The data coverage and bandwidth in Nottingham on Orange has been dismal for quite some time. I actually spoke to them last week about this, and they said that they’re aware of issues in Nottingham, as there are too few 3G masts, and one of them is faulty. They were rather hoping that things would improve once they merged the T-Mobile network into theirs (apparently around the 6th October), though as far as I can tell, it hasn’t.
The photo below shows two speed tests, run on iphones, with the same app, at the same time. The one on the left is on Vodafone, and the one on the right is on orange. You can see the Vodafone download speed is around 3Mbits/sec, which is comparable to ADSL. The Orange speed, on the other hand, is 0.24Mbits/sec; this is around just 6 times faster than a standard dial-up connection.
Total cost of ownership (TCO) is a measure applied to (usually) IT hardware to determine its true cost, and apply that to the cost-benefit analysis. It’s typically used to explain why (for example) a 3-year life cycle for desktop PCs might actually work out a lot less costly than replacing them every 5 years, due to most of the costs being involved in the management and maintenance of the unit itself, rather than the acquisition costs.
The management and maintenance costs usually increase as the hardware ages, partly due to the failure rate, service and support, but also the “cost” of an employee using slow hardware, and the extra time it takes to carry out their work. This cost is really visible when a machine gets so old and slow that the user struggles to operate with it, or the device actually fails. By then, however, the true cost has already exceeded the cost of a new device acquisition, so you’d have been better off replacing it before that point.
What is often missed, however, are the benefits of the new system, as many such benefits are either difficult to measure in financial terms, or they’re simply unknown until the new system is implemented. Of course, if you operate in a very small enterprise, spending too much time calculating the TCO, and attempting to identify the financial benefits of replacement systems could take so long that your time begins to affect the TCO itself, and you’d be better off making very rough estimates. Luckily, it is generally accepted that a 3-4 year lifecycle for desktop machines is appropriate in most cases, and it’s pretty safe to follow that sort of timescale.