WEBVTT 00:00:05.044 --> 00:00:08.784 Welcome to the Conversion Science and Instrument Formed podcast on collaboration. 00:00:09.864 --> 00:00:14.564 I'm Paul Verschurek, and together with my colleague Julia Lug, we speak with Dr. 00:00:14.724 --> 00:00:19.184 Sybrandt de Jong, Professor of High Energy Physics at Radboud University, 00:00:19.864 --> 00:00:22.424 where he's also the Dean of the Faculty of Science and Technology. 00:00:23.904 --> 00:00:27.664 Sybrandt is an expert on elementary particle and astroparticle physics, 00:00:27.884 --> 00:00:30.104 and he's a member of the CERN Council. 00:00:30.544 --> 00:00:35.164 Having conducted high energy experiments at large accelerator laboratories such 00:00:35.164 --> 00:00:40.504 as CERN, Ferney Lab, and the Pierre Auger Observatory, he reflects on how large-scale 00:00:40.504 --> 00:00:44.744 scientific collaboration is carried out. Sebrant, welcome to our podcast. 00:00:45.824 --> 00:00:53.624 Hello, good afternoon. And Sebrant, before we really delve into the questions around collaboration, 00:00:54.064 --> 00:00:58.224 it would be very helpful if you could situate us a little bit in your background 00:00:58.224 --> 00:01:01.584 and in your career path that brought us together today. 00:01:02.584 --> 00:01:06.304 Okay, well, I'm not completely sure what brought us together today, 00:01:06.464 --> 00:01:09.304 but I can certainly sketch my career path. 00:01:09.604 --> 00:01:14.944 So by building, I'm a physicist. 00:01:15.524 --> 00:01:20.844 So I studied physics, but also I have a keen interest in mathematics, 00:01:21.184 --> 00:01:23.384 computer science and astronomy. 00:01:23.664 --> 00:01:27.524 So I also picked up parts of that in my study. 00:01:27.524 --> 00:01:32.644 After graduating at the University of Amsterdam, 00:01:32.924 --> 00:01:39.884 I did my PhD research with NICEF, already on a large experimental collaboration 00:01:39.884 --> 00:01:44.344 in Hamburg, on the Hierar Ring at the Dacey Laboratory. 00:01:45.904 --> 00:01:50.444 Of course, there I was a very junior member of that collaboration. 00:01:51.504 --> 00:01:57.664 From that, after obtaining my PhD, I moved on to CERN. 00:01:58.684 --> 00:02:02.504 I've been for eight years at CERN as a CERN fellow and a CERN associate, 00:02:02.804 --> 00:02:10.304 really working on another large experiment, the OPAL experiment at the lab ring. 00:02:11.310 --> 00:02:15.610 And then after my stint at CERN, I came to Nijmegen. 00:02:17.190 --> 00:02:21.650 Somehow I called it coming back to Nijmegen, but I'm not really from Nijmegen, 00:02:21.670 --> 00:02:23.550 as many people will hear. 00:02:24.670 --> 00:02:30.250 I'm, of course, from the area of Amsterdam. But ever since I've been in Nijmegen, 00:02:30.310 --> 00:02:34.790 well, it first was called the Catholic University Nijmegen, and now it's called 00:02:34.790 --> 00:02:35.570 the Radboud University. 00:02:35.570 --> 00:02:42.850 And there I served in several managerial tasks. 00:02:43.090 --> 00:02:50.370 So I've been a department head, I've been the director of the School of Physics from 2000 to 2004. 00:02:50.650 --> 00:02:55.670 After that, I was the founding director of the EMAP Research Institute for mathematics, 00:02:56.010 --> 00:02:57.510 astrophysics and particle physics. 00:02:58.370 --> 00:03:05.530 After that, I founded the Radboud Pre-University College of Science, 00:03:05.790 --> 00:03:08.090 so an interface to high schools. 00:03:08.910 --> 00:03:17.030 And then I moved on to serve on the CERN Council and the last three years as 00:03:17.030 --> 00:03:18.370 the CERN Council President. 00:03:18.790 --> 00:03:22.910 And of course, meanwhile, I did many things in both together, 00:03:23.130 --> 00:03:28.270 always simultaneously in research and in governance and management. 00:03:29.090 --> 00:03:32.790 So far, I've been able to always combine it. 00:03:32.930 --> 00:03:36.450 So from the 1st of December on, I will be the Dean of the Faculty of Science. 00:03:36.590 --> 00:03:39.650 And now for the first time in my life, I won't be able to combine that anymore 00:03:39.650 --> 00:03:42.610 with doing really science myself. 00:03:43.570 --> 00:03:52.030 So I had to give that up and I'm still sort of in the process of getting accustomed 00:03:52.030 --> 00:03:54.150 to that. Right. Understood. 00:03:54.530 --> 00:03:58.610 Well, that's an amazing career up to this point. 00:03:59.550 --> 00:04:07.870 But then in the context of that experience, how do you define collaboration and what is it good for? 00:04:09.030 --> 00:04:13.530 Well, collaboration already is a very interesting word for those who are not 00:04:13.530 --> 00:04:15.050 Dutch listening to this podcast. 00:04:15.990 --> 00:04:19.950 Collaboratie, collaboration in the Netherlands means quite the opposite, 00:04:20.010 --> 00:04:22.610 I guess, from what it means anywhere else. 00:04:24.370 --> 00:04:31.010 So in Dutch, it's like siding with the enemy, which is, of course, 00:04:31.030 --> 00:04:33.410 more or less the opposite of collaboration. 00:04:35.790 --> 00:04:39.450 I've always been working in large experimental collaborations. 00:04:39.690 --> 00:04:43.990 So typically from ranging from, well, my very first collaboration I worked in 00:04:43.990 --> 00:04:46.430 as a master's student was only like 60 people. 00:04:46.790 --> 00:04:49.750 And then it scaled up from hundreds to thousands. 00:04:52.470 --> 00:05:01.290 So for me, sort of a natural way of dealing with, say, large and complex situations. 00:05:04.180 --> 00:05:11.660 So, in that sense, it's sort of my habitat, my natural habitat is working with large collaborations. 00:05:12.500 --> 00:05:16.500 And these collaborations I do, of course, there's the common goal, 00:05:16.660 --> 00:05:18.360 which is very unifying normally. 00:05:19.440 --> 00:05:22.960 But of course, there's also a lot of fighting, a lot of infighting, 00:05:22.960 --> 00:05:26.280 a lot of fighting between colleagues. 00:05:27.260 --> 00:05:30.840 Because these collaborations are usually formed from like, well, 00:05:30.880 --> 00:05:33.640 I've been in collaborations with over a thousand full professors. 00:05:34.360 --> 00:05:40.040 And of course, they're all sort of king of their own empire. 00:05:41.180 --> 00:05:45.300 But then in a much larger context, they have to work together as equals. 00:05:45.560 --> 00:05:49.900 Okay. And okay, depending on, especially depending on nationality, 00:05:49.960 --> 00:05:51.600 that gives more or less problems. 00:05:52.800 --> 00:05:56.960 So, but then how does that work? So if you look at these larger, 00:05:56.960 --> 00:06:03.260 the ones you just described, the really challenging ones, why would it work and why does it fail? 00:06:03.260 --> 00:06:07.960 Could you could you describe that or give an example the foremost thing why 00:06:07.960 --> 00:06:16.840 it it can work is to have a recognized common goal um typically the cycle in 00:06:16.840 --> 00:06:18.260 a large experiment is that. 00:06:19.220 --> 00:06:22.340 People dream of doing something spectacular 00:06:22.340 --> 00:06:25.420 and then they they they they 00:06:25.420 --> 00:06:28.160 find out okay i can i can't do it alone okay so i 00:06:28.160 --> 00:06:30.960 need companions through this and then they round 00:06:30.960 --> 00:06:33.760 up a couple of friends and then the circle gets larger and larger 00:06:33.760 --> 00:06:36.660 and then at some point you have like the the 00:06:36.660 --> 00:06:39.380 volume to to jump into the big 00:06:39.380 --> 00:06:42.080 enterprise uh of course at 00:06:42.080 --> 00:06:45.400 that point everybody is very enthusiastic about the common goal and 00:06:45.400 --> 00:06:48.320 then usually things like letters of intent 00:06:48.320 --> 00:06:51.100 are written which are of course a letter 00:06:51.100 --> 00:06:54.020 of intent it sounds like a letter okay but usually it's 00:06:54.020 --> 00:06:57.620 like a few hundred page book um with quite 00:06:57.620 --> 00:07:00.300 detailed uh prognosis of what you want to 00:07:00.300 --> 00:07:04.080 do and studies on how you're going to do it etc so at 00:07:04.080 --> 00:07:07.800 that point there's still like one community then um 00:07:07.800 --> 00:07:11.960 the trouble actually starts when these things get approved because 00:07:11.960 --> 00:07:16.660 then suddenly like the first hurdle the first goal has been taken the approval 00:07:16.660 --> 00:07:22.140 is there okay so what do we do now then of course typically there's years between 00:07:22.140 --> 00:07:28.160 a new collaboration approved and a large apparatus really being built. 00:07:28.340 --> 00:07:33.900 And so there's lots of designs in between. And this design phase is usually like a huge fight. 00:07:34.180 --> 00:07:38.500 It's like a huge cage fight because you know you're condemned to one another, 00:07:38.680 --> 00:07:43.280 yet you want to kill the enemy. You want your idea to prevail. 00:07:44.400 --> 00:07:48.480 So it's really like a cage, an enormous big cage fight. 00:07:50.680 --> 00:07:55.780 And then all sorts of things come in so not only the scientific argument of 00:07:55.780 --> 00:07:59.980 what is best but also in the end my ideas may be second best but I pay for it 00:07:59.980 --> 00:08:02.620 so it's my idea that's going to be built, 00:08:04.440 --> 00:08:09.420 and actually this is one of the better arguments there are worse arguments that make people win, 00:08:11.520 --> 00:08:13.720 which I won't go into but. 00:08:16.500 --> 00:08:22.620 Okay and then sort of at some point okay this fighting takes usually a lot of 00:08:22.620 --> 00:08:26.500 time and a lot of energy people do this actually typically for years in these large collaborations, 00:08:27.580 --> 00:08:32.840 and at some point I realized like oh gee gee, we should have started building 00:08:32.840 --> 00:08:37.880 half a year or a year ago to make it for the deadline of, say, 00:08:37.940 --> 00:08:42.420 the accelerator to be finished or the promised start of the project. 00:08:43.000 --> 00:08:47.960 And then sort of there's usually there's first a panic. 00:08:48.540 --> 00:08:52.920 Then, of course, there's a lot of blaming and shaming on who's guilty of this. 00:08:53.400 --> 00:08:58.860 And then people realize like, okay, this doesn't help to actually pull it together. 00:08:58.860 --> 00:09:01.780 We have to really work together again so then 00:09:01.780 --> 00:09:04.680 there's a there's usually a very rapid stage of 00:09:04.680 --> 00:09:07.500 convergence uh typically like in 00:09:07.500 --> 00:09:10.340 months where people realized okay we we just 00:09:10.340 --> 00:09:13.140 and then and then um people just 00:09:13.140 --> 00:09:19.960 tend to be more soft on giving up some of their own ideas etc like okay anything 00:09:19.960 --> 00:09:25.100 okay as long as it's been built and so then there's this phase of of building 00:09:25.100 --> 00:09:31.680 where people are pretty much unified and then you at the point where the whole thing is going to be, 00:09:32.440 --> 00:09:39.000 commissioned uh there's usually like a great group spirit okay but things tend to work together. 00:09:40.300 --> 00:09:45.120 Sometimes miraculously there's also great satisfaction and group identity and 00:09:45.120 --> 00:09:49.980 big parties sometimes parties of thousands of people you i mean you build belief 00:09:49.980 --> 00:09:53.880 in corona times um and uh And then, 00:09:53.920 --> 00:09:56.600 of course, you have to start the experiment. 00:09:56.680 --> 00:10:01.020 And usually the first few measurements are, again, really like the core business. 00:10:01.320 --> 00:10:04.040 So everybody's unified, big success. 00:10:04.820 --> 00:10:12.080 And then you get into the stage where you have to accumulate more data and to 00:10:12.080 --> 00:10:18.420 either supersede the first publications or to do something entirely new. 00:10:18.540 --> 00:10:21.360 You just need so much data that you need to wait for years. 00:10:21.360 --> 00:10:25.020 Years, and then people start to fight again, because there's one analysis against 00:10:25.020 --> 00:10:28.300 the other, because they're basically waiting for data, they get a little bit 00:10:28.300 --> 00:10:34.840 bored, okay, the initial successes are over, and the goal becomes sort of fussy. 00:10:35.200 --> 00:10:39.800 So my conclusion from all of this is that it's all in the goal, okay? 00:10:43.521 --> 00:10:49.101 You can get people working together. It was a long story, but... 00:10:49.101 --> 00:10:51.441 No, no, no, it's very good. The conclusion is relatively simple. 00:10:51.921 --> 00:10:57.921 But, Saber, if we focus on the goal as the unifier, is that really enough? 00:10:58.041 --> 00:11:03.021 Because in some sense, it's also, at some point, also these phases that you 00:11:03.021 --> 00:11:07.281 described, right, of infighting, coalescing again, and so on. 00:11:07.821 --> 00:11:14.621 In some sense, now you also have a limited resource that you know you have to 00:11:14.621 --> 00:11:17.661 share in order to achieve your selfish objectives. 00:11:18.381 --> 00:11:23.321 So that is our new element that comes in. So is the goal and actually still the same? 00:11:23.721 --> 00:11:28.901 Yeah, but that's okay. So again, okay, the resources can play both parts. 00:11:29.041 --> 00:11:32.901 Okay. So of course, when the project is defined, everybody realizes that, 00:11:32.941 --> 00:11:34.901 okay, I cannot pay for this on my own. 00:11:35.001 --> 00:11:38.001 So we will have to share it. And then everybody, of course, 00:11:38.001 --> 00:11:41.121 immediately agrees that, yeah, of course, if we have to to share resources 00:11:41.121 --> 00:11:44.541 we also have to share responsibility and then 00:11:44.541 --> 00:11:48.101 share tasks and and okay fine 00:11:48.101 --> 00:11:54.381 no problem but then especially in this design phase okay when people have already 00:11:54.381 --> 00:11:59.621 committed basically their resources then it's like yeah but you give like a 00:11:59.621 --> 00:12:04.681 thousand and i give ten thousand so i can tell you what you should do so there 00:12:04.681 --> 00:12:06.961 the resources are abused again 00:12:07.101 --> 00:12:10.721 okay and everything has to be built usually yes 00:12:10.721 --> 00:12:13.681 yes before we go to that stage 00:12:13.681 --> 00:12:16.541 i want to back up again to the goal because you're 00:12:16.541 --> 00:12:23.161 talking about recognizing a common goal but how is that common goal constructed 00:12:23.161 --> 00:12:28.221 in other words if you have a collaboration of eventually a thousand people are 00:12:28.221 --> 00:12:35.561 a thousand people involved in constructing the common goal who constructs that common goal okay well, 00:12:36.921 --> 00:12:44.941 again okay so usually of course there's like a couple of people that really pull the cart okay that. 00:12:46.303 --> 00:12:49.363 Really express the the goal and and 00:12:49.363 --> 00:12:52.683 usually those are the people that can sort of make good 00:12:52.683 --> 00:12:55.603 presentations present it in a in a in a 00:12:55.603 --> 00:12:59.403 in a very nice way so that it's convincing but 00:12:59.403 --> 00:13:03.183 usually these goals um that are 00:13:03.183 --> 00:13:06.103 then defined by individuals already i like living 00:13:06.103 --> 00:13:09.683 in the community for quite some time so 00:13:09.683 --> 00:13:12.843 um and that's where people then recognize it 00:13:12.843 --> 00:13:15.483 say yeah yeah i'll i thought the same 00:13:15.483 --> 00:13:19.603 and and then they join this collaboration uh of 00:13:19.603 --> 00:13:22.943 course also projects actually exist where people are 00:13:22.943 --> 00:13:28.043 so much ahead of the troops that they have marvelous ideas okay but if it's 00:13:28.043 --> 00:13:33.103 not it's if it's not living in the community okay then they they consider that 00:13:33.103 --> 00:13:36.723 they have this brilliant idea with the three of them okay and they have to find 00:13:36.723 --> 00:13:41.503 like a half a billion bucks with the three of them and And it's not going to happen. 00:13:42.303 --> 00:13:47.063 But then, Sebron, do you see a stage where the people who might be in front 00:13:47.063 --> 00:13:51.743 of the troops start really an active, let's say, campaign? 00:13:52.403 --> 00:13:56.743 Yes, that's usually what is being done, okay? Okay, so how is that structured? 00:13:59.983 --> 00:14:05.223 Yeah, no, it's just like going for it, okay? Okay, so you have this idea, 00:14:05.423 --> 00:14:10.343 and then typically you try to get conference talks. 00:14:10.643 --> 00:14:16.763 You try to get, and also what is very powerful in our case is you try to get 00:14:16.763 --> 00:14:19.123 invited at certain research institutes. 00:14:19.263 --> 00:14:24.263 You present your case, and especially you go with these people for lunch and 00:14:24.263 --> 00:14:28.143 maybe dinner, and you try to convince them that it's a really good case and 00:14:28.143 --> 00:14:30.803 that it would be really marvelous if their group would be joining. 00:14:30.803 --> 00:14:36.443 And the other thing that is in the first stages is that, of course, 00:14:36.463 --> 00:14:38.363 at that point, it doesn't cost anything. 00:14:39.063 --> 00:14:42.663 So it's practically resource-free because it's just an idea. 00:14:43.823 --> 00:14:48.063 So you join this collaboration, and then it's just a little bit, 00:14:48.223 --> 00:14:50.243 we sort out the business later, okay? 00:14:51.643 --> 00:14:56.883 First, join us, okay, and do a few studies and run a few simulations and see 00:14:56.883 --> 00:15:06.023 what you can contribute to the goals or to the ideas of realizing the goals. And then we see. 00:15:08.578 --> 00:15:14.178 This is usually how collaboration is formed and then it's then sort of it acquires 00:15:14.178 --> 00:15:16.838 sort of it goes over a critical mass, 00:15:17.778 --> 00:15:23.118 but then you start looking for resources and then actually this helps because 00:15:23.118 --> 00:15:27.238 you're already tied into all the other organizations the universities, 00:15:27.678 --> 00:15:32.238 the research institutes etc and then each of them has to fight in their own 00:15:32.238 --> 00:15:39.578 institute for a piece of the pie right But that means you first build a coalition around the idea, 00:15:39.738 --> 00:15:45.778 which is a process that has psychological, social components to it. 00:15:45.998 --> 00:15:52.498 Does that also lead to, let's say, a regression to the mean of the idea? 00:15:52.758 --> 00:15:54.338 Do you see that also as a compromise? 00:15:54.678 --> 00:16:00.918 Yes, of course. Like I said, at several stages, the idea is watered down. 00:16:01.198 --> 00:16:03.278 There's no question about it. 00:16:05.318 --> 00:16:09.878 Usually in the goals, okay, if you're still in the dreaming world, 00:16:09.998 --> 00:16:13.998 okay, you can actually stack goals. So it doesn't matter too much. 00:16:15.438 --> 00:16:20.798 Then when you get into the stage where you have to secure the resources, 00:16:21.138 --> 00:16:24.518 then that's the first time you're confronted with reality. 00:16:25.198 --> 00:16:28.438 And then you find out, okay, maybe you have to descope a couple of things, 00:16:28.538 --> 00:16:30.798 or maybe you have to descope many things. 00:16:30.798 --> 00:16:40.038 And then usually, of course, the people that can secure the money, 00:16:40.178 --> 00:16:45.138 can secure the resources, their ideas will prevail. 00:16:47.618 --> 00:16:53.618 Of course, usually also a couple of good ideas that are in common to everybody will also make it. 00:16:53.698 --> 00:16:58.698 There's no doubt about that. But then some other ideas of people that are less 00:16:58.698 --> 00:17:05.978 resourceful, they will have to go because they are compromised on that. 00:17:06.658 --> 00:17:12.158 And in fact, it's true that sometimes good ideas go. Right, exactly. 00:17:12.618 --> 00:17:19.458 But then, so now you form your consortium in some sense. 00:17:19.578 --> 00:17:23.598 Now they started to compete with other consortia to dominate for the resource. 00:17:23.598 --> 00:17:30.518 But now the whole process is extended over years so what's the mean what's the 00:17:30.518 --> 00:17:32.678 duration of such a process? 00:17:33.258 --> 00:17:39.538 10 years? So in our field the process of doing an experiment is typically that 00:17:39.538 --> 00:17:45.678 you have a phase of 5 to 10 years which we call proto-collaboration so that's 00:17:45.678 --> 00:17:47.038 where the collaboration is formed, 00:17:47.818 --> 00:17:50.378 then typically you have a phase of 5. 00:17:52.001 --> 00:17:58.081 Well sometimes even to 10 years where you try to secure resources it's also 00:17:58.081 --> 00:18:02.581 true that you never secure the resources that you really need so you have to 00:18:02.581 --> 00:18:05.181 decide at what point you're going just go, 00:18:06.801 --> 00:18:14.221 so typically at 50 60 of the resources pledged you just go and you hope for the rest, 00:18:14.921 --> 00:18:18.901 typically by the way this is a strong mechanism because 00:18:18.901 --> 00:18:21.721 um this is like with all sort of large 00:18:21.721 --> 00:18:24.621 as public projects if you go over a certain uh 00:18:24.621 --> 00:18:28.161 volume it's too 00:18:28.161 --> 00:18:30.901 big to fail so typically you will 00:18:30.901 --> 00:18:33.941 get so typically what what happens is you find 50 or 00:18:33.941 --> 00:18:36.901 60 percent and there's not so much problem in 00:18:36.901 --> 00:18:41.161 finding another 30 percent it's usually the last 10 percent which is a problem 00:18:41.161 --> 00:18:46.521 um because then the argument of too big to fail doesn't work anymore and for 00:18:46.521 --> 00:18:51.581 that you have to descope This is typically also a pity because de-scoping for 00:18:51.581 --> 00:18:56.401 the last 10% actually means that typically the apparatus only works half as well, 00:18:57.361 --> 00:19:04.801 which is something that funding organizations do not always realize. 00:19:07.617 --> 00:19:12.277 But yeah, this is how it goes. And then there's the time of construction. 00:19:12.497 --> 00:19:17.037 Again, time of construction in our field is by now five to 10 years. 00:19:17.257 --> 00:19:21.257 And then there's the time of exploitation, which is typically in our field now 00:19:21.257 --> 00:19:23.777 going up to 20 years or 25 years. 00:19:24.497 --> 00:19:26.277 So you're talking about... 00:19:27.937 --> 00:19:33.557 40 years. You're talking about people that actually can spend their entire career, 00:19:34.317 --> 00:19:36.037 doing one project. Right. 00:19:36.557 --> 00:19:41.017 This is important, right, to understand, because now the question becomes, 00:19:41.277 --> 00:19:45.677 if the goal is so critical in keeping this together, 00:19:46.297 --> 00:19:52.317 how do you assure that that goal also stays intact over that whole period, 00:19:52.377 --> 00:19:57.077 is propagated properly, and is also that you don't have mission drift, 00:19:57.357 --> 00:19:59.617 right, so that the goal fragments? 00:19:59.617 --> 00:20:04.777 No, in our case, okay, I've never seen that really happen. 00:20:05.257 --> 00:20:11.177 So the goals are typically so well, like these are typically already questions 00:20:11.177 --> 00:20:15.257 that are decades above the market. 00:20:15.477 --> 00:20:18.297 So really big things that we want to solve. 00:20:18.297 --> 00:20:25.497 Okay um some well what what are the most interesting cases of mission drift 00:20:25.497 --> 00:20:31.017 of course is that you you you build something for a certain goal and then the 00:20:31.017 --> 00:20:34.837 thing is built and then you switch the apparatus on and you look at the data 00:20:34.837 --> 00:20:37.497 and you just find out something completely different. 00:20:38.477 --> 00:20:44.417 We we don't call this mission drift okay it's just like okay you adapt your 00:20:44.417 --> 00:20:50.197 goals would adapt if the implementation stage you find new information that 00:20:50.197 --> 00:20:51.737 will change your goals or bring you... 00:20:51.737 --> 00:20:56.997 Typically, the old goals are still there, but you acquire new goals. 00:20:57.057 --> 00:21:00.077 And sometimes the new goals sort of overshadow the old goals. 00:21:01.657 --> 00:21:06.297 So famous for that is that there was an experiment that tried to look at decay 00:21:06.297 --> 00:21:11.177 of protons, proton in the elementary particle, and it's predicted to decay, 00:21:11.337 --> 00:21:14.577 but with a very long lifetime because otherwise we wouldn't be sitting here. 00:21:16.117 --> 00:21:20.777 And so you need a huge detector, and then you look for the one proton to decay, 00:21:20.977 --> 00:21:23.257 one in enormously many protons. 00:21:23.517 --> 00:21:26.577 So they build a huge detector, they wait for this proton to decay, 00:21:27.377 --> 00:21:32.217 and what they found out is that for a completely different particle called neutrino, 00:21:32.917 --> 00:21:36.677 they actually did some measurements which were done by the PhD student, 00:21:37.217 --> 00:21:41.197 because, well, they have to do something while waiting for the proton to decay, 00:21:41.377 --> 00:21:46.877 and they actually found out that these neutrinos, they change their nature. 00:21:47.417 --> 00:21:51.717 So a neutrino could start as one particle and after some while could end up 00:21:51.717 --> 00:21:55.457 as a completely different particle, still a neutrino, but a different neutrino, okay? 00:21:55.697 --> 00:21:58.637 So they could completely sort of... 00:22:00.814 --> 00:22:03.754 Change. This was a fantastic discovery. 00:22:06.414 --> 00:22:11.314 But the thing is, if it runs over such a long period of time, 00:22:12.194 --> 00:22:14.954 does it also mean you have different generations? 00:22:14.974 --> 00:22:21.134 That's, for instance, the first push even by a senior generation that in the 00:22:21.134 --> 00:22:24.694 end also will hate each other so much they cannot go on, but then there's a 00:22:24.694 --> 00:22:29.534 younger group that comes in to push you to the next stage. That's part of the dynamics. 00:22:29.574 --> 00:22:34.954 Typically, it's the old people just before or just after retirement that dream 00:22:34.954 --> 00:22:39.714 of the new big things that they will actually never see working alive. 00:22:41.434 --> 00:22:46.534 And then in the process, of course, they drag on the younger colleagues. 00:22:46.534 --> 00:22:52.434 And the most interesting thing is that the peak of, say, 00:22:53.034 --> 00:22:59.454 the young colleagues joining the collaboration is typically when the experiment 00:22:59.454 --> 00:23:02.654 is already running for a while. So in the last stages. 00:23:04.714 --> 00:23:09.434 And part of it is because the old guys in the experiment then are already busy 00:23:09.434 --> 00:23:12.254 with dreaming up the next project. Right. 00:23:13.774 --> 00:23:17.754 So we go from old to young in these collaborations. 00:23:18.354 --> 00:23:24.594 Right. The average age goes down in steps, but it goes down along the way. Right. 00:23:24.854 --> 00:23:30.254 So now we looked at the important challenge of goal setting and how this might 00:23:30.254 --> 00:23:32.614 be maintained or might also change. 00:23:32.994 --> 00:23:38.314 But you also mentioned that there are these serendipitous discoveries that were 00:23:38.314 --> 00:23:43.154 not part of the original path. path, on the average, how is that balanced? 00:23:44.233 --> 00:23:50.033 So if you look at the high-impact outcomes of all these huge experiments, 00:23:51.013 --> 00:23:55.993 are most of them on the critical path of the original goal, or are they on this 00:23:55.993 --> 00:23:58.013 serendipitous offshoot? 00:23:58.573 --> 00:24:04.673 No, typically for all the experiments that I know, hardly any of them actually 00:24:04.673 --> 00:24:10.233 have failed to fulfill their initial goal. 00:24:10.233 --> 00:24:17.453 All. Of course, sometimes you're looking for things and the things turn out not to exist, 00:24:18.493 --> 00:24:24.413 which is among the public and it's called a failure, but we think actually this 00:24:24.413 --> 00:24:27.753 is success because that makes a paradigm shift because we always thought something 00:24:27.753 --> 00:24:29.833 was there and now we discover it's not. 00:24:30.893 --> 00:24:36.293 So that is a big discovery, basically, for us, because that means that your 00:24:36.293 --> 00:24:39.733 common theory doesn't work, which is a big, 00:24:40.533 --> 00:24:44.093 overturn of the thinking exactly so 00:24:44.093 --> 00:24:47.053 in that sense in that sense the goals are nearly 00:24:47.053 --> 00:24:49.853 always met like either you find what you 00:24:49.853 --> 00:24:54.033 were looking for and in fact personally okay i always find like you have this 00:24:54.033 --> 00:24:58.813 theoretical idea and you you set up to check it and then you check it and it 00:24:58.813 --> 00:25:04.173 turns out to be true yeah it's like ticking a box okay it's really boring so 00:25:04.173 --> 00:25:08.253 it's much nicer if you don't find it, or you find something completely different. 00:25:09.973 --> 00:25:15.333 So now we looked at the goals, we looked at outcomes, but as you mentioned, 00:25:15.373 --> 00:25:17.833 the process is complex and also contentious. 00:25:17.853 --> 00:25:20.113 There are phases where people, like you said, are in a cage fight. 00:25:21.017 --> 00:25:25.717 Now, another aspect of collaboration is that also the different participants 00:25:25.717 --> 00:25:31.217 at least have a sense of trust in each other and in the process to move forward. 00:25:31.517 --> 00:25:39.057 So how is trust then defined and maintained in such a complex process? 00:25:39.577 --> 00:25:45.337 I don't know how trust is defined, but it turns out that even in these cage 00:25:45.337 --> 00:25:49.817 fights or at the end of the cage fights, people make up. 00:25:49.817 --> 00:25:52.837 So they they still know that 00:25:52.837 --> 00:25:55.737 they are condemned to one another okay there's there's no way 00:25:55.737 --> 00:25:58.597 they can they can have an exodus of 00:25:58.597 --> 00:26:04.217 people like i the famous thing okay we have is that very often we have like 00:26:04.217 --> 00:26:10.837 three ideas on certain detector construction and they're nearly equally good 00:26:10.837 --> 00:26:14.637 and of course there are some deciding criteria and you could sort of say who's 00:26:14.637 --> 00:26:17.537 best okay but but the second and third best are not so far off. 00:26:19.017 --> 00:26:25.837 And so I heard one spokesman say, like, whatever I do, okay, 00:26:25.917 --> 00:26:28.737 I will lose two-thirds of my collaboration in this decision. 00:26:29.817 --> 00:26:33.757 And this is true. But we also know that if you lose two-thirds of your collaboration, 00:26:34.037 --> 00:26:35.677 the whole thing is not going to happen. 00:26:37.357 --> 00:26:43.197 So people also know that they can be sore about things, okay, 00:26:43.197 --> 00:26:47.337 And they can sort of, yeah, what do you do then? 00:26:47.397 --> 00:26:52.477 Okay, you kick in a door or so, and then you... 00:26:54.208 --> 00:26:58.488 Go home and um but how 00:26:58.488 --> 00:27:02.628 could how can we do how do you get a little bit of shopping you return and you 00:27:02.628 --> 00:27:08.368 say well how can i help okay so you give in in the end but is trust not based 00:27:08.368 --> 00:27:13.848 because in that sense physics is a very unique domain right it's one of the 00:27:13.848 --> 00:27:16.288 few if to my knowledge the only area, 00:27:16.928 --> 00:27:22.248 in science where people succeed in pulling off these huge collaborative initiatives 00:27:22.248 --> 00:27:28.408 but But isn't that also not related to the availability of, let's say, 00:27:28.408 --> 00:27:31.548 a theory that everybody trusts in some sense? 00:27:31.788 --> 00:27:39.668 Yeah, that's also true. Like I said, that's also why there is so little goal drift. 00:27:41.788 --> 00:27:44.988 Because usually the goals are already sort of stably defined. 00:27:45.268 --> 00:27:51.928 And it also plays a role in what I said about there are a few forerunners that promote ideas, 00:27:52.248 --> 00:27:55.528 but there is lots of followers that already thought like, well, 00:27:55.628 --> 00:28:00.548 I had the same idea, okay, so I can as well join them and then we'll make it to success together. 00:28:01.648 --> 00:28:07.928 So it means that the meme is already living in the crowds. 00:28:09.108 --> 00:28:14.508 Exactly. And that, of course, has much to do with like an underlying theory, 00:28:14.508 --> 00:28:20.828 which is commonly believed in and relatively stable. 00:28:21.268 --> 00:28:24.788 So it's not like we change theory every couple of years. 00:28:25.488 --> 00:28:29.468 Right. And so what you're saying is that the trust is in the thing itself, 00:28:29.688 --> 00:28:33.848 in the ideas, in the concept, and not so much interpersonal trust. 00:28:34.148 --> 00:28:35.188 Is that what you're saying? 00:28:35.528 --> 00:28:38.828 Oh, there's also interpersonal trust. There's also interpersonal distrust. 00:28:39.528 --> 00:28:41.568 But the interesting thing is 00:28:41.568 --> 00:28:45.868 that even people that have mutual distrust in the end will work together. 00:28:47.781 --> 00:28:55.581 And really cooperate. Is competition in the sense a positive thing for the outcome? 00:28:55.921 --> 00:29:00.521 I don't mean competition over funds, but I mean competition in terms of ideas. 00:29:01.881 --> 00:29:07.381 Yes, I think overall, yes. So what the competition makes is that people are 00:29:07.381 --> 00:29:08.881 just sort of working harder, 00:29:08.981 --> 00:29:16.681 going the extra mile in developing their ideas and in presenting their ideas 00:29:16.681 --> 00:29:18.181 and getting the details right. 00:29:18.301 --> 00:29:24.661 Because they know that if any of these has a little flaw, then they will be shot by the other party. 00:29:24.921 --> 00:29:30.741 So in that sense, it works very much elevating the quality. 00:29:30.741 --> 00:29:34.761 On the other hand okay practically always 00:29:34.761 --> 00:29:37.581 it's like not quite the best idea that 00:29:37.581 --> 00:29:40.801 wins but nearly the best idea but then our my 00:29:40.801 --> 00:29:44.561 my feeling is that nearly the best idea then would still be much better than 00:29:44.561 --> 00:29:48.981 in the case if there would have been no competition right it's actually quite 00:29:48.981 --> 00:29:54.881 subtle the whole thing but now if you have competition about ideas in in in 00:29:54.881 --> 00:30:00.641 science there are various characteristic mechanisms also to express this competition as, 00:30:00.681 --> 00:30:03.201 for instance, in the critical dependence on peer review. 00:30:03.541 --> 00:30:10.421 So do you see that people also use that mechanism to dominate or win in that competition? In which? 00:30:11.341 --> 00:30:15.221 Because I'm trying to figure out what you mean, okay? 00:30:15.321 --> 00:30:20.861 So if one part has to sort of be the referee of the other part. Exactly, yeah. 00:30:21.601 --> 00:30:25.581 There's interdependence, In the background, there's interdependence because 00:30:25.581 --> 00:30:29.721 everyone is, in some sense, a potential reviewer of everybody else. 00:30:30.741 --> 00:30:36.661 Yes, yes. And in our fields, they're more or less permanently because it's sort of in the open. 00:30:37.301 --> 00:30:42.681 So I think it's not that, okay? It's not that that influences things so much. 00:30:43.501 --> 00:30:49.841 What does influence things is that, of course, some people are more powerful than others. 00:30:51.220 --> 00:31:00.320 And uh so there there's there's also a lot of backroom policy so so a lot of 00:31:00.320 --> 00:31:06.100 i mean basically all the important decisions are taken uh in the cafeteria. 00:31:07.540 --> 00:31:13.600 Right so so so there's a lot of wheeling and dealing and hassling okay uh to, 00:31:15.700 --> 00:31:18.800 to set up certain people against other people, et cetera. 00:31:18.980 --> 00:31:24.100 So there's, of course, also physicists are completely human. 00:31:24.840 --> 00:31:30.600 Yes. But now you have been at different sides of this process. 00:31:30.760 --> 00:31:31.840 You have been in the middle of it. 00:31:32.140 --> 00:31:36.260 But also, if you had the board of CERN, you have to manage it. 00:31:36.880 --> 00:31:42.320 So do you then, from that perspective, are you able to really manage and or 00:31:42.320 --> 00:31:46.620 engineer that process by, for instance, bringing people together in certain 00:31:46.620 --> 00:31:49.080 ways or setting up forms of communication? 00:31:49.380 --> 00:31:52.540 That's also where we distinguish things. Okay, so I was not in management. 00:31:52.680 --> 00:31:54.960 I was in governance, which is a different game. 00:31:56.180 --> 00:32:00.840 And in fact, most of the structures are set up in governance, not by management. 00:32:03.440 --> 00:32:07.140 And actually, this is an important distinction distinction. 00:32:08.060 --> 00:32:14.440 Because in governance, there's a little bit more distance because the governance 00:32:14.440 --> 00:32:20.460 is always over a number and maybe even a large number of projects. 00:32:21.700 --> 00:32:29.460 So it's more like a mechanism that would fit all the different things. Okay. 00:32:29.560 --> 00:32:33.040 Well, management is much more trying to tailor 00:32:33.140 --> 00:32:42.980 things to a certain situation and it really helps that the structures are set 00:32:42.980 --> 00:32:45.120 up from a more global perspective. 00:32:45.660 --> 00:32:51.580 Because I've also seen structures that have been set up to tailor the situation 00:32:51.580 --> 00:33:00.620 and they nearly always end up getting messy because they were so much tailored to the situation. 00:33:02.689 --> 00:33:10.029 New events arise they have to be adapted and then you have to add on things 00:33:10.029 --> 00:33:13.669 and you have maybe to scrap a few things and add on more things and it should 00:33:13.669 --> 00:33:18.209 sort of get a biologically grown structure and in the end, 00:33:18.949 --> 00:33:22.329 that is very advantageous for the cowboys but not for the. 00:33:24.029 --> 00:33:28.689 Straight thinking people yes, but also then the management becomes part of the 00:33:28.689 --> 00:33:31.909 situation to be managed yes, exactly Exactly, okay, exactly. 00:33:32.329 --> 00:33:37.789 But then from the governance perspective, what you can do is put in place procedures, 00:33:38.589 --> 00:33:42.969 protocols to try to structure the collaboration. 00:33:43.629 --> 00:33:48.729 Now, for CERN, did you also do that in a very deliberate way that you also knew? 00:33:48.869 --> 00:33:52.789 Yeah, all the things are structured very deliberately. 00:33:53.629 --> 00:34:00.049 So the idea is to have peer monitoring on practically everything. 00:34:01.229 --> 00:34:06.769 So that's an important thing. And so those are people that are in the spare 00:34:06.769 --> 00:34:11.069 monitoring that are very knowledgeable about what is going on, 00:34:11.149 --> 00:34:16.129 but they're completely independent to both the thing they monitor and to the 00:34:16.129 --> 00:34:18.469 body they actually report to. 00:34:19.509 --> 00:34:26.029 So also it's not like the body that is in the governance is actually arranging 00:34:26.029 --> 00:34:31.309 the monitoring, but it arranged it through a third party, which is an independent monitor. 00:34:32.349 --> 00:34:39.649 So they can also sort of say things that are not very much to the liking of the governance. 00:34:41.300 --> 00:34:47.260 And without consequences. And that is a very powerful structure. 00:34:47.980 --> 00:34:55.100 So peer monitoring is a quality control system, but that has to operate on, 00:34:55.140 --> 00:34:58.380 let's say, protocols of, let's say, communication, responsibility. 00:34:59.980 --> 00:35:04.620 So are you also structuring the OSINT in a very specific way? Like hierarchy? 00:35:08.240 --> 00:35:15.140 Well, this monitoring is usually sort of, there's not, okay, 00:35:15.200 --> 00:35:17.760 in our case, there's not so much hierarchy in the monitoring. 00:35:18.620 --> 00:35:25.120 So there's monitoring at, in principle, at a rather global level, 00:35:25.200 --> 00:35:29.680 but they're allowed to actually go into the detail. tell. 00:35:30.560 --> 00:35:35.740 So it monitors a large span of things. 00:35:36.020 --> 00:35:41.540 So part of my career, I've been into one of these monitoring bodies. 00:35:42.380 --> 00:35:50.880 And on the one hand, we were dealing with bolts that you should make sure that 00:35:50.880 --> 00:35:52.920 they were not containing any magnetic material. 00:35:53.360 --> 00:36:01.320 And on the other hand, of the spectrum, we were also dealing with the accountancy of. 00:36:03.480 --> 00:36:11.300 Material requisitions and how this was done and how this was monitored and whether 00:36:11.300 --> 00:36:14.780 all sorts of buying procedures were in place, etc. 00:36:15.680 --> 00:36:18.420 So over the whole span, we were actually ... 00:36:19.480 --> 00:36:21.660 But I'm a bit confused now. 00:36:23.320 --> 00:36:28.820 Because The monitoring I understand, but the monitoring will very often have 00:36:28.820 --> 00:36:32.640 to refer to certain standards to look at different parts of the process. 00:36:32.800 --> 00:36:34.520 And these standards also must be defined. 00:36:34.880 --> 00:36:39.380 And I would assume that's also part of the governance structure. 00:36:42.128 --> 00:36:46.528 Well, okay, that's the other interesting thing, okay? So the real monitoring 00:36:46.528 --> 00:36:49.048 is done by committees which have very little constraint. 00:36:52.868 --> 00:36:58.888 And who establishes the committees? Who establishes the committees? 00:36:59.968 --> 00:37:04.728 The committees are established by the governance structure. 00:37:05.068 --> 00:37:11.648 Okay. So, of course, there's a subtlety there, because usually they are proposed 00:37:11.648 --> 00:37:16.428 by management, but then have to be established by the governance structure. 00:37:17.408 --> 00:37:22.488 But that sounds very loosely defined then, because that would also mean… That 00:37:22.488 --> 00:37:24.768 part is loosely defined, okay? 00:37:24.928 --> 00:37:30.548 Actually, the strict standards are, of course, things like safety, accountancy. 00:37:32.328 --> 00:37:38.408 They are quite strictly defined, okay? But since we are in the process of building 00:37:38.408 --> 00:37:44.208 things that have never been built before, it is very hard to set standards for this. 00:37:44.728 --> 00:37:51.048 And we go for the common sense of the people in the review committees or in 00:37:51.048 --> 00:37:52.008 the monitoring committees, 00:37:52.850 --> 00:37:59.790 to see what is right and what is wrong. But then you don't try to then extract a rule book or so, 00:38:00.410 --> 00:38:05.310 or a playbook from successful projects to say, well, this was the playbook of 00:38:05.310 --> 00:38:08.830 this project and this could be standards for a future project. Oh, yeah. 00:38:08.970 --> 00:38:13.690 But that's again not in the monitoring side, but that is in the implementation 00:38:13.690 --> 00:38:15.710 side, so on the management level. 00:38:16.030 --> 00:38:19.430 And yes, there is also like a strong idea 00:38:19.430 --> 00:38:23.670 of how the collaboration 00:38:23.670 --> 00:38:28.770 is formalized and there are variations on that but if you look at it and it's 00:38:28.770 --> 00:38:31.890 from a distance okay then you see that the overall structure is the same and 00:38:31.890 --> 00:38:36.950 the variations are minor yeah so one of the interesting things about these large 00:38:36.950 --> 00:38:39.970 collaborations for example is they don't have a boss, 00:38:41.170 --> 00:38:47.790 so the actually the the acting boss is called spokesperson so traditionally 00:38:47.790 --> 00:38:52.810 Additionally, that was the person speaking to the management and the press when there was like, well, 00:38:52.930 --> 00:38:57.410 to the management to say how things were going and to the press to say what they discovered. 00:38:58.950 --> 00:39:05.050 And basically, that is the person now who is in charge of a collaboration. 00:39:07.670 --> 00:39:13.970 But that person has no personnel. So there's no hierarchical relation between 00:39:13.970 --> 00:39:15.990 that person and the rest of the collaboration. 00:39:15.990 --> 00:39:20.550 Usually it's somebody from a larger group so he has like some of his group members 00:39:20.550 --> 00:39:25.630 but typically in a large LHC experiment like 3,500 collaborators, 00:39:26.290 --> 00:39:32.790 then the spokesperson has like at most 20 people that he or she can fire and 00:39:32.790 --> 00:39:36.530 all the others okay there's no leverage, 00:39:37.190 --> 00:39:44.050 how is the spokesperson so you have to do it on persuasion and the spokesperson is elected okay, 00:39:45.390 --> 00:39:48.170 okay celebration so and that and 00:39:48.170 --> 00:39:52.090 that actually that is an important thing because that that gives credit, 00:39:53.210 --> 00:40:00.370 So you're elected and then of course your power in the collaboration comes from 00:40:00.370 --> 00:40:01.990 the fact that you were elected. 00:40:02.310 --> 00:40:06.910 You can always say like, okay, you don't like it then you shouldn't have chosen me. 00:40:07.830 --> 00:40:12.330 People also vying and competing for this role of spokesperson? 00:40:13.790 --> 00:40:17.010 Yeah, of course. This is highly prestigious. And again, so 00:40:17.010 --> 00:40:19.610 there's like the formal thing of of uh there's the 00:40:19.610 --> 00:40:22.410 competition and there are usually several candidates and then there's 00:40:22.410 --> 00:40:25.570 the voting and then there's all the the the 00:40:25.570 --> 00:40:30.550 things that you have in a voting process like people try to gain votes and and 00:40:30.550 --> 00:40:37.350 they make all sorts of promises etc but there's also uh and that is maybe different 00:40:37.350 --> 00:40:42.570 from a normal democracy or at least from a national democracy there there's 00:40:42.570 --> 00:40:44.710 a lot of things is going on behind the scenes. 00:40:45.750 --> 00:40:49.050 Right. So there's a lot of trading. 00:40:49.350 --> 00:40:53.490 Like if I become spokesperson, then this and this person from your institute 00:40:53.490 --> 00:40:56.910 gets this and this position in the collaboration. Right. Okay. 00:40:58.150 --> 00:41:00.950 Okay. One thing, okay. 00:41:01.350 --> 00:41:09.070 We, the Dutch, are very not good at this, okay? Mm-hmm. So that's why Italians always win. 00:41:10.230 --> 00:41:14.930 Why? They're much better at these things, okay? We always think that it should 00:41:14.930 --> 00:41:18.070 be fair and honest, et cetera, okay? And they don't care. 00:41:19.470 --> 00:41:24.650 It should just be effective. Exactly. That's also something you mentioned earlier, right? 00:41:24.710 --> 00:41:28.890 That there were cultural differences and how people approach that process. 00:41:29.150 --> 00:41:31.230 And without necessarily wanting 00:41:31.230 --> 00:41:34.890 to hear you say certain stereotypes about different cultures, still, 00:41:35.050 --> 00:41:40.650 it would be useful to understand how these cultural backgrounds make a difference 00:41:40.650 --> 00:41:45.730 in how collaborations get built up and work or fail, right? 00:41:45.810 --> 00:41:48.750 So what are the basic dimensions there? 00:41:49.826 --> 00:41:53.426 Well, again, okay, so in these large collaborations, it actually helps to have 00:41:53.426 --> 00:41:57.726 a lot of culture because some of them are more risk-taking than others. 00:41:58.046 --> 00:42:03.286 Some of them are more solid than others. And if you somehow can deploy all these 00:42:03.286 --> 00:42:08.706 qualities more or less in the right places, then it's a very powerful combination. 00:42:09.626 --> 00:42:13.366 And so, again, okay, this is in the process. So usually, like, 00:42:13.446 --> 00:42:17.666 the more risk-taking people, they're the ones with the big dreams and the fancy 00:42:17.666 --> 00:42:19.186 ideas, and they try things. 00:42:19.186 --> 00:42:24.706 And sometimes they succeed and often they fail, but sometimes you get very good ideas from that. 00:42:25.026 --> 00:42:28.386 And then in the end, okay, you have to have an apparatus that actually works. 00:42:29.046 --> 00:42:35.046 Then there's also a large part of the community, which are just very solid working people. 00:42:36.746 --> 00:42:40.806 And apart from the one or two highlights where we have a fantastic idea that 00:42:40.806 --> 00:42:45.186 works, most of the apparatus is actually built on steady progress. 00:42:45.186 --> 00:42:48.426 Progress and and so but this combination is is 00:42:48.426 --> 00:42:51.446 quite powerful and it's and and again okay 00:42:51.446 --> 00:42:54.346 you never get it perfectly sort of 00:42:54.346 --> 00:42:57.446 everybody in the right position right but 00:42:57.446 --> 00:43:03.706 there is a sort of a natural selection on that because the the the risk-taking 00:43:03.706 --> 00:43:08.806 people they they tend to focus on the parts where they allow to take a risk 00:43:08.806 --> 00:43:14.306 the people that are delivering solid work okay they They would go to places 00:43:14.306 --> 00:43:16.446 where their solid work is appreciated. 00:43:16.986 --> 00:43:26.966 So there is some sorting out mechanism, okay, with the nature of people sort 00:43:26.966 --> 00:43:29.266 of fits with the thing they're doing. 00:43:29.546 --> 00:43:32.706 Right. But then who are the real players in this context? Who are, 00:43:32.846 --> 00:43:38.546 in the end, managing to grab the most control over that complex process? 00:43:38.766 --> 00:43:42.446 Like you say, these are more the Italians or the Americans, the British. 00:43:42.606 --> 00:43:44.766 Who are the strong players there? 00:43:46.045 --> 00:43:50.285 No, in the end, often, okay, so what you see is like spokespeople, 00:43:50.385 --> 00:43:55.585 et cetera, they're very often like more of the risk-taking types and the outgoing types. 00:43:56.405 --> 00:44:00.405 And then there's also a very important role in all of these large enterprises 00:44:00.405 --> 00:44:05.185 is something we call technical coordinator, which is just what it is, okay? 00:44:05.245 --> 00:44:09.685 So it's like the person who coordinates all the technical stuff. 00:44:09.685 --> 00:44:16.965 And therefore, you need a very solid and stress-resistant person. 00:44:19.085 --> 00:44:25.085 And that also selects itself, okay? Because typically, people that are not robust 00:44:25.085 --> 00:44:32.145 and stress-resistant will actually disappear within finite time. Sure. 00:44:32.945 --> 00:44:34.705 But now, so we looked at the process. 00:44:35.705 --> 00:44:41.805 Can you give an example of what is the biggest success of your time at CERN? 00:44:41.905 --> 00:44:44.925 And what was the biggest failure? And what's the difference between these two? 00:44:47.105 --> 00:44:51.085 Yeah, of my time at CERN. I had, of course, different times at CERN. 00:44:51.225 --> 00:44:53.805 Oh, no, no. So you can choose. 00:44:54.485 --> 00:45:05.065 Okay. Well, by far my biggest failure, I consider that I lost the detector. 00:45:05.365 --> 00:45:09.025 So I was responsible for the detector, which actually melted. 00:45:10.385 --> 00:45:16.325 It was not a small detector. It was actually a fairly substantial part of a large detector. 00:45:16.325 --> 00:45:26.625 And that was because basically not all the security that should have been there was in place. 00:45:26.805 --> 00:45:32.345 And then there was a series of human failure in my absence. 00:45:32.465 --> 00:45:37.365 But okay, I still was responsible for it. So that was... How does it reflect 00:45:37.365 --> 00:45:40.625 on the process of collaboration? That was a failure of collaboration? 00:45:41.645 --> 00:45:45.805 Or it was a failure of you just not being there? No, no, 00:45:45.845 --> 00:45:56.725 it was a failure on my part of sufficiently recognizing human weakness. 00:45:57.765 --> 00:46:06.185 And trying to exclude the risks associated to that with the fact that there was like a sequence 00:46:06.445 --> 00:46:13.665 of of human failures so a highly unlikely sequence but and then you can say okay well, 00:46:14.445 --> 00:46:18.425 normally this wouldn't have happened it's not your fault etc but but any of 00:46:18.425 --> 00:46:23.385 the failures that was that was there could have been prevented if the right 00:46:23.385 --> 00:46:26.765 procedures would have been in place, 00:46:27.585 --> 00:46:34.005 right and and so in that sense okay because after that you immediately see okay 00:46:34.005 --> 00:46:38.325 if we had done this and this okay then this would not have happened and it would 00:46:38.325 --> 00:46:41.265 have been presented in at least three or four different, 00:46:42.065 --> 00:46:43.865 things that would not have failed. 00:46:45.127 --> 00:46:50.247 But of course, what you learn is that you can be more careful in thinking out 00:46:50.247 --> 00:46:51.847 all the possible scenarios. 00:46:52.447 --> 00:46:55.887 And even if you think like, well, who on earth would do this? 00:46:55.967 --> 00:46:59.727 Okay, if it can be done, somebody will do it. Yeah, exactly. 00:47:02.467 --> 00:47:10.067 But in terms of the collaborative projects and... What's interesting is that 00:47:10.067 --> 00:47:14.707 in that case again, okay, Okay, so this was a case in which the experiment was 00:47:14.707 --> 00:47:16.367 actually coming to a grinding halt. 00:47:16.427 --> 00:47:22.107 Not only that, it was actually one experiment out of four on the lap accelerator. 00:47:22.527 --> 00:47:26.107 And so the accelerator had to be stopped. So the other three experiments were also victim. 00:47:28.207 --> 00:47:33.007 So it was a big thing because you also have to realize that the running costs 00:47:33.007 --> 00:47:34.947 of these things are about a million a day. 00:47:35.227 --> 00:47:38.807 And this failure took out like one and a half months. 00:47:38.807 --> 00:47:45.947 It's not like small problem so now you're the 60 million dollar man well, 00:47:46.687 --> 00:47:53.827 this was spelled out to me very explicitly when I had to go to the director to, 00:47:55.227 --> 00:47:57.967 explain how on earth this could have happened. 00:47:59.867 --> 00:48:05.307 You feel very tiny at the time but I had nothing but support, 00:48:06.347 --> 00:48:11.467 and even from the directory it was not like okay of course this is really bad 00:48:11.467 --> 00:48:16.127 but the immediately like there's a switch like okay how do we how do we recuperate 00:48:16.127 --> 00:48:21.027 from this how do we how we get back on track right but what do you see as the 00:48:21.027 --> 00:48:25.107 biggest success then of of these large-scale experiments, 00:48:25.707 --> 00:48:33.367 over the last decades what's the biggest success story the biggest success okay so okay, 00:48:34.967 --> 00:48:38.507 Well, the biggest success of that, no, there's several dimensions. 00:48:38.707 --> 00:48:44.307 So, of course, there have been like marvelous discoveries, expected and unexpected. 00:48:44.767 --> 00:48:50.207 And we have a very different view of our universe, actually, 00:48:50.247 --> 00:48:56.247 from the discoveries of the last, well, I would say three decades ago. 00:48:58.447 --> 00:49:03.227 So this is one asset from this enterprises. 00:49:03.727 --> 00:49:07.607 The other asset is that we have this worldwide collaboration. 00:49:08.227 --> 00:49:12.067 So there's like, I don't know how many nationalities involved, 00:49:12.327 --> 00:49:19.067 180 or so, like practically all nationalities in the world, peacefully cooperating. 00:49:19.967 --> 00:49:25.567 I've seen Israelis and Palestine doing shifts together and actually getting 00:49:25.567 --> 00:49:30.487 along, et cetera. So I consider that also to be a big success. 00:49:32.087 --> 00:49:39.847 And then probably the biggest success for society is that we educated many people 00:49:39.847 --> 00:49:45.227 in performing state-of-the-art, 00:49:45.247 --> 00:49:50.807 beyond state-of-the-art things in large collaborations in an international setting. 00:49:50.807 --> 00:49:58.407 And of all the people that come to these large experiments as a master's student 00:49:58.407 --> 00:50:01.247 even, or a PhD student, or a postdoc, 00:50:02.767 --> 00:50:09.287 in the end, 90% ends up not in research, but in some other part of society. 00:50:10.227 --> 00:50:17.047 And there they play a large role. For example, the whole of ASML would not have been possible. 00:50:20.307 --> 00:50:24.967 If not for the science that we do. 00:50:25.347 --> 00:50:29.507 Not in terms of that we invented what they should do, because we didn't, 00:50:29.507 --> 00:50:33.487 but we delivered the people. Right, exactly. 00:50:35.067 --> 00:50:39.087 Go ahead, Julia. Sorry, Paul. If you were to construct, 00:50:40.607 --> 00:50:48.007 a future major collaboration using Using your knowledge of what works very well, 00:50:48.187 --> 00:50:51.927 using your knowledge of the frameworks that you had in place, 00:50:52.027 --> 00:50:59.607 but mindful of the fact that failures are inevitable in areas of science where 00:50:59.607 --> 00:51:04.107 there is not a basis of already understanding mechanisms, 00:51:04.287 --> 00:51:08.907 as you say, in many of the parts of the project. 00:51:08.907 --> 00:51:15.767 How do you try to compensate for the unknown when you are constructing a collaboration? 00:51:17.563 --> 00:51:20.423 Compensate for the unknown when constructing a collaboration. 00:51:21.703 --> 00:51:28.803 I think the best robustness you can gain is by having a pluriform collaboration, 00:51:29.043 --> 00:51:31.543 to have many talents on board. 00:51:31.843 --> 00:51:38.723 That is actually the main mitigation mechanism, is that if something fails, 00:51:38.883 --> 00:51:43.463 you have so many different angles to look at it that you fix it again. 00:51:43.603 --> 00:51:47.883 And also before it fails, to have so many different angles to look at things 00:51:47.883 --> 00:51:50.203 that failure is avoided. 00:51:50.563 --> 00:51:56.183 And there again, this competition within collaboration plays a large role because 00:51:56.183 --> 00:52:02.943 if you're competing for building the same piece of equipment and you see a flaw in the other ones, okay, 00:52:03.583 --> 00:52:06.903 you would not hesitate to actually say this will never work or your thing will 00:52:06.903 --> 00:52:11.023 burn down on the first instance because, et cetera, okay? 00:52:11.023 --> 00:52:21.023 Okay, so I think that's probably an important thing, is this competition and 00:52:21.023 --> 00:52:24.603 collaboration and the diversity of talent. 00:52:25.383 --> 00:52:29.663 But are you also describing a sort of humility in the face of the challenge, 00:52:30.543 --> 00:52:32.363 that therefore also people give each other space? 00:52:32.823 --> 00:52:36.863 Is that also what you were expressing here? No, I don't think it is. 00:52:37.723 --> 00:52:43.483 But yeah, what I say, people only give other people space when they are more or less forced to do so. 00:52:43.623 --> 00:52:47.343 In terms of things are more relaxed, okay, they don't get a millimeter. 00:52:49.003 --> 00:52:56.163 And I think humility is not one of our virtues. Okay. 00:52:57.323 --> 00:53:02.223 But I was thinking more about humility in the context of the theories that people try to pursue. 00:53:02.223 --> 00:53:08.503 Because there, of course, you enter highly complex conceptual frameworks that 00:53:08.503 --> 00:53:11.063 not always everybody understands all aspects of, 00:53:11.163 --> 00:53:18.383 where you also need to have a sense of a common approach towards the more challenging 00:53:18.383 --> 00:53:21.423 scientific objects that you're dealing with. 00:53:23.423 --> 00:53:26.223 Yeah, but I'm not sure that, okay. 00:53:31.343 --> 00:53:38.203 It's not like we're sort of in awe of the big things okay because that would 00:53:38.203 --> 00:53:43.943 be sort of freezing us so in that sense there's no humility okay we think we 00:53:43.943 --> 00:53:49.303 can find out how nature works we really do it's probably not true but we really do, 00:53:50.443 --> 00:53:55.063 so there's a lot of confidence there I understand yeah because why would you 00:53:55.063 --> 00:53:57.743 otherwise embark on such things. 00:53:57.843 --> 00:54:02.543 If you think you're not going to understand it anyway in the end, you better play golf. 00:54:03.727 --> 00:54:08.627 Well, but you could also see it like the constructs are of a certain complexity, 00:54:09.187 --> 00:54:14.067 that I need to be in a continuous dialogue with others to get my head around it. 00:54:16.247 --> 00:54:21.747 I'm not going to solve this on my own. Oh, yeah. No, no. Most people realize that. 00:54:21.847 --> 00:54:24.847 And of course, there are also some individuals that think they know it all. 00:54:25.127 --> 00:54:32.567 But most people realize that they need all this input from all the others. 00:54:32.567 --> 00:54:35.387 And it's it's it's i think by 00:54:35.387 --> 00:54:38.447 now it is probably cannot find 00:54:38.447 --> 00:54:41.527 anybody anymore who can who is 00:54:41.527 --> 00:54:47.147 actually even willing to claim that he understands uh one of the large experiments 00:54:47.147 --> 00:54:54.147 like completely right and and so in that sense but it's not humility i think 00:54:54.147 --> 00:54:59.507 it's it's because many people would like to, but they feel forced to be realistic. 00:55:00.907 --> 00:55:06.987 But understanding the limits of your knowledge is a form of humility that stands 00:55:06.987 --> 00:55:11.547 in contrast to over confidence, right? That balance. 00:55:12.427 --> 00:55:19.367 No, no, okay. I would rather sort of phrase it in terms of optimism and pessimism. 00:55:19.567 --> 00:55:24.627 And there again, in the collaboration, you find these hopeless optimists who think anything goes. 00:55:25.067 --> 00:55:30.267 And then you also have some group of people that are always like, they'll not work. 00:55:31.367 --> 00:55:37.927 Right, exactly. Okay. And yeah, and they both have a function. 00:55:38.407 --> 00:55:44.807 Sure. Okay. It would be unhealthy for a collaboration not to have this group. 00:55:45.107 --> 00:55:47.047 Yep. Now that I understand. 00:55:47.387 --> 00:55:51.747 So before we go to the final stretch, I want to understand a little bit the 00:55:51.747 --> 00:55:54.327 continuity of large-scale projects in physics. 00:55:54.407 --> 00:55:57.427 So is it fair to say that the Manhattan Project was the first one. 00:55:59.163 --> 00:56:04.043 Oh that i don't know that was really like a huge constructed project but of 00:56:04.043 --> 00:56:05.563 course well wait a minute before 00:56:05.563 --> 00:56:09.523 that okay the manhattan project is basically the answer on a sort of. 00:56:10.763 --> 00:56:23.763 Uh you much larger scale of uh german projects uh both uh for building a an atom bomb and also in 00:56:23.903 --> 00:56:26.163 rocket science, to call it that. 00:56:26.943 --> 00:56:29.423 Yeah, but that scale, that's my point. 00:56:30.323 --> 00:56:33.203 So they were also of a certain scale, okay? 00:56:34.283 --> 00:56:37.663 And actually, then the Manhattan Project, of course, was independent, 00:56:37.923 --> 00:56:41.343 and they really made it happen. 00:56:41.423 --> 00:56:44.703 But it was like really, that was really like power. 00:56:45.043 --> 00:56:46.823 It was an enormous project. 00:56:47.223 --> 00:56:51.963 Yep. But you know, it was not the biggest project during the war by the US. 00:56:51.963 --> 00:56:58.583 It was the second biggest the biggest one was for the B-29 superfortress but 00:56:58.583 --> 00:57:01.123 my question would then be, 00:57:02.303 --> 00:57:07.083 was the dynamics and the organization of these early projects like the Manhattan 00:57:07.083 --> 00:57:13.443 Project already anticipating how they run today at organizations like CERN or 00:57:13.443 --> 00:57:14.703 was there some transition point? 00:57:15.883 --> 00:57:20.043 No, that's an interesting thing because of course the Manhattan Project was 00:57:20.043 --> 00:57:21.683 really run as a military project, 00:57:22.383 --> 00:57:27.583 And since then, that has been tried one more time for one of our projects, 00:57:27.723 --> 00:57:35.663 which was the SSC, the Superconducting Super Collider, which ended up in demise. 00:57:37.080 --> 00:57:42.960 Because it was at some point really converted into a military operation, 00:57:43.160 --> 00:57:44.700 really with military leaders. 00:57:45.420 --> 00:57:49.880 And that doesn't work with us. Okay. Again, okay, it's all about the goal. 00:57:50.020 --> 00:57:58.460 So it probably works if you have like a world war going on, but it doesn't work for anything less. 00:57:58.720 --> 00:58:00.920 Right. Yeah. Right. Exactly. 00:58:02.020 --> 00:58:04.440 So how much is your science worth to you? 00:58:06.240 --> 00:58:13.920 Yeah. Yeah, it's a lot worth, okay, but running that as a military operation didn't work. 00:58:14.340 --> 00:58:20.300 Right, exactly. But now, if you look forward, and also if you look at the current 00:58:20.300 --> 00:58:25.060 situation we're in, we have two big crises in front of us, the COVID pandemic 00:58:25.060 --> 00:58:27.460 and ecological collapse, 00:58:28.120 --> 00:58:29.420 sustainability of the planet. 00:58:29.640 --> 00:58:35.860 Do you believe that our society could learn something from the rulebook of these 00:58:35.860 --> 00:58:37.540 large-scale physics experiments? 00:58:38.200 --> 00:58:43.500 Okay. Well, first of all, I'm already sort of, I heard a lot of people doing 00:58:43.500 --> 00:58:44.680 it, so I'm not surprised anymore. 00:58:44.860 --> 00:58:48.380 But I think there's quite a big difference between COVID. 00:58:48.600 --> 00:58:52.880 Of course, COVID is just hitting us now very hard, okay? But COVID is just a 00:58:52.880 --> 00:58:55.940 ripple compared to our ecological problem. 00:58:57.040 --> 00:59:01.520 Okay? COVID will go over. Our ecological problem, probably not. 00:59:02.440 --> 00:59:12.980 So it's of a quite different scale and so what maybe I think people in my field 00:59:12.980 --> 00:59:16.380 and I think many people in my field are of the same opinion. 00:59:18.086 --> 00:59:25.126 If not most. Because I think in my field, people can actually look like 20, 00:59:25.266 --> 00:59:29.366 30, 40 years ahead and look at it in that perspective. 00:59:30.446 --> 00:59:34.606 And also when you talk about COVID, it's not COVID-19, which is the problem. 00:59:34.706 --> 00:59:40.326 Maybe it'll be COVID-22, COVID-25, COVID-27, COVID-28. 00:59:40.726 --> 00:59:45.426 Or a completely different virus, right? No, that's why I label them all by year. 00:59:46.686 --> 00:59:54.426 Right. This is how it's done. But we're not done yet with that type of vulnerability. 00:59:55.626 --> 01:00:00.706 And so that requires something different than what we're doing now in fighting 01:00:00.706 --> 01:00:03.786 this one virus or even this one type of virus. 01:00:04.266 --> 01:00:08.566 So we have to think about how we organize this. 01:00:09.266 --> 01:00:12.446 If it's everyone's question, can we learn from you guys? 01:00:12.946 --> 01:00:17.666 Can we learn from you guys to then respond? Well, exactly this, okay? 01:00:17.886 --> 01:00:24.126 That if you, of course, it's important that next week we cannot go to a cafe anymore. 01:00:24.586 --> 01:00:26.546 Well, is it? Is it really? 01:00:28.086 --> 01:00:31.046 Or is it that over the years, okay, 01:00:31.146 --> 01:00:37.926 we will have a very limited life for 20 or 30 or 40 years, that our children 01:00:37.926 --> 01:00:43.746 will have a completely different life because they will have to isolate one 01:00:43.746 --> 01:00:45.626 from the other, more or less continuously. 01:00:48.676 --> 01:00:52.856 So think a little bit further, okay? 01:00:52.996 --> 01:00:58.716 So think about what you're doing in terms of a much longer timescale. 01:00:58.936 --> 01:01:03.736 And of course, I mean, if there's an imminent disaster, okay, you should solve it. 01:01:05.796 --> 01:01:13.956 But at the same time, you should live in a long-term perspective. 01:01:14.456 --> 01:01:16.676 And that's, of course, in general missing. 01:01:17.456 --> 01:01:20.896 Right. But my question was a more specific one. 01:01:20.996 --> 01:01:25.836 Can we learn from the models that the physicists have developed for their large-scale 01:01:25.836 --> 01:01:31.436 experiments to really also advance collaboration in society, 01:01:31.676 --> 01:01:33.456 responding to large-scale experiments? 01:01:33.456 --> 01:01:40.136 I don't think that in COVID it was because of our organization that they organized things that way. 01:01:40.136 --> 01:01:52.676 But the fact that in a year's time you have a vaccine is due to this focused effort, okay? 01:01:52.816 --> 01:01:56.256 So this is how we would have done it. But it's not because they looked at us 01:01:56.256 --> 01:01:58.396 and they said, oh, we do it like those guys. 01:01:58.676 --> 01:02:04.136 But because there is this huge common goal which just has to be met. 01:02:05.236 --> 01:02:08.356 And then how do we bring that to the ecological challenge? 01:02:08.356 --> 01:02:18.176 Challenge yeah so there my feeling is that apparently people are only compelled 01:02:18.176 --> 01:02:20.116 to actually go for the goal, 01:02:20.896 --> 01:02:26.716 when they really feel it when they recognize it so so. 01:02:27.956 --> 01:02:34.096 As long as as you just drive your car okay you hardly notice from day to day 01:02:34.096 --> 01:02:37.216 that that things are deteriorating I mean. 01:02:40.718 --> 01:02:46.198 Yeah. So it's about times. Again, okay, it's about timescale. 01:02:46.198 --> 01:02:48.218 It's about thinking about the long-term perspective. 01:02:48.518 --> 01:02:54.098 And I think, in general, humans are not very well adapted to that. 01:02:55.258 --> 01:03:01.358 No, we're not. Which is also completely logical in terms of looking at the evolution, okay? 01:03:01.798 --> 01:03:07.538 Because if something is threatening you today, you should immediately respond to survive. 01:03:07.538 --> 01:03:10.558 Survive and if it's threatening you in 20 years you 01:03:10.558 --> 01:03:13.518 can still produce some offspring and you and you survive that 01:03:13.518 --> 01:03:19.858 way but now do you believe humans will be ever able to collaborate effectively 01:03:19.858 --> 01:03:24.558 to answer these kinds of challenges will are we able intrinsically yes in particle 01:03:24.558 --> 01:03:29.838 physics we are in in astronomy we are okay there's more fields like this so 01:03:29.838 --> 01:03:32.878 there are fields in which it has been demonstrated. 01:03:34.978 --> 01:03:41.718 So, yes, it can be done. But I think what may be important there, 01:03:41.818 --> 01:03:44.958 of course, is the subset of people that you... 01:03:48.278 --> 01:03:55.218 It's also selection bias, right? Because it works in physics because it works in physics. 01:03:55.998 --> 01:04:01.298 But maybe in other domains of human endeavor, However, the conditions are not 01:04:01.298 --> 01:04:06.478 conducive to actually instill these kinds of collaboration because of whatever, 01:04:06.678 --> 01:04:09.738 competitive forces, resource limitations, what have you. 01:04:10.398 --> 01:04:12.598 Well, it's more like a short-term win. 01:04:13.878 --> 01:04:18.958 Sure. That is the most destructive force, I think, for these things. Yeah. 01:04:19.358 --> 01:04:23.398 But now, if you could change one thing in humans in order to make them successful 01:04:23.398 --> 01:04:27.898 collectively, also as non-physicists, you can change one thing. What would you change? 01:04:33.098 --> 01:04:36.638 Probably would help if you could kill the instant satisfaction gene, 01:04:38.958 --> 01:04:44.618 alright Cybron de Jong, thank you very much for this conversation ok hi, 01:04:44.778 --> 01:04:47.758 you listened to one of our podcasts in the series on collaboration, 01:04:48.758 --> 01:04:53.558 produced by the Ernst Trommel Forum and the Conversion Science Network you can 01:04:53.558 --> 01:04:55.838 find more episodes on our website.