On Sun, 23 May 2021 08:02:32 +0100, "Brian Gaff \(Sofa\)"
wrote:

Yes one of the problems with annonamising data is that given enough of it it
can easily be mined to find out with a pretty good accuracy, who any given
person actually is, and you can bet there will be a market for that to sell
you even more crap. I am already being annoyed by disability product spam
even though I do not use a wheelchair, since I'm blind, but it does show
that some dangerous stereotypical assumptions are made by these so called AI
data aggregators.
Brian

I can tell you exactly when I realised the power of metadata: it was
Tuesday 1st May 2001 and I think it was at about eight o'clock in the
evening.

Paul Vickers was presenting a programme on BBC Radio 4 called "What Do
They Know About Us?" As an example, he had picked a person at random
and whilst carefully avoiding naming them, showed examples of how the
little bits of their life appeared all over this exciting new thing,
the World Wide Web. I'm not going to repeat exactly what he said
because even without the name I immediately knew that I knew who he
was talking about and the same information would identify them today.
Long story short: I contacted the person, they contacted the Beeb,
Mark Savage (the programme producer) edited that section out of the
programme repeat and AFAICT no actual harm was done. But it remains
that surprisingly little innocuous information from different places
can combine into a very powerful tool.

Nick