 Felly, byddwn i ni i siŵr mae'r ddestun eich ddataeth i sylwlad i ei gweithio i ychydig a'r dd—rwynt i ddarllach i dda i'r ddyn nhw yng Ngolwg hon, ac yn ei wneud i dd—rwynt i dd, ac rwynt i ddyn nhw yng ngorth wath o wneud i'r ddyn nhw geithreidraeth neu rhai. Rwy'n ddweud o'n dod o gweithio i weld fawr i fade i'r dd, i ddim gefnogiadag am wneud allan. Ond rydyn ni'n gweinio'r cwestiwn, rydyn ni'n gwybod, ar y llwysoedd cyntaf yw'r cyfnod? Rydyn ni'n gweithio'n cymaint gydag yw'r cymryd? Rydyn ni'n gweithio'r cymryd ar y 6-10 o'r cyflwyffon a'r cyfnod arall o'r gofod o'r cymryd. A o'r cyflwyffon, rydym yn ei ddweud o gwasanaeth y cyflwyffod yw'r cyflwyffod rydym yn hyn gyfnod ar gael. I came to the Greater London Authority in 2010, and the data store was set up around that time. It receives plaudits. We are the Open Data Institute, Open Data Publisher of the Year. But I really sort of question the usefulness of an Open Data publishing approach in 2015. In some respects, it's a little bit one-dimensional. Again, if I'm feeling critical, I'm going to say there's no point really in pursuing Open Data in a less than informed way. I have no real interest in publishing more Open Data that does not get used, that just sits on the data store and lays idle and redundant. I think we need to think really hard, and government needs to be a lot more deterministic about the uses to which it could put data, and Will has already outlined. A whole number of national cases that also have impact at city level as well. And I think this thing that I call a city data economy requires much, much more. And as I go through my talk in these next few minutes, I'll start to talk about how I think supply should proliferate in terms of the sorts of data that we should actually be looking to be publishing. If you ask me about city data exploitation, so what we do with it, I tend to score it a bit higher actually. I put us on about an eight out of ten. And I'm going to sort of trumpet some of my team's exploits around data. We've got something called the London Schools Atlas. Eight million individual records there that we extracted from the DFE under strong licensing arrangements as was needed. But that was our way really of circumventing the rather difficult relationship that we sometimes have with the London boroughs. And I don't want to go and have that conversation with the London boroughs 33 times over because they'll all move at different speeds and they will all reach a different conclusion, not all of which will be successful in them giving us the data. So getting uniform access to data on strategic issues that affect this city that pay no heed to the artificial boundaries that we impose upon ourselves in London government is a really, really important thing for me to do. And you'll notice that that's a cultural thing and an organisational thing more so than it is a technical or a data oriented thing. Those things matter of course, but first and foremost they are things to do with people and fiefdoms and the like. But that is a really useful service that if you are working in local authority planning schools places, if you are an individual looking where little Johnny and Jane should best go to school, you can go there for an informed view based on population projections, birth rates, where schools are currently cited, where they're planned to be cited, what the level of performance in schools is, which is clearly a consideration in this city, you can do so. But importantly it's a thing that we can now look at in a strategic sense where we couldn't do a couple of years previously. We do things like look at data from different sources as well to understand issues. So we're using A&E data again after long and difficult negotiations with the relevant parties to understand violent crime hotspotting in the capital. So not just going to the usual sources in the crime data, we actually understand who is going to A&E, what problems they are reporting because those are the things that do not crop up through some of the more traditional sources that we would go to. We have something called the London Crime Landscape that we'll be releasing this week as well, which again draws on about 80 separate variables and the relationships between them and we're using that to drive predictive analytics in the capital. Again one of the things that I found interesting in the discussion that we had last week at Windsor Castle was about how you actually protect anonymity whilst also making predictive analytics useful and in fact not dangerous to the individuals and their identity. Loads more work that we do around rough sleepers, financial inclusions, infrastructure in the harder sense. I'll not dwell on them here. This thing called big data, again if your glass is half empty it's sort of a foggy marketing thing. It's becoming more real in our city undoubtedly. Organisations like TFL are undoubtedly, I'm going to say superb at exploiting it. They are world leaders in this. There are 12 million oyster card records that are gathered each day. So there are things going on in this city that help commuters move around it based on the work of the analytics team there. We're starting to do big data on the London data store, what I would call big city data. There are 115 million lines worth of household energy consumption data on data store at the moment gathered from an exercise called low carbon London which was about 5000 households each equipped with a proper smart meter that allows for energy consumption to be transmitted every 15 minutes and stored by us. At the moment it's a static data set but it is a sign of promise about what we can do with the proper safeguarding around the internet of things once it happens in this city. I've already dwelled too long on the school's places so I shall carry on. Where does the impact lie really looking to the future? We are dealing of a time still of financial and fiscal restraint around infrastructure and resources, proper means of exploiting data so that we can get to the heart of city problems remains super, super important. Again, one of the questions that I'm asking in an organisational sense increasingly is how many authorities in this city, how many public bodies can properly say around all of the strategic questions, all of the service delivery questions that they have to answer around that city management table are properly exploiting data to its full potential. I think you could probably count them on one hand. Skills markets around devolution as well, or sorry, one of the areas in which London now has increased responsibility is around skills markets after the autumn statement. Again, there are many other forms of data that we can use to better understand the skills that organisations in this city need to make the economy grow. Of course, they sit in repositories like LinkedIn, monster.com and other things besides, we still need to do more work to understand how that data can be incorporated alongside some of the more institutional datasets. Adult social care, I'm not even going to start, massive issue, hugely diverse economy, loads more that we can do with some of the data. But this really, for me, exposes a few issues that I'm going to round off with. Privacy for households and individuals is of paramount importance here. I've spoken about those things called smart meters already with the digital catapult hosts, and I'm setting up a series of personal data stores or personal data exchanges in which we will put use cases, private sector, public sector, that will allow us to take the debate around data privacy and more positively put how individuals can see their data used to societal advantage to private sector advantage if companies want to use it for marketing. Why shouldn't we offer citizens the chance to sell that data on that basis? And just to improve public services in this city as well. I've spoken about organisation and culture, but I think to address those persistent policy problems across boundaries, this is a broken city. The 33 boroughs cannot go on not sharing data in the way that they currently manage to do, and therefore harmonisation at the point of sharing is something that we should work towards. Each local authority crudely put will hold an adult social care data set. There are a few fields in it that we should see are harmonised so that we can start to sort of address that issue in a coherent way. There's a lack of regulation around certain parts, certain industries as well. I've already mentioned energy companies. What incentives are there for them to share data in a way that helps us understand how we can smooth the peaks and troughs in energy demand, get a better deal for consumers. I've spoken about city data not open data and the proliferation of supply, but I think government should be more deterministic. We need to be much better at saying here are the policy challenges, here are the skills that we think we need to use to address them, here are the data sets that we need to properly deploy and whether they are public sector, private sector, institutional, less formal, we need to better understand what those are. But all of the above really requires more collaborative government and, importantly, monetisation of data as well. We sort of let in government let the horse out the stable door on this a little bit, but in terms of proving the value of city data and how it can be used in a sort of multifaceted market, we need to understand how much it costs us to release the stuff, what societal, economic, environmental benefits can be accrued by the results of exploiting it and the financial value of these things and all of those things add up to us really needing to have a city data strategy, which I'm very happy to talk more about, but you've heard plenty enough from me already. So I'll pass over the mic.