Skip to main content

This isn’t your father’s “cable” industry anymore. In fact, the word cable is no longer spelled out in the name of the NCTA, which last week rebranded as NCTA-The Internet & Television Association.

But that also means that what used to be true about “cable” operators is no longer true today. Rather than focusing on one new product or service at a time, these new-class service operators have to juggle twelve balls at once in order to keep up with consumer demands and the competition.

To get a fix on what’s important as many people in the tech- and engineering-facing part of the industry converge on Philadelphia for this week’s SCTE/ISBE Cable-Tec Expo, Multichannel News technology editor Jeff Baumgartner and Leslie Ellis, independent analyst and regular MCN contributor, caught up with some of the top engineering minds in the industry to discuss their priorities and to drill down into hot areas such as DOCSIS 3.1, wireless and mobile broadband, next-generation video, network virtualization and the shift to DevOps models.

What follows is an edited, transcript compiled from separate conversations with Kevin Hart, executive vice president and chief technology officer of Cox Communications; Balan Nair, EVP and CTO of Liberty Global; and Sree Kotay, who was recently promoted to CTO and EVP of Comcast Cable.

Related: The Long Goodbye


MCN:Let’s level-set the conversation with your technological priorities. What are the top one or two things on your to-do list for Q4 and the rest of the year?

Sree Kotay: Really? Only one or two? One is, we want to continue to lean in to building elegant, differentiated products. We’ve been focused on building better products and services, and that focus has had a profound positive effect. We’ve done a lot of that for TV, and want to do it for other services, across the board. So you’ll see us putting in foundations and building blocks this year, for some big things we’ll do next year.

Two is the customer experience, with a real focus on simplification and self-service. It’s an intense focus on how we engage with our customers — starting from the acquisition funnel, throughout the engagement — whether it’s self-installs or enabling people to resolve issues themselves. It’s an effort we’re internally calling “All Digital.” It’s about giving our customers a powerful blend of self-service, transparency and visibility.

Balan Nair: Right now, it’s mostly operational. If you really want to be a few steps ahead — and all trends indicate this — you have to change. Change is hard, but it’s mandatory. Mark Twain said that the only person who likes change is a wet baby. He’s right.

We’re in the middle of a company-wide transformation that we call “Liberty Go.” It’s about putting the customer front-of-mind, completely. Customer first, front of mind, sides of mind, back of mind. For real — to the point of making the Net Promoter Score (NPS) metric part of performance bonuses.

Kevin Hart: First and foremost is our network transformation plan. We’re putting together a 10-year network transformation plan with the hallmarks being around DOCSIS 3.1, Node+0 architectures — taking fiber deep — and then eventually leveraging Full Duplex [DOCSIS].

No. 2 is our strategy around trying to own the home, better enable the home, better enable the customer experience. Things like our new WiFi offering that we’re launching called Panoramic WiFi. It’s also the work that we’re doing on new Contour, our new video experience, and we’re continuing to make great progress with Cox Homelife, our home security/automation platform.

MCN:Can you elaborate on the foundations and building blocks in the works?

Kotay: On the technology side, it’s obviously all the DOCSIS 3.1 work, looking at broadband and our XB6 [Gigabit gateway], which comes out of a lot of work on RDK-B, where the “B” stands for “broadband.” It’s radically improving visibility as a functionality to move up the value chain. As we think about connected services, that’s one big element. In particular, we’re focused on digital relationships, and evolving to serve the people in the house, not only the household address.


MCN:WiFi: The hype machine around “smarter WiFi” and “intelligent WiFi” is pretty loud. What factors go into building a competitively superior WiFi service offering?

Kotay: If you look up Maslow’s hierarchy of needs, with self-actualization at the top, and air, water, food and shelter at the bottom — there seems to be a new element in that bottom slice, and it’s WiFi. And the thing is, when people say “WiFi,” they usually mean “end-to-end connectivity:” My device is online, it’s working and I can get to the services I want.

A lot of previous approaches have been focused on slivers of that problem. We fell into that category — well, you’re connected with the CMTS, so you should be good! But the reality is, you have to look at the whole chain. It’s not just the access layer, or the in-home things like speed and reach. It’s the data centers, the back-end services, the care and the end-to-end-ness of it.

For instance, if you have a gaming console connected over WiFi to my gateway, I should know it’s a gaming console, what kind of console it is and whether there are any issues impacting its connectivity, whether on our end or somewhere else in the chain. The Internet is inherently a best-effort service. So we have to think about how to make it more than that — more carrier-grade, in a sense. Like broadcast video quality, in that it really, really works, all the time. It’s not just about any sliver of that chain. It’s the whole end to end. Our industry is uniquely positioned to tackle that challenge, but it takes a significant amount of focus and attention.

So the lesson to us was, let’s stop thinking about it as just WiFi. “WiFi” has to mean all of it.

Nair: People say content is king. My boss likes to say that connectivity is king, and content is Kong. WiFi is obviously a big part of “connectivity.” The world is moving away from single-point WiFi access. We’ll have to look at a mesh network in the home, to achieve a very high quality of service. Step 1, we improve quality at the chip level; 2, improve the antenna(s). Next, embed the software that enables us to optimize things like beam-forming. All of that enables us to do a whole-home, holistic WiFi view.

MCN:Kevin, tell us a bit more about this PanoramicWiFi offering.

Hart: Eighty percent to 90% of the devices connected in the home are enabled by WiFi by our customer base. Enabling a better in-home WiFi experience is top of our list. Panoramic WiFi will have a tagline of, “It’s wall-to-wall fast.”

We’ll be implementing some of the best-in-class WiFi gateway routers and will be leveraging over time the RDK-B component within those routers. We’ll be installing extenders, depending on the size of the room and the [performance] of the signal quality, and focus on ease-of-use.

We’re also going to be accompanying that with some additional focus from our techs in terms of the setup, giving them more set-up time to ensure quality of service and signal strength

We’re launching that in the fourth quarter, in our California market, and then we’ll be taking it more broadly in 2017.


MCN:Multiple choice, with LTE-U and its impact on Wi-Fi capacity: 1.) Battle stations!; 2.) If only that were my biggest wireless concern! 3.) Don’t worry, MuLTEFire will fix everything.

Nair: We have to look at this from two sides — as a WiFi provider, and as a mobile business operator. Which means my answer is a little bit of all three. It’s “battle stations,” because if connectivity is king and it goes away, that’s a real problem. It’s certainly not our only wireless concern. That said, I’m confident that our collective work with MuLTEFire and other standards bodies will help. [Liberty Global and CableLabs announced last week they have joined the MuLTEFire Alliance.]

Hart: I think it’s No. 2. I think there has been a lot of good work with the different organizations, including CableLabs and the Wi-Fi Alliance, to sort out some of the interoperability and co-existing requirements of LTE-U. I’m pretty sure that’s going to work itself out over time.

I think trying to drive that in-home WiFi experience and also continue to build out our WiFi footprint where it makes sense are some of the things that are at the top of our list. And we’re also keeping an eye on some of the advances of 5G from a wireless standpoint. It’s both an opportunity for us and also a potential threat.

MCN:Gigs and competition: Two years ago, we asked (some of you, anyway) about Google Fiber’s threat to expand to a bunch more metros. Now, Google Fiber is being dragged through the mud by the Silicon Valley press, suggesting trouble in Google Fiber-ville. How does (or doesn’t) that impact your Gigabit deployment and/or “fiber deeper” strategies?

Hart: We launched our G1GABLAST product a couple of years ago and have had some really great success with that. Obviously, Google’s not the only one building out fiber. There’s AT&T and many, many others. That specifically hasn’t had that big of an impact. We’ve always been committed to investing in our network, building out ubiquitous speeds and making sure we can deliver at peak hours with minimal congestion.

I mentioned the 10-year network transformation plan: Starting right now and into next year, DOCSIS 3.1 is high on our radar, so we’re making great strides with the testing, deployment and trials of 3.1.

We definitely have a fiber-deep strategy and an N+0 architecture that we’re working towards. We’ll be doing that both to keep up with speed and contention and just to deliver a better quality of service. And taking fiber deep is also key to enable Full Duplex, ultimately, which will come after 3.1 to enable more symmetrical Gig speeds.

Yes, we’ll definitely keep our eyes open on the competitive landscape, but we’re going to keep doing what we’ve been doing for a couple of decades and keep investing in our network.

Nair: We’re big on Gigs. Our bull’s-eye product right now is about 200 Mbps. At a traffic engineering level, though, our peak downstream usage is in the 900 Megabit-per-second range in some markets. 920 Mbps is pretty close to 1 Gig. If you consider a serving area with 2,000 homes, at 50% penetration, times 1 Mbps, guess what? That’s a Gig. Which is pretty much 25 DOCSIS channels. If that goes to 2 Gigs, that’s 50 DOCSIS channels. We have 100. That’s why we spend a lot of time on this.


MCN:DOCSIS 3.1: Status report please, for DOCSIS 3.1-based deployments — CPE [customer premises equipment] and CMTS [cable modem termination systems]? Anything notably different from prior DOCSIS upgrades?

Kotay: We’ve launched D3.1 service in Atlanta, Nashville and Chicago, with more coming this year and next. We think our XB6 is the next big vision of connectivity, with DOCSIS 3.1 at the heart of it.

Nair: Short version is, we’re in market and nearly done with CMTS. CPE begins next year.

Hart: We’ve got the 3.1 CMTS platforms up and running. We’ve got probably 50 different types of CPE that are working beautifully with the CMTS. Most of those are 3.0, so we’re testing out the backwards compatibility.

It will be similar in many respects to DOCSIS upgrades in the past, but with the introduction of OFDM [Orthogonal Frequency Division Modulation], there will be a little bit of navigation in terms of the complexity. But we’re feeling pretty good about it and the early results are very solid.

MCN:Advanced Class question: Full-Duplex DOCSIS, or fiber-deeper? Or both?

Hart: It’s both, and from our perspective it’s fiber-deeper first. To enable concurrency on 3.1, we’ll continue to take fiber deeper and then get the full benefit of Full Duplex, depending on the service group size in any given market.

The other component is the remote PHY [initiative] as part of the fiber-deeper strategy and moving those electronics closer to the customer. That provides additional optionality to upgrade to Full Duplex and it also provides more optionality in the future potentially for some wireless [technology] in the remote cabinet if need be.

Kotay: We don’t look at these sorts of technologies as competing with one another as much as we see them as increasingly powerful tools in an increasingly large toolbox. Today, we’re offering Gigabit Pro, a 2-Gig symmetrical fiber-to-the-home service across our footprint, as well as D3.1-powered gigabit service in several markets, with more to come. For green-field development and MDUs [multiple dwelling units], fiber is often the best tool, but DOCSIS 3.1 and the forthcoming Full Duplex provide great flexibility to deliver gigabit speeds using our HFC network.

Nair: It’s not an either-or situation. We like them both. We’ll do fiber — we’re doing fiber — where it makes sense and, as it turns out, it makes sense in lots of places for us. Full Duplex DOCSIS means building for symmetrical bandwidth. Symmetrical DOCSIS is more a question of “when” than “if.” That’s because right now, we’re still delivering much more traffic to homes than from them — it’s asymmetrical. Going to node+0 (amplifiers) aligns us all for Full Duplex DOCSIS. So, whether it’s customer-generated content or the next OTT-like thing, it means we don’t have to swap out that last bit of coax.

MCN:4K, UltraHD, High Dynamic Range/HDR, wide-color gamut, 10-bit interleaving: What of all of this matters to you? How is it best harnessed to improve the life of people watching TV?

Nair: Of that list, HDR is more impactful than just “plain” 4K. However, only 4K televisions have HDR, so they’re kind of inter-linked in that way. We’re launching HDR as something that comes with our 4K service. It came out in an earnings call as “EOS,” but that’s more of an internal project name — it’s cloud-based and will roll out in the U.K. first.

Kotay: We’ve done a lot of testing. You can see our view reflected in our product strategy. Short version is, all those things you listed are good things. There’s just nothing bad about increasing resolution and so on. But the ones we’re most focused on right now from a technical standpoint are HDR and wide color gamut.

We can show you some 1080p HDR video that looks better than 4K non-HDR. Do you remember when you first saw HD, next to a standard definition TV? It was, whoa, that’s much better. HDR is the same. You don’t realize you haven’t seen a blue sky on your TV until you’ve seen it in HDR.

Hart: To me, HEVC [High Efficiency Video Coding], the new video compression, that’s equally important. It’s really being able to move that content around no matter what flavor it is from an efficiency perspective to preserve some of the bandwidth. I think all of those — 4K, HDR, etc. — are going to be important. What the consumers want and what the manufacturers can provide will work itself out based on the content that’s produced. To me, the concern is how to transport all that information in the most efficient way.


MCN: It wouldn’t be a CTO conversation without talking bandwidth. What’s the latest in broadband usage trend lines? (E.g., is the growth of the growth still in decline? Are there patterns worth noting?)

Hart: CAGR [compound annual growth rate] on the downstream, in terms of utilization, is about 49% to 50%, and that’s what we continue to model in our investment plan. From an upstream standpoint, we’re probably in the 27%-28% range. We’ll probably see that move into 30% or so. With all of the cloud-hosted solutions and cloud-managed solutions, it puts a lot more pressure on the upstream.

Nair: We’re seeing peak speeds growing — and they’re growing logarithmically. That keeps our attention. The main driver now is still video. With video, we used to have very large bundles, with smaller content libraries. The world is going to smaller bundles, with larger libraries of content.

Kotay: We double our network capacity every 18 to 24 months on average, and that has remained relatively constant. We can realize some efficiencies by doing less over the backbone and driving more regionalization, but as long as consumers’ demand for rich, powerful online and video experiences continues to grow, and the quality of those experiences continues to improve, we know we need to stay one step ahead of what’s next.

MCN:At what point is entertainment video not the bulkiest passenger on broadband? What other bandwidth-insatiable things on the horizon have your full attention?

Hart: We’ve always been a leader with business services; Cox Business is going to hit the $2 billion revenue mark this year. I think there will be some additional applications and managed services, hosted solutions that will start to eat a lot of bandwidth but deliver a lot of value.

Maybe perhaps longer-term, there’s virtual and augmented reality — those could be a huge driver if they take off.

Kotay: I think the Internet is experiencing a lot more social video as a form of communication. This includes things like Facebook Live, Periscope, Vine, quick clips. I think we’re still pretty far from video not being the bulkiest thing we carry, but we are seeing a rise in other forms. The good news is that compression improvements can help offset that.

Other things we’re seeing, with the preponderance of smartphones and devices, are more firmware updates. They’re bursty, but they can be materially significant. When a new game launches, that can be a big, bursty thing. Video is more consistent. It’s bulkier, yes, but more consistent and predictable.

So if you look at downloads, entertainment video will be a huge chunk for the next few years. However, as far as peak demand, we’ll likely see a continued rise in firmware updates and games — things happening in the background.

MCN:What’s the situation around the possibility of widening the spectrum above 1.2 GHz? Is it time to raise the roof around here?

Hart: I think in the near-term we’re in good shape without having to widen. It’s something we’re always evaluating and reviewing. If you’re going to be making some enhancements to the network, it makes good sense to try to future-proof it, but probably not here in the near term on that particular component.

MCN:Widening the upstream path: How do you think through when it’s time to do so? What’s the situation now?

Hart: The good news is that a fiber deep strategy helps to address some of the upstream bandwidth by decreasing the service group size. That’s going to buy us a little more runway than we had thought when considering mid-splits and high-splits options.

For now, we’re in pretty good shape with our strategy. With the mid-split maneuver, you’ve got a massive CPE implication; that’s where a ton of costs could be. I think we’re in pretty good shape today with our N+O strategy.


MCN:RDK, the Reference Design Kit, has a new profile: Broadband. What should people know about it?

Nair: The RDK started with video. The RDK is expanding, so it got segmented to “RDK-V.” The next one is focused on broadband, which is why you hear about “RDKB.” It’s a little tricky because the letters “V” and “B” sound very similar when spoken!

Regardless, RDK-V is how we got to a user experience that satisfies, on many levels. Comcast started it with X1. We’re onboard, so is Cox, Shaw [Communications] and a growing global footprint. RDK, no matter the letter behind it, is the for-real open source platform behind the platform.

RDK-B reinforces connectivity being king. Everything we deploy will be RDK.

Hart: I think RDK-B has a lot of the same promise [as RDK’s video profile] in terms of speed to market, good economics in terms of the open source and keeping it open to multiple vendors.

The other piece is the functionality and capability to do remote installation and monitoring and repairs, and some of the tools … will be a differentiator. And with the new tools coming out, ease of use is another big benefit of RDK-B.

Kotay: It’s this — broadband is a two-way network, but as an industry we’ve been relatively blind to the connectivity in the home. The feature sets for routers and broadband devices have been relatively static for 10, 15 years.

It’s analogous to video services, in a way: It used to be all about channels. Channels are still important, but there’s a lot more to it. Same thing for broadband: It used to be all about speed. Speed will continue to be important, but it becomes more than just that.

Things like RDK-B will be central to be able to understanding what’s going on, and responding to that understanding. Whole-home WiFi is a good example: It’s about putting a lot more visibility and tools into the hands of our customers.

MCN:Kevin, give us an update on where Cox is with the RDK video profile [Cox’s new Contour platform uses the RDK and syndicates Comcast’s X1 platform]. Any shareable Net Promoter Score or other metrics or anecdotes? In NPS parlance: Would you recommend it to a friend?

Hart: It’s been a huge success for us. We’re, of course, partnered with Tony [Werner, president, technology and product at Comcast] and his team. But it’s also what we’ve done to enable it for ourselves. We probably did two to three times the amount of integration work on this.

It’s definitely been a home run. In the first couple of months, we’ve already [surpassed] 400,000 subscribers. We’re moving quite rapidly in terms of our deployment relevant to our initial plans and expectations. And the customer feedback has been off the charts. The NPS is multiples in terms of some of the other legacy platforms that we have.

The ease-of-use has had a nice halo effect, much like G1GABLAST had for our Internet service. Within just a few months, we’ve been able to achieve the same overall level of performance of the platform around tuning, etc., that it took Comcast and many others many years to achieve. We get some of the benefits of their learnings, but we’ve been able to do that within our ecosystem.

MCN:What’s your approach to devices — those you provide and those customers bring and attach to you?

Kotay: If you went back four, five years — that’s when we started looking at having to build software and apps as more of an opportunity than a burden. Over the past two or three years, we’ve had the same shift with hardware. In our industry it used to be, oh, we have to provide a box, to do conditional access and some navigation, and we have to provide a cable modem, to do HSD [high-speed data].

It reminds me of [astronaut] Alan Shepard, when he was sitting in the spaceship, ready to take off, and was asked what he was thinking. He said he was thinking about “the fact that every part of this ship was built by the lowest bidder.”

The funny part was, while we were thinking that way, a lot of innovative companies were devoting a lot of effort to getting devices into the living room where we already had a foothold. I think it made all of us sit up and say: What if it’s not just about having product for an ISP or a cable provider? What if it’s about making the best product, period?

In the last couple of years, we’ve gotten a lot of religion about that and we’re seeing the rewards, in terms of customer satisfaction and retention with products like X1 and our X1 voice remote. Over the next year, you’ll see us continuing along that path with products like our Xi5 wireless set-top box and XB6 home gateway. The goal is simple: make products people love.


MCN:How do you answer this question when your bosses ask it: What does 5G mean to us?

Nair: It means highly distributed, small cells with a mix of licensed and unlicensed spectrum. In some ways, it’s a credible threat. In other ways, it’s an opportunity. It’s an opportunity because it’s coming out of mobile and all those small cells will need backhaul. We do backhaul. Ultimately, it’s a hype bubble at the moment — we think it’s more of a 2020 thing.

Hart: As you know, the majority of wireless traffic is facilitated by wireline, so there’s a lot of backhaul opportunities for us to partner with our carrier customers. As that becomes more prevalent and that becomes more pervasive, they’re going to need to leverage networks like ours and others in the MSO community.

As it related to leveraging it as a last couple of hundred of feet fixed wireless methodology, we’ll have the same opportunity to take advantage of that, particularly with some of the remote PHY components. We could potentially leverage some of that infrastructure with strand-mounted small cells as well.

There’s definitely a place for 5G and we think it’s both an opportunity that can help with the backhaul but also even potentially leverage ourselves. And we feel that the product that we already deliver is very competitive and will continue to stay ahead of the performance from a throughput and speed and connectivity standpoint.

MCN:Enterprise services: What’s the hot seller to commercial businesses right now? What does it take to go after the large enterprise crowd — to roll up the sleeves against Verizon and AT&T and the like?

Hart: We still have plenty of opportunity within the small-to-medium businesses within our footprint, so we’re continuing to stay focused there. Our IP Centrex hosted voice solutions are selling great right now. We’ve also got great opportunity with schools, government, our hospitality network and convention centers, and we’re looking to go a bit further upmarket over time. In many cases, there are probably opportunities to partner with other MSOs, because many of those larger enterprises span multiple geographies, sometimes not always directly within our footprint.

Kotay: Business services is a major and growing segment for us, and a focus for a lot of really cool innovation on the product side. As just one example that I’m excited about, we’ve been putting a lot of effort behind cloud enablement. Mid-market and enterprise customers, in particular, need bandwidth and solutions engineered to enable their move to cloud services.

Earlier this year, Comcast Business announced that it is providing businesses with direct connectivity to Amazon Web Services, which, as we all know, is one of the leading providers of cloud services in the world. As that group continues to move upstream and serve large, national enterprises with dispersed office locations and data centers, you’ll continue to see them make more news with major cloud providers.

Nair: It depends on the size of the business. In general, small and medium businesses are really interested in a high-quality, dedicated IP interface, in the range of 500 Megabits per second to a Gig. It goes on from there to hosted voice, layer 3 VPN, and eventually to an SDN-type solution.


MCN:What are the important elements in building a video cloud — dos and don’ts? Where is each of your respective companies, with respect to a cloud-based video offering?

Nair: I’ll go with “do” rather than ”don’t.” Do pick the right vendor. Two, make sure the vendor isn’t bullshitting you. Third, do make sure the vendor delivers.

Our path to cloud-DVR started with putting the navigation in the cloud. Then we put VOD in the cloud. Third is putting DVR in the cloud. That’s the progression that’s worked for us.

MCN:Kevin, you have options for the new Contour. Any plans underway to enable a cloud DVR service?

Hart: It’s on our product roadmap for next year. That being said, there’s been some mixed results in terms of customer uptake on the platform. Some of the things we’re working our way through are the content and the copyright components with some of the content providers. We’re still trying to work some of those models out.

The other piece is the economics. Setting up our own specific cloud DVR solution, the economics would be a little steep. We’re looking at a couple of different partners to provide some flexibility there. Some of the economic models from Cisco [Systems] and some other providers are becoming a little bit more nimble, agile and affordable and more “by the drink,” if you will. I think there will be some advances … that will allow us to make more progress on it. We’re evaluating it.

MCN:Where are you with virtualization, generally or specifically?

Kotay: I think virtualization has gotten fetishized. It’s not an answer to all problems. It’s a critical tool in the toolbox, yes. But, for example, some of the stuff we’re doing around containers isn’t virtualized, because they don’t need to be. There’s no elasticity in that workload, so, the less layers and complexity, the better.

Virtualization is important, and we need it in some areas, but it’s not the only tool. Of course, in some areas, virtualization technologies can deliver powerful benefits in terms of network performance and reliability. We’ve been pushing more programmability into the network for years, and have seen the benefit of that approach.

Hart: We are making a lot of progress on our virtualization plan, virtualizing from a hardware and software perspective and software-as-a-service and becoming much more nimble in terms of the things we do on a cloud-based way in which we leverage technology and leverage our products and services.

Nair: At a core level, we’re looking at SDN [software-defined networking] and NFV [network functions virtualization]. Both have significant and practical benefits. Benefits like providing dynamic bandwidth: clearly important. What comes with that tends to be economic and efficiency gains. That said, we do understand the challenges virtualization poses to our key vendors.

MCN:DevOps — What is it, why does it matter, and to what extent are your companies pursuing it?

Hart: We were working some DevOps opportunities last year, but we kind of went all-in earlier this year. We did a bit of an organizational realignment to create a true DevOps team that combined all of the development and IT operations team, and co-located a lot of our business stakeholders.

The speed-to-market, the learning, the quality, the feedback we’re getting from our stakeholders has exceeded my expectations in terms of the way our teams have embraced it and what they’ve learned from their counterparts that are now part of their same team members, and just the pace and scale and quality of what we’re delivering around multiple platforms. I actually thought there would be a bit more of a resistance, culturally, from the teams, but it’s been embraced.

We’re extending it not just to our traditional technology DevOps model, but we’ve had some major data center moves, so we’re applying DevOps methodologies to physical moves as opposed to just some of the virtual and technology development and making it more of a mindset in how we do business.

Kotay: We’re huge on it. There are two or three reasons why it matters. The biggest is, it creates the right culture of autonomy and accountability. Put another way, the people who cause the pain have to deal with the pain!

The rate of change is only increasing, so the whole idea of sequential handoffs — formally document your requirements, formally hand them off to the design group, which hands off to the engineering group, which does the test cases and hands it off to QA, then to integration and test, then to deployment, then to support — there was a time and a place when that made sense. There are still cases where it makes sense, actually — hardware work is still very much a “waterfall.”

But being there on the DevOps side, supporting the triage elements that come with RDK-B, and being deeply involved all the way through product and process … DevOps still blows my mind. It changes everything, with everything that means. We wouldn’t be where we are without it.

Nair: DevOps matters if you’ve adopted an agile software methodology, and you’re building a lot of your own code. In other words, if you’ve decided to own your stuff. If you choose to do your own build framework and you’re writing code — because you want transparency and control of whatever your situation is — then the model works.


MCN:What disruptors or potential disruptors are on your radar right now?

Hart: For the disruptor component, it’s about keeping our eye around 5G in the fixed wireless space and how do we capitalize on that emerging technology, particularly around MDUs and commercial businesses?

It’s not very glamorous, but regarding special projects, it’s about working on process excellence and customer experience and putting the right set of tools and customer self-care and navigational flows within our call center and field services. It’s a lot of process work, but we’ve got to codify that within some of our technology tools. It can help to distinguish ourselves and continue to separate from some competitors.

Kotay: I’ll take “potential disruptors,” in the context of the Internet of Things. I find it interesting that at this point in 2016, IoT [the Internet of things] still has so much headroom to grow. And that growth, when it accelerates, will create interesting dynamics.

If you think about the connectivity axis and look at power consumption, bandwidth and range, there’s an interesting gap. For lack of a better term, call it the narrowband space, for instance. It creates an environment for things with aren’t hard on batteries and very long range, in terms of miles. Their bandwidth requirements are medium at best, like for M2M [machine-to-machine communications]. I don’t know if it’s disruptive, but it feels like there’s something interesting there.

Another I’m still constantly watching are display surfaces. They’re also ripe for disruption, more in the five-to-eight-year horizon. I don’t know if it’s flexible displays or cheaper displays, but it feels like we’ll have a lot more screens in our lives, and not just mobile. Fixed screens too.

Nair: I’ll take disruptors. Primarily, disruptors really are those competitors who aren’t interested in profitability — for five-plus years. That’s a long time.