This month, BT's proposed "Bluephone" project has become clearer. When it was first talked about in 2003, the proposal was to save companies money by having a single handset that could route calls over the fixed or mobile network.

The technical issues are complex, and the commercial side is not straightforward, as it requires a mobile provider to get involved. The details and pricing are still under wraps, but last week BT announced it had attracted Vodafone into the partnership. What remains to be seen is whether it will also wow the potential users.

There will be extensive trials during the summer, and the full service should be launched before the end of the year.

Let's cut the mobile bills
Money over-spent on mobile bills is one of the big savings that vendors are dangling in front of users at the moment. Most people are away from their desk a lot of the time, se we tend to connect by mobile phone, even to people who are in the office. A vast amount of money in a mobile bill goes to pay for calls to and from people who are actually in the office building, but not actually at their desk.

There are answers to this problem, but at the moment they are too awkward for the user or the IT manager to find them worth doing. The simple answer is to replace the desk phone with a DECT phone, which users carry when they are in the building, alongside their mobile. This means in-building calls can be made cheaply, but it requires the user to carry two phones.

A more complicated option is to use a Wi-Fi phone, which can save more money by routing calls over the Internet, via an IP PBX. There are various problems with this (see Voice on Wi-Fi? Just say No), the biggest being that it again requires two handsets, and SIP phones are currently poor on battery life and so expensive compared to cellphones (competitors point out) that the payback would be slow. This hasn't stopped Wi-Fi vendors pushing voice and it is worth mentioning there are improvements arriving - such as the WISIP phone, reviewed here).

But we need to do it on one phone
The biggest problem is that such approaches make users' life more awkward (asking them to juggle two phones) while only delivering cost savings with no positive benefits. Combining both the local wireless connection and the cellphone in one device is the only way to make the idea attractive and this introduces technical and commercial problems.

The technical issues involve roaming from one network to another, without losing a call if one is in progress. The networks also have to be able to route calls to a given number to the phone, whether it is connecting via the cell network or the local wireless. There is also a patent issue to be resolved, as Calypso claims wide-ranging rights over cell-to-local roaming systems.

BT has opted for Bluetooth on the local link, to the surprise of those who only think of Bluetooth as a personal area network technology with only a couple of metres range. In fact, Bluetooth has many profiles, one of which is suitable for a telephone base station, and has a range of about 25m. Some handsets such as the Sony Ericsson P800 already support the required Bluetooth profile. The clinching factor for BT is that Bluetooth is much less power-hungry than Wi-Fi.

A single mobile handset uses Bluetooth to connect to a base station linked to the PBX in the office, and GSM/GPRS to connect to the cellphone network elsewhere, handling both incoming and outgoing calls.

The system will be integrated by Alcatel. Other companies involved include Norwood Systems, which was involved in Bluephone trials in 2003, as well as Motorola and Ericsson. Alcatel will provide the IP security and encryption, Ericsson will provide access bridge and access point technology for mobile access and Motorola will configure the enterprise section of Project Bluephone, provide handsets, and work on a future migration to WiFi.

Other partners, Inventel and IVT will make the portals to link Bluetooth to residential and enterprise fixed phone systems, and MBT will provide system software to link to billing and record systems at the service providers

But who do you pay?
The most crucial partner in the project, however, is the quietest: mobile operator Vodafone is a passive provider of network bandwidth, but it has opened the way to a scheme that will allow BT co compete with it directly. It is ready to let BT's Bluephone handsets use its network, some of the time, while siphoning other calls onto BT's wired network.

The phone certainly won't have a Vodafone label on it. BT and Vodafone have signed a MVNO (mobile virtual network operator) deal, which will give it a strong position in the mobile business, after spinning off its mmO2 subsidiary. MVNOs sell phone services using the big mobile networks - the most famous in the UK being Virgin Mobile.

In fact, BT already has an MVNO deal with T-Mobile, called BT Mobile, which claims to be signing up about 20,000 consumer customers a month. The Vodafone deal is likely to eclipse this, and replaces a business MVNO deal with BT's former subsidiary, mmO2. Including Bluephone, BT has been reported as hoping for £300 million to £1 billion a year from the Vodafone deal.

Both Vodafone and BT are facing up to reality in the deal. Vodafone knows that a deal like this will get its network involved in a scheme that could do a lot of business with enterprises, while BT is embracing wireless in a way that does not just accept the inevitable erosion of wired business, but uses the fixed and cell networks in partnership.