Why Posit Means Growth for the R Community


Whether you heard it in person at rstudio::conf(2022) or caught the news through the internet grapevine, RStudio PBC is rebranding itself to – Posit PBC

As with most changes, this has come with mixed feelings from the R community. Most are excited to expand RStudio products into a language agnostic data science ecosystem. And yet, some fear R is losing out to Python. We think both feelings are justifiable, but as the leading R Shiny consultants and one of the largest RStudio Certified Partners, we don’t fear this change. In fact, we’re excited about it. 

Many of us discovered the power of R through RStudio products like their IDE and open source packages. So if “Posit” doesn’t have ‘R’ in the name anymore – does that mean the biggest proponent of R development is leaving us in the dust?

As a company that built itself around R and Shiny, Appsilon doesn’t see it that way. We see this as an opportunity for the R community to expand, invite other communities through RStudio products, and adopt new technologies into R for data analytics. 

“Posit translates to more accessible, open source and professional software to better serve data science teams.” – Filip Stachura, Appsilon Co-Founder and CEO

And as we heard at the conference, there are big names in RStudio that aren’t giving up their love for R any time soon. RStudio isn’t leaving R behind. And trust us when we say, there are a lot of companies out there exploring R and Shiny to handle their data needs. Long live R!

TOC:


RStudio’s (Posit) rebrand and the future of R programming

However we may feel, based on a quick scroll of LinkedIn, you might be thinking,  “R programming is doomed” or “the Python vs R cold war is still raging.” Now, that’s not to say there isn’t love from the R community either. There are plenty of folks sharing their excitement about new products like Shiny for Python or reaffirming our beliefs that R is the tool to get stuff done. So what gives? Why are folks worried that R will be unseated as the statistical language? 

One language to rule them all

Well, obviously there’s no universal language. No one is using R to develop AAA video games (although the future might surprise us) and I doubt you have a pure Python mobile app on your phone. In short, each language serves a different purpose. As the old but still relevant joke goes, “All programming languages are bad. They’re just bad in their own, unique way.” 

Python and R are both beginner-friendly languages that are great for handling large data and data analysis. But both have their own unique shortcomings and focuses. Python seems to control machine learning and production, whereas R dominates statistical analyses and rapid proof of concept output. Both are great at different things for data science, but neither will ever fully replace the other.

Let’s be realistic, RStudio is just making its tools and resources inclusive for other data science users – not diving in headfirst to overhaul the Python ecosystem. It’s extremely doubtful that the RStudio IDE will be a ‘side’ project for RStudio (Posit) or that open-source R development will be dropped in favor of Python projects. It’s more likely that Python will gain some valuable teamwork tools, commercial support, and improved open-source tooling for general data science.

Graph first published on the RStudio blog, “Why RStudio Supports Python for Data Science

R for the people

The R community is one defined by its open-source and inclusive nature. R users are a diverse group of individuals coming from every field of study you can imagine. Be it Ecologists or Web Developers, Business Analysts or Farmers – R is the people’s language for data science.

R is successful because it’s so easy to get started with. It has an awe-inspiring ecosystem of open-source packages that were developed by the community with some big contributions coming from the team at RStudio. Free publication options for R/Shiny apps from RStudio allow the average Jane/Joe to share their work. And there is no cleaner, R-friendly IDE than RStudio out on the market. 

Graph first published on the RStudio blog, “Why RStudio Supports Python for Data Science

Python for everything else

Python, as Joe Cheng harkened at the rstudio::conf, is the 2nd best language for everything. And it is true that Python is the more popular programming language. It’s easier to master being object-oriented with easy-to-follow syntax. But if it were the best tool for you or for us, we wouldn’t be talking about R right now.

Python is great for solving and creating some pretty cool things, but at its core, it just isn’t the statistical powerhouse that R programming is. No matter how Posit contributes to Python in regards to package development or Shiny for Python – Python just won’t be able to compete with R for statistical computing and data visualization.

Positioning RStudio (Posit) for data science

A few years ago, RStudio announced support for Python in their RStudio products. They did this with clear intent. Their mission from the start was to “enhance production and consumption of knowledge by everyone, regardless of economic means.” From moving to a PBC to the inclusion of Python, BI extensions, and other open source tools – RStudio’s history was always leading to this point – whether or not they or we in the community realized it. They were just making sure R users had what they needed to achieve their data science needs and creating accessible tooling for everyone. And yes, R users do use Python! You can explore the RStudio survey data here.

RStudio’s conclusions:

  • Reject the myth that users must choose between R or Python
  • Embrace Python because half of our community uses it in addition to R
  • Embracing Python means RStudio products should support it too

Posit’s RStudio IDE and R programming

Now, for those of you who are new to the R community, we need to set a little background and history of R and RStudio. Because today, RStudio is synonymous with R programming, but that wasn’t always the case.

It took almost two decades from the inception of R to get the first public beta version of our favorite R IDE – RStudio. We saw the first public release way back in 2011 with v0.92, sometime around the Great Pumpkin. The company behind the IDE was RStudio Inc and it had a mission to create free and open-source software for data science, scientific research, and technical communication. And much to our benefit they did so by focusing on R.

RStudio was one of the first R-specific IDEs that was easy to use and provided a better way for the average user to write code and build programs – without having to access everything through your CLI. 

The RStudio IDE is hands down the most popular IDE for R users. Walk into any university lab – be it quantitative ecology, functional genomics, or financial modeling – and the RStudio IDE is what you’ll see on that screen. Sure, you can code in R with the widely popular Visual Studio using an R extension, but it feels like an afterthought; because it is. RStudio was, is, and always will be the preferred R programming language IDE.

RStudio knows this and they’re not going to give up the gold standard of R programming IDEs to try and compete with PyCharm or VS Code. But what we might see is a push by RStudio to include better extensions and interoperability with a modular platform. Which will benefit everyone!

What Posit means to R Users

We reached out to a few members of the R community to see what they thought of the recent announcement. 

Mouna Belaid, Business Intelligence Consultant at Prime Analytics, Co-Founder at Tunis R User Group and PyLadies Tunis, personal website

How did you react to the news of RStudio rebranding to Posit?

It was unexpected to me. I was happy to know about this consistent rebranding. I believe that it will increase the value of the products delivered by Posit (formerly RStudio). [Posit will gain] more visibility to their capabilities in supporting other open source tools.  

Do you think the rebranding will affect your day-to-day work or client services?

This will be an opportunity to expand beyond R. That will help me to communicate better my findings while working with both R and Python in the same project. It might lead me to develop a new way of thinking to set new strategic goals and objectives if Posit develops more services in support of other open source tools

Veerle van Leemput, Managing Director & Head of Data Science at Analytic Health

How did you react to the news of RStudio rebranding to Posit?

It didn’t come as a surprise. Today RStudio is much more than the R IDE we all know it for. For a while now, RStudio has been offering support for other languages besides R and the name “RStudio” was limiting the company from truly growing beyond an R-only solution.

Some fear it’s the end of R. People feel the biggest supporter of the R language, RStudio, is moving away from R to focus on other things. More specifically: Python. And that feeling of abandonment grew with the announcement that Shiny also will become available for Python. For many, it feels like Python won the game.

But in my opinion, we need to stop making it a competition. Posit expanding and diversifying into, amongst others, Python isn’t a bad thing for R. If executed well this even can be an accelerator– especially within companies that use enterprise software solutions. If a software solution is language-agnostic we can choose the right language for the right job. Sometimes that means language B in favor of language A, but also the other way around. The competition between languages A and B doesn’t need to be there anymore. That’s something R can profit from as well. Because of its strengths in data manipulation, statistics, and visualization, R has a chance to be chosen for the job that it wouldn’t be considered for before.

Do you think the rebranding will affect your day-to-day work or client services?

Not immediately. We’ve been using the professional products of RStudio/Posit for multiple years now and the rebranding isn’t going to change what we do overnight. Working with R for many years, we’ve built our team around R as well. Most of our team members are R and/or Shiny developers. The fact that a language-agnostic ecosystem makes it easier for us to choose Python for certain jobs, doesn’t mean we start using Python tomorrow. It takes time. But making use of language-agnostic software does open doors for the future

It allows us to think about what we can do better in another language. And if that leads to us wanting to use another language, our team can adjust: either by investing time to acquire new skills or by welcoming new team members. Again- it takes time. I think that’s the case for a lot of companies. I’m sure that Posit knows that too, and I’m sure Posit will be improving its services and products to accelerate that transition. So who knows where we are in a couple of years!

Have you tried Shiny for Python?

Yes, I did have a go with the example. If you know R Shiny well, the syntax for PyShiny looks very similar- it gives a nice and familiar Shiny feel. However, they’ve chosen different names for certain functionality in Python. For example, shiny::reactive() is equivalent to reactive.Calc, and shiny::observe() is equivalent to reactive.Effect. 

When you’re coming from R this might be confusing, but on the other hand, the naming does make sense. I think that’s one advantage of bringing Shiny to Python after it being present in R for so long: you have the ability to change things that you can’t easily change anymore in R. That might be for the better! Of course, PyShiny is still in Alpha and far from complete, but I’m very curious to see how it develops over the coming years.

What Posit means to Appsilon

Appsilon develops some of the most advanced R Shiny applications. We’ve worked with NGOs, Fortune 500s, and top research institutions across the world. And we’ve helped them succeed using RStudio (Posit) products. We’ve watched how Shiny has progressed from a fledgling framework into a powerhouse for interactive web apps and data analysis.

With the launch of Shiny for Python, we’re excited at the prospect of a new community joining the Shiny world. We’re also glad to see RStudio firming up its inclusion of other languages. RStudio was well on its way to becoming a “Single Home for R and Python” so this step was a natural progression. 

Although we are experts in R and Shiny, as engineers, we value using the right tool for the job. RStudio’s transition makes it that much easier for us, and other consultants to help provide the best services possible for our clients.

After the conference, we sat down with Appsilon Co-Founders Paweł Przytuła, VP of Engineering, and Marek Rogala, CTO, to discuss RStudio’s big announcements. Here’s what they had to say:

How did you react to the news of RStudio rebranding to Posit?

Paweł – “It’s the right move. There’s no better way to underline your openness to new communities and technologies for data science than by changing the name. New name, same DNA. For some time we will have to use both names RStudio/Posit to avoid misunderstandings. [And if] customers will ask us about the future of RStudio, we will have only good news for them.”

But the rebrand wasn’t the only big news. Like Veerle, we were eager to explore the Shiny for Python example and play around with the new tool. Marek, who’s been exploring the alpha version of Python for Shiny (sometimes referred to as PyShiny) with our lab, was pleasantly surprised with the alpha version:

Have you tried Shiny for Python?

Marek – “It runs smoothly for an alpha version. The no-server feature is great and impressive, and works fast so it has a lot of potential in scaling apps – doing computation on the user’s machines.”

Marek sees it as a “new big thing in Python” and “bringing over something that worked great in R to Python.”

But Shiny for Python is still in alpha so it’s got some flaws to work out. Marek pointed out a few of these: “It lacks a lot of features. It’s harder to build an advanced, great-looking app at this point, but I’m sure this will change. And of course, it needs to be tested extensively before use in production. But in all, the coding experience itself is really nice.”

The big battle for big data: R programming vs Python

This so-called battle between Python and R is played out. At the end of the day, we all work with data and just want to share insights with friends, colleagues, or clients. Data science, if applied correctly can enhance industry and even improve life on Earth. So why become entrenched in one side? Why alienate and shun tools that can help us all?

If one language helps you solve a problem and another helps showcase it – why not use both? If you’re not a developer or software engineer in your day job, you should know one thing: concepts of coding apply to every language. That means no matter what programming language you’re familiar with if you understand the underlying concept you can jump to another language. You’ll just spend 60% instead of 40% of your time googling syntax. #devtruth

Posit unites Python and R programming

R and Python are better off in alliance. Both communities can benefit from RStudio’s expansion with their data science services and open source contributions. RStudio has been successfully incorporating Python and other data science tooling into its products for the past few years. And unsurprisingly, this hasn’t led to the downfall of R. 

On the contrary, we’re seeing curious Python users venture into the RStudio suite. And we’re seeing companies adopt new technologies. All of this means RStudio is able to expand the proverbial ‘pie’ of data science instead of focusing on cutting out a larger chunk for R users. 

As RStudio grows, R and Shiny grow. As R and Shiny grow, the greater data science community benefits. So join us in celebrating change and let’s build better data science tools!

If you’re curious to try out the latest innovations from RStudio (Posit) check out the links below: