Meet Douglas Bates - How a UW-Madison professor helped shape R and continues to advocate for open-source technology

New programming languages are constantly being developed, helping aid and furthering research all around the world. One of the more prominent coding languages R was developed by one of University of Wisconsin-Madison’s own. Douglas Bates is an emeritus professor in the Department of Statistics, his research mainly focuses on statistical computing. Twelve years after retiring, he continues to work with graduate students and faculty.

“It’s a superb situation where you enjoy what you’re doing so much that you just keep doing it,” Bates said.

After starting his work in statistics, Bates soon learned about open source in the 1980s – which at the time was considered free software. Emacs text editor — originally from the Free Software Foundation and now produced the GNU operating system — was one of the first open-source softwares he had become familiar with.

“It was quite radical because the open source version was better than the commercial version, and this just completely blew me away,” Bates said.

Bates later became involved in the development of the S programming language during his time working with Bell Laboratories. In the mid 1990’s, the language eventually was made into an open-source package, the R language. Bates was on the core development team for 20 years, contributing to it until March 2024. He also worked on many Julia packages.

“The R project was what got a lot of statisticians and then people in areas like genomics and the life sciences because of the bioconductor project and many, many other areas like that,” Bates said.

One of the main challenges to developing an open source project is maintaining it. Open source may not be the most lucrative industry, but can be exceptionally rewarding given its adeptness at adapting.

“It’s fine to have a person or a small group of people say, ‘Hey, wouldn’t it be neat if we built it?’ and then they build it. But then it’s handling all of the notation, the questions, the user support, etc.,” Bates said. “It’s still difficult to carry over the long term. How do you keep developing a programming language? Eventually, you want to have some way of making a living.”

Though maintaining a project for long term sustainability can be a prominent challenge in the open source realm, the low barriers to entry that open source has to offer can be one of the most valuable aspects of open source, according to Bates. He follows it with the only true barrier of it is a person’s ability and motivation.

With open source software, it may take some motivation to read the code and discover bugs if they are there, but it can be more rewarding as the contributors and maintainers are able to see their results along the way.

Using the example of monitoring the code and security behind it, Bates goes on to mention how logically there is no reason why someone would prefer closed source to open source.

“It may be difficult to read the code, but in theory, you can do it. You can trace through and see what the calculations are going on,” Bates said. “Whereas you don’t get that in closed source software. open source, everyone can see it. If there was something dodgy, someone would see it.”

When starting an open source project, Bates believes it’s imperative to be idealistic. The time will come when you have to figure out how it will be supported long term and how to support the project.

“There’s so many open source alternatives that we should seriously consider at the university level,” Bates said. “Especially a place like the University of Wisconsin-Madison, which is dedicated to openness.”

UW-Madison has been a great asset for how Bates has been able to continue his work with open source for so long due to their stance on open inquiry.

“[UW-Madison] were willing to support me and credit my research, production, my work on R and later Julia,” Bates said. “I was kind of fortunate in that it was close enough to what would be regarded as research in the academic area that it could kind of slide back and forth.”

The papers he wrote on open source software continue to be beneficial to those in the community, especially with its ability to provide documentation with explanations on how to use the software. Many people were able to cite his own work in their own publications.

“Sometimes you can get caught in an area where you’re solving technical problems, but they don’t look like research,” Bates said. “I was very fortunate that I didn’t peep myself into a corner with that.”

Since the projects were open source and rooted in the spirit of open inquiry, there was no commercialization of it. Anyone with a GitHub account is able to access these open source code repositories, allowing for continuous research opportunities with the code.

With Bates’s contributions along with other practitioners, UW-Madison is able to continue to be home to a prominent open source community. Continuing to support its contributors and maintainers that produce these essential projects can be vital to continuing research and exploration in the community.