As the cost of computation and connectivity plummets, manufacturers are going to put ‘interactivity’ into every device. Some of this will be trivial: my power adaptor knows its charging history. Some of it will be control related: my television will be grand central for my smart home. But at its heart, we’ll be swimming in world where every device will have ‘an app.’
What does this mean for us as designers and curators of experiences? This talk will discuss how the principles of the open web must apply not only to prototocols but to hardware as well. How can we make sure the menagerie of devices has a chance for working together?
Although we have a complex situation with devices (mobiles), the cloud, apps, companies that run them all… We still think about human<->device and don’t go beyond that.
There’s a relationship triangle between user – device – cloud.
We need to remember “smart devices” can be barely smart. People laugh about the idea of a smart toaster… but what if every device broadcast a URL? Suddenly you have a link to all the information about that make and model of toaster, you have access to support for that toaster… We need to realised how valuable simple things are.
If you go up from the $1 chip to broadcast a URL to $5 for control chips, you can start doing fun things like set your phone to play a fun sound when the toast is ready (“it can play spongbob squarepants for my niece!”).
This gives several types of smart device:
|Swarm||Direct swarm: Sensor based things like carpark space indictators, smart lights, iBeacon||Background swarm: all the devices working together – smart light bulbs that do different things at different times of the day.|
|Device||Direct device: everything we know… phones, consoles…||Background device: Lavatron, Nest – devices that sit on your door handle and let you in without a key. Lavatron does this with bluetooth proximity. Nest only becomes a background device over time.|
The main idea here is to have a vocabulary to discuss the modality of devices and how they move between them. There will be user flows that cross over these modes over time or even during a single interaction if it’s a complex enough process.
Interesting questions arise with devices that have no screen – not to mention the complexity that may occur once a single app or page can be used on multiple screens at once… Also ponder the limit of how many apps we can install, how many single-purpose devices we can manage, how many things we can even do everything manually (or with user input to install an app).
Why mobile apps must die? I just wanted another alternative. People can’t be bothered any more – you put a sign in your store “we’re in the app store!” – people just can’t be bothered.
Fundamental rule of UX design: Value > Pain
There has to be more value than pain. You can increase value or decrease pain.
“As a designer, my life is pain. I am a pain reducer.”
Curiously if pain goes down fast enough, value can come down too and yet you still have a successful product. When Google was only on the desktop you used it for important research, now it’s in your pocket you use it for pub trivia arguments.
We’ve seen this before. Yahoo’s directory model didn’t scale, Google killed it. The mobile app model doesn’t scale, something will kill it. “How many people have removed old apps from their phone? Trick question! We all do. We are gardening our phones!”
The web is amazing… but we have attached a command-line interface to the top of it. We type stuff into it. We don’t take advantage of all the sensors in the phone the way apps are doing. The browser should be handling this stuff!
Just In Time (JIT) ecosystem: lots of barely-smart devices broadcasting simple information (a URL) and other smarter devices doing stuff with that.
“The cloud is a complete myth. There are cloudS and they are cranky and don’t like each other.”
Every company sets up their own cloud for every device. But users want their data to be in one place, especially data like personal health data.
There’s not going to be a single killer device, the killer app is going to be a meta app: reasoning over combined data.
Some companies are thinking about this – they’re trying to open up services to collate data. (Spark, PersonalCloud, OwnCloud, Cosm)
People say this can never happen. But if we don’t know what we want we can never move towards it.
Some examples… we forget it took ten years between Netscape and Gmail. Things take time. We forget how crap the web used to be! We need to think a little longer term.
We forget that in the very early days, the walled gardens like AOL were providing a better experience – until the open web blew them away. If we are seeing proprietary devices now, it’s probably just the early days…
Two types of idea:
Malcolm McLean invented the shipping container, which vastly reduced costs and changed shipping forever – then gave it away to ISO, he did end up making a lot of money but not by holding on to a great idea… it was a rising tide that lifted all boats.
If we don’t know what we want, that’s what we’ll get. We need to have a vision for the future and work towards it.