should be able to do it without getting phished.

A philosophical fear

The fears that kids won’t learn about programming or computers because they’ve only got a closed-system computer at home just aren’t valid. I grew up with no computer at home, and somehow managed it. How? Well, there’s this thing called school, and there were all sorts of different computers there, from crapola RM Nimbus machines to the latest Acorn Archimedes, which, I must point out, was in the art department and was pretty much instrumental in me becoming a digital designer. This was in the 1980s in a smalltown school in a rural area in the north of England. For any country where iPads and iPhones are popular I hardly imagine the schools are going to have fewer computers than the one I went to.

I say we need a bunch of different types of closed-system computers for people to just use and a whole load of open-system computers that people can develop on and play with. We’ve got the second thing, we just need to add the first. The iPad is the start of that, but don’t expect Apple to be the only player.