Report on WiFi Hotspot Usage

This report from In-Stat is interesting but not surprising: WiFi hotspot usage is increasingly shifting from laptops to hand held devices:

As a percentage of total connects, handhelds increased from 20% in 2008 to 35% in 2009. By 2011 handhelds are anticipated to account for half of hotspot connects ...

I wonder if we can anticipate a model where handheld devices automatically test WiFi hot spots for data service either in conjunction with or instead of carrier-based services. Using this in conjunction with carrier services would require inverse multiplexing capabilities in the phone and in the service provider, which is more complex and requires more coordination than an either/or type of service. The capability to augment carrier-based services could ease the pressure on spectrum that some carriers are reporting due to the increasing popularity of smart phones.

Prices at different app stores

This item over at GigaOm is interesting. The Apple App Store currently has the cheapest average price if you consider paid apps only, while Android has the lowest overall price. Given the open nature of the Android apps, I would expect this to change as competition increases as the number of Android apps increase.

Challenges with phasing out landlines

This commentary over at GigaOm is insightful and worth reading. In it, Stacey Higginbotham reflects on this AT&T filing at the FCC. Some excerpts from the filing:

Over 99% of Americans live in areas with cellular phone service, and approximately 86% of Americans subscribe to a wireless service. Many of these individuals see no reason to purchase landline service as well. Indeed, the most recent data show that more than 22% of households have “cut the cord” entirely.

Demand for VoIP service – from both cable companies and over-the-top providers such as such as Vonage, Skype, and many others – is also booming. At least 18 million households currently use a VoIP service, and it is estimated that by 2010, cable companies alone will be providing VoIP to more than 24 million customers; by 2011, there may be up to 45 million total VoIP subscribers.

Today, less than 20% of Americans rely exclusively on POTS for voice service. Approximately 25% of households have abandoned POTS altogether, and another 700,000 lines are being cut every month. From 2000 to 2008, the number of residential switched access lines has fallen by almost half, from 139 million to 75 million. Non-primary residential lines have fallen by 62% over the same period; with the rise of broadband, few customers still need a second phone line for dial-up Internet service. Total interstate and intrastate switched access minutes have fallen by a staggering 42% from 2000 through 2008. Indeed, perhaps the clearest sign of the transformation away from POTS and towards a broadband future is that there are probably now more broadband connections than telephone lines in the United States

The public policy problem, of course, is what to do with the 20% that do not have connections. Further how about those people who use the copper PSTN plant for DSL access (whether from a telco or a reseller)?

It seems that the hard problems with public policy is dealing not with the majority but with the marginal.

Singapore-Japan cable system

This article over at Telecommunications Online announced a new cable project, so I collected it here with my other similar announcements. Quoting the article:

Several Asian telcos and Google are to invest US$400million in a new subsea cable linking South East Asia to Japan aimed at providing the highest capacity connection link to-date. The South East Asia Japan Cable System (SJC) has a design capacity of 17 terabits per second with provision to be ramped up to 23 Tbps. As an example of its capacity, the bandwidth allows the SJC submarine cable system to handle 30 million high definition videos simultaneously.

The 8,300 km cable will link Singapore to Japan with branches to Indonesia, Philippines and Hong Kong initially. From Japan, it will be plugged into the recently commissioned trans-Pacific Unity cable to the USA, besides a branching link to Guam which is becoming an alternative cable junction linking Asia to USA.

Comprising 6-fiber pairs, the SJC cable is scheduled to be completed by the second quarter of 2012, and is believed to be modeled after the Unity business concept providing autonomy in operation for partners to the initiative.

The decision to proceed with this landmark project, no doubt, has been prompted by growing demand not only from Internet traffic but also from the surge in telco TV, games and enterprise data.

An earlier report by TeleGeography shows that international Internet traffic has not been affected by the recent economic meltdown. In fact, international traffic growth was up 79 percent in 2009, from 61 percent in 2008.

By end 2009, fixed broadband in the Asia Pacific is expected to grow 17.3 percent to 182 million subscribers clocking billings of US$44.8 billion, according to Frost & Sullivan. The next generation networks in progress in South Korea, Malaysia, Singapore and Australia will further fuel growth in fixed broadband, not counting the phenomenal surge in mobile broadband.

So this cable will cost a bit over $48,000 per kilometer. This is the first time I have seen the capacity of the cable described in terms of the number of HD television streams that can be supported; previously, it was always the number of simultaneous telephone calls.

Test of 3G networks in the US

Gizmodo recently ran a test of 3G networks in the US. The results are here. AT&T had the fastest average upload and down load speeds in the cities where the test was conducted. This is a bit of a surprise given the heat that the AT&T network has been taking recently from iPhone users. Gizmodo wonders whether it is the iPhone or some interaction between the iPhone and the AT&T network, and I can see why. These test results don't corroborate the belief that AT&T's network is (only) at fault.

Follow-on to Mobile OS item

This item over at Ars Technica caught my attention, especially in light of my last post on this blog. Given the importance of a vibrant ecosystem that includes apps, this software from Palm substantially lowers the barriers to entry for app developers for the WebOS. Using this environment, apps can be constructed by relatively unsophisticated (i.e., without particular programming skills) end users and, if done well, even be monetized. I presume the most successful apps would be professionally re-programmed and optimized. Wouldn't this have strongly positive effects for the ecosystem?

Mobile operating systems

There are two articles related to Mobile OS that caught my attention. This one, in Ars Technica, shows that, for the first time, the total number of iPhone users have passed the total number of Windows Mobile users. This one, over at CNN Money, shows the growth of Android-based devices. Clearly the flat Android line in the first article will change as 1.5+ and 2.x-based devices enter the marketplace and start gaining in installed base.

This is significant in a number of ways. First, installed base is a key factor in the development of the "ecosystem" of a device. Apple, which has locked down the hardware as well as the OS on the iPhone, enjoys an unrivaled hardware ecosystem from third parties. By tightly controlling the application environment, they maintain a strong grip on the app store, which enables them to monetize this very profitable aspect of the iPhone ecosystem. Because of the size and momentum of the installed base, they will attract more applications developers.

Android does not have the hardware control and has a more open approach to the app store which does not lend itself to monetization in the same way. However, the more open approach may attract more innovative app developers. There is clearly momentum in the Android OS camp, even if the installed base is lacking at present.

Carriers have an interest in minimizing the number of mobile OSs they support. Adding a new OS is costly because the features and services of the OSs must be tested with the network and supported by the carrier after the device is released. Thus, OSs with little installed base or momentum are at risk of losing their sponsors (the carriers) which leads to a self-reinforcing downward spiral in their ecosystem. Thus, the stakes for Microsoft in its Windows Mobile 7 (due in 2010) could not be higher; for that matter, the same is true for Palm's WebOS. Failure to attract a user base means failure to build an ecosystem.

This creates a high hurdle, but it is not insurmountable. At one time, the Palm OS was the standard with a huge catalog of applications; it is no longer so dominant. Thus, dominance is not permanent even if it can be a huge advantage in the market place

Google and DNS

I found this commentary on Google's entry into the DNS domain the most worthwhile (from this article in the NY Times):

“You have to remember they are also the largest advertising and redirection company on the Internet. To think that Google’s DNS service is for the benefit of the Internet would be naïve,” Mr. Ulevitch wrote. “They know there is value in controlling more of your Internet experience and I would expect them to explore that fully.”

In an interview, Mr. Ulevitch said OpenDNS was steering customers who mistyped addresses to Yahoo search results and ads. “I have no doubt they see that as a competitive threat,” he said. He also stressed that there is valuable information in the DNS layer — like where servers are located, what Web sites people are looking at and how frequently they are looking at them. “I have no doubt they will be monetizing this by increasing intelligence on people’s surfing habits,” he said.

Ulevitch runs the "Open DNS", which has provided this service for some time. He differentiates Google DNS from Open DNS here; you should bear in mind that Google is offering him competition. Finally, I think it is important to remember, as Ulevitch does here, that Google is a profit-making organization.

Importance of framing

No, I'm not talking about Time Division Multiplexing. This NOI from the FCC came to my attention (via this item in Ars Technica). I have addressed the importance of the "framing" of a topic in regulatory reform before (see this item), and I believe that this may be a similar example. Here, it seems as though the FCC is framing the debate about spectrum reform as a matter of broadband policy, which, of course, is related if wireless access is a key component of broad policy. Maybe Tom Hazlett's ideas are gaining traction after all?

Telegraph in the US

Ars Technica posted this article, which is a partial history of the telegraph in the US. The history of the telegraph involves considerably more than what is written and, taken as a whole, provides a graphic illustration of "bare knuckles" capitalism. Fortunes were won (eg. Ezra Cornell, who later founded Cornell University) and lost sSee my book Shaping American Telecommunications for more). What the Ars article does not mention is that while Gould was nipping at WU's heels, the telephone business was just getting off the ground. WU ended up making the business decision to focus on its core telegraph business and sold its telephone business to National Bell. It was a fateful decision indeed.

Mesh networks

Here is a nice article on mesh networking technology from Ars Technica. It includes a nice discussion of the historical context and a very little bit about the economics. What I found most interesting was the following, taken from an interview with NYU's Jinyang Li:

Li pointed out that current mesh protocols don't give nodes incentives to forward one another's packets. This isn't a major problem in academic, corporate or government settings where all the nodes can be assumed to have the same owner. But in open, heterogenous networks, each user will be tempted to "cheat" by modifying his own device to take advantage of other nodes' forwarding services without reciprocating. "It's still an open question whether we'll be able to solve this incentive problem," Li told Ars.

A related concern is security. Users should worry about the security implications of entrusting their packets to random strangers. Things would get even more dicey if routing protocols were enhanced to add monetary incentive schemes such as micropayments; in that case, security flaws in mesh networks could allow unscrupulous mesh participants to generate bogus traffic in order to siphon money away from their neighbors.

These are indeed serious problems that have to be solved before this approach to networking has a prayer of commercial success in anything but niche markets.

Telecom cost estimates

This is my latest installment in my continuing effort to collect information on the cost of telecom system deployment. NATOA filed these comments with the FCC that contain some case studies on fiber deployment to "anchor tenants" in communities. Perhaps you will find this interesting or useful.