Okay, so hello everybody and thank you for coming so early.
And so for those that were not there before, since you came so early, there are a few free
t-shirts if you want to take them.
So I'm going to talk about the story of our company, X-Wiki SS, building the X-Wiki and
CripPad open source software in the last 20 years.
So first a bit of a track about myself.
And so I discovered technology in 1984, like using an Apple II and then I moved to PCs,
I even moved to Windows 95, then I graduated from a good school.
And actually in that good school I was kind of told, so I was actually, it was that we
had a speech at some point telling us, you're a soldier of economic war.
And so that of course resonated in a young person, but it also, I mean later on, it's
like what, I mean why are we doing war?
Like that doesn't make any sense.
We're not fighting other countries, we should work together with other countries.
And then in 1995, I was really very interested by the internet.
I saw people using Mozilla browsers, Mosaic browsers in the school and I really just wanted
to work on internet technology.
And I ended up, so I started, I took one job about the internet at Capgemini, but after
a few months I was recruited by Netscape because somebody from the team had left to Netscape
and I ended up working three years.
So who knows Netscape here, like okay, like so just not so bad.
And so I was a consultant, I became also their Mozilla fan.
I even wanted to work for Mozilla, Oregon, inside the company when they launched it.
They didn't take me, so I ended up working in a French startup.
I wanted to stay in Europe and that startup raised money, went IPO.
I actually even was a virtual millionaire and then there was the internet bubble, it
crashed.
I was not at all a millionaire anymore, just like any other IT guy.
And in that company, we were both by US company and in that company I used Wikis and that's
how I wanted, I found Wikis like amazing in terms of how it brings people and helps people
share knowledge.
And that's how in 2004 I created XWiki.
I was a bit accustomed to open source with what Netscape was doing.
Netscape had a highly transparent organization and a way to share things.
It was really pushing internet protocols standards and then they made Mozilla open source.
And then I was a user of open source in my company as a CTO, like installing using Apache
software foundation code and so on.
And so in Wikis, we're purely coming from the open source world.
It was very natural when I wanted to create a company and create a software to create
it as open source.
But I was not as much aware of the political aspects of open source.
I was really looking at open source from the technical point of view.
So that's how I started XWiki.
I'm going to continue from that in this presentation.
But I'm also now a member, our company is a member of APEL, which is an organization
of companies that do open source in Europe.
In France, we have the CNL.
We have the Herb Open Source, OW2.
I'm also on the board of OpenFoodFacts, great association working on open data for food.
And I'm a small shareholder of Morena that is doing an open source phone.
So I'm welcome to look at this.
I find this amazing work.
So what is XWiki SIS?
So XWiki SIS is a French and European independent company.
So it means we've been self-funded.
And I say independent.
It means it's self-funded.
And I still own majority ownership.
And the very large majority of the share is owned by employees, some ex-employees.
Yes, I should stop.
Okay, great.
It's on HDMI, you put HDMI.
The slides are here.
No, no, you're here.
And you're not seeing them.
That's too bad.
I took this cable.
Sorry.
Okay.
So it means we control the company, which is actually something that's not so easy to
achieve in tech companies.
We are at around 4 million revenue in 2023.
We did 50% growth, which has been really very nice.
And we have 60 people, mostly in France and Romania, but also some people in Germany,
even two people in Brazil.
We do two open source software, XWiki and Kripad.
One is the one we started with that I created.
And then Kripad that was created inside the company in 2016.
We have an international community.
And so we also are very engaged for digital sovereignty.
We think open source is very important for gaining control of software, both for states
but also individuals.
And we have a business model of allowing to have revenue for that software so that we
can build it.
And this is done through services support training, like anything software company
do, but trying to do it in a way that it allows us to fund the open source software.
So we have employees in all these countries.
And so what we're trying to do is enabling freedom, both with the code but also with
the products that we do.
I'll come back to that.
So two software, XWiki.
It's about knowledge management.
It's sharing information.
What's really interesting with the knowledge, Wikis, is that it really allows people to
share and make knowledge available.
So we all know Wikipedia.
But we do it for organization.
Inside that area, we have competitors such as conference or notion or even the Wikis
of Microsoft Teams.
And this is part of, I mean the competition in the end for any open source software has
a high impact on how you can actually fund your work.
Depending on how you compare to the competition, you can find more or it can be more or less
difficult to find money.
We have more than 7,000 installs and more than 400 clients.
And XWiki is now part of the Open Desk project.
Also if you don't know Open Desk, I think about looking at it.
Google it.
We also did CripPad since 2016.
And CripPad is an end-to-end encrypted document editing platform.
Who knows CripPad?
Here?
Okay, good.
And so I'd say competitors to Google Docs.
And the real part is that it protects people's privacy.
So I'll go on, how did we start?
And so the big question is why be an entrepreneur in the end?
Because I'm trying to, so in this talk I will try to focus more about the open source aspect
of what we did.
But when you talk about your company, it's all also about the entrepreneurship and the
difficulties to just run a company.
But so I had this wish to kind of create things and make them happen.
And so that was a bit at the core of being an entrepreneur.
But one really important thing was to try to do something that's useful for people and
have some impact.
I wanted to do it also in Europe.
I've been to the Silicon Valley and I didn't feel I liked the moons.
So the technology was great, but I didn't feel as good about the fact that people were
just talking about money and how they would become rich all the time.
And that really made me think, okay, I don't want to spend my life in an area where that's
the goal.
Like I want to be more in a place where we're talking about culture or whatever.
And another aspect was as an employee in companies, you sometimes feel your managers are not doing
what you want them to do or they are not fair or the company makes decisions that you don't
understand.
And in the end, you can complain and stay as is and keep complaining about what people
tell you around.
But my idea was that my feeling was, well, instead of complaining, just try to do better.
And that's also a reason to become a manager or own company and be the one that has to
take responsibility for what's happening in the group.
A big aspect is about believing in the product and in the purpose of the product.
One of the really important things that motivated us at X-Wiki for 20 years is the fact that
we feel our products are missing or they're not enough existent in the world and they're
useful.
They serve an important purpose.
When I started X-Wiki, I was a big user of Wikis, I was a big user of task management
tools.
And I said, okay, we could do task management tools, we could do Wikis.
And I directed myself towards Wikis because of the fact that they help sharing knowledge.
Task management tools are a lot about efficiency being more efficient in companies and I felt
knowledge is missing more.
Like we're missing more the fact that we spread knowledge and that we educate people more.
And in the end, this has stayed with us for 20 years.
So we have a lot of Wikis inside companies that help people get more knowledgeable about
what they do and about the work in their own company.
But we also have a few public Wikis.
We have the dictionary of the history of Switzerland, which is a public funded Swiss project about
the knowledge about Switzerland.
We also have a Wikis about rare disease.
And you don't want to look at the website too much because it's sometimes really hard
to look at it, what the parents of the kids that have this disease live.
But it's highly useful for this community of parents that are living with the disease
of their kids.
We also have a Wikis for public service in France and so on.
And so from my point of view, if you want to stay motivated about software for 20 years,
you also need to really believe about the fact that your software is useful.
And in the end, in 2016, we created CripPad.
We created the technology, but we decided to make it a product because we really thought
it was doing something that was missing, is protecting people's privacy and that too many
software are exposing the data, are not built to protect the data, and CripPad is a product
that is built to protect the data.
So now the problem is that if you want to do a good software, you're interested in doing
it as open source, then how do you fund it?
There's different ways.
So you can just raise money.
So that works.
You can build, there is a lot of open source software that is built by companies that have
raised money.
Even now, the modern way of raising money is doing some crypto thing and launching a token
and getting millions of dollars.
So it can work.
I'll come back why.
I didn't feel it was a good approach for us.
You can be an open source volunteer and that's great.
But what I tried to do in this graph is measure the sustainability of that action and how
much impact it can have and on the other side, how fast you can develop things, but
also the comfort of doing it.
Because in the end, if you want to do that for a lot of years, are you doing this under
stress?
Are you doing this having a good day and being able to have a good life aside?
And so open source volunteers won donations being an independent professional, like a
freelancer and getting paid for doing service around open source.
That's really good ways.
And bootstrapping a company, this is what we did.
And I feel, and this is what I want to show also in this presentation, is that it's a
good way.
You have a decent level of comfort.
You can have speed because if you hire people, you do more.
There is the sentence that you can go fast alone, but if you want to go far, you need
to go as a group.
And that's what the company allows, being a group that is funded, that has some money
and can go further together.
And you can see in this presentation also the acceleration that we had over the time
between the beginning and now.
So investors, why?
I want to take a little bit more time on the investing.
So it took us a little time to realize it was not what we wanted.
I came from a company that had raised VC money, and I saw the fact that you can create a momentum.
You can have a money hire very highly skilled people, and you can build things fast.
But in the end, the real thing that you need to think about is the day you take VC money
is who is the real boss and who holds the key of the decision in the future.
And whenever I had discussions with some investors, beyond the fact that they tend to like the
salespeople or the business people more than the tech people, and so that might be a reason
for them to not giving us money.
But for us, the problem was, okay, are we agreeing on where we want to go?
And in the end, investors are in for a return on investment, so making more money with the
money that they put.
And as an entrepreneur, that's not what I was in for.
I was in for the human relationship with the employees, running a project over the long
term, and creating open source.
And when you discuss about open source, and in France at the time, it was also quite simple.
They didn't understand open source, so you had to explain it.
Today, it might be better like, oh, open source, great.
Let's do open source AI in France.
They love it right now, and they tell you, oh, it's great.
But what is their goal with it?
Do they want to sustain that open source AI, for example, or do they just want to make
a play to take a piece in the market and then cash in at some point and close the work?
And so this can create good open source.
And that's fine.
But if you have a goal of being, for example, good to your community, not lie to them.
Not tell them that you're doing open source and not have a hidden agenda about how you're
going to make some money.
That's going to be difficult.
And I felt as a CEO, if I raise money, I would start lying to my customers about what our
real goal is with this open source project.
And being independent allowed us to not be that.
It's much slower.
It was much slower.
But in the end, it's more important to do it like that.
In the end, money is a mean not a goal.
That's really a thing to think about.
And so what was bootstrapping about?
And so from 2003 to 2010, it took seven years to get to one million of revenue.
It took a lot of time.
It took three years almost of myself not getting paid.
I found some other ways.
And then any time we would do a little bit of money through service, we would use the
product.
We would use it for hiring more people and growing the product and making it better.
One of the great things with open source is that you can build on other people's software.
And that's magical.
And so you can really reuse a lot.
And that's actually what the proprietary software companies are doing.
Now you have 90% of proprietary software is actually open source software.
And they keep control of the latest piece, trying to cash in or build some business model
about our data.
But 90% of it is open source.
And that helps us also.
That helps also the open source companies we can build on that.
The support of the community is huge.
The service is a good way to start.
It has problem over the long term to do only service.
But it's a good way because you sell time, you make money.
So you don't take risk with service.
It can be something that doesn't have the level of risk.
Another aspect that allowed us to go from zero to one million is European research money
and French credit and brochures.
In France you have a lot of help about research.
So you can, if you do something innovative, you can bring it to the state and get some
taxes back.
So you will have less cost as a company.
This is, for example, more difficult if you are in association.
You can get subsidies, but you won't get social charges back because you're doing research.
It's going to be more difficult.
And then you have European research projects.
You can group with other companies.
And we had the chance in 2007 to join some other companies in projects and get some funding
through that.
In the end, over the 20 years of X-Wiki, I calculated that we received 10 million euros
of European research grants of projects in France and so on.
And that in the end was our VC.
I mean, getting 10 million from a VC is quite difficult.
It took 20 years to get that, but it allowed us to fund the software.
Another thing that happened in that time is that we went to Romania in 2006.
It was initially through the Google Summer of Code.
And we had a student that was in Romania and we gave some projects.
He candidated and he was really great.
At the time, I didn't have money to pay people a lot.
It was difficult to hire a full-time employer in France.
There was competition about the cost.
Romania was really an emerging country in the tech industry with great scientific skills.
And we hired some of the first people.
They all stayed.
The first three that we hired are still working with X-Wiki today.
And we have 25 people now in Romania.
It was initially a cost-driven decision with the opportunity to have people with skills.
And over the time, it's a fully integrated team that is also believing in open source.
And so we hope that we also had this little effect to bring some open source to Romania
because we're one of the rare companies that is doing open source in the city we are
among Amazon, Microsoft and so on.
And we also have, thanks to Romania, a lot of women in the team.
And it may also happen that there were some couples created at X-Wiki.
So as an entrepreneur, it makes you think about the impact you have.
So that's just a graph of finance.
I'm revealing our finances.
So I'm not going to detail them, but it can show you the split.
The most important data here is that when we started, 0% recurring revenue.
And after six years, 20% of recurring revenue supports.
And in the end, that's the goal.
The goal is increase the support revenue.
It takes time.
You need a great product.
You need to reach a maturity in the product.
But it grows over time.
So it's all about the strategy to make that recurrent revenue grow with the users and
customers of the software, whatever the goals are, whatever the type of recurrent revenue.
So it took a lot of time.
It grew to 20%.
So one thing that I really want people to think about is there is no success in open
source without a good product.
There's a lot of people that think that it's open source business model doesn't work, but
in reality, it's just the product's not competitive.
There is a huge amount of products, including in the open source world.
If you don't do a good product, there's not going to work.
And so you also need to think about the strategy to direct revenue towards the product.
So when you do service, that's part of the problem.
You might diverge from the product roadmap to make a great product.
Because you're going to follow what some customers say instead of following what all
customers need.
So you need to think about that.
And one of the things we learned over the 20 years is that it can be a good idea to condition
the service on taking the support, which allows to give extra funding to the product
and dedicate people to work on the product.
And but there is also some companies, for example, NextCloud, one of the things they do is that
they don't give you service.
They don't sell you service.
They make you pay the product, the support a bit more, and they give you the service.
That's also a strategy that is interesting, that is going to raise the product revenue
and really make the company focus on the product.
And then the service will be used to make the product better.
So we need to think about focusing the revenue towards the roadmap.
That's the case, for example, of the research projects.
Another aspect is the community is super important.
It's your marketing.
It's also your insurance.
Customers will find it reassuring that you have open source.
And it's also your recruitment tool.
You'll find developers.
We have hired so many people that came toward the community.
And it's also very important, the community, to be a good open source citizen.
That's also how you look.
You see if companies are really true about open source.
Are they really working with the community?
Is the community open?
If a software doesn't take patches, doesn't take pull requests,
is not discussing with people about how the software should be,
you could question their motivation to really do open source.
So at X-Wiki, for example, even though we don't have that many contributors
because we're moving fast on our end, and it's not fully natural that people come
and give you code, it doesn't happen like that.
It's a challenge to make people give you code.
It doesn't happen on all software.
So the fact that the community is not huge around a product, from my point of view,
doesn't necessarily say that it's not a good open source community
because it also depends on whether the people want to come.
At X-Wiki, we have a fully open development model,
but we don't get automatically people running.
We're using Apache Software Foundation kind of rules for running the community.
You can find our code, you can comment, discuss in the chat, and so on.
So some companies bring their products to a foundation, that's also an approach.
One thing is the relationship that we customer in open source at the beginning.
I realized that you talk about open source,
oh look, it's great, open source, you're going to be more free, no lock-in, etc.
And you talk to some large companies, the thing is they don't give a shit.
They don't care about this.
They just want the best product at the lowest price possible, efficiency.
So some people do care in the end, but you have to kind of find them in companies
and find the people that can be sponsors of open source.
Today you have OSPOS in very large companies,
even in the European community, in public service.
These are the sponsors, but the majority of buyers of software
are looking for the best software at the lowest price.
And that's why you need to be competitive to show them that you have also the best software.
And there is a difficulty with the marketing of the proprietary products.
There is so much marketing of the proprietary products
that it will cloud the vision of the customers.
They will get stuff for free.
They don't look at the long-term price evolution of software.
We lived it with our competition with conference.
Conference recently changed the prices, but for years customers were buying.
We knew it would happen.
We knew that at some point they would cash in as much as they could.
I mean, they would cash in on the proprietary nature of the software
and the fact that they control people's data.
And a good thing with open source is that open source validates your product.
So you can go and show to customers,
look, we have these users, it shows that the software is good.
And that works very well.
And we also have progress in Europe today
because there is an issue of digital sovereignty,
not something that was not foreseeable with the dominance of American companies,
but it's something that politicians or European organizations or state organizations
took time to take some action on.
Now there is a bit of action in this area.
One thing that I learned also through creating X-Wiki
is looking at Floss and free and liberal open source software as a goal.
Initially, let's create good software, let's create a good company,
let's have a good balance with employee.
But in what I discovered is the goal of open source of free software
is giving us freedom, giving us control about software.
It's all the values that are described by the FSFE that are really interesting.
And that they discovered that and that motivates us even more in building what we do.
In the end, we had to find a balance between all these things.
And these are the values that we promoted internally in the company.
This is what makes our company.
We need to take care of our community, we need to take care of our customers,
we need to do a great product.
And the great product is about the domain in which we are,
the knowledge and privacy, the goal that we have for software.
And we want people to be happy inside the company and we want to do open source.
So these are the values that we promote internally.
And what the challenge for CEO and for the group is to find that balance
between these five items.
And for example, we can see that these are the highest ranked reason why
the people at XWiki decided to join XWiki.
This is recent data, this is not data from the past.
And we can see that being open source is a key reason why people want to be there.
But they also want to be there because they like the product that you're building on and
that we're building.
So one of the key things was building on support revenue.
I mentioned the recurring revenue and this was really important to really
make the support revenue accelerate to be able to gain sustainability.
And that's really the challenge for a company that wants to build open source over the long term.
And so from 2010 to 2015, we moved from 1 million to 2 million revenue.
But most importantly, we grew from 250K to 800K of recurring revenue.
In the end, that's what I look more at, like how much recurring revenue we're making.
Because that is what's funding the company.
We failed at building partnership.
We hope the product could be used to build some other products.
But we found it very difficult to find the deals.
And in the end, we found that we were better at creating a direct relationship with customers.
Also explaining them the open source model and what we were trying to do.
And also explaining them the value of our product.
Relationship with direct customers is key in order to build the value of your software.
We also tried to build the first version of SaaS, we call it XWiki Cloud.
In the end, we focused on the main product and the main product's value.
It also allows some simplification.
And so that's the graph, you can look at it.
Recurrent revenue grew to 35% in the time and that's really great to have that.
It's not only about the percentage of recurrent revenue, it's about the amount that sustains the team.
Because even if you stay with a percentage that's 50,
the extra money that you're getting from the service from the research product,
it becomes bonus when the recurrent revenue is enough.
So if you reach a certain amount of recurrent revenue, the rest becomes a bonus.
At the beginning, that doesn't work like that.
You have close to zero recurrent revenue, so you don't even manage to find the team
to continuously develop the product.
And one thing to keep in mind is that close source competition is tough.
Even if you're doing something innovative, you'll launch something new,
such as Enterprise Wiki when we started or an end-to-end encrypted tool.
At some point, if there is a big market, you're going to have close source competitors
that are raising money that are going to come.
And they might grow faster than you for a while.
They will educate the market, which is interesting for you, but
they will also try to take the market and then lay the cash in.
But you can stick and stay true to your goals and wait.
I always tend to say when you're number three and
number one buys number two, you become number two.
And when you're number two, you're the alternative to the number one.
And all companies want a need alternative for competition.
So I hope, I wish that open source would not be the alternative, but
would be the leader that's not always happening and not always easy.
But being the alternative is also something that helps you grow.
And so after that, we had a challenging period.
And so we were growing progressively, but what happened at some point, we flattened.
And that was because of the competition.
At the beginning, we were working mostly on innovation.
We had customers interested in buying what we were doing through innovation.
But at some point, we flattened.
And we didn't have that innovation thing.
We had stronger competition, SaaS coming in and speeding up.
Deployment people would just buy SaaS.
Companies would be less interested in the open source aspect.
So we basically were flat with 35%.
So what difficulties that we had?
Competiteness, SaaS, competition.
Our custom work was less demanded because there were more products that
were doing things in a standard way.
The fact, we had to educate the market about the fact that open source is not
completely free.
So that's also, you tend to put the priority on this when you're not making enough money.
You tend to think it's just a business model.
It's also the business model, but the main thing that we changed in that period,
one of the things that we stopped trying to think like open source startups,
we came back to think like what we had, what was the value that we had,
and it was about our product.
And what we did in the end, we created Task Force to transform the company,
focus again on the product, making the product better.
And trying to convince again people that the product was good.
And it worked.
And we looked at what was missing, what was not so good in the UI,
and really did effort in it.
And the thing is, when you're doing two million revenue,
you do have money to try to fix problems, and that's nice.
And so we relaunched a competitive offering, and we also changed a few
things in the way we were selling to customers to try to improve their
understanding of open source so that they would give us more recurrent revenue
in order to fund more of the product.
And so one of the things we did is rewarding customers, paying the product.
So for example, we decided to build open source paying applications.
So this is quite unique.
So I don't believe in the open core model where you're doing open source and
proprietary on top of it, because it tends to push you towards doing more
proprietary.
At Xwiki, what we decided to do, we did paying extensions that's similar to
open core in the sense that you have to pay for them.
But the code is open source, we just don't make the build available.
So we have the Xwiki core, completely free, and of course,
completely open source.
And we have extensions, you get them as extension, they say pay for it,
pay for it.
In reality, the code is fully 100% available in GitHub.
If people would want to use them for free, they rebuild them.
The thing is, people don't make the effort.
It's a lot of effort for companies to do that.
And by making a little bit of friction for companies to adopt these
extensions, it motivates them to pay, to give them a reason to pay in companies.
And the bad part of it is we like it to be free completely for individuals.
But this means that you would need to find some other ways to make it happen.
But the most important thing in this strategy is that the code itself is
open source.
That means over the long term, it's owned by everybody, not just by us.
So we cannot be the owners of that code over the long term.
So this is a part where open source is not free and it needs to be explained.
We tend to think that everything needs to be free, but
you cannot pay people if everything is free.
So if you want to build it, you have this difficulty.
In 2016, we launched CripPad.
And it was another experience there because we relaunched
a second product inside the company.
It had some useful aspects.
It recreated innovation in the company.
And it helped us gain other research projects because research projects are
highly linked to innovation.
So we had a second batch of innovation inside the company.
And it also helped for the image of the company.
It made us more known by individuals.
And then, oh, you're doing Xwiki also.
Some people, I mean, I don't know.
In people that know Xwiki and CripPad,
how many people didn't know it was the same company that was doing that?
I don't know.
Who knows Xwiki?
Who knows CripPad?
Anyway, so then 2020 happens, COVID, what happens?
So that's a crisis.
One thing to think about is always be ready as an entrepreneur for a crisis.
It will happen if you stay long enough.
We had subprimes in 2009, 2020 we had COVID.
The thing is for us, we were more ready than a lot of companies because we were
already remote friendly.
We were allowed to do, everybody was allowed to do two days of remote in the company.
We just moved it to do whatever you want, just work.
And everybody worked from home.
We had the tools, everything was already adapted to work with remote tools.
That's one of the magic of open source tools and open source development model.
We had the knowledge, we had the knowledge tools.
And it also gave a boost to CripPad because CripPad actually was used by education.
For example, in Germany, we had credible usage of CripPad over a couple weeks during COVID.
So it also gave a boost.
But as a company, of course, it creates a bit of scare what will happen, will customers
go away, will there be a financial worldwide crisis for years.
In the end, we went through there.
One of the things that COVID showed is a challenge for European digital sovereignty.
Politicians realized that supply chains were a problem and that there were risks there.
And this has tainted towards digital sovereignty and software.
And since a few years, we've seen that there is an interest in this area.
But the most important thing that happened for us is Atlassian changing their business model
and saying that people should move to their cloud.
And they should stop using software in their own companies.
And they closed the smaller offers.
And they decided that in 2020 in November.
So I don't know how they found that COVID was a good time to add some stress to their users.
But they did and their customers didn't like it because we received a lot of mail like saying that,
okay, what is their way to replace Atlassian conference with Xwiki.
And so we spent time on improving our migrators.
And we were not necessarily surprised.
We were surprised the extent of the change that they made and what they did to their customers.
But from our point of view, it was something that would happen at least progressively.
When investor backed companies want to cash in, this is a time for the SMAs and
open source companies to really propose an alternative that is more sustainable over the long term.
Open source is more sustainable for other people than proprietary software is.
So for us, it brings some maturity.
This raised our revenue to 3 million 3 of sales in 2023.
50% growth, I said that at the beginning of the presentation.
And in the end, 1 million 6 of recurring revenue with 30% growth on the recurring revenue.
And that has been huge for the company and for allowing to build more software.
So this is a graph you can see last three years, pretty nice.
So when you look, when you are in 2020.
When everything's flat, you feel a bit depressed.
And that's not going well.
But then three great years behind it.
So it's never given.
You can always turn things around.
Not only because of Atlassian or thanks to Atlassian, but also because we
went one project with Digital Sorbitancy in France and Germany and the software was recognized.
And what about the future?
So everybody talks about AI.
For a knowledge company, it's a real question.
So we need to think about it.
One of the things that AI is doing right now is that it's questioning the aspect of open source again.
We saw a lot of big companies, as I said, not caring about open source.
Politicians not caring about open source.
With AI, it's the first time the president of France said the word open source.
Which we wanted him to do for years in the industry saying that it was important.
And he said it for AI.
Okay, what will it change?
We'll see.
But at least it raised the question of transparency again, of the control of code of data.
And that's something that is positive for the future.
But you also need to get prepared because it changes a lot of things.
The architecture of running AI is complicated.
It's much more harder to run it on premise.
And so you need to find solutions for that.
We're working on AI at Xwiki.
We have an extension.
And we also gained a research project to do some search engine using AI.
And I would like to point out the approach of also NextCloud with ethical AI.
We are completely aligned with that aspect.
You cannot do AI today without thinking about whether it's ethical,
whether it's protecting data or not.
One big aspect that I think is really important for the future is software modularity and integrations.
We believe at Xwiki that the future of open source software is allowing to assemble software together.
And making better reuse.
I said at the beginning of Xwiki that when we started, we reused a lot of open source software.
Well, if we want our software to survive in the open source world,
we also need to make sure that it can be reused more.
And this is why we've launched a new product.
We call it Xwiki Crystal.
And it's going to be a new modular AI that will not only work with Xwiki,
but can work with other Wikis and can be integrated in other tools.
And the other thing is that we're part of the Open Desk project,
which is a funded project in Germany to make an open source suite of collaborative products.
And we're very happy to be part of it.
And the other aspect is doing with CripPad what we did with Xwiki.
I showed the financial of Xwiki, the company that included both Xwiki and CripPad.
But what's really interesting is when you run a second product inside a company,
how does the other product look?
And what's really interesting is that it looks a lot like the Xwiki product at the beginning,
20% only of recurring revenue.
And it's difficult to build that recurring revenue.
So if you love CripPad, we are very happy.
We've been able to double the size of the team, as you can see in the funding in the last two years.
So in 2023, we doubled the size of the team for 2024 too.
But it's only 20% of recurring revenue.
And that means we don't have the sustainability yet.
And if you look, the blue and red part is our recurring revenue, subscriptions to CripPad.fr and donations.
And you can actually help us build sustainable revenue by promoting CripPad and allow us to find more users and customers.
But you can also help it with donations.
Any software needs to reach that sustainability through the recurring revenue.
That's really the challenge for it.
Finally, giving back.
So first we give our software because it's open source.
We give our code as a company.
But we also think that it's important that we give back to the other open source projects we do.
We wish large companies would do that.
Large companies that use open source for free a lot or proprietary software company today that are building on open source should give back something to all the project that they use.
And we decided to create a fosfan of 1% of our recurring revenue to give it back to the projects that we use.
We have three years backlog that we're going to give like almost 30K to the different projects that we use.
We're going to give for example to the Matrix Foundation.
We're going to give to MasterDont and to lots of other tools that we're using.
And we're going to continue to participate to industry organization to help it make known.
The conclusion is that nothing of this would happen without the team itself.
And we have a team of 60 people,
more than 200 people over 20 years that worked on Xwiki.
And that's really the kudos to them because you cannot do that without all the people that worked.
And we, for example, at Xwiki we have more than seven people that worked 15 years.
We have seven people that worked 15 years at Xwiki.
15 that worked 10 years at Xwiki.
And this is not necessarily that easy to achieve that for a group of 60.
We have a difficulty of funding all the time.
If you want to join, we have jobs.
And also nothing would have been possible of what we did without the help of European projects,
French projects, BPI, Europe, NLNet.
If you don't know the NGI program, the funding you can get for
the open source from NLNet, go look at it.
It can help you fund your project.
That's it.
And if you have any questions, I'm welcome.
I'm available or I can take any questions.
Any questions?
No, I guess people are just installing here.
I have a question if nobody has a question.
So I was wondering how was the ride between building a company and having a community,
basically, was there any conflict of what to put in the product, what to not put in
the product, let's hide this away so that people pay for it, let's give it for free.
How was this dynamic in the building of Xwiki?
Yeah, and that's the difficult part is what do you do as a pain module?
What do you do as free?
Well, one of the things, so first is really keep an open community.
We are very important and really keep everything open source even if you have pain stuff so
that people can look at the code and discuss.
For the choices of what the features are, well, we try to direct them as much as possible
to the ones that the bigger companies would need, would need most themselves, not necessarily
the individual or the smaller companies.
Because in the end, it's mostly the bigger companies that have the funding for you or
for an enterprise software.
And it sounds weird that the bigger companies are not paying for it.
I think Matrix has a talk just after and I know that they will talk about the fact that
you have huge deployments of Matrix and with zero money and some smaller deployment that
are giving significant money.
And the larger companies that are massively using open source need to participate to it.
And so directing the specific features that they need, for example, audit logs for compliance
reason is something of big companies.
But for example, LDAP authentication or SSO, it's a bit tougher to not give it because
it's a security feature that's really important to make software more secure.
So for example, that has been a difficulty for us.
So we made active directory paying application, but LDAP configuration is still available
in XWiki as a documentation in the open source documentation.
But if they want the simple configuration with Microsoft Active Directory, they pay
the application and we sold a few of them.
Hello, first of all, very nice talk.
Thank you.
What would you have done differently on the Twiki journey?
What would you have done differently on the Twiki journey?
Oh, that's a good question.
Well, the little strategies to make people understand open source better, for example,
making people pay more for service if they didn't take support, what I've done earlier,
the paying application maybe earlier, not so sure because initially you need to build
community for sure first and you need to build competitiveness.
So it's kind of difficult.
So that part, not do the four products on top of XWiki with partners, but at the same
time they gave us some money.
So maybe do less service sometimes, more product.
So these are the things I would have done.
But it's done differently.
Like basically the playbook of how you can fund the product or try to do it earlier.
And then when we learned it, I had other presentation about the different method we found in prior
for them about how to fund an open source software.
So I gave to Kerl, which also has a great experience about how to fund the work in Kerl.
Any other questions?
Nope, okay.
Thank you, Ludovic.
Thank you.
Thank you.
Thank you.
Thank you.
Thank you.
Thank you.
Thank you.
Thank you.
Thank you.
Thank you.
Thank you.
Thank you.
Thank you.
Thank you.
Thank you.
Thank you.
Thank you.
Thank you.
Thank you.
Thank you.
Thank you.
Thank you.
Thank you.
Thank you.
Thank you.
Thank you.
Thank you.
Thank you.
Thank you.
Thank you.
Thank you.
Thank you.
Thank you.
Thank you.