Data Viewer: RStudio vs Positron — A Product Owner’s Perspective

For anyone working in data science or statistics, looking at your data is almost an irresistible impulse — and for good reason. In RStudio, that usually means typing iris |> View().

Over time, I grew increasingly frustrated with the RStudio data viewer. Eventually I resorted to dumping dataframes to a temporary CSV file and opening it externally — a workaround borrowed from a Stack Overflow thread.

Quickly view an R data.frame, vector, or data.table in Excel

Clearly, only one of the many workarounds the community found over the years:

Viewing tables of data in R

How can you visualize data frames in a good way?

Is there a good browser/viewer to see an R dataset (.rda file)

So when Posit released Positron — their new polyglot IDE — I was curious. And as a certified Product Owner, I couldn’t help asking: how would you approach improving the data viewer as a product feature?

---

Start with the User

The right starting point is always user research. A rapid scan of GitHub issues, Stack Overflow questions, and Cross Validated threads reveals the most common data viewer tasks:

#	Task
👁️	Open & inspect a dataframe
🔽	Sort by column
🔎	Filter rows
📐	Navigate wide dataframes (50+ columns)
🔄	Export or reproduce the current view

---

Filtering in Action

Here I focused on filtering rows — one of the most frequent and most complained-about tasks.

Take a concrete example: filter Sepal.Length between 5 and 7, for a given Species only.

RStudio

Fewer clicks. More direct. The interaction is efficient and familiar — if occasionally clunky.

Filtering in RStudio

Positron

More steps to reach the same result — but the experience feels noticeably more fluid and modern.

Filtering in Positron

---

The PO Lens: Efficiency vs. Experience

Then, why is it that Positron feels more pleasant to interact with even if it entails more users’steps?

Maybe this is just an evolutionary constraint? Rstudio and Positron optimized their respective established workflows. RStudio was optimized over years around the R-first analyst. Positron evolved from Microsoft’s VS Code — a different ancestry, a different design language, and maybe a different target user: the multilingual analyst working across R and Python.

Don Norman’s levels of design offer a useful frame to resolve this tension:

• ✨ Visceral (how it feels) → Positron wins. It’s just nicer to be in.
• ⚡ Behavioural (how efficiently it works) → RStudio still leads on raw task speed.

Positron appears to have invested in the visceral layer first — and that’s strategically defensible.
First impressions build trust: while efficiency can be iterated , first impressions can’t.

The implicit product bet seems to be: if we make it feel better, users will tolerate relearning the micro-interactions..

Whether that’s the right call depends entirely on which user segment Posit is prioritising — the loyal R community, or the next generation of polyglot analysts who never had strong RStudio habits to begin with.

That’s the question I’d be asking in the next sprint planning session. 🎯