 Great. So I'll be chairing this session as many of you know Alessandro De Vita, that was a very dear friend and dear colleague. Too many of us, I was killed in October, October 2 of just this past year while going back to London. Alessandro was a professor at King's College in London and also a professor here at the University of Trieste. So several of us will be giving very brief talks with a little bit of a personal element and especially a scientific element just to show some of his scientific results to all of you. I read something very nice about him in the few days after his death. Someone said Alessandro was wired differently from most of us and I think we all think so. I mean, he was really different and I tried to list here something of his quality, of his intelligence, of his knowledge, imagination and his mischievous and deeply generous spirit. So I'll just spend two or three minutes. Ariel yesterday has broken new ground as far as you can go back in time. So a few pictures. This is Stefano de Gironco, his wedding. So Sandra is here. You probably recognize Francesco. You probably don't recognize me. And this is Alessandro Variolla. It was a wedding. Yeah, yeah, yeah. It was Stefano's wedding. Another one. Sorry for this. This was for Alfonso Baldereschi, 60th birthday. Actually, Alessandro was a student of Alfonso Baldereschi just 30 years ago in January 1989. He started his laureate work up here on the second floor. Everything was just the same, apart from the computers and the BT100 terminals. You see also here Christina, Sandra's wife. And here appears also Sandra Scandolo, Alfredo Pasquarello, Alessandro Satt and Silvia Rubini. This is one of the last pictures that I have of him in Claircourt with Carla Moultonian with Caterina Serra. And I decided to give you something somehow personal, but just to show you a little bit of something between the two of us. It was a friendship for 40 years and emails for 30 years of apologies. I skipped something also very precious to me. And this was Arash this summer celebrating. We were in London. I visited King's College for a month. He had just been promoted professor. And I'm really grateful he decided to really open all his most precious bottles. And we had a great day together again with Christine here that is Sandra's wife and Arash's wife and daughter. And this is an email, as I said, apologies if it's personal, but I just wanted to give you a little bit of an insight into Sandra. So it's an email when I wrote for his last birthday in March. I was worried that he didn't reply to me. And so I wrote to him checking that we weren't in the situation in which only one of us was dead. Because I think if both we were dead, we could still communicate. And Sandra replied saying he's not dead. He's just buried. And then he went on one of his literary outreach. This was Borges dear to both of us. He said Aleph's week. And he talks about Beatriz Viterbo. That was the woman loved by the poet in Aleph. And then he mentioned Zia Otto, beloved Zia Otto. And then he gives me some advice. He says, sorry, it's in Italian. He says, don't breathe a nerve gas on public benches. He was referring to the Salisbury poisoning. Don't elect an idiot. I will not tell you who was he referring. Don't sell assault rifles to disturb youths. And you see some social criticism. And of course, choose your parents carefully. That is something that you can't do. And then he goes on to James Fraser and the Golden Bull. Great inspiration for Thomas Elliott, our friend, Cantic Gosh. And I was going to LA to the APS. And he recommended me to not get shot, or in case I was in the middle of a shooting to be careful. And he concludes this took some time, but is an interesting thing. We were both found when we were in high school of metamagical games by Douglas of Stutter. And there was a famous story in which he mentioned 1453. That is the year of the fall of Constantinopolis. And it's a prime number. And this prime also, if you read it in reverse. And of Stutter came to the University of Trieste in 1986. May 1986, we were there listening to him. And Sandra was drawing a caricature of him. And then we go down and start talking with him. And I saw him that surrethitiously goes to his back and sticks the caricature in his jacket. So that night of Stutter went back and found the drawing of Sandra and this. So this was just to give you a little window in some of our many, many years of interactions. And these are the last pictures I took of him. The day we saw each other at a rush. And with this, I give it to Stefano for his talk. OK. OK, I'm Stefano de Gironcoli. And I was actually a colleague of Sandra's colleague at the university. And actually, these are the places where Sandra was born in Udina, not far from here. Graduated with Baldereschi as myself and Nicola a few years in Trieste. Baldereschi was a commuting between Trieste and Lausanne. And Lausanne actually was also one of the places where Sandra worked. From Trieste, he went to Kiel University. Then moved to, for PhD, then moved to Lausanne, working with Roberto Carr and colleagues. And then got positions both in UK and Trieste. And so started commuting. And so for most of the last, I don't know, 20, 30 years, he was commuting between Trieste and. OK, I didn't work explicitly with, I mean, we didn't publish anything together. But whenever he was here, we discussed stuff. He was updating me about what he was doing. We were planning or, I mean, wishing to do something together, but we are a bit different interest, et cetera. So this didn't happen. One thing that in the back in the days was not very clear was how to deal with, I mean, making simulation in the Taliq system with a reasonable number of k-points and how you can do that. So that we are talking about 30 years ago and in an efficient way. And of course, one idea is to introduce some temperature. And but introducing the temperature to spoil the results, especially if you want to do it with the machines that you had at that time. So you need to do very few k-points and therefore very large temperature. One thing that, I mean, Sandra was exploring was the fact that actually we are all interested in computing. OK, temperature or smearing, they look the same so they have similar properties. So we are interested in the low temperature result. But what we can afford is something that is high temperature because this allows to use a few k-points. What you can calculate is the internal energy or you can calculate the free energy. Both are significantly wrong. So what one can do is actually take the same sum, the average of them, and then this gives you a much more stable and close to the limit that you are interested in result. This is something that he was playing around in the 90s. And then the problem is that if you take this semi sum, this is non-variational. So the forces that you can compute from the variational problem are not consistent with the corrected with this. So the energy is accurate, but the forces are not consistent. And OK, but he was saying, OK, but these are metals, what you are missing affects a few electrons around the Fermi energy. How much this could be important? So maybe it's not so important. So I think then the solution, a more elegant solution, involves some modified smearing. But this is not the point. What I want to say is that he was always very happy and interested in trying to solve relevant problems, and even doing something that is not formally correct, but give you some insight, and then maybe things can be fixed. But don't be scared of doing something that maybe is not fully right. And so I mean, one memory that I have of him is that we were discussing about type of physics or type of approach that you may have on your work. And he was saying that, OK, you can play attack, or you can play defense. I mean, you can try to stay safe, or make bold moves. And I definitely think he was in the category of people that makes bold moves and have good insight. And then things will be, I mean, if something good come from that, as for instance, I mean, we had, and there will be a couple of games we'll talk, and other people will talk about this machine learning, learning on the flight, et cetera. I mean, in the past 20 years, we had several discussion, dinner discussion, about what were the problem left. And each time he solved the previous problem, but there were new problems arising, but he's never stopped him to think that he was going forward. So that's what I wanted to say. Thank you, Stefan. Thank you, Nicola, in particular, for having shown this beautiful email message. I think that summarizes really perfectly, as far as I'm concerned, what Sandra was. I mean, this combination of passion for art, for beauty, for literature, and so on and so forth. I think we're trying to describe Sandra with jargon that has been used in the last session. I think Sandra was really lying. I mean, probably six or seven standard deviations away from the average scientist, from the average of us, including myself, certainly. And I was trying actually to try to identify something that could really define what it was. I think what really characterized it was this passion for beauty. Beauty in all its manifestations, whether it was art, it was a book, it was a movie. And whether, of course, it was science. And so I've been asked here to say a few words about this contribution to science. It was an outstanding scientist, of course. This is a picture we took probably a year and a half ago. Well, you can see the wine. He loved the wine, by the way. So this is probably, I chose just a couple of papers just to give a flavor of, actually, the way I actually met him, I got to know him, because that was a time in which we were both beginning our careers. It was the early 90s. We were both graduating at our PhDs and both starting a career in the field of Abinition Molecular Dynamics, which is really the beginning. For those of you who are the younger members here in the audience, that was a time in which a few years after Roberto Carre and Michele Parinello in the mid-80s started this new enterprise of being able to do Abinition Molecular Dynamics for realistic systems. Of course, the beginning was just a toy model, but at the end of the 80s, beginning of the 90s, we're really starting to use and implement these techniques into software that could be used to solve practical and interesting problems. And I think this is probably the first example of a really large-scale simulation. Of course, you shouldn't be, I mean, 100 atoms at the time was really huge by any standard. So this is the first time that actually a chemical reaction was looked at. Chemical dissociation of a chlorine molecule on top of a silicon surface. Of course, this was really starting to hit that very relevant technological issues, which our field that was at the time still struggling with. I mean, we're still struggling with developing simple methodologies that could describe, say, bulk systems, crystal structures, but that was really the first time, at least as far as I remember, in which this methodology that our community is now using, you know, on a daily basis, we're starting to be used for a process that really had a strong technological interest. In this particular case, it was etching the physics of semiconductors, the engineering of semiconductors as well. And you can see here, well, I don't want to go through the details. I mean, you actually run a number of simulations looking at different starting conditions for the Cl2 molecule, and you actually manage to get insight into the electronic structure of the process from the time in which the molecule is still associated to the time in which the molecule dissociates at the surface and forms the bonds. Now, these figures are probably very, you know, common. You find them in any paper, but at the time, it was really a breakthrough. I must say. Let me also say something about a bit of an historical remark. I mean, Sandra was also the kind of person who liked to make connections and networks with everyone. At the time, the community of Abinition Molecular Dynamics was quite polarized. There was the group in the UK doing primarily Borno-Penheimer Abinition Molecular Dynamics and the group in Trieste doing primarily Carparinello-type molecular dynamics. Sandra was actually jumping between the two. It was probably the only one who was mastering both techniques quite deep way. And we mentioned a second, oh, sorry. Yes, this is actually a movie that Sandra and colleagues, I think I have to click on it. Oh, it's going, there's no sound. Okay, it's a pity because the soundtrack was actually Bach's Brandenburg Concerto. So again, it shows the interest that Sandra had. That's okay, that's okay. It's only music. It's actually short. You see the credits here and you see then a movie precisely of the trajectories that led to the results that were described in the paper I just showed. So this is chlorine impinging with an energy of one electron volt into the surface of silicon. Again, these kind of movies where are common nowadays, but at the time I was really shocking. I mean, I remember the first time I saw it, I was really shocked to see that with our techniques we would actually generate insight. And not only insight, but I guess you probably all agree that this is actually a beautiful piece of work just in terms of pure beauty. Let me actually just mention now a second paper. This is actually much closer to what I was doing at the time. This is related to the problem of the graphite-to-diamond conversion of which I've been working. That was actually my first paper in the field of Abinicio, in the Abinicio community. I was trying to look at something which is rather academic problem, looking at the transition between graphite-to-diamond, which is a very well-known, even from a technological point of view process. We're trying to understand essentially the microscopic details of the transition. To a large extent an academic problem, which however took a substantial amount of computer time because I was actually, I think this simulation was, if I remember correctly, 48 carbon atoms. And that was the best I could do. And then a couple of months later, sorry, I skipped this, sorry. A couple of months later, here comes Sandro with a simulation which is of course, or is of magnitude better than what I was doing. Essentially it was looking not just at the dime at the graphite-to-diamond transition, which is of course of sort of academic interest, but it was looking at the reverse transition, the transition of diamond into graphite, which had instead an enormous technological importance because that was the time in which the first examples of diamond growth by chemical paper deposition were published. And so there was this problem that because the temperature was high, the surface of diamond was actually graphitizing during the growth. So the graphite islands in these samples. So it was extremely important from a technological point of view, but it was of course also a tremendous tour de force from a computational point of view. As you can see now, he's dealing with 288 atoms plus vacuum. And to do that, he actually developed a code. I'm not sure, I mean, I think there are actually people that have been using this code, but it was actually the developer, the first developer of low-trek, the code that he developed when he was in Lausanne. It's actually Sandro. So not only was he interested in the application, but also did important, very important contributions into the development of software. That was the first, I guess, massively parallel code for a Carparinello method. And as I said, I wanted to close by showing this picture, this one here. This is actually taken from the chlorine on silicon paper. I like this picture because I think it really summarizes what Sandro was. This is a beautiful example of science. You see electronic structure, you see charged density, but I think nobody would be surprised to see this piece of art in a contemporary art museum, isn't it? It's really beautiful, I love it. So I think this really summarizes his passion for beauty. Hello, everybody. So I sort of represent a pretty large group of people who studied at the materials department at the University of Trieste and who were really truly inspired by the overwhelming enthusiasm that Sandro had for science and for computational material science in particular. Sandro actually got his second PhD at the department of materials in Trieste. Yes, he had two PhDs and later he became a faculty member until his death. The very first time I met Sandro was at the eighth edition of this conference at the, in January 1997. And I was still undergraduate and I was introduced to this young scientist, literally while he was playing guitar at the poster session, while discussing science, while also having fun with this group of friends that were shown at the beginning. And this meeting for me was literally life changing for many things, for many aspects. The very first one is that after a few months we became friends. After a few months he recommended me to my future PhD advisor who accepted me and he's actually the director of this edition of the Total Energy. So this is one very important thing that I owe to Sandro. So coming to science, so this has been shown already but those years, I'm talking about 1997, 2000. So Sandro was making his first footsteps on what later became his Triangle Trail that he kept doing over and over. And at that time one of the things he was studying and was involved in, among many others, was this field that was emerging and it was a hot scientific topic at the time was molecular self-assembly at surfaces. And really this is a seminar work in that area and it's one of the most cited paper in Sandro's publication list. And really with his simulation they were able to nail down the physical origin of the formation of these ordered structures at the atomic level and you see modeling showed that hydrogen bonds caused the observed self-assembly. Sounds obvious now but at that time it was not so obvious and actually he demonstrated this from that time over the next 10, 15 years. He kept working on this idea and elaborating it further and to the point that he added an intermediate stop along the Triangular Trail that was Stuttgart where we met again scientifically. And the reason for Stuttgart is that he developed at the time longstanding collaboration with the group of Klaus Kern. He is a pioneer of that field and over that period of time between 2002, how I would say it, some years ago he became a world, Sandro became a leading figure in this field. He gave many contribution explaining complex and subtle and very important problems in surface science. So here again, just two examples, again working on this concept of hydrogen bonding and the importance of hydrogen bonding at surfaces, something he explained with his hands. I remember he was explaining this chirality of one-dimensional hydrogen bond and assemblies explaining that this with his fingers, depending if they met this way or the other way, I'm not able to do it, but you actually see it there. And he developed this even further. He really was able to explain even much more complicated situation like this one in which there were phase transition going to in two-dimensional over molecular overlays, developing open structure and as a function of temperature more closed to the highest density. Another thing I learned from him and that he was really a peculiar quality of Sandro was his gift for communicating science with really original thinking and creativity and enthusiasm. And I thought that this cover of this cover paper in Angevanth Kemi somehow summarized this. You see, there's so much Sandro in it. There is London Street Life. There is the movie because of the molecular dynamics. There is the simulated STM images. There are molecules who sort of pair. There is the dimerization of Diffenina Levine on the surface, but not only they pair. You see, this is a pair of people. A couple actually dancing. So Estanoce Kiraetango on the surface. And this is another example, live camera action. And this was the sort of things that Sandro liked to explain. So these were the puzzles he saw on the microscope. Well, the colleagues showed him on the microscope. And this was really, yes. He provided the molecular understanding of why this happened, this pairing and the molecules changing the structure and locking into pairs that then formed the left or the right chains on the surface. So I wanted to sort of bring my experience of Sandro's life in this bunch of time. And here's a wonderful picture of him also explaining the soft skills that he had and the way in which he used the soft skill to succeed in science and to have such a good network and connection. So I appreciate the difficulty of switching from current research talk to this, but I will try to do that by, I think the best way to do it is to give you a further flavor of Sandro's enthusiasm for his atomistic work. And he really introduced me to Trieste, introduced me to this community, introduced me to thinking about molecular dynamics. And I want to tell you about one of his passions. He called it learn on the fly, some of the students and postdocs involved in it. It all started actually before I got to know him. That's the title of an MRS bulletin, 197. There's novel scheme for accurate MD simulation of large systems. And it's a great title. It's a title I would have loved to give to any of my favorites and we're still working on novel schemes for accurate, accurate MD. I got to know him when I came back from the US and got a postdoc in Cambridge and immediately was introduced to him by Mike Payne and started working on developing his ideas in this. And I want to point you, and the culmination of the first milestone of that is the second paper where the name learn on the fly was in the title already. I would like you to read this highlighted sentence. It expressed his very deep vision at the time which I took over and completely subscribed to. And it says, we take the view that there is no feasible analytic form for the classical force model, can reproduce simultaneously the reference model's accuracy and the transferability since these two properties post opposite requirements on the force model's complexity. So if you think back to the research talks, that has changed and Sandro has changed but we were all thinking following him that machine learning was nowhere on our radar that this can't be done. And the second picture is really his solution. He loved that figure. Essentially what that figure represents is an MD trajectory at each point of which you have a very simple functional form, a pair potential. And it's parameters keep changing because the real quantum mechanics locally is different and you inform those changes by doing quantum calculations say every 10 steps. And that idea filled my world at the time that all came from Sandro of thinking about MD as information, information transfer from quantum mechanics into the trajectory. And the end of that paper, the figure on the bottom, right, which is a figure of a crack. And again that fracture was really one of Sandro's passions and I got to understand that much later, but that it's really the prototypical problem for which these ideas and these method is perfectly suited and really the only way to do the fracture problem was something like this. And that is because you need about 100,000 atoms to set up the system but the bond breaking that's going on is exquisitely quantum mechanical. I'm going to show you a couple of movies that I and other students and postdocs working with Sandro have made. He all loved these movies and showed it, looked at it many times. So what you see here is that silicon crack, which for the first time we were able to simulate with learn on a fly. What you see on the top, once I started, let me start it and then talk over it. Yeah, so the top is what you'd get if you have a classical potential and don't update it and eventually it'll really go wrong. What you see on the bottom is the blue atoms are the ones where quantum mechanically computed information is being incorporated into the system and that gives qualitatively and quantitatively the right behavior, whereas what happens in the top is lots and lots of defect production and that's really not what happens experimentally. If you crack silicon carefully, it really cleaves. So this made him extremely excited and all of us thought that this is really the best thing in the world. And then as you were playing with this, this is going to be a similar, we now adjust the combined one, the learn on a fly with the red atoms, quantum mechanical and the yellow atoms staying classical, but what's happening here, as you do this very, very slowly, because you can, because you can afford to, the largest quantum calculation we need to do was a hundred atoms, is you see that step? It was actually one of, it was James Kermode who noticed it, the student in Cambridge and later a postdoc with Sandro. And those steps always occur stepping down in this orientation, only one way. And we all thought this was sort of very cute and we didn't realize the importance of this. Sandro realized the importance of this and he was the connection to an experimental group which once we started talking with them and Sandro was the link between the modelers and the experimentalists resulted in, I guess what is the highlight of this piece of work, is this paper in Nature in 2008. Sandro went to talk to the experimentalist, a guy in Israel, Dov Sherman, and asked him, have you ever seen anything like this? And he showed him these movies. And Dov, so scratched his head and said, well, there are these samples in my drawer from five years ago and Sandro loved telling this story which they put away because they couldn't explain what was going on. And those samples are shown there, this is an optical image of the fracture surface on the top left here, that's the rig in which you do the fracture. And they just couldn't explain what those ridges were. And this team organized and convened and organized by Sandro. We managed to connect these two things. So basically the claim is that these little steps are the origin of those ridges, the edges of that V. And we managed to connect, do something very rare, connect the breaking of a single bond, the rotation of a bond and then the subsequent breaking to a macroscopically, experimentally observable phenomena. Which are those ridges. I think it's a bit tenuous, but Sandro was persistent. He loved the system so much. He went on to really prove that this was the connection. If you stick a boron atom into the system, this was, now I wasn't involved in this work, it was James Cromode and Jean Pietro Morris. If you stick a boron atom into it, then that bond break, that bond rotation and step is much, much more prevalent. That boron atom really induces that. And they managed to correlate the density of boron atoms with the propensity for stepping and the appearance of those ridges, which really conclusively proved that this was going on. The next, I went on to do other things, but Sandro's assistance with fracture, he really was the expert in modeling brittle fracture in its all shape or form. This was another one of his favorite systems. This is a piece of silicon in which you inject hydrogen gas at high pressure. And you can control the depth to which the hydrogen is injected, and it's actually a way of, it's a way, it's something that's used in manufacturing silicon vapors, because you inject this hydrogen to a certain depth and the silicon cleaves as a very large, flat sheet off of that, the bulk. And Sandro wanted to understand how this happens microscopically, and he said it's the first, now chemo-mechanical corrosive fracture simulation in which those hydrogen atoms dissociate and thereby advance the crack at a much, much smaller load than the material would break otherwise. This, the last piece of work is more recent and really represents in some sense giving up that early view that potentials cannot be complex. And that's, and although Sandro loved the original idea of learn on a fly, when it was time to go beyond it, he did. And that's something that I keep thinking about and in admiration of him. So he got into the fray of using the machine learning techniques to do the same thing that he loved, fracture and enhancing MD simulations to bring in that quantum information. These are two examples of more recent work. This is a piece of silica, a much more complex system than anything that we did earlier. And they were able to study silica fracture on the bottom right. Again, more recent work on developing machine learn models for nickel clusters. We gave up on clusters ourselves. They were too hard. Sandro assembled a crack team of postdocs and students and showed that these clusters are also accessible. Before I hand over to Arash, I would like to show you the last, the last movie that he showed me. It's not actually, well, it is physics, but it is, he spent ages looking at it. See if that works. So this is, and three hours looking at this, and the last time we met, he insisted that I spend 15 minutes looking at it. And the smile on his face of the beauty of the world that you can do things like that is what remains with me. Can everyone hear me? Yeah. Sandro was one of the three founding directors of the Thomas Young Center, the London Center for Theory and Simulation of Materials, or the TYC for short, and he served continuously in that role since the foundation in autumn 2006. During that 12 year period, TYC flourished as an umbrella for research and education in theory and simulation of materials in London, at the four London universities of Kings, Imperial UCL, and Queen Mary's. Today, the TYC represents over 100 research groups, over 500 researchers from the departments of physics, chemistry, materials, earth sciences, mathematics, engineering, at these four universities. Now, successful cross-departmental networks and centers are rare in themselves, but successful cross-institutional and cross-departmental networks are even rarer and very, very precious. And the success of the TYC was due in no small measure to Sandro's unique personality. People have already talked about his warmth, his humanity, his wisdom, his generosity of spirits, and the esteem in which he was held by his colleagues. His breadth of scientific knowledge and interest was one of the things that really helped bring together this very diverse community in London and to make lasting relationships between the individuals in that community. Together with the other two founding directors, Mike Finnis, who's the chair of this workshop, and Mike Gillan, Sandro really created an environment in London where collaboration trumped competition, where researchers felt connected with one another, and where education of the next generation was really at the forefront of the mission for the center. And I think very importantly, and most importantly, this was all done with a great sense of fun. Sandro always loved to have fun. The word fun would always appear in any conversation that you had with Sandro. Joie de Viva was infectious, as was his mischievous sense of humor, and the almost childlike enthusiasm and wonder with which he devoured knowledge and new experience. Unlike most children, certainly unlike my child, he had a very keen awareness of the preciousness of the present, and he'd always have a camera close to hand to record snapshots of the moments that he was experiencing. Most recently with a Fujifilm X100S, of which he was very proud and about, which he could tell you all the technical details for hours, it seemed. In fact, I've never met anyone with Sandro's capacity for conversation and appetite for conversation. What should have been a five-minute chat would somehow turn into several hours of talking, after which he'd probably persuade you to go for a drink and then dinner, and then he'd crack out a couple of cigars, and this was just the gravitational pull of his personality. When you were talking to him, he was always present, always in the moment. You never felt he had somewhere else to go, a message to check on his phone, someone else to attend to, and the conversation could really go anywhere. He was equally happy talking about science, art, music, poetry, philosophy, for the latest technological gadgets that had come out. He was always an early adopter of gadgetry. I first met Sandro in 2000 in Cambridge, where he was a frequent visitor, and to me, this very enigmatic figure was definitely seemed cut from a different cloth to the usual theoretical physicist that I'd met, and he made a very immediate and memorable impression on me with a story about how Neapolitan's, like their espresso-served piping hot, the Neapolitan slang phrase that he told me begins with three C's, but it's too rude to repeat here. I'll let you look that up yourself. Some of the Italians in the audience might even know what it is. A few years later, Sandro had a pretty direct influence on my life. It was September 2006. Nicole already showed some photos from this birthday symposium for Alfonso Baldoreschi, where I bumped into him, and at that time I was a postdoc, and Sandro was very enthusiastically telling me about this new thing in London called TYC, which was founded sort of that very month, and that the London universities were about to sort of expand activity and theory and simulation of materials by making lots of faculty appointments over the next few years, and that I should really have a look at this. So he persuaded me into looking into it, into applying, and that chance encounter with Sandro in 2006, and his thoughtfulness in sort of telling me about this is what set in motion my move to London the following year, and my deep and continued connection with both him and with the TYC. If you think about it, one of Sandro's lasting legacies in London is his role in the expansion of theory and simulation activity. So for example, at King's College London, in 2006, there were three theory and simulation materials faculty members in the Department of Physics, and that's gone from three to over 10 now, including people like Tony Paxton, Mark Van Schilfgaard, Francesca Balletto, Nicola Bonini, Cedric Weber, who's here and gave a talk yesterday, and George Booth. And in fact, Sandro really understood the importance of connections and connecting people. He always seemed to know everyone, he seemed to have worked with everyone, and you look at his publication list, and actually you realize why he had over 200 co-authors on his publication. So if we would think about a Sandro index instead of an Erdosha index, I think there wouldn't be anyone in this room who has a Sandro index more than three. He was an avid supporter of CCAM and Psy-K, and of the TYC's role in strengthening those networks and centers. He was the driving force behind the establishment in 2012 of the UK's second node of CCAM, the JC Maxwell node that involves the TYC universities, Oxford and Cambridge. He was one of the founders of the UK's Materials and Molecular Modelling Hub, which is a new UK network for materials modeling and high performance computing associated with materials modeling. And he was also an enthusiastic supporter, an encourager of TYC initiatives that provided opportunities for young researchers in particular, including a mobility program for junior researchers in London, and strong support for ACESMA, which I think we'll hear about a bit later, the African School of Electronic Structure Methods and Applications. Heracles of Athens said, what you leave behind is not what is engraved in stone monuments, but what is woven into the lives of others. Sandro wove so much richness into the fabric of the lives of so many. Despite his tragic passing, I strongly believe he will live on in our hearts, thoughts and actions. Rest in peace, my friend. There will be a memorial event, a memorial symposium for Sandro, 30th and 31st of May, 2019. So this is just to save the date announcement. There'll be more details available soon. Be advertised on the PSI-K and SCAM mailing list, so you should all hear about it in due course. So just wanted to put that out now so everyone's aware.