Finding telecoms link in the value chain.

Personally, I think that net neutrality will force operators to innovate a bit — this isn’t about discriminating against heaviest users, it’s about changing the business model from flat rate all-you-can-eat access to one in which heavy usage isn’t a problem that requires random throttling. Operators should offer access that does NOT include video, VOIP, or P2P unless the user pays an extra fee for access to those services (for example).

Qualcomm, the world’s biggest maker of chips for mobile phones, has entered the net neutrality debate in the US with its chief executive calling for heavy data users to be discriminated against as wireless networks reach capacity.
Paul Jacobs’ call at the CTIA wireless industry conference in San Diego came a day after Julius Genachowski, head of the Federal Communications Commission, warned that there was not enough room available on the airwaves for the “explosion” in wireless data traffic.

Mr Jacobs said he had given the FCC chairman his views on “traffic shaping” as one solution to what the FCC describes as a “looming spectrum crisis”.

He described traffic shaping as “the ability to say: ‘let’s be fair, this person’s moved a lot of data, this person’s used a little’, if they’re paying the same amount, then the person who’s used less will get more access”.
John Donovan, AT&T’s chief technology officer, told the conference that smartphones and the applications they ran had caused a 5,000 per cent increase in data usage over three years.

“We will need to manage our way through data-hungry applications or devices on our network that would degrade the experience for others,” he said.

Supporters of net neutrality say there should be no discrimination on a free and open internet.
Internet providers should not block, speed up or slow down web content based on its source, ownership or destination.

Mr Jacobs said this “more radical notion” of net neutrality was born out of the internet bubble and the notion that bits of data were free, when in fact they had now become very expensive for providers.
“Regulators may not know that, the lawmakers may not know that, and so we need to make that clear, and it’s very obvious that we are pushing the limits of the amount of capacity we have.”

Mr Jacobs said it would perhaps be too intrusive to go down the route of saying one internet service was fine on a network and another was not, but operators needed to be able to manage their networks.
In his speech, Mr Genachowski had said that the FCC had not yet decided what measures to take to preserve an open internet. Proceedings would begin this month to establish “rules of the road”.

He conceded that mobile had unique congestion issues and the last thing the FCC wanted was to impose “heavy-handed and prescriptive regulation”.

http://www.ft.com/cms/s/0/cf650104-b463-11de-bec8-00144feab49a.html

That is, in times of high demand, a choice must be made as to who gets connectivity and who does not.

I argue that this will require operators to rethink their existing subscriber model, because a subscriber by definition has a right to access t he network. Put another way, the existing subscriber model precludes new business models because it eliminates operator flexibility in terms of allocation of bandwidth.

The solution is to eliminate the subscriber model, and instead bundle connectivity directly to services. People will purchase a voice service, or an email service, with associated connectivity (and SLA) bundled with it. Ad hoc access for general browsing will be provided only if bandwidth is available after servicing access connected to an SLA.

To be fair to subscribers, and to spur innovation and competition, ad hoc access must then be open — people will connect to whatever network is available at the time. This has a number of implications:

1. Any given subscriber can connect to multiple networks simultaneously, e.g. network A for email and network B for general browsing
2. Operators can find niche service areas — for example, some operators might specialize in providing ad hoc access during busy hours
3. Account management for ad hoc access will require federation. All subscribers essentially become roamers, with the HLR information maintained in a clearinghouse.

I’m curious to see reaction to this piece…

Killer quotes:

[Because of network capacity being sucked up by iPhones] The result is dropped calls, spotty service, delayed text and voice messages and glacial download speeds as AT&T’s cellular network strains to meet the demand. Another result is outraged customers.

Just wait till all operators offer iPhones. And a full netbook line. And more dongles. And enterprise services in the cloud.

AT&T says that the majority of the nearly $18 billion it will spend this year on its networks will be diverted into upgrades and expansions to meet the surging demands on the 3G network. The company intends to erect an additional 2,100 cell towers to fill out patchy coverage, upgrade existing cell sites by adding fiber optic connectivity to deliver data faster and add other technology to provide stronger cell signals.

Wonder what the ROI will be on that 18 billion…

http://www.nytimes.com/2009/09/03/technology/companies/03att.html

Not to beat this drum too loudly, but cloud computing is among the one or two key truly significant changes that affect the telecom industry, dramatically expanding a business’ reliance on the network and dramatically increasing data volume — particularly data volume through cell sites as roaming employees access documents and other business content heretofore maintained locally.

The other significant change, as I’ve mentioned ad nauseam, is telecom’s move completely away from business models based on services (such as voice), toward models based on connecting developers of content/services and consumers of this content.

More data showing showing growth of cloud computing. Most relevant to telecom is this growth in the context of Netbooks, and the resultant deluge of data volume.

HTML5 is a standard that is still being developed and is likely to remain so for several years. Its focus on running applications within the browser is an important driver of interest in cloud computing, where applications live somewhere off on the Internet and are delivered by the browser.

The focus of future browsers will shift from “going places” to “doing things.” This will be a boon to free operating systems, which will increasingly be able to hide themselves under the browser user interface. While Windows and Mac OSX won’t go away overnight, the pressure on them will be to innovate beyond the browser, perhaps through a common set of extensions for HTML5 applications to use.

The key takeaway is that the operating system is in the cloud, in addition to office applications and everything else. Mobile operators are the most important — and to date largely unsung — players here; it is they that have the most to gain (or lose, if they succeed at screwing it up).

You’ve heard it before: remember Sun’s campaign that the “Network is the Computer. ” Back in 1995 Larry Ellison was predicting that Network Computers would replace Personal Computers as the computing equipment of choice. And Netscape said for years that the browser was the “next OS.”

The difference now is that a viable broadband mobile infrastructure is in place. Network technologies have matured, not from a technological standpoint (they’ve been mature from some time), but from a social standpoint. The network is now, finally, ready to realize it’s place as the computer because people’s lives are integrated with it more deeply than even Larry could dream more than a decade ago.

T-Mobile and Orange are both saying that LTE has many technology hurdles to overcome before it is viable for their networks (Verizon and NTT’s aggressive plans notwithstanding).

The operators’ concerns with LTE include: support for voice services; the impact on backhaul capacity; intellectual property rights; the lack of standardization for self-organizing networks (SONs); and the lack of spectrum. And those are just the chief conerns.

In addition, T-Mobile and Orange, along with other European carriers, are scarred by the disastrous early days of WCDMA (3G), when they spent billions of euros on licenses and hyped 3G downlink speeds beyond reality, leaving customers and investors sorely disappointed. They don’t want to repeat those 3G mistakes in LTE, especially in the midst of a global financial crisis.

In my mind, this reinforces my contention that mobile bandwidth will quickly and inevitably become a finite resource in the face of unprecedented demand. Business models will change from a flat-rate all-you-can-eat (or all-you-can-eat-up-to-a-point) to more complex and nuanced charging that includes dynamic tiering, utility-style billing, and so on.

Consider: Right now, AT&T is selling a range of laptops in Northern California at Radio Shack for $49.95 (that’s not a typo — 50 bucks; see http://tinyurl.com/oagx33) with a 2 year contract for flat rate data with a monthly cap. Netbooks are now the most popular computer type on the market (Intel expects sales to double in 2009 http://tinyurl.com/djxdrd).

Too, there’s the recent announcements of businesses standardizing on cloud office applications like Google Apps. Imagine the savings a large enterprise can enjoy by standardizing on netbooks. Laptops so cheap they are almost disposable. All data saved safely in the cloud.

Suddenly you have lots of torrent sessions and Skype calls routinely supported through the mobile network. Huge increases in youtube traffic. Hulu. Even parental control becomes much more meaningful as parents worry about inappropriate content accessed via the laptop much more than a phone.

Suddenly, despite usage caps, the GGSN is supporting data volume like no one could ever dream, with smartphone traffic a minority. Some reports say this is a long range play. We’ll see. I think it’s going to explode as we watch. And vendors that tie licensing terms to usage/seats/transactions or similar, will watch revenue explode with it…

Users of iPhone download games, video and other Web data at two to four times the rate of other smartphone users, according to comScore. Yet AT&T charges iPhone subscribers the same fee of $30 a month for data that it levies on other smartphone customers. And aside from restricting certain activities, like file sharing, AT&T doesn’t limit how much data can be downloaded…

Too, the iPhone is just the tiny first drop of a huge deluge that will be caused by access dongles and netbooks. The situation is pretty clear: network resources are finite; demand upon them is (as good as) infinite.

A message to those folks philosophically opposed to any kind of network management: something’s gotta give. If access remains unrestricted, physics will take over and reduce everyone’s throughput and QoS to a trickle. It’ll be a small comfort to know that the occasional packet you receive is officially “neutral.”

The net neutrality issue as currently framed is a red herring anyway. Subscriber’s don’t care if their bandwidth is restricted, as long as they have control to remove the restriction. The water supply in my sink is restricted, but I can remove that restriction at any time (for a fee) by turning a tap. The more I turn the tap, the more I pay.

And that’s really the key — allocate network resources according to those willing to pay for it. Everyone wins.