Mobile Crunch has a good primer on the standards surrounding UMA, the protocol for managing hand-offs between private infrastructure WiFi networks and mobile cell networks.
But I pose two questions–(1) how important is seamless hand-off to the end-user, and (2) how important is it to the carriers?
1. If you have to “manually” flop to private infrastructure, how much of an inconvenience is that, really? I volunteer that, as WiFi networks become the standard and not the exception, less and less of us will be worried about making sure the carrier has appropriately handed us off to IP. I argue further that eventually CDMA/GSM/GPRS access will be something we only use “when we have to”. Especially in the cities, where pervasive WiFi is happening now.
2. Of course the carriers have no interest in handing off billable minutes to WiFi. Unless they plan on getting into the ITSP game, what benefit is there to a T-Mobile or a Cingular if you’re jumping off their bandwagon to use your choice of WiFi networks? None.
There now, go read about UMA and see if you think it ranks on your importantometer.
Ah… I love a good UMA debate:
#1 covers a lot of points
Seamless handover: Is it a mobile phone or a cordless phone? These are different applications/services. You can buy Wi-Fi phones today with no mobility, but that’s not a service a mobile operator wants to deploy. Cellular is about seamless mobility, your call doesn’t drop between cell towers and you can take your phone to any GSM country in the world.
Seamless handover, part 2: When you “flop†to private infrastructure, how does a mobile call get to you? If someone calls your mobile, the mobile operator needs to deliver that call to you. If your phone has no connection to the mobile network, the call goes to v-mail. Mobile phones are popular because one don’t have to give out a home #, desk #, Skype address,… The mobile network can find me anywhere, and with UMA, it can find me on Wi-Fi for a fraction of the cost.
Using mobile “only when we have tooâ€: I completely agree. I can see an operator rolling out a plan like $10 unlimited calling over WiFi/UMA, with a per minute charge for the GSM network in no time.
On to #2…
Of course operators have no interest in handing off billable minutes: That’s a true statement. And that’s why UMA is actually linked to the mobile core (see handover part 2). The operator many not bill for the minutes on WiFi/UMA, but they certainly have a CDR showing that a call was made. If the call leaves their network all together, they lose the customer. Not cool (if you’re a mobile operator).
What benefit is there to T-Mobile and Cingular: Does anyone disagree that voice is a commodity? And does anyone disagree that over time, the cost per minute of voice is going to approach $0.00 (Skype). So if you’re T-Mo you see this market inevitability. Also you see that anywhere from 30-50% of mobile usage today is in the home or office, in places where Wi-Fi exists already.
So let’s do the math: 1,500 minutes for $40 (an actual T-Mo promotion). Being generous, let’s say it costs T-Mo just $0.01/min to deliver a call over GSM. That’s $15.00 of the $40.00 going to network access fees (and they still have to pay off that handset subsidy). But if they can offload 50% of those minutes to Wi-Fi @ $0.001, now T-Mo only spends $8.25 ($7.50 + $0.75) on access network fees. They just reduced their network OpEx by 45%.
To conclude, if you believe voice is a commodity, and you believe the cost of voice is going to $0.00, it isn’t a question of IF operators will do UMA, the only question is how much do they have to lose before they realize they MUST do it.
If you’re a mobile operator, UMA ranks high on the “import-o-meterâ€. If you’re a subscriber, all you want is your mobile phone bill to head to $0.00. I think everyone is in synch.
PS – Full disclosure, I work for Kineto, the leading supplier of UMA technology.