Blogs

RedMonk

Skip to content

Mobile First and Last: End to End Computing and The Age of Context

Mobile First and Last: End to End Computing and the Age of Context from James Governor

A few weeks ago I flew out to Las Vegas for IBM’s Impact 2014 conference. I am working on a write up of Impact now, but in the meantime I wanted to flag this presentation I gave during the show, alongside Michael Perera Vice President, CICS, TPF & WebSphere System z. Some of you might wonder what the mainframe has to do with mobile apps, but in the end to end context transactions matter, and a lot of transactions still run on the mainframe.*

The theme of my talk is that Mobile First is a real phenomenon, and drives a number of changes that both IT and the business need to accomodate. Mobile is a catalyst for wholesale changes to development methodologies (hello agile and continuous integration, goodbye Waterfall and good riddance!), which then has a knock on effect on middleware platform choices. Customers need easy access to their back end systems of record.

Another key theme is that change in the mobile space is so rapid, that you really need to choose toolsets and frameworks that track platforms so customers don’t need to. For every developer that tells you HTML5 will dominate, another couple of business users insist that all mobile app dev should be native.

Another key theme of my talk was the Scoble and Shel’s Age of Context is a brilliant framework for understand the types of app we can now build to take advantage of all the telemetry in modern phones, mobile and wearable computers. Facial recognition for example is an extremely useful function, but today we’re increasingly going even further. Recognising a face is one thing, but now we have APIs that can identify the emotion of the user. That’s a big deal. Especially if your goal is happy customers ;-)

So mobile changes everything in a way that the last revolutions really didn’t. The user is now at the center of the experience, which means investing in design and user experience skills.

*For mainframe customers it’s also worth noting a new program and toolset IBM is currently building, whereby customers don’t need to pay the same amount for capacity driven by mobile usage. Kind of like IBM’s ZiiP and ZaaP offload engines, the idea is that SOA shouldn’t penalise the customer paying per MIP, especially when in many cases the mobile apps are effectively read only.

IBM is a client, and paid for my travel and expenses to Impact.

Categories: Uncategorized.

Comment Feed

No Responses (yet)



Some HTML is OK

or, reply to this post via trackback.