Giving benefit to people using what you build
We were recently joined by Hadley Wickham, Chief Scientist at Posit PBC. Listen in to hear our chat about building tools (like the tidyverse) to make data science easier, faster, and more fun.
36:57 – While I’m bought into developing open source packages to help deliver better processes, any advice to those of us doing that development in getting their company bought in?
You have to give some benefit to the people using (what you’re building)
You’ve got to either remove pain or add pleasure in some way because if you can’t do that and you’re not someone’s direct supervisor, it’s hard to get people to change.
The way I think about the tidyverse is, how do we give people some sort of quick wins so they can be motivated to do the things that are slower where they’re gonna have to learn some new ideas or some new tools. You kind of build up some equity with that person.
They build trust that you’ve helped them in the past and now they’re willing to invest a little bit more time before they see the payoff. But in the early days, it’s all about delivering payoffs as quickly as possible.
And I think if you’re doing, like, you know “my company’s first R package” – the easy pain points are: make themes for your company corporate style guide, make a ggplot2 theme, make an R Markdown, a Quarto theme. Make a Shiny theme that people can just use to get, you know, something that’s reasonably close to whatever your corporate style guide dictates.
That just feels like an easy win for people because it makes them look good inside the corporation and because you’ve put in all the hard work, it’s like three seconds for them to type the right function name to get the right theme.
I think the other bit is making it easier to get access to data. Set up some wrappers around DBI connections to the most important data sources. Provide some conventions around authentication so that stuff just works so that they’re not struggling with “What packages do I need to install? What’s the password? Where’s the path I need?” Just give them some, like, a list of the top ten most common data sources and people will love you by and large.
Follow-up question: Once you identify the things that you think would be useful for people – do you have a philosophy or a way in which you approach putting things together?
When you’re in an environment of scarcity when you’ve only got so much time that you can take out of your everyday job to invest in writing a package, it’s really tough to balance. Like, how do I add new stuff versus making sure the old stuff continues to work?
I think, again, some of it’s about building up trust. So, give people some wins so that when you inevitably break stuff, you’ve got some kind of cushion so people aren’t going to be really angry with you right away. They’re gonna be like, ok, well there’s a little bit of suffering now, but this person saved me so much time.
But yeah, it’s really hard. And particularly as you’re starting out, like, you’re going to make mistakes. That’s inevitable.
You’re going to do things that when you look back a year later, you’re like, why on earth did I do it that way? You’ll want to rip out the whole thing and ride it from scratch. And I think that if it feels horrible, you have to remember, that’s great. It means you’ve grown immensely as a programmer.
Certainly if you have my kind of mindset, you have to resist the temptation to rip things out and redo them as much as possible and just focus on making the next generation better rather than breaking what stuff people already have.
So I don’t have any great answers here, but I think you just have to think about those tensions of “how do I keep my forward velocity up while getting better as a programmer and evolving over time, but also thinking about how do you make the things you did a long time ago better?”
Featured in this episode
Hadley is Chief Scientist at Posit PBC, winner of the 2019 COPSS award, and a member of the R Foundation. He builds tools (both computational and cognitive) to make data science easier, faster, and more fun. His work includes packages for data science (like the tidyverse, which includes ggplot2, dplyr, and tidyr)and principled software development (e.g. roxygen2, testthat, and pkgdown). He is also a writer, educator, and speaker promoting the use of R for data science.