The Bluetooth brigade are planning a new version of the wireless cable-replacement specification, a prospect that fills me with a little horror.

The idea of a wireless cable replacement is dangerous. Of course, faulty or miss-routed cables are frequently the cause of downtime, but such problems can only get worse once cables are swapped for invisible radio links.

However, my concerns about yet another Bluetooth specification stem from the havoc that results from having several different versions of a specification in use at the same time. Ironically, one of the main justifications for the development of the forthcoming Bluetooth 1.2 specification is to make products easier for people to set up and to use.

Clearly there are problems to be addressed. For example, I recently needed to replace my mobile phone. It turned out that there is no shortage of phones that meet my general requirement of having Bluetooth and tri-band support. I wanted Bluetooth in my new phone so that I could continue to use the same Bluetooth headset that I've owned for a couple of years.

But finding a new Bluetooth phone that was compatible with my ageing Bluetooth headset proved almost impossible. Actually it would have provided some excellent material for a farce.

The manufacturer of one handset said it guaranteed the handset would work only with its own Bluetooth accessories. I can understand what might motivate such a bizarre stance, but it still remains a bizarre stance.

But far from being uncommon, this position seems to be the norm.

The problems of Bluetooth compatibility are only compounded by the fact that Bluetooth devices generally are governed by embedded software that is tricky to update.

In the case of the mobile phone that wouldn't work with my Bluetooth headset, the phone's firmware had been superseded by a new version, which had been published on the manufacturer's web site. Unfortunately I couldn't put the software onto the phone because my phone company does not include a serial lead with the handset.

Of course, I could buy a lead, although this would cost the best part of £50 - which would almost double the price of the phone - but even then nobody involved could say whether the new software would fix the problem I was having. Rather than serving hard-nosed businessmen, my phone company seems to serve hard cheese to everyone.

In the end a colleague allowed me to test my headset with his phone, and as the two actually worked together I ordered a phone like his.

Now, IT departments can probably manage most of these problems with careful testing and evaluation procedures, but the real horror may become apparent when millions of consumers become embroiled in these Bluetooth headaches.

Quite what effect this could have on an already struggling phone industry remains to be seen, but I wouldn't be surprised to find that huge numbers of Bluetooth gadgets are being returned because they don't work properly.

Have your say: reply to IT Week