It rather worries me that people don't give a toss about the fact that once you have more than 20A on the cable supplying a socket you have no guarantee that the socket will be OK.It rather worries me that unsuspecting DIYers are reading assertions, which they may regard as authoritative, suggesting that 'standard practices' which have been with us for decades are unsatisfactory, and may actually be unsafe.
As BS 1363 sockets were developed specifically for this "standard practice" it seems to me such a no-brainer for the test to have specified a higher current that the only conclusion to draw is that their performance when current is higher than that is of no interest. Why on earth would they devise a test to demonstrate acceptable behaviour under conditions which they expected to be exceeded in normal use, i.e. a test which would be of no use in demonstrating acceptable behaviour in normal use?