Thank you Northerner for this very balanced and well thought out response. I think it does have to be realised that "one size does not fit all" and I think some of us, me included, have become a little vociferous in our opinions.I think there is some misunderstanding creeping into this thread. As one of a number of methods of getting diabetes under control there is nothing wrong with it, and it may even be argued that, subject to the right assessment of likely benefit, it might be used as a 'first line' treatment that might help many, many people, as long as they received the level of support required. Sadly, we know all too well that people are often given a diagnosis and then sent away with an unhelpful leaflet or two and a packet of metformin - there are neither the resources in the health service, nor even in many cases the slightest interest in providing this support.
Weight is a very significant factor in the development of diabetes is a lot of people, where they might be fine up to a certain 'tipping point', but then that little extra weight increases insulin resistance to a degree that levels become abnormal, the pancreas tries to respond by producing extra insulin which then causes more weight gain as excess glucose is laid down as fat, particularly around the major organs where its effect is most detrimental. This tipping point may vary vastly from person to person, so a person who is only slightly overweight may develop problems - for others it may require a much greater period and amount of weight gain. We're human beings, and we are incredibly complex in the way we work and are able to deal with things.
There's an important genetic element at play as well (except perhaps in 'temporary' manifestations, such as gestational or steroid-induced diabetes - I don't know the facts). If, as we are told, 60-70% of the UK population are overweight or obese, yet only around 9% of the population has diabetes, so just being overweight doesn't inevitably lead to Type 2 diabetes, you need some other factors to be at play. I think this genetic element is what comes into play where the 20% of people with Type 2 (a not-insignificant 800,000 people) who do not fit the the stereotype become affected. Many people develop diabetes simply because they are getting older and their pancreas/body does not work as efficiently.
Because most overweight or obese people don't get Type 2 diabetes, I think this is why the dire warnings often issued are ignored - 'it would never happen to me'.
My objection is not to this approach as a treatment but to the presentation of it as a panacea when no such thing exists. As @grovesy says, Type 2 diabetes is not one condition it is more like a spectrum - I've read enough 'case histories' here over the past 10 years to know this to be true. If we are really going to turn things around then it has to start with the medical support that people receive right at diagnosis, so people are fully aware of their options and receive intensive support in achieving their aims. That would require a huge upfront cost, but save the cost many times over in reducing medication, complications, and potential loss to the economy.
When I saw my GP as I had gone from pre-diabetic to normal,back up again, he is wondering if I am becoming more insulin resistant with age (I am 65).