There was quite a range there, so, again, I'll go backwards, and if I miss something out let me know. We are actively exploring that. There are some difficulties, because a large amount of the infrastructure that we've put in place has been done with European funding through a state-aid programme, and then having a company that interferes in the market, to use the parlance, on top of that is problematic to overcome. But, we have a team of officials working on how we can maximise the benefit of the public infrastructure that's there—it'd be through the public sector broadband aggregation, for example, so we have a pipe into every public building in Wales. That's not the Superfast Cymru programme; that's Government pipes into every building. What we can do to exploit that and what we can do to bring its benefits to communities, including things like community Wi-Fi schemes and so on, we're actively researching that at the moment. We're hopefully going to put some pilot schemes in operation very shortly around school communities and so on, so we're very much looking at that.

In terms of the UK full-fibre issue, that's very problematic for us, because our biggest problem still in Wales is getting it out to everybody—some kind of broadband out to everybody. Full fibre is about changing the copper network that's carrying the speeds into full fibre. So, in my home, for example, I don't get superfast speeds but I get high 20s, but that's because I'm on the end of a copper wire. The full-fibre scheme is to change those copper networks into full fibre. That would be great—we've got a lot of communities that are hovering around the 30-ish mark. It would be good for them, but it won't help the people who are not connected at all.

So, in terms of priority, I've got a real desire to get the people who aren't connected at all in first, although there's nothing to stop us having full fibre afterwards. And, of course, we're looking to see how best we can take advantage of the UK's money for full fibre, and we are looking to see whether an all-Wales bid might be more successful in phase 2. We've had some success for local authorities, as I know that Adam Price is aware, in the first stage, so we're very interested in seeing whether we might marshal a bid to do that, but it won't help the people who are not on the network already, which is something that's of some regret.

In terms of the communications, I couldn't agree with you more—the comms have been dreadful. I've had endless discussions with BT about why the comms are so bad. They hide behind semantics—they say that the letters say that things that are scheduled are never definitive and so on, but I myself have hung on the end of a three-month rolling programme for quite some time and I understand totally the frustration of that. I put my hands up to that. When we let this contract, that wasn't an issue, because nobody had it and everybody was delighted when it came. Vast numbers of people were being connected. It became a big issue towards the end of the contract because of the way that the contract was structured, which is that we didn't specify any premises in Wales. BT had to get to 690,000 premises, which, in 2011, was 96 per cent of the premises in Wales; it no longer is. So, I accept that the comms are bad.

That's a different issue to the issue I talked about where we've discovered that some communities' fibre isn't properly connected. So, that's not about the communication with them; that's about the way that the claim has been processed and what we've tested. Some communities in the Presiding Officer's community, actually, and in a couple of others, have said to us, 'Well, I should be able to order fibre.' We've said, 'Yes, we should be able to order fibre.' When we've investigated, it's not been possible for various very complicated engineering connectivity issues back down at the tetrabyte hubs, and so on. So, it's a different issue, but it results in the same frustration for people who are being told one thing and it's not true. But it means we've revisited all the claims, just to be absolutely certain that we're not paying for something that we haven't actually got, and that's ongoing, which is why we're still in that process. I suppose it has the same result for the citizen—the frustration—but it's a very different cause from our point of view.

And then, in terms of the money, the three procurements have been put out to procurement in the ordinary way, and we specified particular areas that we want to—. So, we're prioritising people with very poor mobile coverage, no 4G and very bad broadband in one of them. We're specifying business premises in others, and the three lots have gone out in a very standard procurement way. We're waiting for the responses back to see how many premises will be covered, at what cost, and so on. But we've deliberately kept a pot back to look at bespoke community solutions, because a number of communities have come forward and said, 'We would like to do something very bespoke here with this small company. A group of us have a solution', and we want to be able to fund that, so we've kept a pot back for that.

The moneys are fluid. I'm casting these figures around but this is the gain share, Deputy Presiding Officer, so obviously it's moving. So, the more people buy it, the more money comes in, and we've pledged to put that money in. So, it's a fluid amount of money because it gets bigger. As more percentages join, more money gets spent on it. So, I'm saying 'about' for that reason, but it's about three quarters of it, and a quarter we've kept back. We want to see what community projects come forward and what they look like, and, as I also said, we're committed now to reviewing the ultrafast—the business voucher—because of what the UK has just done, because we want to make sure that we're ahead of that curve as well. They've just announced the gigabit fund, so we want to be ahead of that. So, we will review it in that light.

But it's a very complex grant agreement with BT, and we're very keen that they don't get any money that they're not entitled to and that we get as many connectivity outcomes from it as is humanly possible.