My very personal RStudio R&D journey
April 23, 2022

My very personal RStudio R&D journey

Anonymous | TrustRadius Reviewer
Score 10 out of 10
Vetted Review
Verified User
Review Source

Overall Satisfaction with RStudio

I have used the R language since around 2010 and before (along with S-Plus). RStudio as soon as it was available, also around 2010. Example use cases: 1. bionanoengineering - descriptive statistics (describing biological motility or nano surfaces), in parallel with image analysis in ImageJ and MATLAB; 2. bioinformatics - producing descriptive statistics for the motility of Neurospora crassa (filamentous fungus) to prove that how one use statistics matters and how it impacts business decisions; 3. pharma - benefit-risk analysis and data visualizations along with Spotfire 4. healthcare - clinical programming along with Stata and Python (one suggestion: it would be nice to have R interface in Stata and improved R interface in Spotfire); 5 - in product development for creating data monitoring & evaluation apps in RShiny. RStudio has been with me since the very beginning of my professional career. I could easily write up a Ph.D. on the use cases of R in life sciences, pharma, healthcare, and computer science. I would highly recommend RStudio for those who need to deliver fast tailored, customized applications, attractive visualizations or need to use Bayesian statistics, for example, to validate pharmacovigilance scores.
  • RShiny applications that are intuitive and help to communicate in the multidisciplinary teams
  • d3.js based visualizations
  • Bayesian statistics
  • calculating confidence intervals
  • merging tables by using SQL commands
  • using regular expressions
  • some of the machine learning implementations are best in R
  • way more hassle-free than SAS, in my opinion
  • open-source - RStudio does not discriminate against people & businesses based on their financial status, many small businesses cannot afford SAS, in many developing countries young people are willing to learn to program, and SAS platforms or other paid software is absolutely out of the question, those people/young programmers will be not able to afford even free cloud SAS due to the internet infrastructure...some of the best ideas come from those who face serious challenges in life and can speak several languages as their minds are often more creative ("necessity is the mother of invention"). I feel that platforms like RStudio or Jupyter connect me with the World, with other creative minds, and contribute to making the World a fairer, better place.
  • something like IronPython in Spotfire, but R equivalent would be great; the existing R interface is not fully functioning
  • something like Pyhon interface in Stata, but R equivalent would be awesome
  • in the pharma World deadlines are tight, pressure is very high - Stata lets manipulate data super fast compared to R
  • brining R and Python community together
  • in my opinion, Natural Language Processing pipelines are better than in R
  • catching up with some of the machine learning implementations - visualization aids in this field are better in Python, at least that is my intuition
  • I landed a good job every single time I showed my RStudio implementations during my interviews
  • every single time I developed RShiny app, it has been a success business-wise
  • R helps to validate some of the other commercial systems, thus helping to make more informed business decisions, for example when choosing a commercial supplier or making an investment decision or planning R&D strategies...also simply because R helps to bring stakeholders together regarding communication of the scientific results
  • good for advertising: a funny story - at some point in my career I was prevented to show my app to more customers as they started asking whether this is a product and whether they can buy it together with another product (perhaps R could create some type of fast-track legal pipeline for commercialization of the R-based apps)
inter-departmental collaboration - my first choice would be TIBCO Spotfire natural language processing and knowledge graphs - my first choice would be Python information security & visualizations (including d3.js libraries) - my first choice is RStudio
my very personal opinion: regarding validated statistical documentation I trust, SAS is the winner, then Stata however, RStudio is the winner to me when it comes to dedicated Customer Success representatives
1. it does not discriminate against creative minds based on their financial status 2. it is easier to create a pilot tool to show to the stakeholders/customers first, before requesting any funding 3. it is a more reliable, convenient way of running the business, as not linking the access to RStudio directly to money allows for creating good, long-term business relationships (again, my personal opinion based on my individual experience and observations/thoughts)
I have one comment: these are great tools, but some of these solutions are very expensive, very often hard to justify the spending at the beginning stages of the work, without the proof of principle
I think shinyapps.io improves communication, especially through interactive visualization. In my job I am often expected to make very complex, abstract ideas (that take many years to fully understand or develop) simple and "communicable" to the non-technical audiences within minutes (literally, minutes).

Do you think RStudio delivers good value for the price?

Yes

Are you happy with RStudio's feature set?

Yes

Did RStudio live up to sales and marketing promises?

Yes

Did implementation of RStudio go as expected?

Yes

Would you buy RStudio again?

Yes

well suited: creating and delivering apps for multi-disciplinary teams, for example, http://drugis.org/index or https://shiny.rstudio.com/gallery/covid19-tracker.html less appropriate: Kaggle competitions, multi-community collaborations, Google collab...scenarios when the whole communities decide to work on a specific problem in Python and R is left behind, e.g. in 2015 my colleague delivered better results with Bayesian statistics simply cause he decided to go for Python to visualize joint distributions (priors and posteriors) ...even if I had way more knowledge on the algorithmic side, I was simply slower because I chose R; what I have learned over the years is that when it comes to the stakeholders, a good visualization (==communicating the results and effectively advertising) is everything as without it there is no funding and without funding no science, no R&D

RStudio Feature Ratings

Connect to Multiple Data Sources
8
Extend Existing Data Sources
8
Automatic Data Format Detection
8
Visualization
10
Interactive Data Analysis
Not Rated
Interactive Data Cleaning and Enrichment
Not Rated
Data Transformations
6
Multiple Model Development Languages and Tools
Not Rated
Single platform for multiple model development
Not Rated
Self-Service Model Delivery
Not Rated
Flexible Model Publishing Options
10
Security, Governance, and Cost Controls
10