Speaking Engagements & Private Workshops - Get Dean Bubley to present or chair your event

Need an experienced, provocative & influential telecoms keynote speaker, moderator/chair or workshop facilitator?
To discuss Dean Bubley's appearance at a specific event, contact information AT disruptive-analysis DOT com

Thursday, December 11, 2008

The danger of "cutting the cord" - where's the femtocell going to go?

I see many references to people wanting to "cut the cord", and have an all-mobile home. This is particularly the case in the US, where it is positioned as a sort of aspirational goal by some, while in Europe it is (often) just something that applies to people either living on their own, or who are in lower socio-economic groups.

There's also another big difference. When Americans say "cut the cord", they usually don't mean "cut the cable TV cord as well". They just mean the copper telephone line. But in markets which are ADSL-biased for broadband, that's not a realistic option for most household. Instead, they may be able to use an "unbundled local loop", and keep their fixed broadband, but get rid of their PSTN telephone subscription.

However, with the rise of mobile broadband, many operators' marketing teams are trying to get rid of ADSL connections as well. Obviously in markets with fully-converged operators and quad-play, that's less likely, but for mobile-only operators, they're possibly storing up trouble for the future.

At some point, the operator is possibly going to want to deploy femtocells, WiFi, or some other offload approach - especially if you're a heavy "mobile only" user. And at that point, the lack of an existing broadband connection is going to be a problem.

Not only that, but your copper line will be disconnected at the exchange - and so even if they want to offer you a new fixed+mobile package with a home gateway including a femto, someone has to pay for it to be reconnected and tested. It's even possible that they'll need to send someone to your house to check the wiring still works OK. (When I moved into my current house 2 years ago, I needed an engineer to reconnect everything & install a new socket, before I could get ADSL provisioned).

So for operators, although cord-cutting sounds like a great way to get more fixed-mobile substitution in place, there's a longterm downside with regard to future flexibility for macro network offload.


Paul said...

Femto shmemto. 18 months after every CMO/CTO + dog said they would offer 3G Femto, there's 1 Huawei supported launch, and only 4 users on the white list is going to cause white spots as well as interference?

Can you explain to my why a Fixed Wireless terminal and UMA subscription does not offer me all I could wish in 1 tidy box? Voice, internet, home coverage enhancement?

HSPA in, WiFi and POTS (and possible IP FAX) out and probably a print-server too. Run UMA on the box with something like a SE G705u (2.4" HVGA, HSUPA, GPS, DLNA). 5 minutes after getting home from the shop, I'm on-line.

If anything goes wrong with my internet I have 1 provider to contact. If I lose DSL, I have ISP, BT, modem provider and possibly separate WiFi router in the path.

Your talk of having to reconnect copper in the future is the sort of scare-mongering that's kept BT in the doldrums for the past decade.

Dean Bubley said...


Femtos are currently less delayed than UMA was. Nobody I speak to in the industry is at all surprised with the current timeline, except maybe some journalists and analysts who underestimated realistic timelines.

Fixed wireless isn't viable in most places, especially dense urban areas - there's not enough capacity with typical frequency allocations and cell-site locations. On the other hand, I expect it to be deployed in areas with no copper/cable/fibre, either as WiMAX, HSPA or something proprietary.

It doesn't need UMA, either - that's just one of several different approaches to doing VoWLAN. It will be interesting to see how long UMA is supported by handset vendors, given its tiny addressable market of maybe 3-4m devices per year.

In urban areas, there needs to be copper, coax or fibre for any serious massmarket connectivity. There is no alternative.


Paul said...

Hi Dean,

I'm talking here as a consumer - what do I gain with a Femto, that I do not gain with a FWT providing UMA? I gain single handset, coverage and I gain a location-dependent service that can be flat/zero-rated. That location can go with me too, so I can move house or go on holiday without losing service.

For out-of-cable territory, I see it as an excellent Converged offering. To the consumer it looks the same as what you already have. Telling someone that in order to converge they have to use Skype is not FMC/FMS.

As regards handsets, WiFi is becoming more standard these days, and it's a fairly small step to add the GAN stack onto the handset. The BB 8900 Javelin is another new handset supporting UMA.
On the other hand we could wait for femto-optimised R8 handsets, with AT&T proudly anouncing Femto trial next year.

Stefano Quintarelli said...

can you give me your opinion on this ?

Dean Bubley said...


As I said, UMA is just one of many ways to use WiFi. Skype is another, but there are plenty more operator-centric ones.

One of the major failings of UMA in this scenario is that it means that the home phone is now a mobile number. Not an issue in the US where they're indistinguishable, but a big deal everywhere else, particularly because of the high costs of call termination for inbound calls.

I'd expect to eventually see WiFi in about 20% of phones, maybe 30% in a few markets like US & Scandinavia. So ultimately, the femtocell should give a better choice of phones - although bear in mind my views on femto-optimised handsets.

It may not be much to add GAN to the stack, but it's non-zero in times of both cost & time-to-market. For a feature only useful to 0.3% of the world market, it's highly questionable in terms of value.

In any case, your scenario is pretty unusual from a global perspective. Given where you live, and the services available in the US, you may well have the best option at present. But that may just be a reflection of poor competition & choice.


Nadeem Akhtar said...

The use of ADLS or some other form of 'out-of-band' backhaul for femtocells makes it a non-starter for countries where penetration of fixed broadband is quite low. India is an example in this regard because very few people have any form of broadband connectivity. 3GPP has rightly started focussing on relays with in-band backhauls which can serve a similar purpose to some extent.