I do know all that, but the experiment would still be a valid indicator.

Reduce it down to six pictures and decide them based on a dice roll - a fair die would give each number 16.7% of the time, so if you rolled the dice a few times and 1 & 6 each came up 40% of the time, whilst 2,3,4,5 come up 5% each, it's probably a weighted die, and whilst it might be unpredictable it's not random.

Of course to get accurate percentages you'd have to roll enough times - doing six rolls and expecting to get each picture once wouldn't be fair. Six hundred rolls and expecting a hundred (give or take twenty) would be better.

But we're talking about it being random, not fair.

Even over a million rolls it's perfectly possible for any one number to come up a hugely dispaportionate number of times. It may not be likely, but that's a different thing again.

Theoretically, yeah. But if that happened you wouldn't use that die as a means to selecting items that you wanted to appear 'randomly'.

Depends on your expectations ofvrandomness, really.

Fish.

*giggles*