It's that time of year again when the next version of iOS is getting close, and speculation about what Apple might do next starts to hot up. Of course, it's that much more intense this time around, because it'll be the first sight we get of what Jonny Ive has done with his new role as head of software user experience (as far as I know, the first time in his life he's had any role in deciding what goes on a screen as opposed to what goes around it). What's he changed?
The obvious thing to talk about here is 'skeuomorphism' - the recent Apple addiction to making pixels look like leather or wood or bits of torn paper. People in the tech community have got pretty upset about this in the last year, with the dialogue reminding me very much of the way architects a generation ago argued about 'truth to materials', and used to claim that designing a neo-classical structure was somehow 'immoral', or at least dishonest. We've even had Louie Mantia mirroring John Portman in claiming that if it looks 'nice' must be OK.
My own view is pretty undogmatic: I hate kitsch as much as the next man but I also think interfaces should be easy to work out; real-world metaphors help, in moderation. This photo, I think, should guide all such discussions: normal people don't want to have to work ANYTHING out. So make the calculator icon look like a damn calculator.
Closely related to this, of course, is the hunger for the 'new'. After 6 years, some people are just bored with iOS. Rather more people, though, will get very upset if things they've learned how to use change at all. So getting rid of the leather may also be a way to address the former without upsetting the latter.
Frankly, though, this is all a bit of a sideshow. I don't really care if Jonny redesigns iMessage with marbled paper and pretty little fake ink blots so long as it delivers messages across multiple devices in a predictable way, which right now it doesn't.
iMessage, to me, points towards the broader issue facing iOS: growing pains. When it was unveiled in 2007 it had a radically clear and structured vision of how everything was going to work. The trouble is, over the past 6 years a bunch of things have been added. Some of them work well on their own terms but aren't coherently integrated. Others don't work well at all.
My favourite example of where things have gone a little adrift is shown in the two screens below. If you're sent an image as an MMS or iMessage, you get one set of options; in the built-in Photos app you get different options.
This illustrates two different issues. On one level, this is just inconsistent execution. But it is also a symptom of an underlying structural problem: iOS lacks a systematic way to think about storing files and using them in different apps, and it lacks a systematic way to think about social. This, of course, reflects history:
- iOS launched before social took off
- It launched with no third party apps
- It made a deliberate (and entirely correct) decision to dump the file system as a source of complexity and confusion
Apple has added apps, bolted on social and bodged on ways to give apps access to files, sort of, but these changes have sometimes come in inconsistent and unsystematic ways. iCloud solves some problems but not others (and has problems of its own). Facebook and Twitter get one-off, hard-coded integrations (as do a couple of Chinese services), but that can only be a temporary solution. Meanwhile fairly normal use cases aren't addressed at all without third-party apps: how should I tell my mother to put a 50 meg PDF on her iPhone for a trip next week? And why did Apple make a beautiful iPad version of iMovie but provide no way at all to get videos shot on your iPhone onto your iPad for you to edit?
In other words, the clear consistent vision from 2007 has had things added on piecemeal, and the core conceptional model of how things work doesn't always hold true any more. A systematic rethink is needed. In the past there's been too much fake leather, too many clever animations, and not alway enough thought given to the user's flow.
On one level, this is housekeeping: tidying up the accumulated additions to iOS since launch and incorporating them into a couple of coherent new UX models. However, it's also rather more important than that. I would hope that Apple is thinking not so much about how to tidy up what it has (and how to plot out out a three or four year roadmap of discrete new features), and rather more about what a competitive mobile OS will need to be in five years time.
I suspect a key part of that, especially for Apple, is how apps and social services on the phone talk to each other and talk to the user, and indeed talk to Siri. Like Google Now, Siri aims to integrate many different services and pieces of data from different places into a single stream. Right now both Siri and Now are flawed and incomplete, though in different ways, but both of them aim to develop a different model of the way you might use your phone - beyond text web search and individual siloed apps, and towards an integrated, helpful interface that gives you what you want. The pseudonymous Counternotions gives a detailed explanation of how this might work here. Key quote:
Being an integral part of iOS and having programmatic access to third party applications on demand, Siri is fully capable of executing a fictional request like:
"Transfer money to purchase two tickets, move receipt to Passbook, alert in own calendar, email wife, and update shared calendar, then text baby sitter to book her, and remind me later."
by translating it into a transactional chain, with bundled and 3rd party apps and services acting upon verbs and nouns
This, of course, brings us back to the beginning: iOS lacks a systematic way to think about how apps talk to each other. Google can relying on collecting and collating all the data itself - Apple needs to lean on third-parties. The obvious short-term solution would be to give them something like Android 'intents', or a shared file store or something. But that, I think, misses the point. The problems we see with apps, and social, and files on iOS are things that evolved over the past 6 years: Apple now needs to think about laying a conceptual framework for how they'll work in the next 5 years.