What one thing would you change?

[Ben]

Think about your primary mobile network. What one thing, above all others, would you change?

Why?

Perhaps it concerns tariffs, coverage, customer service, or something else altogether, but find the one thing that matters most to you. Don't even think about telling us your network is perfect! ;)

[Ben]

My primary network is still Vodafone.

The thing I would change? Their approach to mobile data.

Why? Vodafone have the resources to be a leader. Their transition to 3G started strongly, and while coverage was slow to spread performance was always sound. On a handset, even GPRS was a welcome improvement over the O2/iPhone data experience. Vodafone boasted of the strength and resilience of their network in the face of O2's problems. Yet since the iPhone, and other smartphones, came to the Vodafone network in great numbers things have deteriorated severely. They appear to have slipped in mindset from one of leading the way, to one of maximising profits from existing infrastructure and leaning heavily on their brand and reputation. 3G network growth feels to have stagnated, and coverage and performance compared to once-unlikely mobile broadband champion Three is embarrassing.

I can see a few possible reasons for their current method of operation. When 3G launched, Vodafone had Vodafone Live! with 3G, the holy grail that was going to have us all spending hand-over-fist in order to access the latest video clips and premium content. That spend would be the return on investment, paying for the 3G license costs and network upgrades many times over. Unfortunately it never materialised. Now they're faced with data demand, but no sign of increased revenue. Once bitten, twice shy, Vodafone now appears to be unprepared to look for new revenue sources made possible by an upgraded network. But it's desperately short-sighted in failing to see the other side of the coin. Better networks will be required to retain revenue, not just grow it. In five years time, a network that fails to live up to the promise of true broadband mobility wont be worth its salt to any but a few. Investment and innovation need to come now, and Vodafone will pay a heavy price if it continues along the road O2 took before its sale to Telefonica; pushing its infrastructure to the brink in order to maximise profitability while failing to plan for the future.

[DBMandrake]

I'm currently with Three (pay as you go only) and the one thing above all else I would change about them if I could would be for them to roll out ubiquitous 850/900Mhz 3G coverage in addition to their current 2100Mhz coverage.

The extent of their 3G coverage on 2100Mhz is quite impressive compared to the other networks, but now that 2G roaming has been turned off in large portions of the country their total coverage footprint has fallen dramatically, and I would wager that from the perspective of voice/sms they now have the worst coverage of all the networks. (Not that the difference is huge, but it is certainly there - there are many places I can't get a signal on Three but can on other networks)

The thing is, I don't want a return to nationwide 2G roaming on Orange - we've been there already, and although the summed coverage of the two networks was excellent (especially due to the redundancy and overlap of the two networks) it does defeat Three's main reason for being - 3G data coverage. (Not to mention costing Three a fortune on roaming charges to Orange)

The inability to roam between the two networks mid call, or even hold an active data session when roaming was a big pain in the butt, so from that perspective I'm glad roaming is down to a minimum now, but it does open up a lot of annoying coverage holes.

In one fell swoop Three could fix this by rolling out 3G coverage on 850 or 900Mhz. With a combination of existing 2100Mhz signal providing capacity in densely populated areas and 850/900Mhz signal providing both increased in-building penetration in populated areas, as well as far greater range in the country, rural areas, and for motorways. Such a network would have better 3G coverage than the current Orange/T-Mobile 2G network on 1800Mhz, and nearly as good coverage as the 900Mhz 2G of Vodafone and O2.

900Mhz is never going to happen because O2 and Vodafone are the incumbent holders of those frequencies and will no doubt hold onto them for grim death, since the reality is they don't have enough even now to do both 3G and 2G on 900Mhz justice, let alone after selling some.

850Mhz is the only other standardized 3G band in the low frequency region which has handset support, thanks to its predominance in the US market as well as some Asia Pacific markets.

The problems I see are:

(1) It's not at all clear to me whether the upcoming "800Mhz" spectrum auction includes the 850Mhz range as a subset, or whether it is only 800-850Mhz. (Does anyone have a link to some hard facts on whats up for auction ?) If it only includes 800-850Mhz, there are little or no current handsets which would support 3G in this range.

(2) If the 850Mhz band is included, it's not clear to me whether it would be auctioned on the sole condition that it's used for LTE, or whether it's up to the winning bidder to decide which technologies to use on it, (2G/3G/LTE etc) and whether they have a license to shift from one technology to another at their discretion.

(3) Even if a network such as Three was allowed to put 3G on 850Mhz as well as or instead of LTE, at least in the short term, would they actually do it ? Would Three just buy frequencies for LTE, start rolling out an LTE network, and leave their 2100Mhz 3G network as is - a reasonably decent network, but one which would then forever have coverage holes and difficulty penetrating buildings.

I know 3G is an "old" technology, but I'd be really disappointed to see Three use any frequencies won at the auction exclusively for LTE, thus condemning their current 3G customers to spotty coverage, rather than putting some of those frequencies to use to bolster and finally "fix" their 3G coverage issues. This is particularly important now that O2 has rolled out considerable 900Mhz 3G coverage already, and Vodafone has the option to do so if it so chose.

3G and LTE can work side by side if the block of frequencies allocated is large enough. UMTS requires a 5Mhz channel to operate, while LTE has a variable channel size that can be adjusted between 1.4Mhz to 20Mhz. Of course maximum theoretical throughput requires a full 20Mhz channel, but it's unlikely that an initial roll-out would use that.

For reference, current 3G allocations in the 2100Mhz band are 15Mhz for Three and Vodafone (3x 5Mhz channels each) and 10Mhz each for the rest. So even if a network was only able to win 15Mhz of spectrum in the new auction, they would have the opportunity to use it either 3x 5Mhz 3G, (unlikely) 2x 5Mhz 3G + 5Mhz LTE, 1x 5Mhz 3G + 10Mhz LTE, or 15Mhz LTE. The split in spectrum allocation between 3G and LTE is something that could be altered at a later date as capacity needs shift - a significant chunk for 3G now when hardly anyone is using LTE, transitioning across to a larger chunk on LTE and smaller chunk on 3G later as the technology becomes obsolete.

Perhaps I'm just dreaming though, we'll have to wait and see what they do, but perhaps it would be worth it for Three to make use of a small amount of gained frequencies to improve 3G coverage, so that they can finally switch off 2G roaming entirely - the cost savings may well be enough incentive for them to do this.