Sri Lanka to Provide Google Project Loon Internet Access

Sri Lanka has become the first country to support deployment of Google Project Loon-based Internet access.

The initiative is reportedly going to offer free access across the entire country.

Whether that is correct or not is unclear to some of us. Some reports suggest a wholesale network is envisioned, with mobile operators and other ISPS able to buy access.

Though that does not necessarily work against the notion of "free access," it suggests access will not ordinarily be provided free of charge by all potential ISPs, though some likely will try to do so.

Many details remain unclear. Project Loon recently has been testing Long Term Evolution 4G access, havind concluded that direct use of the Wi-Fi protocol was impractical.

The extent of 4G across Sri Lanks might be an issue. Involvement of mobile operators therefore would seem to be imperative, but nothing so far has been released about mobile operator participation.

The other logical approach would be to use 3G protocols rather than 4G.

“The entire Sri Lankan island--every village from (southern) Dondra to (northern) Point Pedro--will be covered with affordable high speed internet using Google Loon’s balloon technology,” said Sri Lanka Minister of Foreign Affairs Mangala Samaraweera.

Officials also said local ISPs will have access to the balloons, reducing their operational costs.

According to Muhunthan Canagey, head of local authority the Information and Communication Technology Agency, Google is expected to finish sending up the balloons by next March 2016.

The agreement between Google and the Information and Communication Technology Agency of Sri Lanka (ICTA) did not immediately detail any other commercial agreements, such as whether the services will be sold at wholesale to retail ISPs and mobile service providers, and if so, on what terms.

But it will be tough to compete with “free,” if Sri Lanka’s government itself provides free access at speeds comparable to mobile Internet access.

There are 2.8 million mobile Internet subscribers and 606,000 fixed line Internet subscribers among Sri Lanka's more than 20 million population.

Up to 33% of All Cable TV Locations Might be Able to Buy Access Exceeding 1 Gbps by 2017

It sometimes is hard to immediately grasp the importance of new protocols in the broader telecommunications business, and DOCSIS 3.1 is no different.

The latest generation of a standard used extensively by the global cable TV industry supports access bandwidth up to 10 Gbps over standard hybrid fiber coax networks used by cable operators.

New research by IHS Infonetics suggests cable operators globally will have at least 33 percent of residential subscribers able to use by DOCSIS 3.1-enabled headends by April 2017.

That means an ability to provider bandwidth exceeding a gigabit, where the headend deployments are matched by modems supporting the standard, and where sufficient bandwidth is available to do so.

Facebook to Discuss Faster Internet Access At Spectrum Futures

Facebook will provide its views on the importance of rapid increases in Internet access globally, as well as what it is doing to support faster Internet access across the South Asia, Southeast Asia and global markets at Spectrum Futures, to be held in Singapore Sept. 10-11, 2015. 

Chris Weasler, Facebook global head of spectrum policy and connectivity planning, or Kevin Martin, former chairman of the Federal Communications Commission, will provide the update. 

The M Hotel Singapore  |  10-11 September 2015

Why Facebook cares about Internet access across Asia REGISTER NOW Spectrum Futures 2015 will look at platforms and policies to connect the next two billion Internet users in South Asia and Southeast Asia. Spectrum Futures 2015 brings together regulators and service providers from throughout the Asia-Pacific region to allow the exchange of ideas about key policies to help emerging markets like India, the Phillipines, Thailand, Indonesia, Cambodia and Myanmar connect to their populations to the Internet within the next decade.

VISIT THE SPECTRUM FUTURES WEBSITE FOLLOW US ON

Hard to Tell How Much Market Share in Cloud Computing the Big Four Actually Have

Will the cloud infrastructure or cloud apps businesses be characterized by the same “winner take all” pattern that seems to be true of most consumer Internet apps? Perhaps not, as cloud computing is an enterprise service, purchased by businesses and other entities, not mostly by end users directly.

Perhaps the question is more narrowly whether the cloud infrastructure market will have a winner take all structure over the long term.

The answer might be harder to glean than first appears. On the surface, four providers seem to hold commanding market share leads, namely Amazon, Microsoft, IBM and Google.

The problem is that each firm counts different revenue sources within its “cloud revenue” total. Amazon likely is the biggest supplier of pure-play infrastructure as a service. Others might be bigger on software as a service or platform as a service, particularly Microsoft and IBM.

That is why some believe Amazon actually has a bigger market share than generally believed, in the core computing as a service market.

IBM has cloud revenue of $7.7 billion, but a big chunk of that comes from "hybrid cloud," which involves companies buying both hardware and software. Its more comparable "as-a-service" business is will only pull in $3.8 billion.

Microsoft boasts cloud revenues representing a $6.3 billion revenue run rate. Most of that revenue likely comes from Office 365, however. While that clearly is SaaS revenue, it is not computing as infrastructure revenue.

“The cloud infrastructure services market is quite clearly bifurcating with a widening gap between the big four cloud providers and the rest of the service provider community,” said John Dinsdale, a Chief Analyst and Research Director at Synergy Research Group. “

source: Business Insider

Carrier-Grade Wi-Fi Might Have Greatest Revenue Value as Wholesale Platform

Extensive “carrrier grade” public hotspot networks now are seen as providing substantial value for mobile and fixed network operators, but the perception of value is shaped by the legacy strengths and weaknesses of the contenders.

It easily can be argued that wholesale access and data offload will provide the most-concrete value, for some time, and perhaps over the longer term as well. The former generates new revenue, the latter offers cost reduction.

Excluding wholesale and offload, survey respondents to a Maravedis survey indicate that carrier-grade Wi-Fi has the biggest immediate potential for advertising revenue, followed by access fees.

Over time, location based services and video content are seen as bigger revenue opportunities. Support for carrier voice seems to be a widespread expectation, but the relatively low expectations for revenue might flow from several obvious aspects of the voice business model.

First, voice is driving a declining amount of revenue, so voice over Wi-Fi might not actually boost revenue for many, if any, contestants. On the other hand, cost reduction, especially for mobile virtual network operators, is a more-tangible possible benefit.

In fact, churn reduction might be the most-direct form of benefit, the survey, sponsored by the Wireless Broadband Alliance, suggests.

As Access Speeds Climb, Middle Mile, in Many Cases, Remains the Chokepoint

Revenue scale, or market potential, has been, and likely always will be a major issue for Internet service providers in areas with low population density, especially when population areas are highly isolated.

That basic problem exists for remote villages across South and Southeast Asia, Africa and elsewhere, including remote locations scattered across many parts of the Americas.

Bandwidth availability, at lower cost, essentially is a barbell, concentrated at the in-home or in-building level on one end, and then the major wide area network backbones on the other end.

In the middle are the access and regional networks, where ability to share costs is lower, and where free to low-cost local distribution (Wi-Fi, Bluetooth) is not possible. 

Those issues will not change fundamentally as Internet service providers ramp up local access network speeds.

EPB in Chattanooga, for example, says it eventually will offer 10 Gbps over its network, something the DOCSIS 3.1 protocol already incorporates as well. That suggests we are heading for eventual local access bandwidths up to 10 Gbps.

While that might allow a greater match between access and in-building distribution speeds, and where it is possible for backbone capacity on the major routes to scale rather gracefully, the middle mile, where business models are stretched, will become even more stretched.

As always, any significant bandwidth change anywhere in the end-to-end chain has repercussions in the rest of the network.

In-home or in-office distribution has not been a major issue for many decades, as local distribution tends to be the part of the network with the greatest capacity and the lowest cost to supply.

Backbone wide area network bandwidth on the major routes, historically, has been easier to supply, as the costs are shared very broadly across the complete set of potential customers. And sharing means lower cost.

All that means the access, middle mile and regional networks will continue to face upgrade issues, as we might well assume that local distribution within a premises or building will not be an issue, while the cost of adding lots of capacity to wide area networks is relatively simple.

As always, the access plant --with the least possibilities for sharing--is where the majority of cost lies.

Middle mile communications network costs are similar to costs faced in the airline industry, where the big city hubs are most efficient, and small towns the most-costly type of facilities.

The same economic problem exists in both industries: some lower-density locations represent difficult business models, since the destinations lack user scale, and therefore revenue scale.

Is Post-PC Era of Computing Now Creating "Post-PC" Device Profiles?

Tablet shipments (perhaps not directly “sales”) declined seven percent, year over year, in the second quarter of 2015, according to the latest analysis by IDC.

Tablet shipments of 44.7 million units also represented a decline of nearly four percent sequentially.That might indicate a faster rate of decline in the recent quarters.

Perhaps the "post-PC era" really means post-PC, with tablets being a form factor derivative of PCs, or acting as PC substitutes.

Perhaps smartphones really are the emerging example of human interaction with computing in the post-PC era of computing. Eventually, a wide range of devices and sensors likely will tell the story, with "cloud computing" enabling connectivity as a core feature of a modern appliance.

"Beyond the decline, we're seeing a profound shift in the vendor landscape as the top two vendors, Apple and Samsung, lose share in the overall market," said Jean Philippe Bouchard, IDC Research Director for Tablets.

Top Five Worldwide Tablet Vendors - Preliminary Results for the Second Quarter of 2015 (Shipments in millions)

Vendor	2Q15 Unit Shipments	2Q15 Market Share	2Q14 Unit Shipments	2Q14 Market Share	Year-Over-Year Growth
1. Apple	10.9	24.5%	13.3	27.7%	-17.9%
2. Samsung	7.6	17.0%	8.6	18.0%	-12.0%
3. Lenovo	2.5	5.7%	2.4	4.9%	6.8%
4. Huawei*	1.6	3.7%	0.8	1.7%	103.6%
4. LG Electronics*	1.6	3.6%	0.5	1.0%	246.4%
Others	20.4	45.6%	22.4	46.7%	-9.3%
Total	44.7	100.0%	48.0	100.0%	-7.0%

Source: IDC Worldwide Quarterly Tablet Tracker, July 29, 2015