I was under the impression that they were so configured as to indicate where the user was, rather like telephone numbers, unless the ISP were temporarily routing via another country for some technical reason
No.
Trying to keep things simple :
A global organisation called IANA allocates primary ranges to regional registries. These ranges are typically /8 in size, i.e. ranging from, for example, 2.0.0.0 to 2.255.255.255.255.
The regional registries then allocate sub-ranges to ISPs, the allocations are typically much smaller, for example /21, so you'd get 2.0.0.0 to 2.0.7.255.
ISPs then make much smaller allocations to their customers.
There are two fundamental obligations on the ISP as far as the public databases go :
(1) Keep their own contact information up to date
(2) Record details of sub-allocations they make to customers
As far as the sub-allocations to, there is no requirement to log full details of customer name or address. Just the size of the allocation you've made to them. The argument being that if law enforcement want to know more, they can knock on your door with a warrant.
In simple terms what is the difference between IPv4 and 6
The primary reason IPv6 is coming is because we're running out of IPv4 addresses. So the main difference is a significantly larger range of numbers.
IPv4 is 32-bit and hence limited to 4294967296 addresses.
IPv6 is 128-bit and so provides 3.4×10^38 addresses.
The visible difference is that IPv4 addresses are easy to read an remember, for example 10.250.240.130
IPv6 addresses are much longer and are expressed in hex form (i.e. can use characters 0-9 and a-f), for example a full IPv6 could look like:
fd01:abcd:ef01:feed:cafe:babe:dead:beef
(Obviously real IPv6 addresses will be a little more "random" and so harder to remember !)
IPv6 does allow for consecutive series of "0000" to be abbreviated as "::", for example :
fd01:abcd:0000:0000:feed:babe:dead:beef could be written as
fd01:abcd::feed:babe:dead:beef
There are a couple of additional features of IPv6 too, such as SLAAC (Stateless Address AutoConfiguration) which enables IPv6 devices to automatically configure themselves when connected to a network. There are also various features of IPv6 that make them able to be more efficiently processed by routers.
The biggest downside of IPv6 is that the geniuses that invented it decided not to make it backwards compatible with IPv4. They are completely incompatible, and as a result you have to implement one of a number of hodge-podge workarounds in order to be able to access IPv6 from IPv4 or vice versa. The easiest, cleanest and best is dual stack IP (i.e. run both IPv4 and IPv6, most PCs are capable of it these days).... but that assumes your ISP gives you both an IPv4 and IPv6 allocations.... sooner or later, they won't have any IPv4 to dish out any more and so you'll have to use less desirable workarounds.