The Digital Agency for International Development

Google Broadband

By Chris Wilson on 09 September 2008

According to FT and other sources, Google has announced their support for a new initiative called O3B to "bring internet access to 3bn people in Africa and other emerging markets by launching at least 16 satellites to bring its services to the unconnected" in 2010.

"They will... order 16 low-earth orbit satellites... as the first stage in a $750m project to connect mobile masts in a swath of countries within 45 degrees of the equator to fast broadband networks... the project could bring the cost of bandwidth in such markets down by 95 per cent."

This will probably be the largest single investment in developing country network infrastructure in history. The TEAMS submarine cable for East Africa (which is not yet active) will cost $82m to lay according to Wikipedia, while SAT-3 (West Africa) cost $280m. However, the comparable but larger Iridium satellite service, with 77 LEO satellites, was estimated to have cost US$6 billion and filed for bankruptcy in 1999.

According to the FT, "wireless spectrum required for the service had been secured through the ITU", but this seems unlikely as the ITU works by consensus and not coercion, and it has no power to override local governments' license demands.

In Kenya for example, a deregulated market, an international gateway operator must pay Ksh 15m ($210,000) for a 15-year license, plus up to $70,000 per year, and probably the same again for VSAT licenses. If this is scaled up from Kenya's 37 million population to the whole of Africa's 955 million, O3B might have to pay US$ 18m per year for licenses alone, adding $360m to the cost of the project over the 20-year lifespan of the satellites.

The technical model is interesting. By not dealing directly with end users, but being a bandwidth provider for communications providers, O3B enters a market with Intelsat, Iridium and other satellite operators, providing expensive bandwidth in places where none is available. Their service should have lower latency than the usual geostationary satellites, as their satellites orbit closer to the Earth in MEO (NYT claims only 120 milliseconds, compared to 500 for a geostationary satellite). However:

  • latency will still be worse than land-based connections (I don't buy NYT's claim that 120 ms is "close to fiber")
  • bandwidth will still be limited by available frequency space
  • bandwidth will have to be shared (to some extent) between users of a single satellite
  • ground stations communicating with moving satellites have spectrum efficiency problems (due to the Doppler effect)
  • ground stations with moving (tracking) antennae must switch from one satellite to another every 10 minutes, causing a short dropout
  • satellite communications require more power than microwave links, and in this case are likely to be located in rural areas with no electricity grid
  • the hidden node problem makes satellite more suitable for leased line substitution than Internet access
Latency is likely to remain a problem for voice customers, both mobile networks and Internet access, for which bandwidth demand is largely being driven by VoIP (estimated at 70% of bandwidth use in some African countries). Satellite VoIP providers apparently use different codecs with lower bitrates than standard VoIP, so popular services like Skype may not work well over satellite.

Google and other partners have so far invested $65m of the total $750m sought, and venture capital is being raised for the rest. This proves just how much this bottom of the pyramid Internet market is worth to Google.

Satellite bandwidth is already highly commoditised, and O3B's plan to reduce this from US$4,000 per megabit per month to $500 requires compelling evidence. In any case this is wholesale bandwidth, not for end users, who will still pay whatever the telco wishes to charge.

I also think that their US$ 750 million investment in satellites will be useless within 20 years of launch due to degradation (NYT claims a 10-15 year life for MEO satellites), whereas building more land capacity in Africa would have much longer-lasting benefits.

Om Malik says: "I’m intrigued by this startup because it does make sense to offer connectivity in remote areas. It also makes sense because Africa is one of the booming cellular markets and one where there is a need for cellular backhaul infrastructure. In remote areas, voice is going to be the killer app for a long, long time. The problem is that this company will always compete with fiber networks in terms of pricing, and that might put them on the back foot."

Google clearly wishes to use this project to enable broadband Internet access in developing regions, but many other things must be in place, including fixed power infrastructure, PCs or OLPCs, technical support and skills, and demand and useful content and services for areas with lower literacy, before that can happen.

Hopefully this will at least increase the spread of mobile and broadband networks by adding backhaul options (competition) and reducing the financial barriers that telcos have to overcome to deploy these networks, but this is far from certain, especially as the telcos would like to offer such services themselves, and may view O3B as a competitor more than a supplier.

See also:

Sept. 9, 2008, 10:35 p.m. - Ermanno Pietrosemoli

Probabbly some of the factor mention will be difficult to surmount. However, the latency of 120 ms is perfectly acceptable for people in developing countries, even those of us who have access by fiber are currently seeing latencies greater than that to most sites. Keep in mind that the number of hops is much more important than the fiber propagation delay in many cases. By using Low Orbit Satellites, the propagation delay will be much lower than with Geostationary ones, and there could be a very limited number of hops to a major POP. I also disagree with the limitation mentioned regarding the powering issue. The service is meant for cellular or other telecom service providers, and those will need a considerable amount of power to operate anyway. Although claims have been made of low power GSM stations that can operate economically with solar power, this has still to be proved and in case this happens, the same photovoltaic approach could be used to power the satellite link. The earth antenna might not need to be steerable to link to a low orbit satellite.

Sept. 9, 2008, 10:44 p.m. - Ermanno Pietrosemoli

I am very much excited about the possibility of using this transport technology in combination with community wireless access (WiFi extensions or even proprietary techniques) to provide a complete communication solution for developing countries at affordable cost, bypassing traditional carriers, provided that the frequency availability issue can be solved.

Sept. 9, 2008, 10:46 p.m. - chrisw

Thanks for your reply. In my view, the latency of 120 ms may be acceptable for Internet traffic, but remember that it's on top of the country's international gateway latency, which is probably also by satellite, probably geostationary, and much higher than a terrestrial link. For telephone calls within the country, it's likely to be even more noticeable. VNL claims to have developed a solar-powered base station for India: . While this could also power the satellite link, as you point out, the additional power required is likely to significantly increase the cost of the base station. You're right that I misread the FT article and the ground stations will not necessarily have moving antennae, although if they don't then the risk of them interfering with each other is much higher.

Sept. 10, 2008, 2:19 p.m. - communicall

The VNL tech site claims 100W per cell, which is certainly doable with solar, although not dirt cheap in and of itself (PV+battery+power management). What is not clear from the page is who is expected to run the 'business', which in India (and many countries) is completely licensed out (ie, the spectrum, aside from other technicalities) to large telcos. I understand the situation in many parts of Africa is equally grim. As far as the Google LEO goes, let me quote verbatim what I posted to our local community radio list, where someone had suggested this service might be good to use as an unlicensed FM community radio (FM broadcast licensing is incredibly difficult in India): "120ms for the first hop does not seem like a very sensible proposition, unless the second hop is straight to a global backbone, in which case there will be no scope in fact for local ISPs (who till now are not exactly known for their earnest efforts to reduce costs for poor Indian consumers). As far as countries like India are concerned, it (imho) makes a lot more sense to work closely with NIXI (the local IP exchange) to ensure a pure local loop for community networks sharing (open source) audio material between themselves, and thereby avoid as much of both delay (latency) and expense as possible. btw, a single 120ms hop means 240ms for direct access between any two points in India, and longer if either or both then have to traverse more networks to pass any single packet between themseves." I am afraid I do not know what the current situation is regarding supply of international bandwidth to India, but I fear the situation is pretty well locked up between three or four very large players. I would be happy to be corrected. However, having said that, the service seems directed towards Africa, so perhaps this discussion is not relevant.

Sept. 10, 2008, 8:23 p.m. - chrisw

Roland H. Alden writes: This sounds like Teledesic over again (on a much smaller scale). The subject "Google Broadband" is misleading as Google is but a small investor in this startup company which is headed up by Greg Wyler who apparently has a "mixed" track record in Rwanda (he bought the PTT and not everyone was happy with the outcome). Still he is a brave man. Just like Bill Gates and Craig McCaw before him (they were behind Teledesic, which failed). (Sorry about the title - it was hard to think of a catchy way to describe this, and I think it is Google's ultimate aim - Chris) According to the O3B website these satellites will deliver (vs. conventional geosync satellites): 5x to 10x better Internet access speeds 5x to 10x lower costs 5x lower latency So. Does anybody think this is good enough? Back when Teledesic was trying hard nobody thought so. I have recently been working on a financial justification for a submarine cable project for some mid-Pacific islands and I have two problems with the O3B proposal: In an island/state served only by satellites, and not many of them, the telco is paying ~ 850 times more for Internet bandwidth than would a terrestrial customer in a competitive market. A 10x improvement would lower that to 85 times. That still leaves a "digital divide" that is enormous. Telecom is a major input into a successful economy; who can thrive with an 85x disadvantage relative to much of the world? For Africa this seems especially absurd. Africa is not a tiny island in the middle of an ocean where this might be the only viable option (apart from a geosync solution). Africa is a huge continent with a huge coastline. There is no reason (other than Africa's own internal chattering and bickering) why Africa should not have 2-3-4 submarine cables and many terrestrial fiber projects to distribute and connect them. (Terrestrial fibre projects are not making much progress, hence Google's frustration - Chris) My second problem is that such announcement, using the Google brand as a promotional factor, given the disastrous past track record of such projects (Iridium, Teledesic, and even Project Oxygen's effort to serve developing markets with submarine fiber), is, in a nutshell, vaporware. Yes this network might get built someday and if all the numbers work out it might actually sell bandwidth for 85x what it costs in most of the world. My worst nightmare is that this announcement puts a freeze on some fiber projects that might otherwise get built. To be fair to O3B and satellite services in general they do have a role to play. A project like this one is a game changer, but in the geographic dimension only. IP EVERYWHERE is a big deal for certain applications. But let's not fool ourselves. This is high cost bandwidth that can be rapidly deployed to any location (within the equatorial belt the network serves). It will be good for humanitarian works, disaster relief, military adventures, cell phones for the super rich, cell phones for the super poor, but what it will not do is utterly transform an economy by making it a true peer in the global information space; it will not bring the kind of bandwidth that lowers the marginal cost of internet access for a school or hospital to virtually zero. Fiber is a game changer in the capacity dimension. The capacity of a fiber connection is simultaneously vast by any standard and yet perishable. Unsold capacity is "lost" in a sense that is not the case with a LEO type satellite network, where capacity that is NOT sold over Kenya is capacity that might be sold over Uganda or Dem. Rep. of Congo. Once a fiber network is built costs should drop to the point where much of the capacity is being utilized. (I thought that SAT-3 was under-used, but Anders Comstedt says it's full or nearly so - Chris) Once a global satellite network is built it will tend to seek and serve the customers willing to pay the highest prices; there will be leftover capacity that can be sold at lower prices but it is a different dynamic than fiber. You still have the militaries and global NGOs in the market and invariably raising prices for everyone. (International competition for that bandwidth may be an important factor, depending on whether the satellite network can actually be used for long haul by hopping between satellites - Chris) There is a tendency for projects like this to be promoted as "silver bullets" that will solve the entire "problem" with one simple platform and investment. They can't live up to those expectations. This project is not going to lift 3 Billion people out of poverty. I hope that when that becomes clear Google and others will get back to laying more fiber. There is no substitute for it.

Sept. 10, 2008, 9:37 p.m. - chrisw

Anders Comstedt writes: Well, seems like the entrepreneurial Greg W is at it again. This is probably of most interest for landlocked countries with toll gating neighbours. And Greg no doubt has some perspective on that. No, this is not even close to a replacement for fibre. That is just a false and misleading hope. It is just what they say on their site: A bet on fiber roll out being delayed to justify a gap filler like this one in dev country markets now taking off. Just compare the system speeds with any of the new cable systems. Orders of magnitude. So I cannot help wondering if this is not a little bit late, just as terrestrial GSM phone systems outflanked low orbit mobile phones a decade a go? But all the arguments on the site are reflecting the remaining potentially viable position for satellites. Still, within the commercial centres, like capitals and district centres, satellites will rapidly fade away as electrification will bring fibre. Otherwise you simply cannot run a modern power grid. Simple. Costs for the hopeless power gen and transmission we see today in many places will be un-sustainable. Shaping them up will provide piggy backed fibre for free. The circumstances for this process need to be discussed more. It will, however, be interesting to observe the effects any low orbit high speed satellite system will have on the various lock-in/lock-out games played on backbone connectivity in Africa. One could actually expect even more "Open Access" interest among submarine cable partners (= pushing cost sharing of the giant fixed cable CAPEX & running cost to the max) to avoid volumes (and ~@~@) being pulled into alternatives. Being stuck with a higher cost base for the submarine cable volume left in "our" cable project and not having others locked out from the end market is not a good prospect for anyone now leading the race. But the far biggest factor in this game will be the other sub-marine cable "next door". Some predatory pricing will no doubt pop up to complement the de-facto licensing shelter the current licensees enjoy. That brings us back to the InfoDev paper. Like many of the papers coming out from that mill it is carefully avoiding drilling through the ice in its analysis. Too bad as the authors are no doubt clever people with an intellectual and policy background capability. But if you claim in the title that you are going to cover the policy aspects on why things are the way they are you simply need to puncture the outer shell.... Or are there a PartII coming: Conclusions, Reasons for slow change, Recognising Bogus Arguments and Acid tests on Policy Change?  In a growing African telecoms market, with telecom companies turning to be the single biggest taxpayers in some places, demand, not least from fast expanding Internet traffic, is now commercially justifying leaving VSATs and microwave links behind for fibre. Financing is no problem. Policy/political risk is. Not least is this true for the small operators overshadowed by the established, politically well connected players. Yesterday it was the PTTs, now it is a handful of pan-African mobile phone operators. Interestingly enough, the report makes reference to important policy discussions elsewhere, like OFCOM spectrum liberations, where their officers expressively say on the starting point that is "Not our objective to maximise Exchequer revenue" but "maximise consumer and citizen value (social welfare)." How often is the opposite not the case in African discussions? The report should have poked more into crucial indicators to serve a much needed purpose of helping local journalists and others understand the policy issues, some of the newspeak and what perspective that really controls change. Looking forward to see a 2.0.

Sept. 11, 2008, 4:26 a.m. - communicall

wrt Roland's comments: As much as the technicalities, I perceive the most significant drawback of such proposals as the lack of local involvement. Which has also plagued the existing Worldspace radio/data network, originally intended to 'worldchange' information delivery in totally underserved areas, and today mainly serving 'elite' audiences. This is exactly the problem with deployment strategies for EVDO/UMTS and WiMax. All are high-ticket technologies, whose implementation can only be handled by major asset possessing entities. Almost by definition, that does not include local players. On the other hand, technologies like WiFi have evolved from primarily small local area networking to include 'respectable' distance (15-25km very doable in flat terrain with low learning curve) point-to-point broadband, thus making it adaptable for more than first-mile delivery. Deployment can easily be handled by local small businesses or communities. Of course the total cost involved for complete continental coverage will rival or exceed that of a LEO project (but will that be true if the SAT ground nodes are fine-grained across everywhere people live?). However, the important difference is that the large bulk of that money will be spent upfront locally, and, critically, ongoing maintenance costs will be almost totally local. In contrast, O3B is going to earn its keep outside Africa.

Sept. 11, 2008, 2:03 p.m. - chrisw

Anders Comstedt writes: SAT-3 may be filling up actually. At least there is no room for price reductions as price elasticity would oversubscribe capacity in a snap. Together with other people we have looked into that for comparable markets. Q: So why is Telkom not rolling out a new cable ASAP? A: Politics in RSA + maximising value as long as possible. they would not be alone in a new cable. (I heard they were upgrading the existing cable instead, which implies that there's unused capacity, and that profit rather than capacity is being maximised, as we'd expect from any good capitalist - Chris)

Oct. 30, 2008, 4:10 p.m. - O3b Networks: The Other 3 Billion - Good cause,...

[...] broadband venture InternetHypergate (1999) and the recently announced O3b Networks’ plans. This blog, by IT specialists Chris Wilson and Jim Jepps, hits the nail quite well on its [...]

Jan. 28, 2010, 7:56 a.m. - Dorthey Schadegg

Very informative post. Thanks!