I only made the Friday of this event (there were sessions on the Thursday too) and this was my first time at UX Live.
I can see why this conference is a fairly big deal for people interested in tech product leadership and UX - the speakers here were pretty big hitters, with speakers from Adobe, Netflix, RBS and Amazon Alexa to name but a few.
I'll outline my favourite couple of talks from the day below.
Jenny Gove, UX Research Lead @ Google
Jenny's talk was quite pertinent to me as I've recently been working on a couple of projects which have native apps in their scope.
While the breadth of functionality available when creating standard web tech is certainly moving closer to that offered by native apps, as Jenny pointed out - there is definitely still a significant gap.
The ability for browser based web apps to tap into more of the inherent functionality of mobile devices (push notifications, location services etc) has been a hot topic for a long time. The avenue that Google appear to be focusing on is PWAs (Progressive Web Apps) which essentially occupy the mid ground between basic browser applications and fully functional native applications.
On the surface it seems to make so much sense - getting the commitment from a customer to download a native application to their device is a pain point for many companies. The idea that they could offer the same kind of UX to users directly from a web URL without the need for any 'app store' based hurdles for the user to jump is clearly quite appealing.
From the user's perspective too, the idea that they would be able to simply visit a site and immediately get the benefits of a web application being able to access and use the various services and sensors built into their device could lead to a richer overall experience with the web app. They could even use a 'save to home' option to generate an icon shortcut on the device, and depending on how the PWA has been structured, there could potentially be some offline capability available to users - essentially a direct challenge to one of the key USPs which native apps currently have the monopoly on.
Clearly there is some way to go - the power of PWAs is based on building web applications to a particular recipe - lots of caching and service workers, unsurprisingly - which (to me, at any rate) means that retrofitting the functionality to existing sites is probably a no-go. Greenfield projects and major site overhauls would be better candidates for those who intend to adopt the approach.
As for the native app projects I'm working on at the moment, unfortunately I don't see either as suitable candidates for this right now. But this is technology I think we'll be seeing a lot more in future, so I'll definitely be investing some of my own time in putting together a quick proof-of-concept PWA to satisfy my own curiosity - depending on how interesting it is/how well it goes, it may even warrant a blog post of its own.
Tom Castle, Director of AI @ RBS
It's not often you get to hear about what what huge entities like RBS are doing with AI directly from the horse's mouth so to speak, so I really appreciated Tom's talk, and in particular his frankness about the limitations and constraints inherent to the current generation of AI services available to developers.
I really agreed with one of the points that om made, which was around the fact that people seem to have a surprising amount of tolerance for the fallibility of AI services at the moment. It's something that has baffled me for a while now - when we compare the relatively precise and exact user experience that we can get from a well designed website or native application, it strikes me as interesting when I watch people happily struggle to get their 'smart' home assistants to understand their routine, or when they simply shrug it off when the advertising suggestions provided by services which tout their 'AI-driven' credentials turn out to be laughably off-target. It kind of amazes me a bit that people don't seem to care much about this relatively low rate of success, but there's a big part of me that's happy that people seem to be showing so much patience - ultimately, making better AI depends upon vast numbers of people using it and providing the behavioural data which will allow for incremental enhancements to the underlying algorithms and calculations which make the technology work.
There was also some really important, and topical content in the talk about ensuring that AI is transparent. We're already at the stage where some AI-backed services can fol users into thinking that they are real people (think chat bots embedded into websites, or digital assistants which can make telephone bookings on your behalf) so it's becoming very important that we start ensuring that we're clear about where AI is being used ion the products that we build. To do otherwise is simply to erode trust between the product and the user.