INSIDE EOSC 08 – Bob Jones
Andrew Dubber
Hi, and welcome to Inside EOSC, a podcast all about the inner workings of the European Open Science Cloud. I’m Andrew Dubber. I’m a senior researcher at the Industry Commons Foundation and a consortium member of an EOSC project called LUMEN. And every month I introduce luminaries, people who are central to the LUMEN project and to the wider EOSC ecosystem.
And Bob Jones first arrived at CERN back in 1984 as a technical student, writing COBOL programs for the engineers digging the tunnel that would later become the Large Hadron Collider. He was still there a few years later, working in the room where the World Wide Web was invented. He led the European Data Grid, the LHC Computing Grid, CERN Open Lab, and Helix Nebula. He sat in the private meeting with Commissioner Moedas, where the European Open Science Cloud was first imagined, and then he served two full terms as a director of the EOSC Association.
At the time of recording, he was in what he called pre-retirement at CERN, and he had a new title. Bob Jones, it’s very nice to have you with us. Your title is Special Envoy to the EOSC Federation. What does a Special Envoy do, especially?
Bob Jones
Well, I’m still trying to find out what a Special Envoy does. But it’s essentially I’m taking a sort of leadership role in the build-up phase of the EOSC Federation. So I’ve been working with the association for a number of years now. And we’ve gone from the planning and the policy aspects to the actual concrete implementation of the Federation. And I’ve been very involved in that and sort of very much attached to it. I like the whole idea of having something concrete there.
And so when my time as a director of the EOSC Association finished, because I’d had two mandates, when that came to an end, they asked me to stay on as a special envoy.
Andrew Dubber
Right. Okay. And so your role is to find out what that means.
Bob Jones
Yeah. Yeah. Yeah.
Andrew Dubber
But you’re a long serving and senior member of CERN since, am I right in saying 1986?
Bob Jones
Yeah.
Andrew Dubber
That’s a long time.
Bob Jones
Yeah, that wasn’t the first time I was at CERN. I was there in 84 as a technical student. At university, I had to do a placement, and I started by doing a one-year placement at CERN, working in the sites and buildings. Enrica this morning showed the tunnel, and when I first went to CERN, I was writing COBOL programs for the guys that would dig in the tunnel.
Andrew Dubber
Tunnels are a very modest word for what it actually is. The Large Hadron Collider is quite a feat of engineering.
Bob Jones
Yeah. At the time, the tunnel was for the LEP, the Large Electron Positron. So that was the accelerator that was in the same tunnel. And then that was dismantled after excellent production, excellent service. It was dismantled in order to make way for the LHC.
Andrew Dubber
Right. And just so I get this correct, your job was to write the software for that while you were a student.
Bob Jones
Yeah. Yeah. So software was being used by the civil engineers. So yeah, that was fun times.
Andrew Dubber
Wow, amazing. And you’ve seen a lot happen over the period since then, obviously, the development of the World Wide Web coming from, did you sort of rub shoulders with that?
Bob Jones
Sure, it was in the same group with Tim Berners-Lee. I worked doing some work for Tim as well at the time. Everybody talks about TCP/IP, and we have porting the protocol to TCP/IP, because that was against the ISO seven-layer model, and that was the thing that took off. But at the time, that wasn’t what was actually being used at CERN. We were using DECnet. We were using the networking capabilities that came with the VAX VMS cluster of machines.
And if we wanted to use it, somebody had to port it to DECnet. So I had to work with Tim to port the World Wide Web’s protocols to DECnet.
Andrew Dubber
Wow. And so you would have worked with Robert Cailliau as well.
Bob Jones
Yeah, Robert Cailliau. I know him as well. I’ve been in contact with him as well.
Andrew Dubber
All right. Okay, fantastic. Yeah, it’s a really interesting group of people. But the idea of that as a model for how open science works, I think it’s really interesting because it’s fair to say that that’s had some sort of impact on the modern world in which we live.
Bob Jones
Oh, yeah.
Andrew Dubber
So do we imagine that CERN is now engaged in a similar kind of endeavor to create, for instance, with the introduction of the quantum initiative?
Bob Jones
Yeah, so CERN’s involved in trying to put together a organization, a sort of virtual organization at CERN, to help ensure that the development of quantum technologies, not just quantum computing, quantum communications, other aspects as well, those are for benefit of the wider society and community as well.
Andrew Dubber
Right. Possibly a difficult question to answer because there isn’t something to compare it to necessarily, but can you try and talk about what it was like to be a software developer at CERN in those days?
Bob Jones
Yeah, so I think it was also a time when there was, let’s say, the work was less structured, and there was more freedom in what could be done. There were less processes involved. And it was very much, I say, a time of optimism as well. Whereas if you look at the general outlook at the world at the moment, everybody’s afraid that things are going to go pear-shaped and the future doesn’t look bright for us because of the climate, because of the geopolitical situation.
That wasn’t the situation then in the 80s at CERN. Everything was a bright future, right? And you could basically do whatever you wanted as long as you could fit it into the overall scheme of what was being asked by the sort of the hierarchy. And we were given quite a lot of freedom to do these things. I think that freedom is what allowed Tim to actually develop the World Wide Web.
Andrew Dubber
Right. People still talked about progress in those days. Is that still on the agenda anywhere?
Bob Jones
Progress in what sense?
Andrew Dubber
Well, technology leading humanity into a new era of prosperity, those sorts of things.
Bob Jones
Yeah, yeah, I think that was the case. I think you could argue that that’s still the case today. Some people see artificial intelligence as also being a source of progress that could help us in some of these directions. Accelerating science, as Robert, our keynote yesterday, highlighted. So, yeah, I think there’s certainly, I think there’s a lot of optimism. I think that was a sort of a brighter outlook, which perhaps the generation today, they don’t benefit from that optimism so much.
Andrew Dubber
Do you retain some of that optimism?
Bob Jones
Yeah, in my day-to-day work and also thinking of when it comes to what we can do with things like the European Open Science Cloud. I’m an eternal optimist there.
Andrew Dubber
Right. So we’re going to talk about the European Open Science Cloud because you were there at the beginning. You were there in the opening conversations of we should have European Open Science Cloud. Was the name grid ever considered? Because that was kind of more the dominant frame at the time, I understand.
Bob Jones
Yeah, so structures such as the LHC computing grid were already in place. They’d been set up by CERN and the physics community, and it was in there, and it had been taken from the sort of computer science labs and operationalized and put on a global scale.
So that was understood. But I think it was also recognized that we weren’t just talking about a grid here. There were other elements involved. And perhaps a name that was more fitting to that approach was required.
Andrew Dubber
Right. But this is pre the era of cloud computing. So did you kind of just stumble upon that name?
Bob Jones
Well, I don’t know. No, I think cloud computing was already in place. The elements were there. There were also some of the leading cloud providers were already in place. So I don’t think it was visionary in that sense that the cloud is already becoming a dominant model for having access to IT resources and services. So I think the idea was more that this would be a way of accessing any of these resources without strictly specifying the exact technical model under which it would be delivered.
Andrew Dubber
Right. I’m aware that we’ve leapt from the mid-80s to the mid-2010s without very much in between. A lot happened in those times, and you’ve led on initiatives, the computing grid, OpenLab, Helix Nebula, these kinds of things. Do you want to tell me a little bit about what that journey was?
Bob Jones
Yeah, so I started at CERN doing programming and so on. And then I moved into what became the IT department. It wasn’t called that at the time. It was the data division. So it showed the importance of data. They recognized that that was the important thing there. And I started working on some of the software that was being used for the LEP experiments then. So the data acquisition systems for those things, working on that.
I then moved from there to another department working in preparation for the experiments for the upcoming LHC accelerator. And I eventually joined ATLAS Experiment. And there I…
Andrew Dubber
ATLAS, sorry, ATLAS is one of the big…
Bob Jones
One of the four LHC experiments. It’s the one closest to the CERN main site I may run. And I was working there. I worked there basically until the year 2000, overseeing parts of the control structure for the data acquisition, and running a group of software developers there, architecting that work.
And then I moved to the IT department to start on something called DataGrid, which was a European Commission-funded project, the first European Commission-funded project, led by Fabrizio Gagliardi. And that was to put in place a number of data centers which would help the processing of the LHC. It hadn’t become the LHC computing grid yet. That came along later. But that was the sort of start of much of the work that was needed.
Andrew Dubber
What was the function of the software? Was it envisioned as being part of the LHC or was it just essentially operational?
Bob Jones
So that was also a part of the philosophy that was around at the time was if we look back to the previous accelerator, the LEP, there each of the experiments were very much independent and there wasn’t so much shared between them. Some elements were, but not so much.
Whereas when the LHC computing grid model came into place, the whole idea very much was to have a shared infrastructure, so the same data centers and resources being used by multiple experiments rather than being separated out. And so the software we talk about here, I think you could summarize it pretty much by saying it’s middleware, right? Okay, and that it’s a common set of middleware that is used by all of the experiments and the virtual organizations that are working on LHC computing grid. And then each of the experiments themselves have their own libraries and structures that sit on top of that and exploit the grid. But much of the middleware is in common across all of the experiments.
Andrew Dubber
Where does the idea of openness come from within CERN, within EOSC? Where is the ethos of that drawn from?
Bob Jones
Well, there’s the open science movement itself where CERN has done it, but often not in name. It’s been sort of inherited inside the CERN culture because it’s always been recognized that by its nature, many of the instruments that CERN builds cannot be built on an individual institute or national basis because the scale of what’s involved. So once you’re doing a joint undertaking for these sort of structures, then also naturally you have to share the software, the data and so on that goes with it. And I think it came as a natural progression there as well.
Andrew Dubber
Is it philosophical or practical?
Bob Jones
Like many things at CERN, I think it starts out as being practical. And then there might be a philosophy that develops afterwards. I think CERN’s a lot better at getting a job done in the most practical way it can. And then they’ll worry about the policy afterwards.
Andrew Dubber
Right. There does seem to be a very kind of, I want to say fanciful. What I mean is experimental aspect. I’ve heard CERN described as very smart people flinging very tiny invisible objects at each other very fast just to find out what happens. And there is something that isn’t practical about that.
Bob Jones
So, well, if you took it from a physics point of view, remember the theories, first of all. And so everything that happens around the accelerator is experimental physics. And it’s put there to try and prove or disprove those theories. So, saying that it’s, and then we’ll see what happens. Well, no, there is an initial goal. And there is, when these large scale investments being made, there is a load of checks and balances and a whole process to verify that it makes sense scientifically, that it makes sense financially, that it makes sense from an engineering point of view. So all these steps take place in putting together an experiment or an accelerator.
Andrew Dubber
Right. I’m not myself a theoretical physicist, and so there’s a lot of terminology that I don’t know. But I find that really interesting that there are people in positions of responsibility at CERN who are not just not theoretical physicists. They are not theoretical physicists. That is not part of what they do. And I think that it’s interesting that CERN identifies somebody who is not kind of a local originally and who has ended up there becoming absolutely central to what CERN does, but whose specialism isn’t the mission. Is that also philosophical or practical?
Bob Jones
I think that’s more practical. Again, the CERN recognized that it needs certain skills and attributes. And that from its origin, it’s been an international organization. So looking far afield for these people is part of what’s been necessary. It tries to attract people from all over the world, in particular from the member states.
Andrew Dubber
How were you found?
Bob Jones
Well, as I said, it’s because I started as… It was very strange. I had to do an internship during my first, my bachelor’s degree. Yeah. And at the time, we’re talking before much of the IT you see today as being available, you had to write your name on a paper list on a board.
Yeah. And the lecturer who was dealing with these placements had started taking the pieces of paper off. And then I asked him about CERN. He told me a little bit about it. And then I said, well, that sounds interesting. So he wrote my name on the list. And then eventually got a letter later saying, yes, you’ve been accepted.
So I came for one year as a technical student. Loved it. Loved the environment. Loved the exciting work I was allowed to do. Very much hands-on work. And then afterwards, when I finished my degree, I applied again. And I was accepted to come back to CERN as a staff member.
Andrew Dubber
Were you aware at the time of the significance of the project that you were working on?
Bob Jones
Depends. Each of the projects, some of them have been more visible than others. The one everybody, of course, hears about is the World Wide Web. I think the first time I remember seeing the first time Tim Berners-Lee wrote WWW on the blackboard.
Andrew Dubber
Right. Wasn’t even a whiteboard.
Bob Jones
No, no, it was a blackboard in a room that still exists, but it’s been modernized, the IT auditorium in the IT building. And I remember the first time him writing that on the board and there was a discussion about it. But I think it really hit me that it’d gone beyond research and labs when I saw it written on the side of a bus. Then I knew that it wasn’t just something contained to the research community.
Andrew Dubber
How far are we from seeing EOSC on the side of a bus?
Bob Jones
Ah, well, there’s already a growing number. If you look at the number of users who come along with the EOSC Federation, that’s attracting a lot more interest with it. So that will grow.
I think we have to look, first of all, at different types of usages. So it’s very much at the moment focused on publicly funded research. The other steps that we have to look at is going beyond that, thinking, well, there’s not only research in publicly funded activities, but there’s commercially funded research. And if we think of some of the European policies of better links with innovation and industry and so on, expanding the European Open Science Cloud so it can perhaps work via the data spaces to work with some of these other sectors, that will help it grow. Then there’s this whole other questions as well about citizen sciences and the undertakings there as well.
That will grow. But it’s still the research domain. I think what would be perhaps the areas where it would seem to affect the general population is when we see some of the things on the other fields sort of with a social impact. So things like the climate, medical aspects and so on, that’s when it’ll happen. But it’s destined to stay unnoticed because it’s essentially a sort of a data infrastructure. So I’m not sure it’s going to rise to become visible in everybody’s home like the World Wide Web.
Andrew Dubber
Right. I’m glad you brought that up, though, because I think it’s an interesting thing. I was co-author with Michela Magas of a 2020 Commission study called Expanding EOSC, which was very much led by your initiative, I understand, but it was about looking beyond academic research to citizen science, other sorts of institutions, SMEs, et cetera. How important is that outward-looking application of publicly funded scientific research into other domains in the EOSC kind of?
Bob Jones
Well, I think it’s essential. Science has to, even fundamental science, it has to be able to justify its activities by the impact it has on society. And working with companies to see that some of the developments that are in the research labs become innovations that can affect the population and other sectors are essential. That’s to show that the actual investments were made by funding agencies, that there is some payback for that and some reward for it.
Andrew Dubber
Right. So one of the recommendations of that report was that we pair FAIR data technologies with JUST data practices, JUST being an acronym for judicious, unbiased, safe, and transparent. Now, since 2020, AI has come a very long way and has basically swallowed up the whole internet. How much has the boat kind of sailed on this idea of JUST data practices and the impact that that can have?
Bob Jones
Well, I think not just JUST, but also if we think about the whole area of trustworthiness, I think that still remains a very important aspect. And we’ve seen great examples of it here at the symposium over the last couple of days where one of the important things to the European Open Science Cloud is to say that it’s making highly valuable, high-quality curated datasets more generally available for larger reuse in other disciplines and other sciences. And that, of course, means that we’re giving good input to AI engines and tools. So if you put good data in, the chances of you getting more trustworthy results and outputs from the AI tools increases.
Andrew Dubber
Yeah. Okay. So you were the lead author and the rapporteur, I understand, for the EOSC Sustainability Working Group’s foundational straw man report. Do you want to tell us what that was and what the issues that it dealt with were?
Bob Jones
So this was a report from a sort of working group that existed before the EOSC Association was formed. And it was to try and understand what are the key elements that we’d need to put in place so that EOSC would go beyond just being another project. And a lot of it rotated around the idea of coming up to see what is the most important points.
And the key element that came out of it, I think, on the initial findings was that it’s all in the data. The data is the most important element, and hence we have the web of FAIR data and associated services. So recognizing that you have to have high-quality data that is more accessible by the FAIR principles and can be reused, having this high-quality data available, and then having the tools and the services that can analyze and process that data, made a compelling combination which could have value in different sectors.
And then also about the architecture of how we can put that in place. What are the essential elements that we find which have been carried over into what we see as the basis of the architecture now for the EOSC Federation as well?
Andrew Dubber
Is sustainability an enormous challenge for something like this?
Bob Jones
I think it is because, first of all, we see today there are questions about do we still manage to keep access to important data sets? Are some of them being taken offline for whatever reasons? How do we ensure that those remain available and those are trusted? So the sustainability of the data, first of all, is very important. And then that brings the whole question of, well, if you’re going to make it sustainable, you’ve got to have the appropriate standards and interoperability elements that work with it.
We can go for the lists, but those are big field and putting in there. And then also ensuring that if we want to get the research community to take the European Open Science Cloud seriously, it has to be more than a flash in the pan. That there has to be something, if we want them to have it integrated as a cornerstone of their daily work, it has to be something that they can count on five years, ten years from now. We talked a lot about CERN. Planning an accelerator takes decades, right? You’re not going to build that on the reliance of a structure, something like EOSC, if it’s only given me around for one or two years.
Andrew Dubber
Right, right. You kind of gestured towards not just financial sustainability, but potentially political sustainability for open science data. Is it a good time for open science?
Bob Jones
Yeah. I think there are people questioning about is this the right moment or should we be closing the doors and protecting these things? I’m not sure. I think if we look at it sort of strategically for Europe as well, it’s an opportunity for Europe to highlight that it will take a middle road in terms of collaborating and cooperating people that have the shared values that Europe embodies. And sharing and collaborating on the important things for the whole planet, the climate, the sustainability of those aspects are essential. So I think we can bring those together.
There’s the question of competitiveness as well. But that competitiveness for me is an area where you can still agree on what you share, even if you’re going to develop competing solutions afterwards.
Andrew Dubber
Is this how you try and navigate the tension between openness and sovereignty?
Bob Jones
Yeah, so I think we can interpret sovereignty by not necessarily saying closing the doors to others, but interpret sovereignty of saying that we will ensure that what we have developed or produced remains available and is not taken offline or closed away. So that element of sovereignty, ensuring that the dependencies we have in what we produce inside Europe, we can ensure that they remain available over the longer term.
Andrew Dubber
Right. I allow myself one stupid question per interview. I edit all the other ones out. I’m going to ask you my one for this interview. What’s a data space?
Bob Jones
Okay. So it’s put into context is the idea that in different domains, there’s a need to collaborate in order to share data to achieve some objectives. It could be a supply chain. Think of Airbus. It has a data space with its suppliers of all its components for its aircraft, where they have to exchange data in order to come to the common product at the end. So there’s a data space there that they all have to share material in that sense.
Now, you can do that not only in an industrial context, but you can do also in a research context. It can be in the European health data space, sharing medical data in that sort of context. So it’s this environment in which you can collaborate to share data along with well-defined policies and rules for doing it. That doesn’t mean everybody can see everything all the while. There may be, in particular in situations such as medical domain where there are sensitive data, there may be very well justified reasons why only certain groups or individuals can have access to certain data. And that’s respected within the context of a data space.
Andrew Dubber
Right. One of the intended outputs of the LUMEN project, which this podcast comes from, is the creation of an EOSC innovation center. What do you see as the relationship between EOSC and innovation?
Bob Jones
Well, innovation can take over the innovation lifecycle, all the different steps involved. I can imagine that EOSC being involved in the earlier phases of innovation, where as well, we again, we come back to the whole concept of the data space and the ability to share data between different parties, according to agreed rules, that will help accelerate the innovation process.
Andrew Dubber
Right. Which is the reason I asked about the data space question, because one of the ideas that you’ve had is the idea of a data space for social sciences and humanities, which I think is a really, it’s a fascinating opportunity for something like the LUMEN Project, which comes out of social sciences and humanities. Can you talk about what you imagine an SSH data space to look like or function like?
Bob Jones
Well, first of all, we recognize that there are lots of different data sets in the social science and humanities. And some of these as well have restrictions. And because of potentially the contents of some of those data sets, who has produced them, being the audio recordings and so on of certain individuals and so on, there may be limitations on who can have access to it and what can be done with that material.
There’s also the whole idea of the annotations that are often very present in that domain as well. And being able to manage all that together and allow people to cooperate to augment those annotations and that material. So for me, having those sorts of that corpus of data and data sets available and recognized and findable makes a good basis for social science and humanities data space.
Andrew Dubber
Fantastic. You’ve had kind of a successful history I guess in fostering public and private partnerships. I’m thinking of the Helix Nebula as an example of that. What do you think is the key to that sort of successful relationship between those different worlds?
Bob Jones
Yeah, I think recognizing the objectives and the priorities for the different stakeholders involved. The researchers are often interested in getting a publication out of their work because it advances their career and the progression they will have themselves personally and for their field.
Whereas the sort of private sector involvement, they’re looking to develop some product or some service that they can potentially then commercialise. And that doesn’t mean that they’re necessarily enemies. There is ways in which they can work together and share elements of that. And finding the common ground, I think, is an important element of that.
Andrew Dubber
How do we know when it’s been done well?
Bob Jones
So I would say when the people come back and they’re ready to work together again. If there’s actually been some mutual benefit coming out of it and they’re ready to repeat the experience, and I think that’s a good outcome, even if there isn’t an actual product that’s come out of the result. But if there’s some mutual benefit, I think that’s a good element of it.
Andrew Dubber
Supplementary question. How will you know when you’ve done well? How will you know that special envoy has been a thing that has been accomplished successfully? Do we know yet?
Bob Jones
I think if we get to the end of the build-up phase and we have an EOSC federation, which is larger than what we have in the first wave at the moment, and it is trusted and used by a large body of researchers, not only in the public sector, but in other sectors as well, then I think we’ve done our job of having the European Open Science Cloud up and running.
Andrew Dubber
It strikes me, and it seems like a nice way to kind of round this conversation out, your career has been a singular one. It’s not just that any computer programmer could have walked into CERN and had the career that you had, not just done the job that you did, but had the career that you had. And obviously you have some managerial skills and you have some sort of insight and all the rest of it. But there seems to be something quite singular about that. And it feels like there is something about your character or something about who you are, which is an additional element to that, that you don’t necessarily see from the outside. If you were to identify that, what would it be?
Bob Jones
So I think a couple of things have helped me along the way is curiosity and empathy and a mixture of those two things. In the sense that, having respect for my colleagues, the work they do, and not automatically assuming because they have a different opinion from me that they’re wrong, but working with them, trying to understand their point of view and see how that can be incorporated into what we’re trying to achieve.
Andrew Dubber
I’m not trying to impose a causative narrative on this, but knowing that you come from Birmingham kind of makes sense of this because it’s such a diverse context. And you do rub shoulders with so many different kinds of people with so many different worldviews. Do you think you’ve carried some of that with you or is that just speculation?
Bob Jones
Yeah, I think so. I think to some extent you’re right. It was very much a mixed environment when I was growing up. And I think I transported some of that to the work at CERN. One of the reasons I liked at CERN is because, again, it was very diverse, very international, lots of different cultures, lots of different nationalities and languages. And, yeah, I enjoyed that environment very much.
Andrew Dubber
Right. It seems to be a successful scientist at CERN, you need to be good at people.
Bob Jones
I think to be a successful scientist in general, you have to be good with people. I think we’ve gone from the time of the turbulent dictator scientist to where now we have to have a more collaborative model based on the problems trying to be addressed.
Andrew Dubber
And we are trying to be a collaborative space, and particularly with EOSC, in a very top-down kind of time, it seems. Does it add to the struggle or just make it more important and underlined?
Bob Jones
No, I think, as you say, top-down is certainly one way of looking at it. I think one of the strengths that EOSC has as well is that there’s a bottom-up push from the community that recognize that there is an opportunity to do something better. And sort of melting that in with the top-down model for a drive, having some positive result out of all these investments is a good way of bringing the two elements together.
Andrew Dubber
Before we started this conversation, you posited the notion of pre-retirement. What’s post-retirement going to look like? What do you want to sort of look back on? Is there a legacy that you imagine?
Bob Jones
Yeah. So for me, I think I see the European Open Science Cloud as an important objective. And I think getting through this buildup phase and seeing it continue under its own strength and continue to grow, I think that would be a great legacy.
Andrew Dubber
Bob Jones, thanks so much for your time. Really appreciate it.
Bob Jones
Thanks.
Andrew Dubber
That’s Bob Jones, Special Envoy for the EOSC Federation, recorded at the EOSC Symposium in Brussels last October. Now, since we spoke, Bob has also formally been appointed Chair of the new EOSC Node Coordinators Committee, which I guess is just the next phase of the kind of pre-retirement someone like Bob will likely enjoy for some years to come.
And that’s Inside EOSC, a podcast from the LUMEN Project, which receives funding from the European Union’s Horizon Europe Research and Innovation Programme. I’m Andrew Dubber, and I’m back next month with another EOSC insider. You can subscribe to Inside EOSC wherever you listen to podcasts, leave a review, give it five stars, or pass it on to a colleague who you think might find it interesting. Thanks so much for listening, and I’ll talk to you soon. Cheers.