Interview: Rory McKenna & BT's open platform
One of the unmissable themes of the Telco 2.0 event was the trend towards partnership and the need to build technology platforms that enable this.
We were pleased to have Rory McKenna present at the Technology Insiders' Workshop. His job title is as unfamiliar as that of the subject to most telco execs: Director of the Web 21C SDK. In English, his job is to create the tools and interfaces that let third party developers, big and small, access BT's core network and IT facilities. He also brings a refreshingly sceptical view of the world from outside of telecoms.
This reinforced BT's message from Paul Excell (pictured below right) on Day 1 of the event: partnerships and an open platform are key components of BT's business strategy.
Rory has kindly agreed to answer some questions on the initiative, but first some basic facts from his presentation:
- They are exposing seven sets of functionality: Voice calling, Conference calls, Messaging, Authentication, Location, Information about me (profile), and Contacts
- Their web services application program interfaces (APIs) work with all the major development environments (Microsoft, open source, etc.)
- They signed up over 1700 individual developer accounts, with UK based developers clearly predominating.
You can learn more here.
Q: What's the status of the project in terms of commercial deployment? (Development, testing, live, etc.)
You always ask the hard questions.... So from day one when we started on this path we have always strived to maintain the mantra of keep it simple, build the community, prove the concept and then add complexity when the market demands it. So we release end of April with a very simple commercial model based on buying credits to use the current set of services. We would be naïve to think all of the commercial intricacies of billing are covered for all possible scenarios but this does lead nicely into our approach to stay at the level of selling services and letting the application developer worry how to handle the commercial aspects for their specific offering. Then learn from the community what else they want from us.
So at the end of April we will have two service offerings, the Beta platform that we are currently running, free for very limited use and a simple banded offering for small, medium and heavy usage of the services. The important thing for us to prove is that there is a market out there that will pay for this, and as we move into a 2.0 world, whether it is branded Web,Voice,Telco it is the software developers that will have more influence on how these services progress.
Q: What's the commercial imperative for this activity? (e.g. Direct generation of revenue from API access, increased user adoption of basic access, product ideation, PR, exploratory R&D, or something else?)
Probably all of the above, leads back to the sector the applications are being built for. We've proven that it works across all customer segments, Enterprise to Consumer. So we are truly ubiquitous with this model. Imperative is a resonating word with me, especially in Telco land. Convergence is interchangeable with imperative now, in my own opinion, we need to maintain the innovative element and allow co-creation with as wide an audience as possible and we finally need to provide a sandpit for developers to test applications and show that BT can help them in that product creation cycle.
Q: What kind of IT and business development resources does it take to enable 3rd party access?
I'll let you accept questions on this, that's such an open question, Do you mean as a big company or someone wanting to use our services? Everyone can be a reseller now, especially in the 2.0 model, so are you a provider of one or more services, how complex are those services, what service wrap is required to support your service. We have tried to prove the one developer in the basement approach, building their application and servicing their single customer. If they scale we scale with them, if you can capture that market then you open up innovation to a massive audience.
Q: Who sponsors and champions your work from the commercial side of the BT organisation?
Al-Noor Ramji, BTs CIO, championed this from the beginning. It has been a low risk investment though in terms of opex and risk, other industries adopt this approach and a few of us have come from the investment banking angle where this open API implementation is de facto if you want to survive. What I am seeing now though is uptake both internally and externally to utilise the SDKs, it appeals to the ISVs(Independent Software Vendors) serving the SME market and we have a number of forward thinking business partners that are actively participating.
The business partners are bringing their experience of market evaluation and commercials which helps considerably in positioning BT as a SaaS provider.
Q: What internal barriers do you face? Are you finding strong acceptance or resistance to the idea of becoming an open platform?
Hard question number 2... Any big company will have issues on opening the platform, I think the big win here is that this is not Powerpoint, you can kick the tires and see what it means to you as a business. Security for sure was our biggest hurdle, once you conquer that milestone for services you offer, the opportunity for businesses to experiment in a secure environment removes the concerns. Everyone from the business side looks to IT to provide that ability to create and innovate, if you can't provide that in a corporate environment then you will be disenfranchised by leaner suppliers externally. But when it comes to implementing solutions on a global scale, BT is uniquely positioned to solve those issues.
Q: How much of a barrier is it to you that other telcos have different partner sign-up systems and APIs? Do operators need a common API set and partner business processes - even going as far as the kind of clearing houses that banks operate?
You mentioned my sceptical view on the telco world at the start. It isn't really sceptical, I want reuse for a market beyond the pure telco play. Making phones ring is actually quite difficult, what happens behind the SDK exposure is complicated but if you use the sdk you realise you don't have to be a domain specialist to make this stuff work.
So in answer to your question I don't know if there should be a clearing house, it makes sense to me, but it has to be geared towards execution whereas today we seem to be geared towards standards and design. We're not imposing a standard, we want multiple exposures that make sense to a consumer, that will drive some sort of commonality. I remember the early days of FIX and that is 9 years ago, it has evolved iteratively, we as telco need to learn as players and vendors that we need to collaborate to identify what is commodity and move to more commercially valuable services.
Q: What partner applications have surprised or interested you most?
We know the internal applications that are being developed against the services, they tend to combine a collaborative Web 2.0 concept with the Telco2.0 services. It is exciting because I believe we are demonstrating how you can make the web apps more customer focussed and making that contact from a web page out to the real world. Externally is a difficult call, we know what applications use which services, but in a white labelled world do you care, or do you offer an avenue for the application developer to request new services from you. We shall have to see how that evolves.
Q: How do you see the programme evolving over the next 12-18 months?
So we have primed the pumps. We will continue to target the development community, we are at Java One and TechEd Orlando in the next 3 months. We will be providing more services that appeal to the community. We have a number of internal engagements that are using the SDK services and this is gaining apace. In summary we will continue to innovate, prove revenue and keep one step ahead (which is difficult).
Q: What are the key lessons for other operators following the same path?
Nice one, finish with hard question number 3...
Listening to the presentations at the event I reckon the lesson is that the IMS/SDP platform is a Telco focussed play, it has to happen but it is just layers of abstraction in my eyes though. The SDK we are providing is at a higher level of abstraction. Maybe standards will evolve, lets let the market decide what that is. If all you want to do is send a message then surely it should be sms.send("message", "recipient"), a call should be makecall("callee", "caller"). Who knows, but provide an open platform that allows people to do that cost effectively, securely and with guarantees you might actually appeal to a wider audience.
Many thanks to Rory for taking the time to share the lessons of BT's open platform initiative with Telco 2.0 blog readers.