5 ms·

Or they could really go for it and buy a carrier. Perhaps Sprint might start looking tasty? What I would really love is if they innovated in the Wifi / unlice

by zmmmmm 16y ago

Or they could really go for it and buy a carrier. Perhaps Sprint might start looking tasty? What I would really love is if they innovated in the Wifi / unlicensed spectrum space and made a p2p type mesh system. Then even if you don't have Wifi you can piggyback on nearby handsets to "hop" to a Wifi point. This would require deep integration into the handset and opt in to make it totally transparent to users. Skype seems to be about half of this equation and maybe deep collaboration with Nokia could be the other half. Having said that, people are so used to Wifi working well that they don't realize what the limitations are once you try and scale it into different situations. Start moving more than 5mph and you'll suddenly see what all that technology in "real" cell networks is there for.

ohyes 16y ago
Using the wifi/unlicensed spectrum is also a lot more efficient use of the spectrum. You can broadcast shorter distances and therefore squeeze more users onto the spectrum. It would be logical to see major cell companies doing something similar (more towers with shorter ranges) in urban areas. It would give them much greater available bandwidth as they try to scale to a situation where more people have data plans. (I don't think that they will, because the telecoms can simply charge more for less using weasel words in contracts, most people won't know any better).
- rbranson 16y ago
  It's already hard enough to run a reliable mobile data network built with industrial-grade systems on exclusive spectrum. Unfortunately, moving to a network like you describe, built on commodity gear doesn't translate to quality, redundancy, and scalability like it does with distributed systems. Major cell companies are already doing what you describe. Example: both Verizon and AT&T have a fleet of mobile cell towers mounted on the back of trucks they park at large events if they think capacity is going to be an issue. If you look around at building tops in cities, you'll see arrays of panel antennas on many of them. These are for cell networks. Every site is intentionally designed and every single antenna is specifically angled and tuned, and their network engineers collect quality statistics from every call at every site, which are fed into massive data warehouses for quality analysis. They literally have fleets of people like "the Verizon guy" continuously drive testing their networks. If anyone knows how to run mobile data networks, it's them. Seriously, network engineers at companies like Verizon are treated like software engineers at Google and Facebook. They keep the core business running and they are the rockstars. Everyone else, including the jerk in customer service, is basically a nobody. From what I can tell, there is a sort of strange, pervasive myth on the Internet that the large cellcos have their heads up their asses and that somewhere, somehow, there is a kind of wild band of rebels with hacked 2.4GHz/5GHz mesh network gear and cantennas doing it for dirt cheap. Please show me even the smallest evidence that anyone, anywhere is running a true mobile data network using a mesh network on commodity gear. This means continuous coverage for a metro area with low power, cheap, antenna-integrated MOBILE consumer gear, seamless site-to-site and channel-to-channel handoff from users moving at 80mph, high quality indoor coverage in large buildings that is basically taken for granted, ramp-ups/downs of 10k+ customers within hours (sports events, etc)... and on and on. You'll find a few WISPs doing fixed site-to-site links and (poorly) blanketing areas with WiFi coverage, but nothing on the order of what cellcos do. The fact that one can hold a uninterrupted voice conversation using a $5 phone with no visible antenna while driving down the interstate at 75mph for hundreds of miles is a pretty awesome modern marvel.
  - zmmmmm 16y ago
    Yep, I think that the prospect of trying to achieve real cell phone like qualities out of mesh networks is a no hoper. The real question however is what the end consumer will accept. Often times products are over-engineered / overpriced (whether deliberately or accidentally) beyond what consumers really demand and it's not until a low quality competitor shows up that everyone realizes it. Think how the vast majority of consumers turned out to be happy with low bit rate mp3s after recording studios spent years pushing CDs, and how very few people really care about HD vs SD very much (at least, not on smaller screens). So perhaps there's room for a network that just doesn't work while travelling at 75mph (or even 30mph). So what? Maybe consumers just don't care that much if they are getting it for free.
    - rbranson 16y ago
      The 75mph isn't really the hardest part of that. Really, it's indoor coverage with dirt cheap consumer gear that's hard, followed closely by call and data connection handoff. Even if it's free, there's a minimum amount of service quality to even be considered useful, and I think handoff and indoor coverage are both essential for mobile wireless to work. In modern networks, handoff is also deeply tied to availability in high density areas.
- yxhuvud 16y ago
  I know for a fact that several european carriers have started to implement this. However, the technology to do it isn't mature. Hell, we won't be able to deliver this exact thing to a certain carrier I won't mention until late autumn if everything goes well. Edit: Oh, and by the way, the term for it if you want to google is 'Mobile Offloading'.
  - timthorn 16y ago
    UMA has been live for several years and is already quite widely deployed in the UK, allowing transparent offloading of voice to WiFi.
brudgers 16y ago
>"Or they could really go for it and buy a carrier. Perhaps Sprint might start looking tasty?" Purchasing a carrier doesn't scale. It doesn't scale to Europe. It doesn't scale to Asia. It doesn't even scale to Canada or Latin America for crying out loud. And scale is why the Skype purchase makes sense to me - it scales into all those markets (and more) in a way that Google Voice does not appear to be headed.
- sahaj 16y ago
  Imagine for a second that Google did buy Sprint and then brought it up to speed as far as wireless coverage that matches ATT/VZ while offering the best in class service for the price. Masses move over to Sprint in a few years. Sprint becomes bigger than ATT & VZ combined and now others around the world have a model they can follow. Too far fetched? Maybe. Costly? That too. I guess I'm just a dreamer.
dsmithn 16y ago
Nobody wants to give away their handset's limited battery life. Imagine when you have one handset within range of the AP, and a number of handsets trying to connect through it. It's bandwidth and battery will be zapped up.

by zmmmmm 16y ago

ohyes 16y ago

Using the wifi/unlicensed spectrum is also a lot more efficient use of the spectrum. You can broadcast shorter distances and therefore squeeze more users onto the spectrum. It would be logical to see major cell companies doing something similar (more towers with shorter ranges) in urban areas. It would give them much greater available bandwidth as they try to scale to a situation where more people have data plans. (I don't think that they will, because the telecoms can simply charge more for less using weasel words in contracts, most people won't know any better).

rbranson 16y ago
It's already hard enough to run a reliable mobile data network built with industrial-grade systems on exclusive spectrum. Unfortunately, moving to a network like you describe, built on commodity gear doesn't translate to quality, redundancy, and scalability like it does with distributed systems. Major cell companies are already doing what you describe. Example: both Verizon and AT&T have a fleet of mobile cell towers mounted on the back of trucks they park at large events if they think capacity is going to be an issue. If you look around at building tops in cities, you'll see arrays of panel antennas on many of them. These are for cell networks. Every site is intentionally designed and every single antenna is specifically angled and tuned, and their network engineers collect quality statistics from every call at every site, which are fed into massive data warehouses for quality analysis. They literally have fleets of people like "the Verizon guy" continuously drive testing their networks. If anyone knows how to run mobile data networks, it's them. Seriously, network engineers at companies like Verizon are treated like software engineers at Google and Facebook. They keep the core business running and they are the rockstars. Everyone else, including the jerk in customer service, is basically a nobody. From what I can tell, there is a sort of strange, pervasive myth on the Internet that the large cellcos have their heads up their asses and that somewhere, somehow, there is a kind of wild band of rebels with hacked 2.4GHz/5GHz mesh network gear and cantennas doing it for dirt cheap. Please show me even the smallest evidence that anyone, anywhere is running a true mobile data network using a mesh network on commodity gear. This means continuous coverage for a metro area with low power, cheap, antenna-integrated MOBILE consumer gear, seamless site-to-site and channel-to-channel handoff from users moving at 80mph, high quality indoor coverage in large buildings that is basically taken for granted, ramp-ups/downs of 10k+ customers within hours (sports events, etc)... and on and on. You'll find a few WISPs doing fixed site-to-site links and (poorly) blanketing areas with WiFi coverage, but nothing on the order of what cellcos do. The fact that one can hold a uninterrupted voice conversation using a $5 phone with no visible antenna while driving down the interstate at 75mph for hundreds of miles is a pretty awesome modern marvel.
- zmmmmm 16y ago
  Yep, I think that the prospect of trying to achieve real cell phone like qualities out of mesh networks is a no hoper. The real question however is what the end consumer will accept. Often times products are over-engineered / overpriced (whether deliberately or accidentally) beyond what consumers really demand and it's not until a low quality competitor shows up that everyone realizes it. Think how the vast majority of consumers turned out to be happy with low bit rate mp3s after recording studios spent years pushing CDs, and how very few people really care about HD vs SD very much (at least, not on smaller screens). So perhaps there's room for a network that just doesn't work while travelling at 75mph (or even 30mph). So what? Maybe consumers just don't care that much if they are getting it for free.
  - rbranson 16y ago
    The 75mph isn't really the hardest part of that. Really, it's indoor coverage with dirt cheap consumer gear that's hard, followed closely by call and data connection handoff. Even if it's free, there's a minimum amount of service quality to even be considered useful, and I think handoff and indoor coverage are both essential for mobile wireless to work. In modern networks, handoff is also deeply tied to availability in high density areas.
yxhuvud 16y ago
I know for a fact that several european carriers have started to implement this. However, the technology to do it isn't mature. Hell, we won't be able to deliver this exact thing to a certain carrier I won't mention until late autumn if everything goes well. Edit: Oh, and by the way, the term for it if you want to google is 'Mobile Offloading'.
- timthorn 16y ago
  UMA has been live for several years and is already quite widely deployed in the UK, allowing transparent offloading of voice to WiFi.

brudgers 16y ago

>"Or they could really go for it and buy a carrier. Perhaps Sprint might start looking tasty?" Purchasing a carrier doesn't scale. It doesn't scale to Europe. It doesn't scale to Asia. It doesn't even scale to Canada or Latin America for crying out loud. And scale is why the Skype purchase makes sense to me - it scales into all those markets (and more) in a way that Google Voice does not appear to be headed.

sahaj 16y ago
Imagine for a second that Google did buy Sprint and then brought it up to speed as far as wireless coverage that matches ATT/VZ while offering the best in class service for the price. Masses move over to Sprint in a few years. Sprint becomes bigger than ATT & VZ combined and now others around the world have a model they can follow. Too far fetched? Maybe. Costly? That too. I guess I'm just a dreamer.

dsmithn 16y ago

Nobody wants to give away their handset's limited battery life. Imagine when you have one handset within range of the AP, and a number of handsets trying to connect through it. It's bandwidth and battery will be zapped up.