I think it is important to specify the objective of the coding. The objectives can dramatically change the best practices and what is required from R code.

For example, if you are coding to perform data analysis for a project, then the only requirements are that the code works for your specific data and that it be organised in such a way that you or someone reviewing your code can easily understand what's going on months or even years into the future. That usually implies lots of comments and explanations and so forth.

However, if you are coding to develop an R package, readability becomes less important and efficiency and robustness become far more pressing concerns.

I definitely think that ecological coding and statistics teaching at the undergraduate level leaves a lot to be desired. Of course with the rise of AI, learning coding is becoming easier than ever! The key is not to teach students to blindly memorise complex formating and syntax but to promote understanding of the deeper processes and reasoning behind the code. Once that is achieved, best practice becomes evident, I think.

Christopher Haddon Garrett

Excellent and very important topic! I completely agree that the time has come for greater standardization in how code is written within ecological (and scientific) disciplines, especially considering that R has become a dominant tool for statistical analysis and modeling in ecology.

Although researchers are not programmers in the traditional sense, the reality is that we increasingly rely on code as a core component of the research process. In this context, good coding practices shouldn’t be reserved only for the software industry. Proper code organization (e.g., modularity, clear comments, consistent naming), efficiency (e.g., avoiding disk-intensive operations like using write.csv() inside loops), and reproducibility (through scripted workflows and structured project organization) should become part of standard scientific training.

Introducing basic software engineering principles into undergraduate and master’s curricula—especially through practical, team-based projects—can significantly improve the quality of research, make collaboration easier, and increase the overall efficiency of the scientific community in the long run.

We might not need to implement every industry-level standard, but the basic culture of writing good code should be universal—whether we're developing commercial software or analyzing data for scientific publications.

Yes, I believe that as R becomes more widely used in ecological research, there is a growing need to establish better coding standards. Many researchers, especially those from non-coding backgrounds like ecologists, use R as a tool for statistical analysis without necessarily adhering to best practices in programming. This can result in code that is inefficient, hard to maintain, and difficult to reproduce.

As you mentioned, teaching undergraduate students not just how to use R for analysis but also how to code efficiently and with best practices is crucial. This could include:

1. Writing modular code: Encouraging the use of functions and splitting code into smaller, reusable pieces can improve readability and maintainability.

2. Efficient data handling: Minimizing the use of disk I/O in loops and ensuring data is handled in memory effectively can save time and resources.

3. Following established coding conventions: Adopting conventions for variable naming, commenting, and organizing code can make it easier for others to follow and reproduce results.

4. Prioritizing reproducibility: Encouraging the use of version control (e.g., Git), proper documentation, and frameworks like RMarkdown for creating reproducible reports can improve the credibility of research.

By introducing these practices early in students’ training, we can help them not only use R for analysis but also build good coding habits that will serve them well as their research becomes more complex.

The growing use of R in ecological research suggests that researchers will need to adopt more rigorous coding standards to improve the reproducibility and efficiency of their work. There is an opportunity to foster a culture where coding is seen as an integral part of ecological research, similar to how laboratory methods are treated.