Remember C. P. Snow, he of the Two Cultures lecture (to read it would be to condone it, according to F. R. Leavis)? He used to say that, to be hailed as forward-thinking in your own lifetime, the trick was to be only a couple of seconds - and certainly no more than a minute or so - ahead of anyone else. Any more than that, and you look like a crank.
For people like me, trying to generate ideas and make a living out of it, this is a fearsome problem. I am constantly either too far behind or too far ahead of the curve.
I spent six months, ending this year, sitting on the Lib Dem public services commission, which will report in time for the party conference. I think parts of the report are pretty good but I worry we weren't quite far enough ahead, especially in our injunction for service integration.
I was thinking this reading the influential NHS blogger Roy Lilley yesterday, who was being generally flabbergasted that Health Secretary Jeremy Hunt was backing US-style Ascountable Care Organisations - and was clearly right behind the vertical integration of services.
In fact, one of the oddities about the NHS at the moment is that it appears to be going in two directions at once. On the one hand (if you read the campaign blogs and the political press) then it is being sold off, disjointed and disco-ordinated (is there such a word?). On the other hand (if you are in the Department of Health) it is being integrated. On the one hand competition, on the other hand integration.
It is true that you can have both, but there is an accommodation to be sought and I'm not sure where that compromise is going to be - probably different in different places. What is definitely true is that the NHS is turning out different to what Andrew Lansley designed in the early years of the coalition - and it is peculiar that this is barely rceognised in the political media.
I hope that the alternative providers will not confuse the integration (they needn't). I hope the integration won't streamline so much that it will get in the way of individual difference, yet will allow the system to focus on the places and the people which where costs are mounting the most.
I'm not one of those who believe that the NHS would be best served by going back to what it was in the 1970s. It needs other kinds of service, community support and a range of other things too, some of which can only be provided by other patients (see what I've written on co-production in my book The Human Element).
And here is the point. There is no point in vertically integrating the NHS if you don't also integrate the service sectors it also needs - from social care to alcohol services, not to mention education and the police.
NHS integration really just begs the question. How can we knit all the services that people need together in such a way that we don't have to deal with every one of them individually, with different protocols and systems?
And please don't tell me this is an information problem as if it can be solved by an app. Nor is it an outcomes problem, as Roy Lilley implied in his praise of SMART commissioning - the trouble with commissioning for specified numerical outcomes is that it misses all the stuff between the definitions, narrows the outcomes and undermines the coherence of the service.
No, in the end, integration has to be very local. It has to be human-scale and de-professionalised. It needs to deal with people's complex lives as they are, not as the service designers would prefer them. And we are still some way from that and, generally, hurtling in the opposite direction. And that direction is ulimately extremely expensive.