Wifi insights
Follow Me

Wi-Fi: Convergence or disruption?

By Claus Hetting   /     Jun 08, 2015  /     Uncategorized  /  

That’s possibly the trickiest Wi-Fi industry question to answer right now. I’ve long held that Wi-Fi is the only tech that can keep pace with the free falling (and unstoppable) price/performance curve that the tech industry has been racing down for 50 years – in other words, Moore’s Law.

If Moore was right, then it’s inevitable that mass-market public Wi-Fi will happen. And if that happens, it’s only a question of time before mobile is relegated to keeping folks online during the 5% the time when they’re not in Wi-Fi range at home, office, and in public spaces. That means in our cars and outside cities. That idea of course is at the heart of the Wi-Fi First movement, which thus far is happening only in the US.

The question is then how all of this will translate into profitable mass-market Wi-Fi and not least who will do it. Wi-Fi being anybody’s game complicates crystal ball predictions immensely. Here we go.

What the mobile industry has tried to do

Mobile has taken stabs at incorporating Wi-Fi into network fabrics with sporadic success. There are 20-30 major mobile carriers today using 3GPP-type offload and not many adhere to strict standards that the industry is used to. At one point tunneling Wi-Fi traffic to core networks was the Holy Grail of telecom-grade service control. But that has yet to happen except in test cases.

So-called ANDSF device clients for offload were all abuzz 1-2 years ago, but little ANDSF deployment has materialized. Instead OTT clients of all sorts have popped up to get devices to make smart network selection decisions, probably because the ANDSF standard is lacking. OTT clients are needed as Wi-Fi still suffers from basic issues like ‘sticky AP’ that keep you connected for too long to APs that are too far away.

While homespots & public Wi-Fi have nearly become a best practice for cable carriers, MNOs are a long ways short of embracing Wi-Fi – at least in terms of own Wi-Fi builds (with a small handful of outliers). So is this likely to pick up? Despite massive price/performance benefits and all the technical pieces in place, Wi-Fi is still shaky ground for mobile carriers.

So why not Wi-Fi?

The biggest reason for the reluctance is structural. No MNO really wants to acquire and manage sites by the tens of thousands or even millions to make up a meaningful Wi-Fi footprint. It’s a daunting task when the option is to buy more licensed bands and hang more equipment off towers. Add to this that Wi-Fi sites are not about simply acquisition. Venue owners expect to get something out of it, too.

The option of course is to find Wi-Fi aggregator partners to work with such as Boingo Wireless or iPass. One reason why Boingo is doing well on the NASDAQ right now is exactly that – they’re in the right position as one of very few with a solid and consistent carrier Wi-Fi network. Boingo is also an obvious carrier acquisition target, by the way – and for the same reason.

Instead MNOs for the most part continue to spend billions of small slices of spectrum that are – in Wi-Fi terms – so small that they can hardly carry a single channel. And thus far most investors are still buying into that story. But for how long?

If mobile technology is so outdated in 2015 that it can’t sneak past a few tens of Mbps in most places, then how attractive is it to consumers who want more for less? This harks back to fundamentals: If mobile detached from Moore’s Law years ago, it also departed from what the market wants and expects.

The politics of a trillion-dollar market

Qualcomm & Boston Group recently said that the revenue of the global mobile market is 3.3 trillion dollars. With such a gigantic pile of money at stake, no wonder the powers that be shudder at the thought of Wi-Fi disruption.

Another reason why MNOs reject Wi-Fi is purely political. I’ve alluded to this before: If big MNOs were to go all out on Wi-Fi builds, their lobbying power for more licensed bands would erode. As Dr. Michael Marcus said at the WIS2015 in San Mateo: US mobile carriers have resisted every allocation of more unlicensed bands since the birth of ISM. Going unlicensed is just not what they do.

Mass-market wireless access is a battle that MNOs are not likely to win without waving goodbye to same-same and business as usual. 5G (if it happens) will likely be too little, too late. It will take the 3GPP 6-8 years even just to standardise 5G, whatever 5G may turn out to be. Meanwhile, big vendors like Ericsson are jumping on the IoT and connected car buzz. Both are sketchy for short-term replacement revenue. There are just too many ecosystem and tech pieces missing from that gigantic puzzle.

Instead the mobile industry is throwing a spanner in the works by trying to invade ‘Wi-Fi bands’ with LTE-U/LAA. The problem with LTE-U/LAA is not that it will wreck Wi-Fi services (it won’t because it’s not likely to happen) but that it has MNOs derailed pondering alternatives to Wi-Fi. If regular small cells didn’t really work out, then why would LTE-U/LAA small cells do better? I don’t think they will.

Real convergence in Europe, Asia, & Latin America

But it’s not all doom and gloom. A more enlightened approach to convergence is beginning to happen with carrier Wi-Fi as an important component in Europe and in Asia, where political hang-ups seem to be fewer.

Liberty Global’s Ziggo (Netherlands) and Telenet (Belgium) are excellent examples of pulling out all the stops in mobile, cable, and public Wi-Fi. Vodafone’s possible tie-up with Liberty Global has yet to materialise but it is another sign that mobile’s dominant position is on the wane.

The value of Wi-Fi is well understood also by Dutch telecoms giant KPN: It lies in its versatility to deal with many applications & use cases. Big former PTTs like KPN may in fact be best positioned to cash in on the shift to convergence if they can accept the writing on the wall: That both Cloud service and (for the time being) ad revenues have sunk irreversibly into the pockets of Google, Apple, Facebook, and their kind.

In Asia, KDDI – the largest mobile carrier in Japan – recently reported offloading 57% of their mobile traffic to Wi-Fi in 2014 and that they want that to reach 65% this year. This elicited probably the longest ever debate stream on the Wi-Fi NOW Forum on LinkedIn – a testament to how controversial the idea of ‘offload’ still is to many in the telecoms industry.

The Coca-Cola network of Brazil – created by Oi & Fon – is still small but another good example of carriers reaching out to small venues with an offer of ‘free Wi-Fi’ for their customers. We’ve still yet to understand how the billion-dollar potential of ads & analytics-based Wi-Fi monetization (in this case around Google, Facebook, etc.) can be unlocked in earnest.

Disruption… and so on

So much is written on disruption that it is hard to open one’s mouth without boring people. But remember this: Every business will eventually be disrupted regardless of size and might unless the civilized world goes under. It’s doesn’t matter how many billions you’ve got in the bank – although it will help you to fend it off for longer, of course.

The disruption faced by carriers is likely not the big & dramatic kind like Nokia being all but wiped out by Steve Jobs and his ‘Cupertino fruit company’, as one Nokia exec put it. It is probably more of a gradual decline into margins as thin as a razor’s edge. The smart ones understand this challenge already, of course.

Back on the American side of the pond, connectivity entrepreneurs are as usual not sitting still. If FreedomPop is ‘the Ryanair of mobile networks’ then perhaps Scratch and Republic Wireless are the Easyjets and jetBlues of the same. And of course Google’s Project Fi has laid down the blueprint for what – I believe – many mobile services will become within perhaps 5 years or so.

Where there are people there will be Wi-Fi

Wi-Fi is anybody’s game and one thing is certain: There will be Wi-Fi wherever there are people and at the end of every capable broadband wire. Except for venues that have long since been secured by the likes of Boingo and some other carriers (airports, stadiums, etc.) the rest will either be self-deployed or Cloud-managed by carriers or the venues themselves or both.

Getting to tens of millions of prospective venues with Cloud-based management however slick is a gigantic task even for talented start-ups such as Purple Wi-Fi, Cloud4Wi, Zapfi and many others. But that doesn’t mean it can’t be done. Purple Wi-Fi recently tied up with Telstra and other startups with a strong monetization pitch – such as Euclid Analytics and Frontporch – are making headway with fixed carriers in Europe and America. Going it alone – outside of carrier involvement – is a lot harder.

What the Wi-Fi industry needs to fix

The raw performance evolution of Wi-Fi needs no fixing – in fact, it’s by a long shot the fastest ramp-up of any wireless tech in history thanks to the remarkable foresight of the folks in the IEEE 802.11 workgroups. Just keep pushing the envelope, IEEE folks. It’s working.

But we still have other important things to fix. The first is quality – whatever that means, I hear you say. And you’re right. It’s a big subject. As well as Wi-Fi works when networks are designed and deployed right; the service can be (and often is) more than awful when done badly.

Attempts to define a Wi-Fi ‘carrier grade’ standard for Wi-Fi have been several – one headed up (but for the time being still in the works) by the Wireless Broadband Alliance and similar groundwork by CableLabs. In the end the fixing of Wi-Fi services may not come from ‘carrier grade’ but by evolution of 802.11 standards and some of the freethinking vendors out there. One of the most promising new add-on IEEE standards is 802.11ai aka FILS – which has received remarkably little mention anywhere (but covered well by CableLab’s Vivek Ganti here).

The second fix is security. At one of my recent conferences I had a senior Orange executive start off his presentation with a picture of the Twin Towers. We may not have seen security disasters of this magnitude yet, but they could happen. I don’t know what the risk level is. But it is true that a lot of public Wi-Fi is notoriously porous to attack. We will be diving into the security issue (and many others) at Wi-Fi NOW in Amsterdam.

The Hotspot 2.0 (NGH) standard is the watertight fix on both security and seamlessness but the jury is still out on whether or not NGH will roll out fast enough before something else takes over. That something else could be something akin to Google’s Project Fi, which relies on any Wi-Fi out there that works well enough.

NGH is a carrier play and for the most part tied to the industry politics of big carriers. Outside of Boingo and US cable operators (and a few city networks) the incentive to deploy NGH for public venues and businesses is not obvious because it costs real money in upgrades. This could change if carriers are successful in addressing the massive global SMB market for managed Wi-Fi – and include NGH.

Are City Wi-Fi networks the future?

Here’s a story on what independent networks can look like: A good friend of mine is rolling out city Wi-Fi networks in the South Eastern US where broadband is scarce and pricy. Run by a couple of seriously visionary folks, the company has rolled out 9 square miles of Wi-Fi covering an entire city of 50,000 people.

The result is 35 mbps speeds at distances of 1,500 feet and much more of course when you’re closer. The price is 1 dollar per GB. Google likely gave up on their old Mountain View network because it wasn’t performing. We now have the technology to build city Wi-Fi networks that perform better than 4G at a fraction of the cost. This is a tectonic shift in technology that I believe will have a big impact.

If city Wi-Fi business models are scalable – and I believe they are – how long before Google, Facebook, Microsoft, and many others plug into partners like that (the list would have to include Boingo, iPass and a few others) into their Clouds?

OTT access disruption has only just begun.

Post a Comment