Software Defined LTE 5G Radio demo at the Ubuntu booth
At an event the size of Mobile World Congress, the last day is filled with the hurried rush to get everything in that you might have missed. The major topics of the day were startups and upstarts, and upcoming mobile protocols standards and implementations. There were also going away parties, like the dance club atmosphere in Intel’s booth in the afternoon.
It seems that ‘born on the web’ companies are almost old news these days. Born-mobile upstarts are demonstrating just how compelling an app-based delivery model can be compared to delivering the same service or information on a web site. Kakao Talk, seemingly just a media and social mobile chat app, is expanding their product portfolio with things like Kakao Taxi, a mobile-only application for getting a taxi in Korea.
If you’re on the web, but don’t have an app, the pressure is on to get one. Just Eat, the largest online food home delivery service, launched in 2001 but didn’t go mobile until 2012. To their CEO’s own admission, they should have sooner. He said that Just Eat app users have greater loyalty and a higher retention rate. Strategically, they want to get more customers using the app, as they believe it just delivers a better overall experience.
Mobile apps also decrease time to response for end users. In an always-connected economy, response times are becoming more important. Airbnb’s VP of Engineering, Mike Curtis, said that for every hour longer that it takes for a host to respond to a guest, there’s a 2% less chance the the guest will end up staying with the host.
As a consumer of mobile-first and mobile-only services, I certainly see and appreciate the benefits of being a mobile-only business. But, as a technologist, thinking long-term, I’d want to make sure I don’t make technology choices that limit my market opportunities.
It’s obviously necessary to deliver your services on the most popular current platforms, but it isn’t so obvious that an organization should design their systems and architectures to be able to pivot to another platform in short order. In the span of the last 10 years we’ve gone from an almost all-Windows end user consumer base, to all-you-need-is-a-browser, to mobile-only app-based delivery models.
What we’re all witnessing is that no end user platform, no matter how popular and seemingly powerful, is unassailable. The market can change in a matter of weeks or months. Your consumer delivery model of today could actually be irrelevant in weeks or months.
Forward looking organizations will design their back-end systems to be agile. Architectures that can survive the pace of change will use technologies like OpenStack for its ability to use plugins and deliver virtualised technologies, like LXD machine containers and VNFs. That way they can easily swap in and out new technologies based on software, not hardware. They’ll use service modeling and application abstraction tools like Juju to model and deliver any emerging new application, and not be constrained by their legacy architectural models.
The discussion around next-gen 5G and unlicensed wireless standards was a completely different theme, but followed the same overall pattern as the above. MulteFire was quite a hot topic on the panel of experts, and it’s barely existed as a proposed technology for a few months. LTE-U, the idea of using LTE speed and technology in unlicensed spectrum (your Wi-Fi router, for example, is considered unlicensed, your mobile phone provider use licensed spectrum), also got lots of interest.
What everyone wanted to know about all of the technologies discussed, though, were: when will I be able to use it, and will it be compatible; will it not conflict with everything else I’m already using?
The Wi-Fi Alliance CEO, Edgar Figueroa, emphasized the need to maintain compatibility of all Wi-Fi devices, whilst still delivering the absolute latest in data speeds and application capabilities. If you work in technology, imagine being asked to use a PS/2 style keyboard with your tablet or maybe, for example, working in finance and trading stocks on NASDAQ with pink slips.
From consumer delivery platforms, to how the data gets there, the pace of change is rapid and seemingly unpredictable. That’s why Canonical focuses on technologies that enable rapid scaling, dynamic interoperability and compatibility with interchangeable standards. We actually run our own systems that way. It’s been so successful for ourselves that we even sell it as a solution called BootStack.
MWC 2016 is finally over. Our entire team was totally invested. We had an absolutely fantastic week. Mark Shuttleworth was brimming with excitement and our team echoed the sentiment. It was very cool to feel relevant, to be on topic, to have laser focus on exactly the kind of solutions that organizations, one-by-one, asked for all week long. From Ubuntu OpenStack, to NFVi, to converged mobile platforms, and an IoT operating system for the modern era. It was just a spectacular show.
Intel’s club-like scene at the end of MWC 2016
I’ve made a lot of new friends this week. That was awesome. I hope to see you all again next year, and I hope to meet more people next year. We’ll also be in Austin, Texas, USA for OpenStack Summit at the end of April, so if you weren’t at MWC 2016, maybe we’ll see you there!