To be pedantic, they do require a third party, but it is a distributed third party rather than a centralised third party. The value of this difference causes great debate.

This is not true, at least for TCP connections. While many implementations might have this limit for simplicity sake, there is no actual reason why you can't use the same local port for different destinations, e.g. TCP port 1000 could have active connections to www.slashdot.org on port 80 and to www.microsft.com on port 80 and there is sufficient information in each packet to work out which of those two connections each packet belongs to. You see this type of multiplexing in reverse on servers where port 80 might have thousands of connections to it active at once.

Intel is already doing this to some extent with their Turbo Boost technology. You can run one core at the fastest clock rate allowed by the heat envelope of the CPU, two cores slightly slower or four even slower cores. It certainly isn't taking it to the extent that you suggest, but it is heading in that direction.

With their next generation Sandy Bridge architecture they are taking it a bit further and if the cores are cooler because they have been running idle for a while then it lets them run even faster for a short period of time until they head up again.

It is all about heat these days and in order for all those slow cores to be helpful, they need to have significantly better computation per watt ratio than the big fast cores AND we need OSs and/or hypervisors that understand how to use a mix of fast and slow cores.

Maybe - see http://www.computing.co.uk/business-green/news/2231469/wind-turbines-mess-weather for example.

However, the effect is debatable, with no consensus on what it will be and whether it would be negative. If we were getting to 50% wind power then I'd want to see more research on the possible effects, but I can't see it having any noticeable impact one way or the other any time soon.

I think that you overstate the effect and the chance, but yes, this is indeed a cost that needs to be taken into account. However, what we are comparing to is coal-fired power plants. These also do massive damage - it is just less obvious. I won't try to put a number on it, but greenhouse effects could easily cause a loss of well over 1% of GDP even in a moderately bad scenario. I don't want to think about worst cast. Then there's the mining, particle pollution and mercury that coal has to deal with.

Yes, decentralized generation can be good when it works, but why do we need it? You say yourself that high-voltage transmission is very efficient, so why?

I don't feel that coal is a choice. It is too dangerous. In my opinion, it needs to be off the table as an option. If it did not exist, we would find a way and we would be fine.

It would be great if solar and wind could produce all our power requirements and maybe one day they will, but I feel that that's a long time away and if that leaves us with a choice between coal and nuclear then I'd definitely support the nuclear option.

That's because the point of technology is not to enforce the levelness on the grid. The idea would be to anticipate the time of peak demand, cool all of the fridges down before this time starts, so that they can all stay off during peak load. Alternatively, if the peak load spikes unexpectedly or power production drops unexpectedly then the fridges could switch off to allow the available power to be used for other purposes. This would mean two things. Firstly, less peak capacity would be required which reduces infrastructure costs. Secondly, it allows renuable energy sources to be used to a greater extent in the grid. Generally with renuable energy you have no control over production, so being able to control consumption instead means that a greater percentage of renuable sources can be used instead.