Why Closed Works: Moving Past Steampunk Thinking About The Future Of Computing
I have been arguing for some time (see Tim O’Reilly on Web 3.0) that the metaphors of computing user experience (or UX, as we tend to say these days) are holding us back from new ways of structuring our interaction through computers.
First of all, way back when we were debugging fire with oscilloscopes, and inadvertantly invented personal computers, no one considered that the primary dynamic of computing would be social. We invented the web to chat among ourselves, to write for each other, and share photos, and argue. We invented a web structured in such a way that all its roads lead back to us.
But, when we were cooking up Unix, DOS, Windows, and Mac OS, we thought people would be ‘working’. And working meant fooling around with documents (files), since that what we thought was the logical atom of information: files.
But it’s not.
In fact, the layout of information on my hard drive bears little resemblance to what I do all day, since the overwhelming amount of my time is taken up in communication with others, buried in email, Twitter, blogs, and various other coordinative tools.
Sometimes I am forced to know where a bit of information is embedded in some document — against my will — and to the extent that is possible, I try not to know such things.
More importantly, I am not alone: more people all the time are using their computers as social interaction devices, and files, folders, and the desktop are just a launch pad for that.
I believe that we are seeing the first glimpses of a break with the past, when we will not care if and where information lives in files on some physical level, either in the cloud or on a local hard drive.
If I am sharing a video with a bunch of friends does it matter if it is located in one place on the web, or distributed in twenty servers and streamed on demand, or sitting on my local hard drive and streamed to them? I don’t care.
But the thing that is blocking us from moving forward, to a better user experience centered on social interaction and not physical data, are the existing metaphors of OS’s. Since we are living in a world of general purpose computers running Unix, Mac OS, and Windows — and we need to have them interoperate — we seem stuck in the 90’s.
The only way forward is to build a new user experience on top of the physical hardware and software that form a platform for it, and conceal its nasty details from us.
This is one aspect of the genius of the iPhone and iPad generation of devices: we don’t need to know about the files and folders. We don’t need a desktop with data bundles lying in piles.
Instead, we have a suite of applications that serve up myriad useful metaphors. Photo apps deal with photos, not files. Expense report apps deal with expenses. Twitter apps reveal the twitter stream. And so on.
To have a break with the past, and to make the past a platform, we have to push it under and not pretend that all its constructs are desirable. We need to push files, folders and the notion of a desktop under the surface of a better user experience, and keep it under. Let a new generation of user experience shield us from that drudgery and detail.
This break with the past is made faster and less difficult if the new system is closed: if we can’t escape its premises, and start rooting around in the basement, moving files and putzing around. Apps have to be limited in what they can do at the physical level, because the whole edifice can come down if apps can punch a whole in the foundations.
I expect that Apple won’t be the only one to devise new OS’s that head in this direction, and perhaps Chrome OS or a new Mac OS kernel might become the predominant infrastructure for a presentation layers like iPad and iPhone. They will become more powerful, faster, and will blossom in new directions, like augmented reality.
I know there are a lot of techies out there that idealize a hypothetical future that’s a kind of steampunk mixture: amazing new functionality — like full augmented reality, digital currencies, and crowdsourced sousveillance — but somehow all strapped to the back of operating environments they are comfortable with, like Linux and the desktop-oriented user experiences of the ’90s. Sorry, folks. No steam-powered rockets to Mars in this episode.
To move ahead, something has to give. To get more abstract, the foundation has to be more closed. And that’s one of the reasons that the iPad/iPhone devices represent a breakthrough.
