Again, not to get into the politics of it but I've read a few different things about the verification methods. The statement about the email validation being a deterrent is absolute nonsense, clearly but they do have other methods of sorting and validating the data which seems more efficient, using the postcode entry.
There's a variable population density calculation used which defines the maximum number of people resident in an area defined by one postcode. The petition list will only 'count' entries submitted for any specific postcode up to the estimated population amount. So, if they receive 1000 petitions from a postcode which should only have 100 people living there, 900 of the submissions aren't counted in the final tally. For a manual check, they'll use local government electoral register details and carry out random cross checks that the name and postcode matches a resident's details.
Although, 3.8% of the votes come from overseas which is where the postcode system falls down. Technically, they're supposed to come from ex-pat citizens who have the right to vote but, again, the captcha and email verification isn't going to weed that out- I was wondering if they may have done something similar by allocating numbers of emigrants to country IP's?
Be interesting to see if there's a way of dealing with the postcode check, though. Something that occurred to me was that new housing developments are being built all the time and allocated brand new postcodes. Anyone who's lived in a new development in the UK will know the pain of the post office or automated systems not validating your address due to the inefficiency of the postcode information being propagated. So, maybe something could be done by using postcodes that don't yet exist but are numerically consistent with other postcodes in the same area, making it look like votes were coming from several new developments?