I've checked the ECMWF cloud amount for London on Saturday and it shows variations from 1 to 6 to 3 oktas (09, 12 and 15h), suggesting Cu/Scu spreading out around middle of the day. MO global gives 6, 8 and anything between 2-6 oktas later - rather cloudier overall but both indicating variability from hour to hour. Of course, at this time-scale these figures are only indicative of the whole day really - MeteoGroup's data at the BBC Weather sites shows the sunny intervals symbol throughout the day and (thank goodness.... this tallies with the text 'sunny intervals'.

I agree that at shorter timescales it is important to differentiate between cloudy and sunny areas of the country - at this time of year and in the winter it can make a huge difference to the feel of the day. Pointless to pretend we can do this 2+ days ahead with cloud varying semi-randomly.

Julian