[rab] Noooo! You were great. I guess it must be difficult to explain mathematical modelling to laymen, and for a presenter to pick up the gist of a study involving lots of people over several years and squeeze the right questions into a five-minute interview. I guess you're dealing with a lot of variables, but you summed it up pretty well. If I do my job properly, I get to put a lot of Woodland Trust people on the radio, and usually guide the presenter into asking the right questions: "You might like to ask why XXX happens, and what it means in the long term" etc. I didn't realise universities had press offices and the fact that it got picked up by so many publications means yours wrote a good press release and sent it out to the right media, with a view to possibly attracting the interest of potential funders...

[pen] The press officer I spoke to - for about an hour - is an ex-journalist, so he distilled it into a newsy-sounding story. (The version on the BBC website is basically the release verbatim). The presenters' briefs/scripts seemed to be cobbled together through informal chats with researchers/producers a few hours ahead of broadcast. 'Twould have been nice to have seen a copy myself beforehand, but I guess you can't have everything.

You're right about funding though. Probably the most useful thing about all this is that I can stick it all in the case for support to demonstrate interest beyond stuffy academia.

[rab] Cool interview and work - well done, and well explained. New Zealand is nice and small, making it a lot easier to study than its behemoth of a neighbour, even though exactly the same process happened in Australia (albeit some half a century earlier) and it's been developing in interesting ways ever since. I'd be interested in the publications to date, if you've got a list somewhere. I have an interesting documentary about the australian accent, too, if you're interested.

[flerdle] So that Google doesn't link my uni page to this one, perhaps I should suggest you type my name into the venerable search engine, click the first link (at least, the first link in our part of the world), and towards the bottom of the page you can find a link to the relevant preprint. I heartily recommend the books by Gordon and Trudgill if you really want to get into the details of the empirical data, as opposed to our modelling.