Florent Cayré – Co-founder of COMMOWN

Interview with Florent Cayré – Commown

Walid : Dear listeners, good evening, it’s June 21st, the day of the Fête de la Musique, and it is with great pleasure that I welcome Florent Cayré. Florent is one of the founders of a cooperative, a SCIC (cooperative society of collective interest) called Commown. I must say that I am one of the members, even if I am rather a sleeper person. A few years ago, I met Commown, I thought the concept was great and so I became a member. And it’s also, we’re going to talk about it, thanks to Commown too that I discovered /e/OS which is an OS that we’ll talk about later, which is a degoogled OS. I really wanted to talk to one of the founders of the cooperative. Florent, listen, welcome to Projets Libres. I hope you’re doing well and thank you so much for taking the time to talk with us.

Florent : thank you. It’s great to have the floor.

Presentation of Florent

Walid : Look, I tried to look on the Internet for interviews, I haven’t found many of you. Can you start by introducing yourself a little bit for our listeners?

Florent : Yes, indeed, I don’t do a lot of interviews, I’m not very visible in the media, but I still try to be present enough at a certain number of round tables. I’m Florent Cayré, I’m an engineer by training, an aeronautical engineer at the base, not even a computer scientist, but I took the tangent of computer science and especially free computing, having a computer under Debian since 1994. I’m a bit in the old idiot gang. And as a result, I quickly did computer science in my profession, since in aeronautics, I started by doing a lot of numerical simulations in fluid mechanics, which is my scientific specialty at the beginning. And then, quite quickly, I didn’t like this environment so much, especially in terms of atmosphere. It’s still a very vertical management style. These are former military companies. Moreover, we were also working on military projects. And so, I was passionate about aeronautics, little by little, the passion diminished. The IT sector remained. And I joined a small group of classmates who wanted to set up a company. I said to myself: Come on, let’s go. I am going into this with no entrepreneurial spirit.

Both my parents are civil servants. It’s not a family culture at all, so it was quite complicated at the beginning. On the other hand, I found a little more meaning in it. I started using a lot of free software and then contributing to it. And little by little, I used some pretty innovative technologies that were being developed at the time by Logilab, which I ended up joining after using the cubicweb framework a lot, a lot for my projects in my previous company.

Here, at Logilab, I worked a lot on the scientific computing part to try to put Web technologies in the management of scientific data, so to manage clean scientific data to make beautiful calculation chains, an issue that I had addressed very well in the field of aeronautics. Little by little, I started to contribute a lot to free software, to not really be able to do without it. I have participated in great projects at Logilab, especially in the field of energy production. I ended up leaving, especially for Strasbourg, leaving Paris and leaving for Strasbourg. And then, I started looking for a project that was really very aligned with my values.

So, I wanted free software, I wanted cooperation too, because one of the great frustrations I had working at Logilab was that we had so much work that it was quite difficult to collaborate with other companies. So, I tried to work with a nice company from a friend who did numerical simulation in fluid mechanics. And it was very complicated because we couldn’t find ways to cooperate and interests in cooperation strong enough for the relationship to last a long time, and it was a great frustration. I started looking for a project that combined free software, cooperation and ecology. And in fact, while wandering through Linux user groups, I met Élie. We sat next to the order, we said: What do you do for a living? What do you do? We told our stories a little bit and we realised that Élie, who was carrying out an associative project at the time, which was that of Commown, was looking for partners to move to SCIC. One of the roles he missed was a technical role and a lot on the free software aspects.

I really fell in love with the project. The next day, I was a volunteer for the association and then, a few months later, after completing the team, we started in SCIC at the beginning of January 2000.

Walid : What does free software really represent? What makes you can’t do without it?

Florent : For me, it’s simply knowledge. That is to say, I can’t imagine keeping mathematical theorems secret. For me, software is the same. It is pure knowledge. It is a commons that is a public good, which is a good of general interest. I don’t want it to be privatized at all. And then, it seems totally counterintuitive to me to use something that you don’t know what it’s doing. So, proprietary software is exactly that. You install something on your computer that is not even really your computer anymore when you install non-free software on it, since you don’t know what it does. I am willing to give my trust to things, but that I can open the hood is strictly necessary to give my trust. This is what free software allows. In addition, it has an enormous emancipatory role, especially in third world countries, developing countries. Today, without free software, I think they would have zero computer skills. Today, I think it gives a chance to people who had none.

Commown’s business model

Walid : You mentioned the name Commown. I also introduced it at the very beginning by introducing you. Could you start by explaining to us what the Commown project is? I’m going to call it a project at first. What was your initial idea? Is the original idea still the same as it is now, or has it evolved over time? When did you set it up?

Florent : We set it up in 2018, the SCIC, in the form of a company, but in reality, it’s a project that Élie had already launched a year and a half before, in the form of an association. His initial idea really stayed at the heart of Commown. Obviously, it has matured a little bit and everything, fortunately, but in reality, everything was already extremely detailed, extremely well thought out. The project is essentially articulated around three really main pillars. The first is an economic model. In reality, that is to say that the economic model was chosen not at all by chance, based on the observation that the obsolescence of electronic equipment, which is really a disaster, because producing electronic equipment causes a lot of both human and ecological damage, and therefore having obsolescence on these products, is really a phenomenal waste of resources, mainly natural resources. You have to know what a mine is to get an idea of what it’s like to put a phone in the trash, it’s really catastrophic. So, the starting point is: we have to fight against obsolescence to produce much less of these devices, to stop throwing them out the window. And so, for that, we have to go back to the root of the problem.

The root of the problem is that the people who produce equipment today, they sell it. And since they sell them, they have an interest in selling more. And to sell more, there have to be some that go in the trash somewhere.

And so, when you’re a big player, in particular, you have no choice but to have obsolescence strategies to survive. And so, obsolescence strategies are extremely varied and extremely clever because in any case, the survival of these major players depends on it. It is not only technical planned obsolescence that regularly makes the headlines thanks to the association Halte à l’obsolescence programmee (HOP), which has really kicked the anthill. But there are plenty of other mechanisms, including advertising, marketing, which doesn’t stop all day long from you slip in your ear that after all, you really have a device. You absolutely have to change if you want to look cool. So advertising is really the biggest obsolescence mechanism that can exist. And the second one, really very closely, is the software. And in particular, the software is non-free. Software is really the thing that allows… Even free software, in fact, even if it hurts me to say this, is really the thing that makes devices obsolete because of a lack of updates, essentially, security updates, since security updates are expensive. And the players have no interest in it, once again, in particular Qualcomm, Mediatek manufacturers, chip manufacturers have no interest at all in keeping the software in place for too long because at that point, their chips, they can’t replace them, they can’t sell new chips. Commown starts from this observation and says: We have to stop selling. When we said that, we didn’t say anything. A whole bunch of things still need to be built. In fact, in reality, it has already been 30 or 40 years since people have made this observation, including Rank Xerox for example, which started selling not photocopiers, but photocopies. And so, the solution to that is to sell the use of a device instead of selling a device, which forces the supplier to repair his device instead of selling a new one, because in reality, he no longer has an economic interest in putting a new device on the market.

It’s better to repair the device because it costs less, since what we sell is the use. In fact, it’s a kind of rental.

So be careful, there are some rental players who really don’t do the job. We are transforming the model so that the ecological interest and the economic interest coincide. The ecological interest is to make the device last and it’s exactly the same for the economic interest. The longer it lasts, the more we will have made the investment in expenses profitable.

Walid : So, it’s a rental without an option to buy.

Florent : That’s it. That’s the second thing, which is quite important, is that indeed, from the moment we give a purchase option, we fall back into the mechanisms of obsolescence, we just delayed them, that’s all, but we haven’t defeated the evil at the root. We even find completely hallucinating rental offers today, like, I think it’s Samsung that does that. It is written on our website somewhere. We give you a new phone every four months. We are almost the exact opposite of the objective, while the economic model is still officially rental. In reality, obviously, there is a lot of resale behind all this. So, it just becomes a purely financial game. From the moment you buy and sell, the rental stage is almost financing. In reality, and so we are turning into a bank, which is obviously not at all our business. We do a lot of services to make the devices last as long as possible by advising people, by repairing the devices, by making sure to protect them, by updating them, to upgrade them if they need to be upgraded, by increasing the RAM, this kind of thing, increasing storage space, helping people transfer their data, not lose it, etc. That’s what makes the devices last, to keep the devices in the hands of their users.

Walid : I’ll make a first digression before forgetting, because when I hear you talk, I feel like I’m at work. We work in household appliances and we work on reconditioning. Not too long ago, the government came up with something called the reparation bonus. I just wanted to know, because it’s one of my current topics, if it was something you get the repair bonus from or not. Are you entitled to it or not? What do you think? Basically, the idea is that the government applies a bonus. That is to say, basically, when you have your appliance repaired, if it is broken and if it meets a certain number of criteria, the person who repairs it and who is approved will apply a discount and then will contact an eco-organization to recover the bonus for himself.

Florent : ok. Clearly, we don’t benefit from it and we can’t benefit from it. The cooperative remains the owner of the devices it purchases. I say the cooperative and not the employees or not the co-founders. There are now about 600 shareholders in the cooperative, which owns the entire fleet. So, clearly, we cannot benefit from this kind of mechanism. What do I really think about it? I’m quite skeptical about this because I think that it will essentially produce a windfall effect and that repairers will just increase their prices by the amount of the subsidy. But it’s the experience that makes me talk, mainly because I saw exactly the same thing when I was a student and they said: Look, we’re going to increase the APL. Rents have increased by exactly the same amount as APLs almost instantaneously. So, in reality, it’s a bit of a disguised subsidy to a certain industry. After all, it’s not the industry that I dislike the most. Repair, I find it an extremely healthy activity. I would be very happy if a large number of repairers were created. I prefer that we subsidize this kind of thing rather than polluting industries like Total, in, I don’t know, me, for example, by not charging the price of aircraft fuel at their fair price.

Walid : We agree. I close this parenthesis to come back to Commown. I got to know Commown in 2019 and at the time, from memory, you were very focused on smartphones, on phones. And then afterwards, you diversified too. But at the time, it was very focused on smartphones and Fairphones, in particular. That’s also how it caught my attention a bit. Is it still one of the main activities, the smartphone part of your business?

Activities around smartphones

Florent : Yes, it’s still a main activity. It remains the main activity. Indeed, today, we have diversified a little in many directions. In particular, we make other electronic devices, computers in particular, laptops as well as desktop computers. And also, now, we’re working a lot on the audio. But there are plenty of historical reasons for starting with Fairphone. One of the reasons is that Élie has one of the very first Fairphone 2, really “One of the first“, it’s written on it, on his phone. He quickly realized that it was still an ultra-innovative phone and as a result, that there are still a number of disadvantages due to this youthful design, and in particular a lot of small false contacts, a lot of subtleties that make it a difficult device in reality for the general public. He said to himself: This is precisely the right device to show the added value of a service like Commown’s. Because it’s not easy to keep a Fairphone 2 for long. You have to have the spare parts, you have to know all the little defects it has, you have to be able to know the device well. And as a result, Commown could gain this expertise and we gained this expertise quite quickly.

We helped a lot of people to keep Fairphones 2 for a long time. And then, obviously, the other really good reason is the basic one, which is that Fairphone is the far, very innovative and virtuous player in the manufacture of electronic equipment. They have a very important transparency. If you read the reports they make, they are extremely detailed, extremely precise. They are working hard to convince the rest of the industry that we need to be cleaner, that we need to do better in all areas: design, manufacturing, transport, even packaging. He was a natural actor. We don’t want to distribute devices that are made in deplorable conditions by kids and possibly with a lot of deaths in the mines. Obviously, Fairphone was the first manufacturer we wanted to work with. Since then, it’s not easy to choose new manufacturers because up front, in Fairphone, they are all much worse. But we still try to find criteria that allow us to open up to other people. And in particular, we try to find players who have a reasonable size that allows us to have an influence on them and to tell them: Look, we have to increase the warranty periods, we have to improve this design point, we have to allow free software to be installed on it.

This kind of criteria. We have a whole bunch of internal criteria to choose equipment that remains for us the best or with the best players on the market. This openness is still very controlled, this openness in terms of products. In terms of the market, we have also opened up to companies since then, which we didn’t do at the very beginning. We only did individuals. Now, we do business, we have a lot of corporate customers, it now represents more than half of our turnover. And then, we also have local authority customers. And then, we are trying to develop this market. It’s much more difficult, much longer, but we think it’s interesting.

Walid : Before we go into more detail, because there is a subject that really interested me very much, it was how to manage a fleet of aircraft and how to make it last the longest. But first, I’d like you to give us a little more detail about the economic model.

Florent : The business model, it’s quite clear, we start by buying devices. We are the ones who make the investment. We borrow money, essentially. We also get help from a whole bunch of members like you, for example, who help us finance our devices. How do they help us? By lending us money, by taking part of the capital, which allows us to come in and say to the banks: Look, we have a lot of capital, you can lend us a lot of money and for cheap. Overall, we are fairly well followed by the banks, not too much by venture capitalists, except for individuals. The first difficulty is finding money. Once we have found money, we buy these devices, we rent them with service, paying close attention to many things, including that they are up to date if possible, before sending them. We update them, we install a screen protector on them. We try to make sure that the device arrives in good conditions and if it is ever a little roughed up, that it is still alive for as long as possible and as little damaged as possible by its user.

For example, screen protectors are things that are often damaged. We change them for free, it’s included in the service because we have an economic interest in it.

In reality, once again, we also have good will, but we also have an economic interest that coincides. Which means that it’s sustainable, even if the co-founders end up leaving the project one day, for example going to retirement, even if it’s a very long time away. But if one day all this happens, in reality, the economic interest will remain there and as a result, the project will be sustainable. That’s one of the important things. This economic model, for us, is extremely stable. One of the subtleties we’ve brought relatively recently to all of this… No, I am not finished on the customers. Rents, as it is in our interest for the customer to keep his phone as long as possible, we make decreasing rents. That is to say, the longer the customer keeps the device, the less he will pay. This is logical, on the one hand, with regard to the accounting depreciation of the devices, but it is also logical that it is in his interest to keep it a little longer. If the rent has dropped considerably, which is the case with us, I will give figures afterwards, obviously, the customer will find it difficult to say: Hey, I’m going to take a brand new phone and pay four times more than before.

To give you an example, today, if you go to our website, you’re going to look at the Fairphone 4. You have a minimum one-year commitment. The monthly payments are €29.60 and after five years, you drop to €11 per month. The economic model encourages people to keep the aircraft and each time, we add discounts if there has been no breakage, for example, if there has been no theft, that kind of thing. That is to say, if, roughly speaking, you take care of your device, the price decreases a little more. This is the economic model. Afterwards, we added a little subtlety, but which, for us, carries a lot of meaning.

Today, we have an agreement with Fairphone, which we talked about on our blog by the way, which consists of donating part of the rent to Fairphone.

Why is this a bit structural and it is really, for us, the ultimate model? In reality, Fairphone themselves, they have a contradiction since they are selling. They don’t stop working for the durability of their devices by making modular, repairable devices, they do an extraordinary job.

In reality, one day, if they really have the success they deserve and it is happening, they will face the same difficulty as the others, that is to say that their business model of sales will contradict their objective of making the devices last. As a result, we’ve been aware of this from the beginning too, but they too have a great maturity with regard to these subjects. We have teased them, not to say harassed them for a long time to set up this kind of economic model where in reality, they have an interest in the device lasting a long time, including financial, since we pay them part of the rent, so they will earn with us, with Commown, they will earn more money than if they had sold their device to an individual or a company thanks to the pension rents that are paid to them. The final objective, our absolute dream, would be that in reality, the money earned by manufacturers could be used to do R&D to further improve their devices. So that they don’t make money through sales, but they make money through sustainability, over time, have an extremely stable source of income through rental.

And in that case, he could focus on one business and do it well, which is designing phones and the software that goes with them and maintaining it, because it’s a phenomenal cost center today, when in reality, it should be more of a gain and not a burden. In our model, making the devices last is a gain. We are trying to transfer some of that to the manufacturers. And we set this up with other people like why! open computing, for example, which provides us with a good part of our Linux laptops. We have the same kind of mechanism in place and they have exactly the same awareness of all this as Fairphone. They have also understood the interest of this model in the long term.

Walid : It’s interesting because having also spoken at one point with the people about Fairphone, the software, it requires an incredible number of engineers and they follow the versions of Android. This is something that is extremely complex. It’s always chasing behind. At one point, they had made their own version of Android. I think that afterwards, afterwards, they abandoned it, in the end, because I suppose that they must not necessarily have been strong enough to be able to maintain it, which was already de-googled at the time.

Florent : Open OS, Fairphone Open OS, yes, absolutely.

Hardware and spare parts

Walid : At the time, I had a Fairphone 2 with it, it was pretty cool. You started with telephones, then you made computers. Now, you’re typically on headphones, that kind of thing, that’s pretty cool because when you have a headset, I don’t know, a Bluetooth headset, for example, and you have to disassemble it, often, you break it. And then, it’s not something that is put forward at all by manufacturers, repairability. There, it’s also, I suppose, a permanent work of sourcing new equipment that you’ll be able to put in, since in any case, there is one time or another, we arrive, no matter how much we do planned non-obsolescence, there is a time or another, it’s just like, I don’t know, me, a Fairphone 2, there are no more parts. There, I guess you live with your pieces.

Florent : That’s really interesting that you address this issue of spare parts. In reality, we still have plenty of Fairphone 2 pieces. Why do we still have plenty of them? Because we still have plenty of phones. Unlike the manufacturers, quite simply, Fairphone themselves, they no longer have parts of Fairphone 2. They have less than us. They could ask us for it. We’d keep them to ourselves, by the way. We jealously guard them because that’s what allows us to fulfill the promise we make to our customers to make the devices last as long as possible. One of the things we do and why we limit our diversification in terms of the number of models is that it is in our interest to have the most uniform fleet possible so that we can take spare parts from our own fleet for as long as possible.

Our own fleet becomes a reservoir of spare parts.

When you can no longer buy spare parts from outside, from the manufacturer in particular, you dismantle devices. There are always devices that have a given fault, but the vast majority of components remain perfectly operational. We can disassemble them and reassemble them on other phones that are broken, broken, etc. And we extend the life of old devices in this way.

Walid : On this purely hardware part, do you do any repairs? I don’t know, welds, stuff like that, etc. Or do you just have a set of parts and then there’s one that doesn’t work, we change the camera module, and then we put another one in its place?

Florent : That’s a great question. Because we’re in the middle of working on this subject. I don’t know how you do it, because it’s really extremely topical for us. In 15 days, I’m even going to a congress in Grenoble called Sustainable ICT, which talks about sustainability in electronics. We will meet a great electronics teacher from Grenoble, Vincent Grennerat, with whom we will discuss the possibility of increasing our skills on repairs that are a little low-level, I want to say. Today, our skills are extremely limited in this area. There are not yet any big names in electronics in the team who would be able to develop repairs, debug difficult problems, etc. Honestly, it’s still a bit early for us. We don’t have a sufficient volume of devices with breakdowns of this type to justify a full-time job in this field, clearly not. But suddenly, we start by cooperating with other people whose job it is. So, from time to time, we subcontract a repair. Clearly, in the long term, we want to increase our skills because it will be part of the basic job.

We also wait until we have the financial strength to be able to pay someone to do this more or less full-time with us. But clearly, we have that in mind. Today, we only know how to do repairs, extremely simple welds. And even then, it disrupts operations a lot today, so we tend to subcontract them to people we know well and who we know will do a great job for us, such as the Déclic Eco Teaching association. In short, we are trying to find a situation that is stable enough to both repair our devices and increase our skills at the same time.

Free software

Walid: I still have tons of questions about the hardware part, but since I’m doing a podcast about free software, we’re going to come back to the soft part a bit. Here, what’s really interesting is that we end up with Android phones, Linux computers. Now, we understand where this influence comes from. I would like you to tell us about your meeting with, I don’t know if we say /e/OS.

Florent : /e/OS, yes absolutely.

Walid : just for listeners who don’t know, /e/OS, it’s an OS based on Android Open Source Project, on which all the Google layers have been removed and which was built by someone quite famous called Gaël Duval, who is someone who set up Mandrakesoft at the time, so Linux distribution, that we knew when we were young, and who did a lot of other things. I tried it for the first time in 2019 on a 2015 Moto G and my phone was given a new lease of life, I hallucinated. It also explains how you can make a phone last long enough by the fact that by changing OS, the phone regains decent performance.

Florent : Yes, absolutely. There are many things in what you have just said. Indeed, we started by making Fairphone OpenOS. This was before the Foundation existed. Today, if you want to find them on the internet, it’s better to type Murena, which is the company’s trade name. /e/OS, it’s impossible to find on search engines. So, type Murena like a moray eel, but with an A at the end. Today, we are partners, which means that we have even set up a collective together. We co-founded a collective called FairTech, both with Fairphone, TeleCoop which is a very interesting and very important player for us too, the e foundation and Commown. Then, we also have other cooperative telephone operators in other countries, WEtell for example in Germany. It is indeed a partnership that is really interesting for us. How did we meet? Simply when we saw what it was like to maintain Fairphone Open OS. It was a purely OSP-based OSP, so very basic, with very few advanced features. The GPS didn’t work well. There were a lot of difficulties in using Fairphone Open OS pure.

On the Commown side, we had built an image with extremely weak technical resources, knowing that we have no or very few skills in mobile software development in-house. We had made an image by installing a whole bunch of software that we wanted to see on phones to make them more practical. We had installed microG.

Walid : Wait, microG, can you explain what microG is? Because it’s a bit of the centerpiece.

Florent : It’s definitely the centerpiece, you’re right. microG is basically a software that will emulate Google’s APIs, that is to say Google’s services, which will emulate Google’s services in such a way that even the applications that are installed on the phone do not realize it, do not realize the difference. That is to say, they think they are using a Google service like: Locate me or there are many other things in microG, but I’ll stop there and then, I’m not necessarily very competent to go into detail. And as a result, microG hides the fact that Google is not there from applications. This means that a certain number of applications that, before installing microG, crashed instantly because they used a Google service that they couldn’t find, thanks to microG, they already don’t crash. Or even, they perform in a strictly similar way and with the same kind of performance as with Google’s services. And so that’s the centerpiece because it allows you to install perfectly standard apps on a phone without Google. So, /e/OS also started from there: Lineage plus microG, plus a large amount of work, gives /e/OS.

We realized that it wasn’t easy to distribute Fairphone Open OS to our customers, but it was a phenomenal success. We didn’t expect that at all. It must be said that we had good press coverage, we had a great article in the DNA (Dernières Nouvelles d’Alsace), with even an insert on the front page of DNA, the emblematic newspaper of the Grand Est region. It was a great advertisement for us and at one point we had a third of our orders that were under Fairphone Open.

Walid : I never would have thought. I thought it was going to be a geek thing.

Florent : Yes, at the beginning, we said to ourselves: Come on, let’s give it a shot. Anyway, we’re free software players, we want to do it, let’s go for it. And we were very surprised by the success. And not only us, that is to say that these are stats that we sent to Fairphone, they were amazed. And so, we hope that thanks to this, that it’s a little bit thanks to us, that they then set up a partnership with /e/OS. We must not delude ourselves, we had a rather weak influence on this. I think the main influence is Agnès Crepet, who is a great French developer who joined Fairphone, who probably already knew about Gaël Duval or at least, it must have made the first contacts a little easier. In any case, they have set up a partnership today with /e/OS and /e/OS was almost released on Fairphone 4 from the beginning, after two months, something like that. We can see that they have been working for a long time, that this partnership, it works like hell. Within FairTEC, there are meetings almost every week. It’s really great. Now, we decided to let go of our incompetence on Fairphone Open OS, in favor of the great competence of the Murena team, which is really, in addition, a completely hallucinating strike force compared to what we would have been able to develop.
And then, top skills in the field of mobile computing development.

Walid : What is your strategy in relation to this? The idea is to have a part of your fleet… They’re mostly Fairphones, I guess?

Florent : Yes.

Walid : you’re going to get There’s a part of your fleet that’s in stock Android, well in Android, Fairphone, and then a part that’s with e. How will it go for you?

Florent : The way it works today and for a long time now, and it works very well, is that we receive all the phones with the manufacturer’s OS, obviously. Often, anyway, we are forced to update them because after a while, we, as we buy them in large quantities, inevitably, the OS, it has already received two or three updates before we put them on the market, sometimes. Anyway, we’re going to flash phones. We offer our customers the opportunity to say: I don’t want Fairphone OS, I don’t want you to update Fairphone OS. Yes, I want /e/OS. We charge €25, which is not very expensive, for people who have not yet ordered their phone. We install /e/OS and send them the direct phone.

Then there are also people who say to themselves: I want to switch to /e/OS and I already have a Fairphone from you, but it has been under Fairphone OS. Can you help me switch to /e/OS? Here, we offer another service that may seem completely crazy to some, but it’s: We take a phone, we install /e/OS on it, a phone identical to theirs, we send them and we get theirs back. We give them 15 days in their hands so that they can play with both, transfer a little bit of data and all that, and presto, they get their hands on /e/OS and then they send us the phone back to us under Fairphone OS and it’s settled. We have no problem with that. In any case, our customers can install the software they want on Commown’s hardware. It’s the same on computers, if they want to install something other than Ubuntu, they have every right to do so. After that, it can cause some problems for the support. In practice, this is not a problem. We just ask people to tell us: If you have installed a different OS than the basic one, tell us. That way, at least, we won’t waste too much time explaining something that doesn’t belong on your OS. And then, if the OS you have installed, we really don’t know it, you will have a lower quality service, that is to say that we will be able to help you less on the use.

But on the other hand, we obviously keep all the services related to the equipment, changing the battery, when they are worn, etc. Support for breakdowns, breakages, etc. It works very well. We are happy to promote free software. For us, this is our DNA. And in reality, today, for example, Fairphone 2, we did a new operation not too long ago when we called FrankenPhone to revive phones that died for everyone, or almost, but not for us. So, we took Fairphones 2 that were still under Fairphone OS, so Googled and all that. We fired Google because it didn’t work anymore and it had become really too heavy for a Fairphone 2. By installing /e/OS on it, the device came back to life. This is exactly the experience you described earlier. It’s so much lighter than a basic Android, that it works much better on less powerful devices. So, it extends the life of the devices in a way…

Walid : Actually, what’s interesting about your model is that if you go to the Murena website and look, I chose my phone because I wanted to put /e/OS on it. I took a phone that was supported, I bought it refurbished and I put the thing myself in like a big one. But you, as, let’s say, the manager of a fairly large fleet, you can in advance, basically, know and potentially have a little weight to have the support of a certain number of equipment that you would have. And the other question is, because I didn’t dig into it, is what is the policy of /e/OS? I think I’ll try to invite Gaël Duval one day to talk more about it…

Florent : yes, that would be great.

The duration of the support

Walid : What is the policy in terms of support duration? Because for you, it’s still the sinews of war. If you have Fairphones that are already five years old or I don’t know how many and you want to make it last another two, three years, but the OS itself, it can no longer be updated, it’s the OS that becomes limiting, it’s no longer the hardware that becomes limiting.

Florent : That’s exactly it. Today, for us, the problem is the software. That’s why Fairphone’s efforts to maintain… Now, they still stopped supporting the Fairphone 2, but after a considerable number of years. I mean, they’re the only ones by far to have made all these efforts.

Walid : Because there are no more parts…

Florent : There are no more parts, but above all, what is the financial interest? As you said, it’s a huge cost to port new versions of Android to old devices because in reality, there is no cooperation from other players. Fairphone had to fend for itself. It’s a tiny box compared to Qualcomm. Still, they are the ones who did all the work of reverse engineering some of Qualcomm’s stuff to be able to port it to their Android 11 device, I think, to Fairphone 2. They have made an effort that is absolutely delusional compared to the rest of the industry. Eventually, you could invite Agnès Crepet, who has a lot to say about this and who is really great. We benefit from their efforts. That’s why we don’t want to buy phones from just anyone, because people who make efforts like that, they have to buy the equipment. They are rare. In reality, we can’t do much about it. That is to say that we do not have the possibility to develop an OS today. Then, in reality, there are people who do it very well and who do it much better than we would.

We are a cooperative, let’s cooperate. We cooperate with the best players in the sector. Today, it’s Fairphone and Murena, clearly.

We also have a lot of discussions with other manufacturers. We now know that Crosscall is still… Which is another type of device that we provide, which has other characteristics than the Fairphone. In particular, it focuses its durability a lot on robustness. They are also very interested in trying to provide software support in the long term. We don’t yet have the significant benefits of what Fairphone is doing, but we can see that there is interest. In France, in addition, there is the repairability index and the durability index which still put a lot of pressure on manufacturers. It helps us. So I say “It helps us”, in reality, we don’t look at it in a passive way. We are very active in the working group on the repairability index then and the sustainability index today, because for us, that’s the key in reality. We fought to get the free software community and even some government departments that were completely neglected in this matter to intervene so that free software would be included in the rating of the sustainability index, that is to say that manufacturers, ideally, have obligations, but also an interest, an incentive to allow the clean installation of free software on their hardware to be able to make it last a long time.

We have Adrien who works extremely intensely in these working groups, always in cooperation with the major players in free software and the great experts in free software who help us to provide and build arguments, etc.

Open Hardware

Walid : Two questions. There’s one that’s on the hardware side, one that’s on the software side. On the hardware side, to my knowledge, there is not really any open hardware, open hardware things that would really hold up for everyday phones on a daily basis. As for computers, the problems are a little different since a computer, apart from the transition from 32 to 64 bits which means that there are OSes that no longer run. For the rest, I have computers that are 10 years old, they are still running. You can always change the hard drive or the RAM, it goes well. As for phones, it’s still very different. The first question was: Are you looking at open hardware? Have you spotted any things that, potentially, could be interesting tomorrow? And the second one, on the software side, is: Are there free tools, free services that you lack to better manage all this fleet of hardware that you have?

Florent : On the first point, open hardware. We are deeply convinced, of course, as free software professionals, that we need open hardware in many areas and that in reality, it is one of the key solutions to fight against the big boom that will come to us with the climate crisis, which may prevent us from manufacturing new equipment, or at least centrally. There are a lot of issues around open hardware and the climate crisis. We try to promote everything we can in this area, but in reality, there are enough initiatives. As far as I know, there is almost none. Fairphone had still released a number of very interesting things, especially about the Fairphone 2 at one time. Today, I believe that they no longer do it, essentially, in my opinion, more because of lack of time than anything else. It’s a bit of a shame. Maybe we’ll talk to them about it soon since it’s the 10th anniversary of Fairphone. They invited us, so we’re going to prepare a whole bunch of small meetings and try to give them feedback. As soon as we can, we give feedback.

On these aspects, on the telephony side, there is not much going on. On the computer side, we participated a little bit, even a lot, at one point, in the Cairn Devices adventure. Cairn Devices, which is also a Strasbourg company, is really our neighbours, who have the ambition to manufacture a very modular laptop, components that are extremely easy to change, etc. They have a lot of difficulty starting this kind of activity, which is really very difficult and very capital-intensive. It’s very difficult in France to set up this kind of company. Today, they have, in quotes, fallen back for the moment on the development of an open hardware keyboard. This project is still really interesting. Here, we put a little effort into it. We hope that the keyboard will be released soon. I think that here, we are really close because they managed to get it produced. There are initiatives, we follow them and we try to support them with our means. Obviously, we are Commown, we are not a multinational. If, technically, we are a multinational, we also provide our services in Belgium and Germany, but we are still only based in France, we are not a multinational and we have no financial means.

But in reality, we manage to set up projects to get subsidies together. We did that with Cairn Devices. We managed to recover a subsidy of €30,000 for them and for us, to try to move their project forward and ours at the same time. So we help each other and we hope that it will give results. Today, there is a framework at the computer level that has killed the game a bit, that is to say that for us, their product, it is absolutely extraordinary and unique in the world for the moment. We would like them to go much further. And here, I think that on the other hand, we will have an extremely limited influence on them because it’s still a very, very… set up by former Facebook employees. In short, it’s not at all the same world as ours, clearly. For the moment, they are not yet selling to companies. We have been in discussions with them for some time. We’re going to try to buy their equipment to test it on a fairly large fleet to see if it holds up in the long term, but in any case, the concept is great.

There is quite a bit of open hardware in there. They published a number of their works, including the chassis. You can easily build a chassis, they have worked in this field and it’s cool. Now, it’s not just open hardware that matters. To make the devices last, there is nothing better than being able to repair them and have a good repair doc, diagrams, boardviews, etc. Today, there are very few manufacturers that provide quality equipment to help repairers work. Let’s start at the beginning. It would already be nice if all manufacturers did that.

Software tools

Walid : The second question was about software. Are you missing any … Hardware, you just explained a little, but on the software part, are you missing any services? Do they have free software to really plug holes in your offer or in general, to offer a rental service? Are there things that are really missing?

Florent : Full. We even started from scratch. That is to say, today, we, as we are very culturally focused on free software and not only culturally, we have good reasons, which I tried to explode, to be. Unfortunately, we have plenty of customers who have to go and find where they are, and in particular, precisely, companies. Our offers for companies include Linux. Of course, we would be very happy to provide a lot of Linux to companies, but in reality, today, most of these companies take Windows machines. So we provide them with Windows machines. One of the things we want to do is really transition counseling in a really qualitative way. Today, we don’t have the means to do that. That is to say, we would like to be able to say to all these people: What do you use as software on Windows? That, this and that. We have free replacements. We already know how to do that. And we’ll help you set them up at home. That, on the other hand, we do less. We would like to set up partnerships, but due to lack of time, we never succeeded. But we would like to have players who help us take our business customers by hand and try to get them to use more and more free software without necessarily making them switch to Linux all at once.

But software by software, little by little, make them uninstall all non-open source software from their Windows. And then, one day, take the plunge. Because once you only use free software, changing your OS is super easy. There are even only benefits to doing it since it’s much better done. Honestly, under Linux, all objectivity of course. We miss it a lot, but it’s not quite software, it’s more about competence, advice, a service offer that should be put in place. Then, the next step, even if technically, we want it to be before and I am the first, we would like to set up outsourcing, as we say, under Linux, that is to say to be able to provide really high-level services to our customers under Linux. Typically, software cable distribution, software remote configuration, the implementation of their entire information system. What I imagine today is to use technologies that are really made for that, to configure machines. It can be very sober, it can be script that we execute remotely via SSH bridges, via appointment servers… anyway, we can do some nice things.

Walid : In 2023, you have conf management tools that allow you to do that. But managing a Linux Desktop, compared to managing a Linux server, having seen my colleagues, a very long time ago, manage the Linux Desktop of the National Assembly, that’s a job.

Florent : Yes, absolutely. It’s a job, but it’s the one you actually want to do. That is to say, today, our support is what it does in practice, but it does so with technical means that are limited. We do small remote control takeovers, etc. We still have skills in this area, but in reality, it’s good for small companies or for individuals, but we’re starting to have customers who are a little bigger or who have needs, who need it to work at the top during desktops on Linux and we want to provide them with better quality service. Yes, clearly, there is Ansible, Salt, etc., a whole technical stack that I use every day anyway since I also have my computers in-house to manage. We still have almost thirty employees today. We can no longer manage this with bits of string and we have never done it anyway. Obviously, we still have a certain technicality. Except that you have to build a whole team to be able to do that. Today, we don’t have the financial means to recruit.

But we recruited young people when they were still really young. We tried to take some good players in the making and they were good. We soon have people who are going to be computer engineers, who are going to have master’s degrees in computer science and all that and who have a lot of experience in how support works at Commown and as a result, could set up this kind of high-level service or in any case, acquire the skills that are necessary to do so. So that’s all clearly the direction we’re going, but we’re going slowly, unfortunately. Or fortunately, because it may be the best way to do it. In any case, we are resolutely going for it.

Commown’s next challenges

Walid : That’s good because it introduces my next question, which is: What are the next challenges that await Commown?

Florent : That’s an interest at the moment too. The next challenges are basically to become extremely efficient at refurbishing our own devices and renting them out, especially computers. Laptop relocation is really complicated because a laptop that had a first life, it really has to be in very good condition to be able to rerent it like that without doing anything on it. That happens, but it’s mobile devices and as a result, it falls, it takes hits, etc. That’s why we’re trying to favor desktop computers, by the way, because they’re really, really, really durable. Laptops are much less durable. So, we recover laptops that are sometimes in a very bad state. We haven’t put in place all the architecture we need today to manage them effectively, i.e. our online store, it doesn’t easily allow us to set up refurbished computers with different grades, to manage the stock of refurbished computers, even if today, we know how to manage new stocks, but not too much refurbished stocks on the online store. We still have a whole set of tools, a whole database that is well done.

We know all our devices, we know where they go, we manage them perfectly in our ERP, open source by the way, Odoo, but we underexploit this stock management for the implementation of refurbished computers on the online store. One of the big challenges that we face is this whole IT and logistics chain of efficient and rapid management to put on the market as quickly as possible – so that they can be used again and not sleep in a closet for too long – devices, refurbished computers, and in particular mobile phones on which, Frankly, economically, sometimes it’s complicated to find your way around. Changing a chassis, for example, takes far too long to be economically profitable. So, it is better to rent the computer really cheap if it is in very bad condition. All this remains to be refined a little, let’s say. Obviously, we have a whole bunch of ideas, we already have practice because in reality, with phones, we’ve been doing this for a long time. But in reality, the telephone is easier. They come back to us in less bad condition or in any case, we have an ease to change the equivalent of the chassis, changing the shell of a phone is generally easy and fast, while changing the chassis of a computer is very expensive and very inefficient and very heavy. We are less advanced on computers than on phones.

Walid : Yes, I feel like I’m at work and hearing my colleagues talk about the reconditioning cycles of large household appliances. I totally understand. Before I conclude, I wanted to address one last question. This is the software where you were in-house. You use a software that I like called Odoo and in which you can manage all the refurbishing part with the beautiful repair module, which I had already done elsewhere too. What do you use just like that, as a range of in-house software, a little iconic, but a little representative?

Florent : Not much, because today, Odoo does a lot of things. That’s a lot of things in our house. You should know that we have developed many, many, many Odoo custom modules. A little too much by the way, because unfortunately, it’s often easy at the beginning, when you don’t know the thousands of modules that exist in the Odoo open source community. That said, our activity is still quite atypical and as a result, we still have trouble finding general, generic modules that correspond to our needs. We really do a lot of things in Odoo today, both support, the online store, accounting, invoicing, everything, just about everything, project management, well… We really use a lot of things from Odoo.

Walid : You don’t use the production modules to generate manufacturing hours, repair hours, stuff like that?

Florent : I did a big experiment on the Fairphone 2 with it, which lasted a long time. I had set up all the modules above that are related to this to assemble the Fairphone 2 at home, to have traceability at the part level. I didn’t tell everything because it would be endless, but when we received the Fairphone 2, when we started Commown, the Fairphone 2, we dismantled them all one by one. All the serial numbers of all the modules of the phones were recorded. We had a complete database of all the modules of each phone.

Walid : Can’t you get that through software?

Florent : Even Fairphone doesn’t actually have it. We were able to track down the modules by module and say: I dismantled this module from this phone, I reassembled it in another to repair it. Or: I disassembled this module from the phone, I sent it to a customer who sent me back the equivalent module that was broken and that I send back to the manufacturer under warranty. We were able to manage all that. In reality, today, we don’t need all that anymore. It was an extraordinary piece of machinery. It was very beautiful technically, but we had a lot of data to recover which, initially, was managed with spreadsheets and the recovery of this data was hell on earth for a long time. It was at a time when our business was exploding. I was the only computer scientist and I exploded in flight on that project because it was too ambitious to do right away. So, maybe we’ll do it again one day, it’s not certain. Today, we no longer trace at the module level and as a result, we no longer need assembly modules, manufacturing modules and recipes, BOMs, the famous manufacturing BOMs to trace at the module level, so we no longer use them.

The cooperative model of a SCIC

Walid : We’re going to stop this subject because I think we can talk about it for hours. We are coming to the end of this interview. I wanted to give you an open platform. If you have a message to convey before we separate, what would it be?

Florent : That’s interesting because there’s a hole in the racket, because I’m a little too talkative. One of the subjects of Commown that we haven’t talked about at all is the fact that we have a SCIC. This is the governance of a SCIC. In reality, when I started with the three pillars at the beginning of the interview, I only gave one. There were two more left. It was the choice of equipment, ethical, all that, building to last, eco-designed, etc. We talked about it afterwards. The third point was: We are a SCIC. And “We are a SCIC” is no small thing.

Walid : wait, SCIC, is it a common company of collective interest?

Florent : It’s a cooperative society.

Walid : cooperative, sorry. Cooperative.

Florent : I’m just going to talk about this model which, for me, is an extraordinary model and a model for the future, especially in terms of dealing with the crises that are coming our way. First, it is a cooperative. What does that mean? This means that it is not capital that governs. And that’s really important.

That is to say, it is a model that is not a capitalist model. So it’s not one euro equals one vote, it’s one person equals one vote.

Whether a person has the means to put €5,000 on the table and the means to put €100, he or she has the same value in terms of governance for us. So, that’s really important. That is to say, it is not money that governs. It’s really important from many points of view. One of which is the stability of the company, because if someone comes up with a million euros, they can’t say: It’s okay, I’m buying all your shares, go and show yourself. In any case, he will be able to buy all our shares. It will be with pleasure that we sell him, but he will have a voice like everyone else. On the other hand, we will welcome his capital with open arms because it will allow us to buy equipment. And that’s always welcome. On the other hand, what goes with the “Get out of the way and I’ll change your project” is not possible. So, a SCIC, you can’t change your project just like that. And so we have statutes in which a lot of things are set in stone, including the fact that our objective is to make the equipment last. And so, we won’t be able to change that, even with all the money in the world.

The second point is that it is a cooperative, but of collective interest. Here, we touch on a very interesting point, that is to say that a classic cooperative is that, roughly speaking, the employees who can become cooperators, more or less, it’s not quite accurate, but really not far off, and therefore only the employees participate in the governance. In a SCIC, this is not the case at all, since employees are a component like any other. We have many other colleges and in particular in a SCIC, it is legally mandatory to make it possible, even easy, for users of the service to take part. So, customers must be able to become shareholders and as a result, gain power. That’s really important.

This means that users can run the company and get it where they want it to go. And so, the company, it necessarily puts itself at the service of users and not, as we see with a lot of large companies in particular, I mention the GAFAM because that’s what annoys us the most as free software players. Basically, they are not doing their users a favor, they are doing themselves a favor to them above all. And so, even if it means stealing data from their users without telling them, etc. Obviously, we can’t do that, and in a structural way. That is to say that it is not just our good will, it is legal, we cannot do otherwise. And that’s really important for us too, for the stability of the project. That is to say, the day we leave, we have an accident, a thing, something, it’s not a big deal. It’s not a big deal, thank you, for us, but not for the company. So, the company will not change, it will remain stable. This SCIC model, which must record potentially divergent interests, is really interesting because it forces us to build an economic model that satisfies everyone in reality, and not the one who has the power at the time.

We, within SCIC Commown, have customers, it’s mandatory, employees and project leaders. We are in the same school, we have the same weight. We, the co-founders, have the same weight as an employee who would be in the same college as us. We have producers, hardware manufacturers, so why! Open Computing, for example, is part of the SCIC and has a say in governance. You can see that just by mentioning these three, they potentially have divergent interests. Employees will say: We would like to have a higher salary. Users will say: You’re nice, but if you have a higher salary, prices will go up. We want prices to go down, on the contrary. And the producers, they’re going to say: Yes, but you’re nice, if you put pressure on us to buy cheaper computers, how do we live? Same thing, I forgot about the financiers. And the financiers, it’s the same, they have a really important role. Since we have to start by buying devices, they too have an interest in being in the SCIC and we have a certain number of investors in the SCIC who allow us to have money.

This model of a cooperative society of collective interest is a model that is a non-capitalist model whose effectiveness we want to demonstrate. I think we’re getting there today because we have members who are all happy, there are a lot of them. Overall, the model is acclaimed by all members at the same time. We would like to increase the visibility of this model and encourage the participation of our members. We are working it, hoping to show that we have made a lot of progress in this area for the next AGM and that we are going to give even more power to people from outside the company to have expertise, to have divergent interests and to innovate in terms of model.

Conclusion

Walid : That’s a good conclusion. This is a praise for the SCIC. We haven’t talked about the Licoornes , which is a grouping of the different SCIC, but we can talk about it in another show at another time because it’s a really fascinating subject. And also the same, we haven’t talked about Telecoop, but we could talk about it again too.

Florent : That’s a lot of people to invite, it’s going to be great. You’ll be able to have a lot of great shows with Licoornes.

Walid : I’m going to do a second podcast which will be called SCIC Project or Licoornes Projects, we’ll have to do that. Florent, thank you very much for taking your time to talk with us.

Florent : thank you.

Walid : I fell in love with Commown the first time I saw Adrien talk. We talked after the presentation I had seen and I thought it was great. I really wanted to introduce this model because I think it’s very interesting. And then your approach to free software, your approach to fighting planned obsolescence by free software as well. In any case, I’m super happy to have been able to discuss this subject with you. I’m going to thank you once again and I’m going to say for our listeners that I hope they also enjoyed this exchange and that if so, that they don’t hesitate to share this episode with others on social networks, to give us comments also on streaming platforms since the podcast is available on all good streaming platforms. We have other episodes coming to talk about other jobs and other projects too and that it’s going to be very nice. Florent, listen, thank you very much and see you soon, I hope.

Florent : a big thank you to you. Keep up your show, it’s great. I listened to a few of them, so it’s really nice. Good job. Good luck to you to continue. Merci.Au pleasure.

Walid : See you soon!

This episode was recorded on June 21, 2023.

Transcription by Raphaël Semeteys.

License

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