F2C: Open mobile and wireless

Rich Miner

Google Android. Group manager mobile products. Open handset alliance of 34 partners to help build Android, an open handset platform, based on Linux and write programs. Also third party dev environment, along with $10m competition to award prizes to best applications built for that environment.

Mobiles – smartphones – have power of a desktop PC in 2002. Mobiles fall far short of the things we were doing on 2002 in terms of ability to connect and delivery communications and multimedia experiences. Huge gap between what the hardware can do and what the software delivers. Mobile ecosystem is very closed, in terms of platforms which are closed, and even the ‘open’ OSs are closed, like Symbian or Mobile Windows. Called open because it has APIs, but if you’re trying to create applications then it’s not open. Level of control these platforms maintains prevents carriers, individuals and others form innovating.

There’s typically an arcane process between developers and the users of the applications, so huge hurdles that the dev has to get through to get his app to consumers. If you’re a software dev and you realise that your destiny is controlled by onerous testing requirements of the platform owners. Same with content.

If you want to distribute video of a cat playing with yarn, you can do that online. You can be votes best video, and you can rise to the top. In mobile industry, there are walls that prevent people making stuff available.

Google knows that mobile is important. Many mobiles are people’s only computer, they may never own a laptop or desktop. Problem with existing platform is that the process of developing and distributing apps is too onerous. So building own platform, Android, which will open the mobile space and will provide significantly more innovation.

Built it so that there’s a large ecosystem, brought a large group of people together to open source the platform. There are other open platform initiatives, but realised that none of them were unifying everything, or providing everything you’d one for a complete phone. Google’s invested significant resources.

This isn’t about a single GooglePhone, but about an entire range of phones, it’s not to hold up a phone to the iPhone, but it’s a platform that hopefully the manufacturers will build phones on. People also want to build on top of the Android platform. Goal is to break down these closed platforms, and that you enable a more open handset ecosystem.

This will allow people to put applications out there and people will be able to choose what they want, and rate how good they are. Want more innovation.

If you look at what Linux delivers, but to say you’re going to take Linux and build a mobile consumer device isn’t the right way to do it. Linux isn’t really even a consumer computing OS yet. Need to put a bunch of dedicated software layers on top of Linux to do that. Linux is a lower level, but want high-speed 2D/3D graphics, tools for Java apps, all the layers of software you need on top of Linux. This software’s either written by Google or written by experts in the community.

Our goal is not to inhibit innovation, so using Apache 2.0 software licence, so that brands can build software and not have to open source it. Don’t want to put them off.

Android is still under dev, but handsets will ship second half of this year, and then the software will be open sourced. So it’s not being developed as an open source project, but is going to be made available when handsets are.

Carriers look at Google as competitive, working with them to get them to understand that they can make them lots of money as a partner, and are not competitors. Carriers can take the OEM and brand it, do whatever they like with it, can build a tightly branded handset. Can even build locked phones on it, and that’s ok.

But since announcing Android, the message of openness is resonating in the industry. There are media battles to be seen as more open. And once consumers understand what being open means, they won’t accept it when that’s taken away. They will value openness over closedness. Google happy that message of openness is so well received.

Michael Calebrese
Debate of consumer’s rights to run whatever they want on wireless. Open access to the airwaves needed. 700 mhz spectrum auction just finished. More spectrum means more competition? No, oligarchy. No new entrants. DSL duopoly got exclusive rights to >90% of the spectrum, and got the best bits too.

Licences can be conditioned by gov’t. Propose conditions that might guarantee new entrants, so that new ISPs etc. could get good quality bandwidth. FCC adopted more limited open access condition. Verizon fought conditions, and then announced open development, thus qualifying to take valuable chunk of network.

Larger point is that general public won’t have an alternative to the carriers unless we have more open spectrum. Need to go back to early days of Hoover, in early 1920s, before he imposed licensing for CBS and RDA, before the radio act, most small radio stations were sponsored by local groups and they tried to share the airwaves. But interference caused regulation.

Exclusive licensing and auctions has resulted in capacity being wasted. Auctions assume we have scarcity. But what’s scares is government licences, not spectrum. 95& of spectrum is not being used at any given time.

If we want open wireless, we need open spectrum. smart radio tech, cognitive radios, could utilise underused spectrum. 49 channels are reserved for TV, even though 7 are used in any local market. Even after the digital TV transition, there will still be a lot unused.

But access to TV whitespace is a starting point. Concept of whitespace needs to be broader. Much more wasted capacity wasted in bands that are licensed to governmental bodies, such as forestry or military. Military are much more open to sharing, because they know that they can use whitespace to sniff out spare spectrum anywhere in the world.

Soon, edge devices will be able to share any unused capacity. Need to think beyond passive sensing tech, particularly when looking at gov’t bands. Last world radio congress, asked for study about how beacons to broadcast data about spectrum environment to make it easier for devices to grab unused spectrum quickly and efficiently.

As TV whitespace becomes available, cost of spectrum becomes no longer a barrier to entry. Will also provide rocket fuel for community wireless networks, such as mesh wireless.

Although we initiated this debate, what will be critical will be degree of open spectrum.

Richard Whitt
Google This is a really important proceeding at the FCC right now, unique opportunity, supporting the tech and public, to create momentum for public use of unlicensed spectrum. Lots of spectrum. Taking on entrenched incumbent, broadcasters,m who don’t want to see anyone come into “their space” and utilise these spectrum band. Lots of misinformation, but they are a powerful and pervasive presence.

Another camp entered, the wireless carriers, CTIA, including T-Mobile and Sprint. Third front is the wireless microphone chaps, who think there will be interference problems.

Testing phase for devices from people like Microsoft and Philips, but test devices not functioning properly. Most recent tests shut down, so whilst that’s really supposed to prove the concept, but the opponents are ;seizing on that as proof that it can’t work. Battle of scientific fact, but also it’s a PR battle, and a political battle. Like to think that the good guys can wine, but if you can get involved, send and email, send letter, write to your representatives, got to the FCC. This has to be a grassroots effort, not just a top down. If you think this is the right cause we need to pull together. Need to make whitespace something everyone can use.

Brett Glass
Runs a wireless ISP in Wyoming. Founded Lariat, wireless broadband ISP, and did it because he had to. Wanted to live in the countryside, and university was the only thing in Wyoming that had decent connection to the internet.

Got together with businesses, pooled money, got a T1 line. World’s first wireless broadband provider. Started as a non-profit, and were besieged by members to take it private because they wanted to have investment. Wanted someone else to take care of their ISP needs. Now a rapidly growing ISP. Serving areas unlicensed by other service provided. Mount a radio on the roof. Cable won’t go to these places, neither will ISPs. But they will.

Fibre isn’t economical, it’s got to be wireless. Why aren’t indie wireless ISPs more well known? Reasons not growing so fast – upstream issues. prices are increasing for any small carrier to do business with the backbone operators. Backbone operators turning into local monopolies, which won’t co-operate. Triple price of the bandwidth, which triples price to customers.

If gov’t wanted to do something, it would be to require the backbones to open up in smaller locations.

Also problems with wireless spectrum auctions. Insufficient granularity – have to buy half a state not a county. Timing issues. Big upfront payments. Prevent new entrants, prevent small carriers getting in there.

Net neutrality stuff could also make them not competitive anymore.

And P2P concerns. Why is it that Comcast are suddenly throttling back P2P. Can get a file either from the client server, goes to ISP, then to local loop to user. ISP backbone connections are pretty expensive, but have to pay it to get content to the customer.

P2P clients start using bandwidth from ISPs, which is more expensive than it would be if the content was sat on a colo somewhere.

Does restricting P2P limit free speech? But any content or service you can get through P2P you don’t have to get through P2P, could provide it another way that doesn’t hit small ISPs.

In Japan, increased bandwidth available to the home. P2P grew to be more than 70% of traffic. Adding capacity doesn’t solve the problem, because it will just fill up. Need to manage network and make it more efficient.

Glen Strachan
Has worked in Romania and Uganda, connecting schools with funding from US Aid. Macedonia, wanted to provide internet to all schools in the country. Went to Macedonia, found that broadband were minimal, only available in major city. 120k account counted as ‘broadband’. 2mb for €10,000.

Not enough to provide connectivity to schools, but also need to think about regulatory environment. Monopoly part-owned by government. Opened up market using schools as anchor. US gov’t paid to go to run open competition for an ISP, could only use a wireless ISP. No money could be spent on a monopoly so had to be wireless.

Market opened in Jan. Four vendors bid, winning vendor and signed a contract. paide them $2.5m for connectivity, just bought services, not equipment. They built out their own network, and had to put up same amount of money. Incentive money only for rural areas. Network had to be built between April til September when Schools opened. Was up by August. Covered all schools.

Goal was to create a competitive atmosphere that would reduce the prices so schools could afford it. Internet penetration was 4%, now is 34% three years later.

Macedonian gov’t has purchase 180k computers for schools. Price for connectivity is €10 for 9gb, €25 for unlimited.

Now doing same thing for Montenegro. Smaller country. Finished up in Montenegro. Working on Senegal, who want to use the Macedonia Connects model, but Senegal has no regulatory reform so will play a vital part. Also lots of corruption in Montenegro so reform was required there.

US Aid has allowed the creation of a model that can be replicated in developing countries.

[Again, brain too slow to deal with the questions. Sorry. Heath Row is taking fab notes too, rivalling me for speed and verbatim transcription. He’s also got the open fibre session notes that I missed. I should find him and say hello just to compare notes on taking notes.]

Technorati Tags: