Interview with Karin Broeking
Matsters Thesis: Open Source Communities

Karin Broeking 

Project Manager at initOS. 

Present yourself, tell us more about you!
My name is Karin and I recently finished my masters degree in computer science. Since I started my master studies I have worked with initOS where I started as a developer and became a project manager over time.

Because I started as a developer, I mainly got to know Odoo from the technical perspective in the beginning. The first Odoo project I managed had a lot of customized code, so that I only recently started to get to know Odoo’s standard functionality and modules. My current goal is to deepen my functional knowledge about Odoo and become a functional expert. Furthermore, I plan to become sort of a community manager in our company to coordinate and improve our contributions and engagement in the OCA.

How long have you been working with the OCA and what do you do here?

My first touching point with the OCA were the OCA Days in 2022 and the interviews I conducted for the master thesis. Before that I have worked with Odoo in my company, but never directly contributed or engaged with the OCA. However, I have contributed indirectly to the OCA with my work team by working on projects in which we push code to the OCA. As mentioned above, I want to increase my involvement with the OCA by becoming an OCA community manager in our company and in the future want to contribute by adding translations and doing functional tests.

During the past months, you interviewed several people from the OCA community. Why and what did you expect?

The Interviews were conducted in the scope of my master thesis. The goal of the master thesis was to create a knowledge acquisition reference model that helps specific types of stakeholders to identify, structurally obtain and evaluate relevant knowledge about an open source ecosystem with the example of Odoo. To get to the end result of the knowledge acquisition reference model that is called KARM I conducted several interview phases in which I interviewed mainly experts with affiliations to the OCA. In the course of my master thesis there were three rounds of interviews with different goals and expectations.


Interview phase 1:

This interview phase had the two objectives to confirm the identified problem of unstructured and different knowledge acquisition and to prepare for the identification of relevant subject areas of open source ecosystems. To gather this information, the interviewees were presented with two problems in the Odoo context and should tell me how they would solve the problem and which information sources they would use.

 The interviewees of this phase were all interviewed at the OCA Days 2022. The first interviewee was a project manager who contributes to the OCA together with his team at work. His approach to acquire knowledge to solve problems focuses on preventing to reinvent the wheel (doing work that has already been done by someone else). He knows the communication channels of the OCA and would be willing to use them to gather knowledge from the community. If he does not know the communication channels of an open source community, he uses the website as a first touching point to get information.


The second interviewee was a contributor and developer who has been part of the OCA since it was founded. He mainly acquires knowledge by using a search engine. If he cannot find an adequate result, he solves the problem himself by writing or modifying source code. He sees potential for improvement regarding the accessibility of information in the OCA. Because there is a big amount of modules and repositories, matching requirements to existing solutions is difficult in his view.

The last interviewee was a functional expert and project manager who had no former affiliation with the OCA and attended the event to get to know and evaluate the community. He was still getting acquainted with the OCA's infrastructure and organizational structure. When facing a specific problem, he would first use a search engine to find a solution and has not considered asking for help in the community. One reason for that is that it was not clear to him where to start and how high the threshold for using the community's resources is when he had not contributed yet. To make the first interactions with the community easier, he would appreciate an onboarding or induction process that could also be implemented with a chatbot in his opinion.

Interview phase 2:

This interview phase had the purpose to evaluate the identified relevant subject areas with experts from open source communities and to get more information on different stakeholder perspectives and priorities. Therefore, I presented a graph with relevant subject areas of open source ecosystems to the experts and evaluated them and also asked which of the aspects are important to the experts personally.

The first interviewee of phase 2 was a contributor in the OCA who is also engaged in several other open source communities. The knowledge acquisition process of this interviewee can be divided into two phases. There are aspects and information sources that she looks into prior to contributing or while learning about the open source ecosystem. This is her active knowledge acquisition for which she uses the website, coding guidelines, the documentation, the corporate influence and in some open source community's chat clients. The other aspects (such as community structure, project transparency, project culture, licensing etc.) are not actively looked at by her. However, even if she does not actively acquire knowledge about these aspects from the start, these aspects are not irrelevant for her. If she notices that one or more of the aspects contradicts her principles and preferences her motivation to contribute to the project would decrease or stop. The more she is involved in an open source community, the greater is her desire to be able to influence it.

The second and third interviewee were persons that have no technical background and no affiliations with the OCA or Odoo. They have the perspective of a user of open source software that has to choose the best suitable solution for specific requirements. Consequently, they have not been directly engaged in an open source ecosystem or with its community and look at open source software as a software product. They both coordinate the selection and implementation of the open source software solutions for schools in their respective regions. When they choose the best suitable software for schools, their focus is on their functional requirements. To be considered as a software solution, an open source software has to be stable. If the functional requirements are met, the software is tested at one or several specific schools and there is an evaluation afterwards. There is no predefined process or set of criteria when a software is selected or the criteria are set and checked by a consulting open source service firm. Often, the chosen software is a single vendor commercial open source project which makes the support dependent on the single vendor company. Additionally, it can be difficult to find companies and experts with the needed expertise to support the implementation of a chosen software solution.

The fourth interviewee of phase 2 is a functional expert and an OCA board member. In the beginning of their work with the OCA, she experienced that it was hard to find her way as a functional expert in a community that mainly has developers as members. The first information source she used were other community members, GitHub and the contributor mailing list. She very actively tried to acquire as much knowledge as possible. The community structure was explained to her by other community members. The first criteria for her and her colleagues for choosing to offer services for Odoo was its functionality and the second criteria was the openness of the community. She also actively acquired knowledge about the OCA governance. The handling of disputes in the OCA could be improved in her view. The project transparency and openness, especially the ability to influence the community, are very important criteria for her.

In summary, the interview phase 2 evaluated the identified relevant subject areas of open source ecosystems and the feedback from the interviewees could be used to improve the structure. Besides that, insights on the knowledge acquisition process, criteria and perspectives of different stakeholders could be derived from the interviews

Interview phase 3:

In this interview phase, KARM was evaluated by experts. Because of the large scope of KARM, it was only possible to evaluate parts of KARM with the different experts. The experts used KARM in the context of Odoo and the OCA and afterwards, the feedback of the experts was used to improve the model KARM. The interviewees in this phase were all contributors of the OCA (developers and one functional expert and translator).

Did you interview people in open source community outside the OCA?
Because my master thesis was conducted with the example of Odoo, almost all my interviewees were involved in the OCA. Nevertheless, there were some exceptions that were not affiliated with the OCA or who were also involved in other open source communities. I interviewed two persons that are not involved in the OCA, but are also not directly involved in other open source communities as they only choose open source ecosystems for schools to use. Besides that, I interviewed one person on the last OCA Days that was not involved in the OCA at this time but was interested in getting to know the community. The person was otherwise involved in a political party and a community for product management (which are communities, but not open source communities per definition). Additionally, one person I interviewed was involved in several other communities besides the OCA and even coordinator for one.

Did you find the same conclusions in OCA and other open source communities?
Because few of my interviewees could draw a comparison I did not have many results regarding this question. However, I’ve learned that the OCA’s governance and culture is very transparent compared to some other open source communities and the principle of consensus is well liked by one interviewee. Regarding the relevant subject areas of open source ecosystems, the results of the interviews indicated that they are transferable to other open source ecosystems than Odoo.

What did you learn that surprised you and that you want to share?
I was surprised that a lot of interviewees felt that the barrier in the OCA to ask questions in public channels like the mailing list and Discord is too high. This indicates that the communication channels are probably used only by some people from the community. 

The interviewees tried to ask people directly or solve the problem by themselves instead. In some cases and for very simple problems/questions that might be better, but it can also be assumed that often other persons from the community would profit from more open questioning and discussion. 

Furthermore, I suppose that if people are asked directly and not publicly some experts from the community repeatedly have to answer the same questions. Because it became clear in the interviews that the barrier to ask questions is much lower in channels/forums with a smaller number of participants (that ideally speak the same native tongue), one solution for more communication could be smaller, specified and maybe even localized communication channels. However, the advantages and disadvantages of such a solution were not evaluated in my master thesis and could be further investigated.
Another thing that surprised me was that some interviewees (with the user perspective) had few or nearly no criteria besides functionality and no structured process when choosing an open source ecosystem.

Did you find new processes or ideas that could help the OCA to grow, get more members and keep the new contributors on the long term?
One pain point for one interviewee is the information accessibility in GitHub, especially with the focus on business concepts and processes. He proposed to “funnel down” the search for functional aspects, eg. in stages that get more detailed. He also would like to have documentation for business requirements and business processes.
The other point that could be mentioned here is the access and first touching point with the community. Most of the interviewees had colleagues already involved in the OCA which made the first contacts easier. For the easier access to the community for those who don’t know anybody from the community, one interviewee proposes a first touching point like a person or a chatbot that directs people to the expert they want to talk to. He states that even if there is sufficient information on the website, people learn in different ways and some need to be able to ask questions and discuss with people. 
Therefore, he is missing some sort of on boarding or induction process. Such a process could include speaking to specific persons who explain frequently asked questions and show the infrastructure of the OCA.

What's the thing you prefer about the OCA community?
I don’t have the comparison to other open source communities, but with all the knowledge of open source ecosystems that I gathered in the context of my master thesis I am able to say that I really like the way in which the OCA is governed and the community culture. The OCA (and its processes and decisions) are really transparent and open and offer the community members a lot of possibilities to engage and make a difference. Furthermore, although I am an introvert and it is not that easy for me to get in contact with people I don’t know, I found a lot of supportive people in the OCA that volunteered to give me interviewees and help with my master thesis, which I really appreciate.

Cover Photo by CoWomen on Unsplash


Interview with Karin Broeking
MoaHub, Rebecca Gellatly 30 September, 2023
Share this post
Archive
Sign in to leave a comment
OCA Days 2023 - Functional Track
What you need to know...