A few words to introduce yourself?
Hello, my name is Sarah Bauduin, and I am a research officer specializing in large carnivores. I work at the OFB, the French Office for Biodiversity. My work focuses particularly on research projects concerning conservation and conflict management, mainly involving wolves and lynxes in France.
The ERC-Lynx project, ERC standing for Avoid, Reduce, and Compensate - has the main goal of developing a decision-support tool to predict lynx viability based on different developments, such as road networks, landscapes, or even the management of different populations.
Why did you choose R Shiny?
Since I have knowledge of R. I first built the entire lynx simulation model in R to predict lynx viability. To create an interface, the easiest way for me was to use an R package, so R Shiny was a natural choice. Since I didn’t have other tool development knowledge for me it was easier to learn a new package rather than learning a completely different software to create a tool.
Why did you reach out to Data Champ’?
I reached out to you because, in our team, we had the skills for model development and and interface development. But the step we were missing was really the deployment of the online tool and user access. We had tried several methods to make a tool available to the public, and online deployment seemed like the most optimal solution. But this was precisely where we lacked knowledge: deploying the tool online, ensuring different users could access it, and also everything related to computing servers to run the large simulation model of this tool.
Did you consider other solutions?
Yes, we initially considered creating a local tool, like a software application that everyone could install on their computer. However, given the computation time which is really very long - about two days of calculations on my computer - we thought it wasn’t feasible to ask a user to leave their computer running for two days waiting for results. So, we ruled out the idea of a local software solution. We then thought about online deployment, with the bulk of the model running on online computing servers. We consulted our IT department, but they didn’t have the structure for it, they didn’t have powerful enough servers for this kind of calculation. That’s why we switched to an external service provider.
We contacted different providers, and Data Champ offered the best proposal - both in terms of service quality and your responsiveness when I reached out, and also in terms of pricing which was really the best by far among all providers.
How was the collaboration from your point of view?
For me, the service went really well. Your responsiveness was great - every time I wrote to you, I received very quick replies. The work we requested was done really well, and everything works perfectly. We are really happy. I also really appreciated the feedback report you made me, explaining all the steps you took and the work you accomplished. Lastly, I really lifed the extra tips you gave me at the end about how I could have done things differently. I know there are some small things you gave me that I’m going to try to apply in future projects because I know it will really be a time saver and more efficient for future projects. I really appreciated this little extra that you took the time to put in writing and to explain all these little things that weren’t actually part of the requested work. It was just a little extra, and it was really appreciated.
What is the future of the ERC Lynx tool?
For this ERC-Lynx project, I would like to develop a second version of the tool, both in terms of content and design. From a content perspective, I’d like to improve the simulation model. There are some simulation processes we’d like to improve, to better reproduce lynx movement particularly the dispersion process. From a design perspective, I’ve realized over time, and through our exchanges, that the tool’s interface might not have been the most intuitive for an external user. So, I’m thinkingof doing an update to make the layout more fluid and intuitive for the user.
Comments