Photo by Tim Mossholder from Pexels
Since I started to work at TotalCross I realized that selling a framework to developers would be one of the most difficult and exciting things I had ever done in my professional life. We had a validated product with clients worldwide (including one which has used our framework to build a bomb detection App for the US Government) and we had an experienced tech team to keep improving our product.
So, we decided to create an outbound marketing team because it seemed to be the best strategy to increase our sales and to reach the desirable exponential growth. We followed the “Predictable Revenue” book method, building a selling machine with 4 steps: lead research, prospecting, lead validation and closing. Six months later we could tell our process reached a high level of performance, except for one step: closing.
We used to have dozens of meetings with developers really excited with our technology. They could see how it would help them in their daily work, delivering Apps for all operating systems in a fast and easy way. But They definitely were not willing to buy it. We thought we were not reaching the ICP (Ideal Client Profile) and we started validating all kinds of markets: IT consulting, consumer goods, insurance, agriculture, food and beverages industry, etc. We never found our ICP. But was finding the ICP the real problem?
Once I went to an event and heard someone saying:
“If something is pretty hard to do, maybe you are doing it wrong”
That made sense to me. Shortly after we started researching, we held meetings and workshops to figure out what we were doing wrong.
It was a tough time, but rewarding as well. If on one side we were living an uncertain period with anxiety and lots of unanswered questions, on the flip side we had many reasons to enjoy it. Once we were not only looking for a new business model, but also creating a new selling point, we needed to bring new minds. So we started to make new connexions, going to events and meeting lots of special people.
In Brazil we have met people from the embedded systems world, like Toradex global team and Raul Rosseto from Foundries, who showed us how TotalCross could fit into this market.
At FOSDEM (the best free and open source software conference in Europe) we have met guys from the open source world , like Gabriel Engel (Rocket.Chat CEO), Jon "maddog" Hall (Chairman of the Board of Linux Professional Institute) and Cédrid Thomas (OW2 CEO), who taught us how open source business works.
All of them helped us a lot by bringing new perspectives from the software development tools market. They also led us to three main conclusions:
First, you must only charge what people are willing to pay and developers do not want to pay usage fees (i.e. developer seat subscription). If you charge developers to use your framework, they will remain your clients only until an open source alternative appears.
Second, if you are not a huge corporation it is almost impossible to maintain and improve a complex technology quickly by yourself. In our case, TotalCross has more than 700,000 lines of code, most of that in C programming language.
Third, defining a product roadmap is crucial and it is really hard to get right. When you have thousands of people using and contributing to your product, it becomes much easier.
When we started to talk about becoming open source it seemed a daring and huge step. But taking these three points under consideration, the decision was not as hard as we thought, we started seeing it as common sense and the right way to succeed. Our product adds value to developers, being open source brings their willingness to contribute to its improvement and to build together the best roadmap possible.
Since March 2020 TotalCross is a free and open source SDK (Software Development Kit). People from all over the world help us with our source code, product roadmap, revenue model and so on.
“If you want to go fast, go alone. If you want to go far, go together.”
I know it is cliché, but it remains truer than ever.
Top comments (6)
That's great, every new FOSS project is good news!, how was the process to choose a license?, which one did you adopt and why?, did the fact that is an SDK and not targeted to end-users influenced the choice?
That's a very good question, Roberto. All variables have influenced the choice.
We've studied a lot this subject and we talk a lot with specialists.
First of all you need to set your target. What is your target to become open source?
In our case, we had three main p targets:
So, we decided to choose LGPL V2.1 then users can build their solutions on TotalCross without worry about developer seat license, runtime fees, open their source code and so on. Because of that they will be willing to contribute, but they are not allowed to copy and modify without returning the source code to us. It is a way to protect our business due to big companies need to pay us if they want to build their own solution based on TotalCross.
given that you provide a tool to build projects of others, does the GPL force users that build their projects to also be GPL? or it just apply to code directly related to the SDK and not the work made with the SDK?
It just apply to code directly related to the SDK. If you only use the SDK to build your project, you don't need to open your source code. That's why we have chosen LGPL, because GPL requires to open the source code for both situations.
Congratulations! My dream has always been to work with IoT devices...I started my career there and destroyed a great opportunity because I was young, reckless, and incredibly stupid. It's awesome that you are chasing your dreams and proving that OSS can truly be a product. I feel like I may have missed my chance for good, but it's nice to see others getting their chance and running with it. Kudos!
Thanks!
All of us were stupid when young. lol
But we always have time to change the game.
Be free to reach us for a talk or if you are looking for an opportunity!