 Thank you, Tom. Is it okay if I speak like this? Can you hear me at the back? Right. Okay, so I'm going to be quite fast because I would like to cover a lot of things. I'm going to tell you about the results of a project that finished a few months ago. And basically the motivation was this. We found, or as probably most of you know, we went to disseminate archaeological information. And the previous presentation concluded precisely with a point about this. And in addition, we wanted to get the feedback from the citizens who are looking at that information. And we wanted to integrate different datasets. We wanted to disseminate information that comes from very different places made by different people that haven't talked to each other at all. Okay, so this was a bit of a challenge. How did we do this? Okay. Basically, you can think of an organization producing archaeological data, another one, another one, another one. And we will end up with a lot of different datasets. Okay, hence the heterogeneous datasets. Our aim was to somehow integrate all those into a single dataset, at least as far as the citizens are concerned, and present that to the citizenship through a unified interface. So people using mobile phones or computers at home or whatever could browse and get to that information and perhaps give us some information back about anything. Right, so this is a bit of a dream we had at the beginning. Okay, this had to be a proof of concept. We didn't aim to produce a tool that could be usable by anyone. And we didn't aim to, on the other hand, to produce just a theory of some description on paper. I wanted to create some actual software that could work as a demonstrator of this infrastructure actually working. So I'm going to tell you quickly about the datasets we work with, how we integrated them together, how we tackle this issue through conceptual modeling technologies, and finally what kind of system we built and how users interacted with it. I know it's a lot, so I'm going to be fast. Okay, we started with two datasets in Spain. One was created by us, the one on the right-hand side, by the institution I work with. The other one was created by some people about 1,200 kilometers to the south in Andalusia, describing a very different kind of archaeological situation. Both datasets included a lot of structural information and some structured text descriptions of sites and objects and everything, and also a lot of multimedia details as well. We have lots of pictures and other multimedia elements. The technologies we use were these two. Basically, we applied Charm, the cultural heritage absent reference model, which we've been working with for a while already. This allowed us to have a comprehensive model of both datasets and then have specific models for each of them. I will come back to this idea later. And we use Konimela's conceptual modeling language because of its expressive power, especially in terms of multilinguality, subjectivity modeling, temporal modeling, etc. Okay, so basically we started with Charm, which is, as I said, a reference model, a very abstract description of what the archaeological record looks like without making any specific commitments about types or specific kinds of things. And from Charm, we derived a model that was specific to our project, a model that's addressed the very specific issues of interaction with the citizenship and getting feedback, right? And then from that model, we refined it a bit further into two different models, one reflecting or capturing the structure of the first dataset, the second one reflecting the structure of the second one. Okay, so with this in mind, we have to say that these three models that we created specifically for this project are the models that, in some fashion, give us the structure, the information that the citizens are going to interact with, okay? The fact that we are using Charm is important from a technical point of view that it's totally invisible to the users interacting with the system through the mobile application. Right, I don't assume you will read that. This is just a quick overview of what one of the models looks like. So this is how we created the knowledge models of the datasets. This is one of them, ours, and actually. Basically, I don't know if you can see that there are two colors. Some boxes are green, some boxes are yellow, more or less. Well, the green ones are stand for concepts that are taken straight from Charm, from the standard, whereas the yellow ones are concepts we added as part of the refinement process, right? Okay, we used some tool sets we developed in-house to actually put this into computer form and created instance models that contain all the information. You could say that we actually translated or converted the original datasets, which were in different formats. We had extra spreadsheets, we had some database records, we had Word documents, everything together, and we converted all the sources into a comprehensive repository of information. Then we created a software system. For the technically inclined, this is how it looks like. We work with a data server, in-house data server, that holds both support database as well as two different datasets with a modeling engine, supported by a modeling engine. This is used by a web server that exposes the information to the public through a totally standard web service front-end and also exposes an administrative website for us to monitor what citizens are doing, how information is going, browse, etc. This means that any user in the world, just using an application, a mobile app that can be freely downloaded from the Apple Store, can connect to this website, sorry, to this web service front-end and get to the information. No, I'm fine. Thank you. What can the users do? Basically, we implemented flexible search features in the mobile application so people could find any archaeological element either around themselves or within a specific area or a specific category or just by looking for a text. If you know the name of the place, you can just type it in and the system will give it to you. The application also implemented some mechanisms for making comments, rate size, rate things, interaction, a sort of mini Twitter, if you want, embedded within the application. Because, as I said at the beginning, one of the aims of the project was to get some feedback from the citizenship about the information being exposed. And finally, we also implemented a feature for the users to be able to suggest missing information. Many times you find that the local people who are living next to a site know a lot about the site and sometimes they are able to give you some feedback about what the site is looking like right now or what kind of information is incorrect or just missing from the system. OK, so this is what the application looks like. It's not in English, it's in Spanish. So, well, that's tough, I'm sorry. I will quickly give you a review of what you can do. Here you have a home screen showing a list of sites that are around you or are suggested by the system. You can touch one of them. You can search. For instance, this screen shows a search for a specific category. So only show me entities of whatever category you want. If you go into one of them by touching it, you see a complete description, including pictures, location maps, texture descriptions, etc. This looks very simple, but behind the scenes, remember that what this is doing is directing the queries of the users to one data set or another, depending on what information required is, integrating that and giving back to the user as a single homogeneous data set. You can, of course, play with location maps, click there and explore the information as you wish. I have the application on my phone, if any of you is interested later. As I said, the interesting thing of the project is not the application itself. This is just if you want a means to develop an end, which is what I was showing before in previous slides. After doing some interaction with users, we basically... Well, we wanted to interact with users in different amounts. So first of all, as I said, we uploaded the application to the Apple Store and people downloaded it freely. Unfortunately, we didn't get that many downloads. Some people did, but I mean, this was a research project. We did have very limited funds, very limited time. And the dissemination we did of this was limited as well. Still, some people downloaded it and used it. Then we selected some used groups for focus testing. We sent people to the field in both areas, up north-west, down in the south. And we did some field testing with users and asked for explicit feedback. And thirdly, we also sent some work to do some people to do some field work interviewing pilgrims on route to Santiago de Compostela and who presumably would be interested in using the application. Conclusions about this work was, well, the technical aspects, the capability of these technologies to integrate heterogeneous datasets and federate queries is good. In that respect, we are happy with the technologies employed that have proven that they actually work nicely when they are actually implemented in an actual software application and put to work in the real world, if you wish, outside of the lab. We still need to do some improvements. We are working currently in some very deep improvements to the modern technologies we used. I'm not going to go into this in detail, but if you are interested in what kind of improvements, what kind of weaknesses we have, how we are planning to solve them, just please let me know and I'll be delighted to talk to you. We got some valuable feedback from citizens, although not a lot, as I said before, this was a limited resource, a limited time project. And we, well, we had inevitably to end up asking ourselves how interesting is archaeology to regular citizens, actually. We often assume people are very interested in our cultural heritage, and that is fair enough. But when we actually try to get some people to give us some time and spend some time with us and look at this information, give us some feedback, not everyone is openly available and willing to do it. So we need to perhaps challenge us a bit more and stop assuming that everyone is going to be willing to engage in that kind of thing. I know this is a very... A lot of nuance can be added to this challenge, and I'm also very happy to talk about it, but this is something I believe we have to do. Thank you very much.