GLPI turns 20 #1: the community project by INDEPNET

Introductory remarks

Projets libres ! would like to wish a happy 20th birthday to GLPI, our IT asset management and helpdesk software.

We take a look back at the history of the project, with its successive maintainers (including the author of this podcast).

GLPI is an example of software that has lived several lives, reinventing itself as it goes along, both in terms of governance and technology.

In this first episode, we talk about the birth of the project and its community life, then the transition to publisher mode at Teclib.

GLPI Part 1: Community Life

Walid : Hello and welcome to Projet Libre. Thank you all for being here. Today, it’s going to be a special episode, it’s going to be a two-part episode. We’re going to talk about a project that I worked on for 12 years of my life and on which I learned pretty much everything I know about free software.

So this project is called GLPI and it is 20 years old this year. And so for these 20 years, it was a project that was close to my heart, I wanted to find a way to be able to look back on it a little bit and talk a little bit about what GLPI is, a little bit about its specificities and also show a little bit what it is like to develop software over the long term. And to do this I am not alone, I am surrounded by two of the historical and volunteer maintainers of the project, Julien Dombre and Jean-Mathieu Doléans and also by Alexandre Delaunay who was for many years my colleague at Teclib. So thank you all for being here.

Presentation of the guests

Walid : To begin with, I’m going to ask the three of you to introduce yourselves. Julien, it’s up to you.

Julien : I’m Julien Dombre, I’m 45 years old, I have two children. I’m a computer scientist and we do a little bit of all jobs in IT. And I have a career mainly, not mainly, exclusively in the civil service, in different structures and specifically in the field of education.

Walid : Okay. Jean-Mathieu?

Jean-Mathieu : So Jean-Mathieu Doléans, I’m going to keep my age quiet because I’m the oldest in the team so I’m not going to shoot myself in the foot by exposing my decades of activity… as I come from the co-founders of the GLPI project and then I was the president for many years of the Indepnet association which carried out this project. Personally, I have been working in the digital sector and professionally for a very long time, perhaps too long. And I have had a certain passion and above all a strong conviction for free software and knowledge sharing for many years. That’s it, and in the professional context, I have worked in a lot of jobs, a little in the private sector but also a lot in the public service, whether state or in local authorities.

Walid : Great, Alexandre?

Alexandre : I’m Alexandre Delaunay, I’m 40 years old. I am the current maintainer of the GLPI project and have been working at Teclib for some time. Previously I had a career as a web developer in different companies, but I discovered free software at Teclib and GLPI software at the same time. I am the one who normally takes care of this software.

Walid : I’m going to introduce myself too. So I too worked on GLPI from 2006 until 2018. We’ll come to that later. I was one of the developers and one of the maintainers of GLPI.

The birth of GLPI

Walid : Now that the introductions are done, the idea is to present GLPI very succinctly so that the listeners know a little bit about what we’re talking about. Alexandre, can you introduce the GLPI project in a few sentences?

Alexandre : Yes, then GLPI is free software. Its name is an acronym that stands for “Free Computer Asset Manager”. He has several hats which are on the one hand the inventory of computer equipment and on the other hand the management of the Helpdesk… and finally a ticketing system that allows you to provide assistance to users within it. And in addition to that, there is a whole set of additional features such as project management and various tools that will be grafted onto one or the other of the two main parts to lead to what is called a tool, an ITSM in fact, which allows you to manage an IT department as a whole.

Walid : Julien and Jean-Mathieu, can you explain to us a little bit about the birth of GLPI?

Jean-Mathieu : We’re going to do a bit of archaeology. In fact, it’s the story of an encounter above all, a little fortuitous by the way, more than a technical story. I was the president of the Indepnet association, which we had founded at the time with Basile Lebeau and other friends.

It was an association that worked in the promotion and development of free software, but especially around associative actors who were a little far from these concepts. And particularly on, at the time, something that we really liked, which were content management systems, that was pretty new at the time. And for the associative actors, it was a technical way to be able to have a presence on the web, to express themselves easily without prior technical knowledge.

At the time, we were doing a lot of promotions, training, installations and deployments of a fairly famous content management system called SPIP, which is still called SPIP, which still exists. And in fact, Basile, during his activities, worked for a structure that needed an IT asset management tool. And so he had done a bit of a tour of the free software that existed, he had found nothing very conclusive, except a piece of code around a project called Irma, which was a project that was no longer maintained.

Its original creator no longer wished to follow up on its production. And when Basile was working on this code and the features, he asked me to take a dip in the interfaces, etc. And in fact, quite simply, we got a little passionate about this little initial project and we especially felt and sensed that there was something to do around it. Because by taking a little look at what existed at the time, and in particular proprietary software, we realized that it was park management software that was sold at a high price at the time, with extremely high and even prohibitive costs for associative actors or public buyers, quite simply. So many did not practice park management in a computerized way and were satisfied with spreadsheets and other tools which were a bit prehistoric tools.

And so we said to ourselves that there was a need that this project could meet. So we repackaged it, corrected it, interfaced it, documented it, etc. And so in 2003 the first stable version of GLPI was released, version 0.2. That’s how the project was born.

Walid : Actually, basically, this project before which was called Irma, it has been renamed to GLPI. So you were the one who chose the name GLPI?

Jean-Mathieu : Actually as we had completely repackaged it, modified it, translated, redone the interface, refactored the code, etc. We did a rework, a complete redesign, it’s mainly based on the features.

We decided to give it a new name, a new identity and to promote it to users who might need it.

Walid : And so in 2003 when you came out… so already you put it under the GPL license, I’d like to know a little bit why, and then when you release it, where do you make the code available? What is there in 2003 to put the code as a software forge?

Jean-Mathieu : There is nothing at all! Well, yes, it’s not true. Initially, there was nothing, at least, that we know of. Because, once again, it’s a project that was born from a chance encounter, it’s a human project before being a technical project. And so we also learned by building. Initially, to distribute the software, we set up a website, on a SPIP by the way.

We packaged, we made an archive, we made a documentation, and at the time, there weren’t a plethora of forges, in fact we didn’t get to work on it until much later. For a year or two, we worked without the tools that everyone knows today and with which no one was present, such as deposit platforms like Github.

Walid : And why the GPL then? What made you choose this license?

Jean-Mathieu : So first of all because the initial code was under the GPL license, so anyway there was a legal obligation, but even if it hadn’t been the case or if we had re-developed completely, anyway everything we designed, created within the association has always been under a free license, whether it’s the docs or the code. There was a desire to share, to disseminate, to distribute and to allow people to take ownership of things. In fact, when we initially released GLPI, we didn’t plan to maintain it afterwards. We said to ourselves, here we have done something nice, it can be useful to others and then others will certainly take it.

Walid : Okay and you Julien, when and how did you get on the GLPI project?

Julien : I arrived in 2004 in view of the history of commits on the project and in a very interested way at the beginning very clearly. At the time, I was an engineer in a research laboratory, a computer equipment manager, and I was looking for a tool. And from my research, in the end, the only free tool available was GLPI. So in the end, I tested it, it suited me and it met a large part of my needs. I had problems with this tool, especially on the authentication part, it didn’t integrate properly with the lab’s information system so I started to offer patches. And unfortunately or fortunately in the sense you see, I started to put a foot in it and then I jumped in. So then I have two memoirs I met in 2004 also at Solution Linux, Basile and Jean-Mathieu and there I was lost for the cause I think.

Walid : Was the legal form that supported the GLPI project there before the project?

Jean-Mathieu : Absolutely, it predates GLPI. The Indepnet association existed before.

Walid : Okay, okay. Does it continue to exist?

Jean-Mathieu : No, it was dissolved a few years ago now.

The beginnings of the project

Walid : At the very beginning, Julien, you, it was interested. Jean-Mathieu, when it started, did you have a professional interest? Did you use it professionally at the beginning when you started contributing or is it just like you say because you fell in love with the project a little bit and you continued like that?

Jean-Mathieu : There was no link with the jobs I was doing at the time and no professional interest in the GLPI project. There is with my professional activities. It was more of an ideological, intellectual, human challenge that interested me.

Walid : It’s 2003, you put the source code on the internet, are there people who show interest? How does it actually work? Did you have a mailing list?

Jean-Mathieu : We had made the code available in the form of an archive, the doc, etc. And then we set up a mailing list and then very quickly we were contacted by users who downloaded, who had questions, who had suggestions for improvement, proposals, etc. What we had initially envisaged, namely to make it available and then for others to take advantage of it, in fact, is not really what happened, it was rather that very quickly it had a small success, all in all modest but still present. And it circulated quite quickly afterwards by word of mouth more than anything else since we didn’t have the marketing means to go and promote the project but there was a little snowball effect and so in fact we got our finger caught in the gears and then in front of the enthusiasm we answered the questions, We made corrections, we started to release additional versions when it’s not really what we had envisaged.

Like Julien, we were a bit caught up in the success, the challenge and then the nice side of the project and then the birth of a community that had high expectations. And then it was quite encouraging, motivating and rewarding to collectively build something and its work for the general interest.

Walid : So at the beginning the community is mostly French-speaking?

Jean-Mathieu : It was exclusively Franco. It’s only a few years later that we had contact with anglophones or bilinguals who told us but my company would be interested, etc. It would be of interest to my structure, it would be of interest to my entity. And so some have started to offer translation beginnings. That’s how GLPI actually started to internationalize.

It’s not a project that was initially designed for the international market, but once again, at the beginning is technical with a developer who is already aware of everything that exists as a communication support, community development, etc. Well, I repeat that at the time there was almost nothing. And who says right away, well here it is, I’m going to build my code in such and such a way, my project in such and such a way, I’m immediately aiming for the international market, I’m already going to develop immediately in English so that… That’s it, it wasn’t the entrance, that’s it. So afterwards we made the necessary efforts to respond to the requests, but internationalization came first on the translation aspects since we had a lot of contributions quickly for the translations of language dictionaries.

The OCS Inventory NG and GLPI combination

Walid : I arrived on the project in 2006 and when I arrived, we were talking about JLPI but we were mainly talking about OCS and GLPI. And in fact what I would have liked you to explain to us, maybe Julien, is your meeting with the OCS team, how it came about and how you started working together at the time. Explain a little bit about what OCS was.

Julien : The meetings with OCS, initially, I admit that I don’t really remember how it was done and how the first hooks were made, I don’t know if you remember Jean-Mathieu?

Jean-Mathieu : It was the Gendarmerie.

Julien : It was the Gendarmerie, yes….

Jean-Mathieu : Initially, I think you weren’t in the loop, Julien, on the very first meeting. I don’t remember, but I know that we were invited by the OCS team, who worked at the gendarmerie, in the Paris region, to talk to us about their project that they presented to us. They wanted to build bridges between their OCS and GLPI solution.

Walid : So let’s just remind you that OCS is in fact another tool, another French free software whose goal is to create an automated inventory system. No fleet management but automated inventory, at least at the time, that was it.

Jean-Mathieu : Yes, absolutely, and we had met the very nice team of OCS developers, there were three of them at the time, and we immediately got on well and wanted to collaborate to allow our community of users to have an automated inventory tool: it seemed extremely interesting to us as a complement. For the OCS team at the time, it also meant that they had to focus on the inventory engine and not on the whole issue of interface management etc. It allowed them to focus on the technical parts and there was already a lot to do on the issue of inventory and inventory agents.

Julien : I remember we met them when they already had an interface, in fact, they had a graphical interface to control their agents and they were wondering about going further in their interface. That’s right, yes. And the link with GLPI meant that they didn’t get into it because GLPI met that need.

Walid : For years, in fact, what made GLPI so successful was OCS plus GLPI in the end. It was these two software programs that were a bit of a couple, I don’t want to say inseparable, but which finally meant that there was a free open source solution that was in competition with proprietary tools. That’s right.

Jean-Mathieu : yes and no, I’m going to qualify a little bit, GLPI worked very well without OCS, there was a large community of users. But let’s say that the arrival of the OCS – GLPI duo has been a booster, especially for the big structures. Because the question of agents in fact arose especially for people who had structures, who had large parks. All the small structures up to I’m going to say anything but a hundred PCs to manage, it was not a problem for them to do CSV imports, good deliveries and others to have their inventory or to manage it even by hand. What was problematic was that we were reaching large fleets and it effectively served as a booster for the adoption of fleets since there we had a complete package and which therefore competed much more easily with the proprietary solutions found on the market.

The first external contributors

Walid : I would like us to talk about the first external contributors, that is to say not Jean-Mathieu, me when I arrived, well Julien and Jean-Mathieu you were both already there, you were already well established and you had the thing well in hand. Do you remember when the other contributors arrived, who are the people who stayed years ago to work on the project?

Jean-Mathieu : who wouldn’t remember it? (laughs) So after Julien, who was a main contributor and who became a member very quickly and because of his great competence and the sympathy that linked us, he joined the main team of developers very very quickly. After that, in fact, it was meetings.

There was the collaborative work with the CNAM , which had a large fleet management project and was therefore very interested in GLPI but had needs not covered by the solution and wanted to see how it could be part of a positive and constructive approach to collaboration and contribution around the project. And so that’s when we met our servant Walid tonight who, at the time, was working for an IT services company at the CNAM as well as his colleague Gonéri (Le Bouder). The fact that we worked together on the issues of development and integration of contributions into an existing community project with lines that were already relatively structured at the time, fairly strong guidelines that were rarely departed from, but which allowed us to stay the course and be consistent with our users and with our philosophy. It was a good match, let’s say also with Walid, who had a rather specific break-in phase because when you come from an IT services company, it’s not necessarily very easy to build bridges with a community development structure, especially when you have a client behind it, who also has deadline requirements, of production, of kinds of things, so you have to understand how you fit into that. Wally, you understood it quite quickly and easily, but in addition you had to work on a question of interface or mediation let’s say between the free software community and then the needs of a customer who was, all in all, quite big anyway.

Walid: I’ll come back to that later, because I’d like to give a little bit of my arrival on the project, because I think it’s quite typical of the opening of a community that welcomes new contributors. I arrived before, for example, Xavier Caillaud (Editor’s note: his pseudonym is TSMR).

Jean-Mathieu : Xavier, he’s going to be mad at me, he arrived before you indeed. He was mainly a contributor to bug reports – he was not a developer at the time, mostly a user – so bug reports, proofreading docs. So he did contribute quite a bit but on less technical objects, that’s why I focused him on the contributions aspects in dev. But yes, Xavier did a lot of beta testing, etc. He also became a Helper very quickly on community forums, very quickly. And then later developer, and contributing developer on plugins (plugins) for GLPI.

Then there are three fairly important people who arrived afterwards. So, Remi Collet and Nelly Lasson arrived through the Health Insurance because they worked there and started to contribute themselves, which is quite interesting that a client finally starts contributing on his own.

Jean-Mathieu : Absolutely. Julien, do you want to add something about the arrival of these different contributors?

Julien : these contributors who helped establish the notoriety of GLPI through the growing community. And above all, to give the project the ability to move forward faster. In particular, the collaboration with the CNAM really made it possible to have a product that gained in technical support for large groups. The Health Insurance at the time, as well as the referencing they had, was more than 100,000 positions managed. It was huge compared to what we were considering even at the beginning of the project. We were happy when we had a user who used GLPI to manage these 200 workstations. They arrived, that’s it, more than 100,000 jobs, that’s it. With the mechanical needs and optimization that were necessary to manage this volume. And so mechanically it has made GLPI progress enormously on all technical aspects.

Walid’s arrival on the project

Walid : I can explain a little, from my point of view, one day my boss came to see me and my colleague, Gonéri Le Bouder, and he told us that we were working at Atos, a big company, we were in the open source center. One day our boss came to see us and said: “If you’re interested, there’s a mission, we have a framework contract with the Health Insurance, you can go and do fleet management and automated inventory”. Well, we didn’t know anything about it, we didn’t even know what it was.

We get there, we look at the projects, Gonéri knew how to do Perl so he goes to OCS, I was doing PHP and LDAP, I said to myself ok GLPI, and there the first contributions are already to go on IRC, start to introduce yourself a little bit, make 2-3 small contributions, send a few patches, arrive really very discreetly and humbly, because in fact we don’t know anyone. Seen from my window it went well at the beginning, at one point we had to go further and then we organized a physical meeting precisely to be able to meet because we were going to have to work together a lot in the end. How did you get that, Julien, for example? How did you get that when we have the science of disease arrived with its big hooves to come and work on GLPI?

Julien : The arrival of disease science in the project for me was the easiest arrival because it was done as you humbly said. Compared to you who arrived humbly. We’ve had other arrivals of contributors who have arrived with their big hooves, but from memory the arrival of the CNAM by you and then by Remi and Nelly, it has always gone very well.

We were always able to exchange calmly, we could have divergent points of view but we were always, it always went very well in the discussions.

Jean-Mathieu : Walid was referring to a famous meeting with a few feathered chiefs at the time of the CNAM. That’s it, I think it’s more like that. Indeed with the contributors we worked extremely well, you are right Julien to point it out. There was a need to clarify a little bit the questions of expectations, objectives and posture and strategy. And I remember a rather peculiar meeting where we met feathered chiefs of the CNAM(TS) who had approached us a little from above at the time and then insisting quite heavily on, I remember the question very well, is “are you happy that we are using your project?”. And I remember shocking a little at the time by saying “no I don’t care if you use the project or not, I don’t care, what interests me is whether you’re going to be a contributor or not”. That’s the reason we were at this meeting at the time, it was to say we came to see you because you told us that you wanted to use the GLPI, so much the better, but above all you wanted to contribute. So that’s what interested us and that we were in a win-win logic, that is to say that we eventually agreed to work on or adapt the roadmap (Editor’s note: roadmap) by responding to needs, perhaps more quickly, to immediate needs for the CNAM, even if it was included in the GLPI roadmap. But it could be modified, adapted, facilitated, and then perhaps integrated contributions more easily and quickly in order to be able to respond to the schedules specific to the structure’s project. We were ready to do it, but if we were in a win-win exchange and therefore there was feedback in the form of contributions to the project. And it’s true that naturally, in fact, we were approached as if we were a proprietary software publisher and that we were indeed happy to break out the champagne because we had won a contract with a large public entity, except that this was not the case. There was no contract, no hard cash. These were the contributions that the CNAM would be able to make via its contributors, via its service provider.

Once all that was said, it was the contributors who talked to each other, the developers to each other, and we were able to move forward in the most intelligent, constructive and productive way possible for the good of the project. This allowed GLPI to grow very quickly on the management issues of large structures, the performance issues that Julien was talking about, who spent nights and nights solving these problems of application performance on very large fleets, and also a more intensive professionalization of GLPI. In particular thanks to the contributions of Remi and Nelly on ITIL issues as well, who were a force of proposal and very strong contributors on ITIL integration issues. It is a contribution that has been extremely fruitful for several years.

Walid : I’m just going to remind you for listeners and publishers who don’t know that ITIL is a set of best practices for organizing an IT department. On GLPI, it had a big impact on the Helpdesk part, and therefore the Ticketing part.

The organization of the project to welcome new users

What I also wanted to know to finish on this point was until the arrival of the Health Insurance, you didn’t have a full-time contributor, developer. From the moment we arrived, there began to be full-time contributions. Then, later, there was David Durieux who also arrived, who started to make contributions. How you organized yourself precisely to facilitate the fact of being able to contribute with people who are paid, which was not your case to work on GLPI.

Julien : How did we organize ourselves? We didn’t organize ourselves I think, we spent our evenings, our weekends and our holidays there.

Jean-Mathieu : We spent a lot of time, it’s true, but first of all, Julien was extremely efficient, it must be said. He may still be, I don’t know, but at the time he was extremely efficient with a pretty impressive level of productivity. We were part of a best effort logic, received contributions, made it a point of honor, to respond as quickly as possible, to integrate as quickly as possible and as well as possible because it’s a question of respecting the contribution and that’s also why we got it a lot, because we didn’t let people wait ten months for us to study either their patch or their proposal or to respond to them. I’ll qualify it a little, but we still organized ourselves in terms of structuring, processes. We wrote a lot about forges at the time, the forges that we set up ourselves, so we equipped a lot, we had automated roadmaps, associated ticket management.

Today it makes everyone laugh because it’s available to anyone but at the time we had to set up servers, install forges, we changed I don’t know how many times because the project was born and lived and died. We managed CVS with SVN, Git, but we also structured and equipped a lot in formalized processes.

So when a contributor arrived, we easily directed him to the community tool which managed the aspects of forging and where everything was explained. How we made a contrib(ution), what process did it go through. How we released a new version. How we put elements in a roadmap. How we validated specs, how they had to be written to be understood by everyone, etc. In fact we worked a lot on the structuring of the project, not the code, but it’s a lot of time and a lot of reflection and a lot of neurons and it allowed a much faster integration of new contributors.

Walid : There were two things, there was synchronous communication on IRC globally, asynchronous communication, there was a mailing list indeed, there were all these specs etc. From time to time, there were meetings in the evening because you had the meetings in the evening, which was when you were available, so we also had meetings like that, etc. It was actually a mix of the two where, while you were at work, often you had a head on IRC, you answered and all that, and it still didn’t surprise me but it always amazed me to see that you were able to be there on IRC and answer during the day, to look at what happens while you had a job.

Jean-Mathieu : Yes, we spent a lot of time on it, whether weekends, holidays and others, lunch breaks, I remember eating a sandwich and then I would answer on the forum (laughs). Because we also had a community forum with a lot of users to respond to. These are things that we did at one time because we were passionate, because we believed in it, because there were a lot of requests, because we also had very motivating and very positive feedback from people. There you go, we were in something very emulating.

Now, one thing is certain, and history shows it, and that is that it cannot be done indefinitely.

Walid : Alexandre, when did you get to the GLPI project? In what year?

Alexandre : So exactly in October 2010, with my arrival at Teclib, and being hired as a developer. I discovered GLPI at that very moment. I already knew a little about free software, I was a lot of user in my previous jobs. Very quickly I was put on production projects, developments, integration projects, let’s say GLPI, with Teclib’s customers. And on which it required a lot of development.

And so in 2010 I think that my, well I met you two, Julien and Jean-Mathieu, in January 2011, since we did the FOSDEM together at that time. And I had produced at that time, in a few months, my first plugin, which was a mobile interface. It was around those years, 2010-2011, that I started my first skirmishes on the software.

The implementation of partnership contracts between the association and the companies providing services around GLPI

Walid : So there’s a subject I’d really like to talk about because I find, at least in the French ecosystem at the time, it’s very atypical. I think it was around 2009, something like that. At some point, you started to work, to talk to us, to work on a partnership system between the Indepnet association and the various integrators, since there were several. Xavier worked for an integrator, David Durieux was himself an integrator, we were integrators, there were others who arrived, etc. And I think that system was quite unique, in the end. I don’t know where you got it from, but in any case, I don’t have the impression that at that time, other projects tried to follow this path in the same way. Can you explain to us a little bit, either Julien or Jean-Mathieu, a little bit how it corresponded?

Jean-Mathieu : My memory of the date was around 2008 when it was formed, but in fact the reflections were earlier. We often had exchanges with Julien about the fact that we were regularly solicited by users for requests for services and development, which we did not do, since the association did not intend to offer services on this software.

We were always in a community development mindset and in any case, we had the job. So, at the beginning, we simply directed people to companies that had collaborated or contributed to the project. And from my sick brain came the idea that we could perhaps formalize this, so that we could clearly display to users a network of qualified partners, but above all beyond qualification, of partners who agreed to support the project. Because I thought it was important when a client, a company or a public buyer had to choose a service provider that he knew which service provider was going to play the game or not of free software and contribution. Because we had seen and observed several times that there were people without conscience or shame who had looted or dressed more or less skilfully GLPI, she had sold to a client who found himself a few years later unable to make a single evolution or update because everything had been tampered with, developed not necessarily very cleanly and then without long-term perspective.

They hadn’t made plugins for example, a specific development packaged in plugins, it didn’t prevent them from updating GPI. There we had structures that had contacted us, some of the times with tears in their eyes saying but we can no longer update, we can no longer do anything, well no, they were stuck. So we wanted to make things much more readable and healthy for our users and then at the same time create a network of partners who would support the association to ensure the evolution of the software.

Since we hoped in the medium term to be able, why not, to hire full-time developers to work on the maintenance of GLPI.

Walid : Do you remember the terms a little bit or not? Overall I don’t know anymore.

Jean-Mathieu : Terms?

Walid : I don’t remember exactly what you’re committing to. There were still some things that were a bit restrictive for you who were a volunteer.

Jean-Mathieu : Yes, we were mainly focused on a best effort logic on the fact that we were committed to processing contributions quickly, to also answering our partners’ questions, to informing them of the evolution of the roadmap.

It wasn’t so much contractual commitments because we couldn’t… It was rather moral commitments to make life easier for the partners who were committed to supporting the project, for whom we were perfectly aware that we still had to respond to customer constraints. So we understood the constraints, we showed an understanding and consideration of these constraints in our operation as well. This reassured everyone a little bit about how we could collaborate. After that, the network of partners was formed with people who had already been contributing to it for some time, so it was above all a formalization of the commitment of these partners to support the project.

Walid : What I wanted to know Julien is when it was set up, you started to have the first partners, what were the first effects in fact? What did these financial contributions and these contributions in general do to the project?

Julien : On the project in terms of contribution, it hasn’t changed anything. As Jean-Mathieu said, it is rather the validation of contributions and partnerships that were already existing. But above all, it has made it possible for at least five years to be able to finance, even if we have never managed to finance a full-time developer, to finance my work for a day. So for five years I started part-time for GLPI and only for GLPI.

Walid : Even if I guess, but I’d like to have your feedback on this, even if I guess it didn’t necessarily achieve the objectives you would have potentially imagined, the effects were still concrete, you know, real.

Jean-Mathieu : Of course.

Walid : I don’t know, did you have any objectives and I was going to say a little with hindsight on the side of volunteer community management, what do you actually get out of these partnership problems with hindsight?

Jean-Mathieu : In fact, I realized that the term you are talking about partnership problems, for me there was no problem. We initiated the network of partners in 2008, we built up a network that has been enriched gradually. This has freed up funds to be able to finance development time on the side. After that, if we had given him more time, perhaps the network of partners would have continued to grow to allow him to reach a sufficient level to have global funding for a full-time position or several, we don’t know. So for me there was no problem, it was really a win-win system for everyone. After I don’t know if you’re referring to the fact that…

Walid : I didn’t say problem, I said program. It wasn’t the same.

Jean-Mathieu : Oh program, I was the one who said that. So there you have it. I told you my old age means that… So now I’m also becoming deaf.

Julien : I’m much younger than you but I also understood the problem.

Walid : Really? Well it’s okay. Oh well, maybe I said problem then, sorry. What interested me was really to know the feedback. What do you actually get out of it from what you have put in place with hindsight? I don’t know Julien?

Julien : For me, it was only positive, fundamentally, but then it could have taken on another dimension, but we hadn’t done it either, we didn’t have the means or we made the effort to go and canvass people to be part of the partnership. That wasn’t our goal either. The partners came to us. We didn’t go looking for them.

We didn’t have the strength to do it, and it wasn’t the goal at all either.

Walid : And you’ve never thought about trying to set up a company? Is it something that has crossed your mind at one time or not?

Jean-Mathieu : Yes.

Julien : I remember that we talked about it several times. We were both in our professions, and at the time in public structures. And I’m fundamentally attached to the civil service in fact, so giving up my job was complicated – for me at least – it really wasn’t… And I think that the moments when we really asked ourselves the questions were mostly … It was a bit pivotal moment in our careers and where it was not at all possible in fact.

Jean-Mathieu : And it’s true that we discussed it several times because when we saw the success of GLPI we could question ourselves. Being an entrepreneur and a builder of a community project is not the same thing, it’s not the same job. We managed a community but we didn’t have HR issues (Editor’s note: human resources) that an employer knows when dealing with employees. We didn’t have any questions about business model, break-even point management, investment, financial risk. We devoted our time to the construction and life of this community project without having all the besides entrepreneurship and it’s not the same job. Besides, you can be an entrepreneur, including in software, and then know nothing about it and not touch a line of code.

What interested us was to be jack-of-all-trades, on all subjects, we were on the communication of the project, on the technical life of the project, on community life, on exchange. We met a lot of people. And we weren’t talking about turnover, profits, capital and others. Anyway, it’s not the same, it’s not the same job either and we found that there were some who did it very well so we let them…

So we asked ourselves the question at one point, it’s true, we said to ourselves, well, now that we also had partners who did the job very well. We wouldn’t have too much of what we would have brought either. And like Julien, I am very attached to the civil service and the general interest.

Julien : I think one of the things that also scared us is that mechanically we would have had customers in front of us and we were afraid that the customers we should have served would divert us and take us out of the path we had traced for GLPI and the path on which we did not want to branch off GLPI.

It could have led to things that we would not have wanted and that we would have been forced to do when, in the community framework, we would not have done them.

The partnership seen from the side of a service company

Walid : And in fact, what I would like to say about this is that if I place myself on the side of a service company, for us, it was really interesting because we were necessarily partners and we were contributors. We were part of the development team, even though we were never a member of the association and it wasn’t something we particularly wanted. We were there on GLPI, it suited us very well.

I think we were all there for different reasons. I’m a bit like you, I didn’t know this environment, but when I started in fact I fell a little in love with it and then I continued because I wanted to be able to see my customers go from a proprietary solution to a free solution and be able to convince people that “stop paying for your big rotten proprietary software and we do the same thing with a software free software and come and help us contribute to better software”.

I think that between the time I arrived until the moment you handed over to us, I think you really passed on to us the spirit of what you wanted to do at the community level: in the reflections we were able to have, the way of thinking, of making the specs, etc. I think that – this is my transition – I think that when we took over the management of the project, we just tried, of course in our own way, but to continue what you have done. Of course not the same because we’re not the two of you, but you’ve really passed on the spirit in which you’ve worked all these years on the software. I don’t know if Alexandre you want to say something about the IT services part, well about the partner company part.

Alexandre : No, because at that time in particular, I was still relatively new both on the project and even in my career and so I’m not sure I have any particular memories, whether from an economic point of view or that kind of thing. I really enjoyed the beginning of the adventure with the whole group, was exciting and me as a new developer it was something than my previous jobs… I wanted to work in a community of free streams or that kind of thing and it was really an extremely exciting moment whether it was with the people outside of Teclib within the internet community but also the people of Teclib who carried me, whether it was you, Walid or Gonéri that we mentioned earlier, It was really quite exceptional at the time. But otherwise, as I said, I was quite new at the time and my opinion is less relevant I think on this part.

The assessment of the project and the beginnings of the change of maintainers

Walid : So now I would like us to spend this third part, the last part of the interview, on this transition period. At what point do you realize that in fact basically things should certainly be changed? How did the reflections that would then come about in 2015 lead to the fact that you finally handed over to the company Teclib?

Julien : I have a bit of a goldfish memory in this whole story, I don’t remember the meeting with the CNAM without being told about it again. I wouldn’t know how to make the connection… we arrived at a point in GLPI where we had built GLPI for 10 years trying to refresh the history but having a legacy (Editor’s note: history, technical debt) on the technical aspects and so we asked ourselves the question at some point of a complete rewriting which could be necessary because there you go, we were dragging a certain number of things, we were on PHP from the 2000s. Then came what was the MVC model on which we were not at all there in GLPI at the time. So we asked ourselves this question, we exchanged a lot, in particular with all the contributors, with David Durieux, with you Walid and with other people, on what was relevant, how we could do it. But I’m not sure that this desire or idea to completely rewrite has an impact on the decision to hand over the reins. Jean-Mathieu will give his opinion later, but I think that we had above all arrived, at least for me it was the case, at a stage in my personal life that made me aspire to something else. And certainly a form of model fatigue.

Even if still at the time, of the same order, I was still at 80% and I still had a day available for GLPI, so that was very very pleasant for me because I really had a day every week to free myself from work and really give 100% for GLPI so it was very very comfortable. But I think that personally I had reached a weariness and especially a turning point in my personal life that made it time to hand over the reins. A turning point in my personal life, typically that’s it, we handed over, six months later, my first son was born. So there you have it, it was…

Jean-Mathieu : We hadn’t spoken with Julien about that time. I listened with great emotion to what he said, but I think that there was a conjunction of elements that led to these decisions to pass the torch to a partner of the GPI project. Indeed, Julien had said it well, but it was a bit underlying in my previous explanations, when I said that we work a little in turbo mode, we can’t work indefinitely, but it’s true. There was… They are not in order, but there was a state of fatigue. We had been carrying the project against all odds for ten years, day, night, weekend, vacation.

Because, we didn’t say it, but to keep the community alive, we also organized delocalized seminars, we invited everyone. I also remember Coding Weeks with Julien, we spent a good number of holidays until 6 a.m. doing developments etc. But hey, it’s a time, we really had a lot of fun and enriched. Like everything, it comes to an end. At the same time, there is an intellectual, psychological, physical fatigue, and then indeed an aspiration also to other things from a personal point of view, we can agree.

Afterwards, there was also, in relation to the project itself, not to mention the Dev Team, the project itself, we felt that we had done a little bit of the trick and that in order to move forward, we had come to the conclusion that we had to rewrite GLPI to meet the aspirations or the guidelines we wanted to form for the ten years following the next life for GLPI in fact.

And then when we looked at each other with Julien and we said to each other “oh there is all this to do, we said to ourselves we won’t have the energy to do it”. So we said to ourselves “are we still the right parents, in quotes, to take care of GLPI? Are we still the right people to take care of GLPI, take care of it and take it further?” Naturally, we came to the conclusion that it was not. And so we had to consider passing the torch in a more formal way.

Afterwards, we would have liked that naturally, contributors would have gone from Padawan to Grand Master naturally but it didn’t happen because in fact, probably by force of circumstances, people didn’t necessarily want to question our legitimacy or I don’t know, or it was also maybe for comfort reasons too, to say to ourselves here we are still dealing with the great sages who are the guardians of the temple and then here is it also gave more freedom as a contributor. I don’t necessarily have a clear and unequivocal explanation on the subject, but the state of affairs meant that we didn’t naturally have people who took over the project in a natural way.

So we said to ourselves, well, we have to make a big big bang, and then we have to contact our partners to offer them to take over the project, but by explaining what we wanted for GLPI and by checking with everyone that they were able to take it over according to the guidelines and founding of the project. Otherwise we could have simply closed the doors and left the code available, and then of course there would have been someone who would have done it, but we also wanted to safeguard the interest of the community and ensure things over time.

Walid : I realize that I forgot to mention someone who plays a fairly important role in the rest of the story, that it’s an oversight on my part, it’s Pascal Aubry who is the founder of Teclib and who has always believed in GLPI for a very long time. Who was a bit of the architect of the change in the governance of the project, that’s it, and no one with whom you had any kind of memory, very good relations if I remember correctly.

Jean-Mathieu : Pascal was one of the first to work with us on the network of partners, he had understood both the logic and the mutual interest that there could be. Pascal has always been part of a win-win logic so we got along very well from the start. When we proposed the takeover of the project, there were other candidates, we were given the choice of the partner who offered the maximum guarantees for the continuity and sustainability of the project, the commitment to respect the philosophy and the license of the project and then also the ability of the partner to put in the means. Because if we were passing the torch, it was not so that the project could survive, but it was to give it the means that we were not able to give to the project. That’s how the transition happened.

Julien : I think it’s really important, we gave up, because there was a weariness, there were all that, lots of factors, but the main factor was that we could see that we couldn’t really give GLPI the scope we expected for the project in the model we had put in place.

Walid : It’s quite clear and we, having experienced it from the inside, could feel the weariness. And I am, I say it every time, but I am extremely admiring of the time you have spent as a volunteer. I don’t know if I could have spent so much time volunteering.

The change of governance in 2015

Walid : So in 2015, we made a joint press release to announce the change in governance of the project, the fact that Teclib, that is to say basically Alexandre and I, we were going to take over. What happens to you from that moment on? After that, I’ll explain on our side, it wasn’t easy. But what I would like is for you to explain how you experienced it in fact. What happened once we published this statement?

Jean-Mathieu : We still had the balls. It was not easy to leave your baby after 13 years of breastfeeding, day and night. It was painful. It was a bit complicated. Afterwards, we committed to ensuring the smoothest possible transition for the project, for the teams that took over the project, because we had perfect control of all the governance tools, but that’s not necessarily the case for those who took over. We made sure the transition was as smooth as possible for the community too, we tried to leave something as clean as possible. We had balls, it was a bit painful the detachment, but at the same time a relief too, for my part.

Julien : for me the transition couldn’t be smooth because I went from GLPI diapers to my child’s diapers, so there you go, it wasn’t… I just changed the baby. And humanly, a lot has happened.

Walid : How did you experience people’s reactions after the announcement?

Jean-Mathieu : More or less well. We had a reaction from everything, Julien. We had very positive reactions saying, “Yes, we understand, you did well, we respect, thank you for everything you did.” Let’s say that this was the majority of the reactions.

And then we had a bit of weird reactions from some, some people who literally yelled at us… We were called bastards, because we abandoned the project and we potentially abandoned them. Reactions that were a little peculiar, a little weird, I think it was also sometimes a way of expressing a certain sadness, but in a slightly aggressive or clumsy way.

We appreciated and took the positive reactions and more or less well received the slightly negative reactions. I’ve always been outspoken, so when I felt that people were going a little overboard, there were a few brutal reframes. Finally, we felt that we owed nothing to anyone but ourselves initially. After all, everything we did was out of goodwill and respect for others, but basically we didn’t owe anything. Except that some people thought we owed something (laughs).

Julien : Yes, the reactions were diverse but we tried to explain as much as possible why we had done it and above all the choice was really guided towards the sustainability of the software and the best solution from our point of view for the users in the long term. That was really the goal. And that was the watchword behind GLPI and all the work we were able to do was always for the users.

It was the slogan that we conveyed all the time. We always do something for the users and everything we do, we do it for the community. And this handover was also made in this sense. Even if it may have been painful for us, it was necessary, I think for us personally, but it was also necessary for the community.

The resumption of the project and the transition to editor mode

Walid : I think that with Alexandre, we will be able to explain a little bit how we experienced this. I think we were talking about it just before, we weren’t ready for that at all. When it happened to us, we were super happy, but we were like, oh there, crushed by the task, so many things to do.

Alexandre : And it was done honestly quite quickly it seems to me, in a few months we learned the news, well I learned the news and yes a bit, a quarter later the thing was done so yes we had to get into battle order very very quickly from that point of view.

Walid : Among the things that surprised me the most in the end, even if I understand afterwards, is with Alexandre and my other colleagues, we spent a lot of time on the phone or physically going to see our customers, our good customers, to explain to them what is… to explain to them, so the people we’ve been working with for years, to explain to them what we have… What was happening and what were we going to do.

And in fact we had a lot of reactions in a community way or… Well, or people who weren’t necessarily close to us, or even others who were close to us, who basically accused us of going to close the project. Something that at the time I didn’t understand at all, since we had been working for years and we were all very happy to make GPL code and we were all very happy to make a community, so I had a hard time understanding actually.

I can understand that partners who were at the same level as us, at the community level at the time, and who tomorrow find themselves at the point that, well, in fact there is a competitor who becomes the publisher of the software… I can understand that but in fact we had to spend a lot of time but a lot of time having to explain to people that no we were not going to close the project, no we were going to continue like this but just give it more means and improve it as we went along etc. It was actually quite a long job.

And the second thing is, well, that’s basically the project, now you’re the one who manages the project and there, in fact, that’s when you go to the other side of the fence and you realize that it’s indeed very complicated. There are a lot of things to do, there was already what you, you had discussed with us what you wanted us to do, so the fact of rewriting the application etc. And we can talk again between what you thought and what we were able to do at the time. Because we had to redo everything, well I mean, there wasn’t just code, there was also a whole commercial management, there were a lot of things to put in place.

So it was a pretty formative but quite complicated period, I don’t know Alexandre, what can you say about it?

Alexandre : There are two points that come to mind quite easily. We had already mentioned it several times, it was the need to rewrite the software. Honestly I think I still agree with this state of affairs and it is clear that we have it, it’s been eight years now, we’ve never done it as a result. By technical constraints, by the size of the team, by the fact that we are a company and that we have customers in front of us and that we don’t necessarily have a… You said it yourself that there is necessarily the fact of being in a community, you have the choice of your roadmap a little freer. On our side, there were also financing needs on our side on which to be part of a complete rewrite was considered but as a result was never possible. In fact, after these eight years, globally, GLPI almost doesn’t look like it did back then: there have been a lot of changes, especially in recent years.

We finally managed to put together a team in the end, since at the moment there are about ten people working on the source code. But at the time, there was Walid and I, we had a young developer with us, we tried to recruit additional people and in the end we realized that it was not possible to take care of that. So we worked by iteration at the beginning rather lightly. So that was the first point.

The second point is above all that there was indeed a community as you were talking about, but there was also a set of partners, commercial companies of which we were a part. And so we were, I don’t know how to put it, chosen to do this cover for various reasons, but suddenly there were the companies on the other side who found themselves in competition with the publisher. There was a choice we made from a commercial point of view which was to say “we no longer integrate with the software and we leave all of this revenue to existing or future partners who will join us and we will focus on paid services, so improving the roadmap and or support”.

And so at the beginning it was very difficult, but in the end it more or less worked. And I imagine we’ll talk about it in the next episode.

The first transition of maintainers

Walid : There are things that we have kept very clearly, the whole philosophy of the software on the functionalities etc. There was a legacy that was excellent. We just tried to look for additional funding. There’s a second thing that we haven’t talked about, which is that contrary to what everyone thought, the source code didn’t change licenses, it remained GPL. The source code was still available.

Alexandre : He changed recently though, but he stayed on a GPL.

Walid : There are transitions of maintainer that are more or less successful or not, we have placed ourselves in the continuity of what you have done, Julien and Jean-Mathieu. We didn’t reinvent the wheel in terms of philosophy and way of doing things, we just did it differently because we are a company, we had customers, we had opportunities, there are things we were able to do, there are things we wanted to do too.

But overall, we tried to keep this philosophy and make a smooth transition because it was easier. There are projects that have rewritten everything, I think. So in particular I think it’s Nuxeo, I don’t remember which GED (Editor’s note: electronic document management) that was created in Python and that was completely rewritten in Java. There are projects that are being rewritten but in the end I think that this is not the majority for part of the reasons that Alexandre explained, I think it’s because after the economic reality I guess makes it very complicated to have two different teams working on two different technologies. Julien, Jean-Mathieu, what is your feedback on this? How do you see things in hindsight? And compared to what you had imagined?

Julien : We passed the torch, but we didn’t pass the torch by saying we have to rewrite.

We passed the torch, then you guess the guardians of the temple. And we, this complete rewrite, we led the project for 13 years by making things evolve smoothly as you have since done too. And the code of 0.2 has absolutely nothing to do with the code of the last version before it is passed by hand. So we did that for 13 years and maybe that’s also what scared us as it scared us the complete rewriting.

Alexandre : Actually, I think we were both very enthusiastic about the idea of doing a complete rewrite of the software, we would have liked to do it in real life… But hey, it didn’t work out like that in the end. I have memories of the discussions we could have at the time on the complete rewrite, what posed a real problem for us in the associative model was to say “if we completely rewrite, we need at least a year, that is to say that for a year, we block all the evolutions for the users”.

And so we risk having a loss of community for a year when we were a little out of our guideline of making GLPI evolve for users.

Jean-Mathieu : Julien said a little bit of everything. So, I would have liked to see GLPI NG (laughs) if only out of curiosity. That’s it, this rewriting project, we had made a phase advance, we even had a POC (Editor’s note: Proof of Concept), so we still had nothing.

I would have liked to see out of curiosity if, out of technical curiosity, if the choices we had made, the precursors we had seen and the functional advances we considered were relevant or not. There you go, it would have amused me to see him.

I’m like Julien, we didn’t impose on the partner who was taking over the project to rewrite GLPI, we explained what we had done and where we were. We completely understand that there was an existing one, there was a community to satisfy and then then in the context of a business model, there are constraints both financial and customers and financing and others, there is no problem.

With hindsight, I have the feeling that the moral constraint was respected. The software is still under a free license, there is still an existing community, after that there have been evolutions in the economic model but even in the positioning of the partner since it has gone from integrator to publisher, properly speaking the solution and the integration part is entrusted to the partner so it’s model evolutions. What makes me happy is that we are celebrating the 20th anniversary of GLPI this year, that GLPI is still there, which obviously continues to satisfy many structures.

I met IT specialists from a large local authority who told me “we’re going to deploy GLPI here”. So there you have it, so it proves that it continues to expand and then to meet needs. I would say that we will be full.

And then I can still see myself in 2002, when we released the first GLPI archive, I never imagined one day that we were going to celebrate GLPI’s 20th anniversary.

Walid : I don’t think anyone imagined the turn it was going to take. It’s still a great adventure. There’s one thing I take away from it, which is a big lesson I learned, which is that in fact, even if you have a project, it’s first and foremost men, it’s people. It works well when people like each other, they respect each other and they manage to work well together. That’s really the thing I’ve always felt with GLPI, it’s the fact that there was really respect and that we were all able to work together.

If I take the case of OCS, for example, there was a moment when the people at OCS told some of the people who worked with them, so Gonéri, David, etc. “Well guys, it would be better if you went and did your project on the side.” And so often that’s how forks that go off are because there are people problems, in the end. It’s not even that we didn’t like each other, it’s just that we had divergent interests in the end.

And on GLPI, we managed to have that, a common goal and all that, and working well together, it’s really something that I remember and that I’ve always tried to keep afterwards.

Jean-Mathieu : thank you, it’s nice, we’ll take it.

I always insist, GLPI was not a technical project, because technically there was nothing complicated about making a GLPI. It wasn’t a diva’s project either. I’ve seen a lot of projects like that, or even sometimes people who came to us saying “your project is shit, it’s not even made in, I don’t know what, the latest fashionable language, it’s not even hosted on this platform because it’s the trendy platform, And then your thing, it rox this not, it is not international, it is not this, it is not that, that’s it. And then I’m going to do one again because it’s not complicated to do again technically.” And then after two years, we didn’t hear about this project at all because the diva had moved on, or she had jumped on a new techno that was even more fashionable and nicer and funnier.

We have always been aware that our project was not technical, that it was above all functional and that after the stakes, the real challenge was the community aspect. It was the strong bond we had with our users. The communication we were able to have with them, the exchanges, etc. We thought it was above all a human adventure more than a technical one, so we always rode it like that.

I have absolutely no regrets about the past 13 years even if they were extremely time-consuming and energy-consuming, it was a great moment in life and a very beautiful human adventure.

Acknowledgments and other self-congratulations;)

Walid : yes, if I was going to say, I’ll give the floor to each of you to make a final word since we’re doing this interview. Julian?

Julien : No, but I confirm, it’s clearly a very good year of my life, so I had a lot of fun with GLPI. I also learned a lot about many things that I use today in my work. GLPI was a community with very varied, very diverse people, so we had to manage to get everyone to work together, to get the best out of everyone. Each with its own qualities and defects. It wasn’t always easy, but I think we got there.

One of the elements that perhaps marks me the most in this adventure is the meeting with Jean-Mathieu, because that’s perhaps what I get the most out of. Because he’s already an exceptional person. He’s the kind of person you don’t meet often in your life. We may meet one or two of them. I don’t even know how to describe it because here it is, it’s someone who brings you this… and therefore who has taught me a lot during all these years.

Jean-Mathieu : Do I have the right to cry (laughs)? Thank you Julien, in any case, I was very touched by what you just said. Because I know that in addition, it’s not necessarily things that you say easily. It gives more value. But I would also say that I had the chance to meet exceptional people as well and you are one of them because without you, GLPI could never have been what it is.

And then on a personal level, I didn’t just meet people who contributed too and I made friends. It’s worth its golden pleasantry.

Walid : Alexandre, do you also want to give a final word about it?

Alexandre : I’m going to continue on the thanks, since we’re in that phase. Overall, I thank the three of you. Julien and Jean-Mathieu, already for your welcome at the time, your extremely good advice and your support at the beginning of the takeover, even before your support in my early career around GLPI, it was something quite driving force for me, whether personally or even professionally.

Of course, Walid at the time, who was more or less my mentor. I was really a young developer at the time, and that allowed me to take off very strongly. In any case, thank you… I insist, I thank all three of you for the project, but also for your welcome and for yourself.

Final words

Walid : I’m going to conclude by thanking myself too. When I arrived on GLPI, I was 29 years old, I was totally immature, I had no idea how free software worked, and you took the time necessary to make sure that I was able to contribute and insert myself into the project. And that was really great because that was what I wanted, to do free software.

Through GLPI, I actually had, basically, I had a lot of opportunities, already personal, since in the end, my best friend is Gonéri, we still see each other and we still talk to each other every day and we got closer thanks to the project. I spent great years with you, with Remi, with Nelly, with David, etc. We all had a lot of fun.

We were still lucky enough to experience some great things. I’m thinking for example of a fork with a project launch at FOSDEM, quite incredible things that we don’t do every day. That was great.

And really, I think everyone considers themselves to be each other’s mentors. So I’d say that Julien and Jean-Mathieu, I really learned a good part of the job with them. So thank you very much.

It makes me really happy to see that GLPI is still here too and that it continues and it’s one of the words, the last word that I wanted, that I wanted two words from the end.

The first is that we are in a world where we have to do everything right away and we can see very well that in this context things take time and that this is normal and that it allows us to do things properly.

And the second thing is that I did a lot of research, I read a lot and in fact there is not much literature on the good transition of maintainer and yet in projects when it works well it is that often there are good transitions of maintainer and here we talked about one, tomorrow we will talk about a second one.

Thank you very much to all three of you and for the listeners if you liked this episode in a few days the second episode will be released. If you liked it, don’t hesitate to talk about it around you, to share it on social networks, and to leave comments on streaming platforms. The podcast is available on all major

This episode was recorded on July 12, 2023.

License

This podcast is published under the double license Art Libre 1.3 or later – CC BY-SA 2.0 or later.