You are here
The Bloor Group: R is for Analytics – Revolution Analytics
The Rise of R
Programming languages proliferate. In fact there are now more programming languages than there are human languages, which may seem strange, but many programming languages are as dead as Latin, often having fallen out of use because the computers they ran on are no longer used. Nevertheless, every now and then a programming language will suddenly acquire a multitude of users because it proves to be extremely useful. This happened with the C language and it happened with Java. It is now happening with an unusual language called R.
We think of R as unusual because it was not created, as most programming languages are, for technical programmers. It was created for statisticians, with the goal of allowing them to build and implement statistical capabilities, including modeling techniques and standard statistical tests and also graphical visualizations of statistical results.
R is not just a programming language. It could more accurately be described as a statisticians development environment that includes data handling and data storage capabilities, many embedded data analysis capabilities and a suite of operators that can carry out calculations on arrays and matrices. To this can be added data visualization capabilities and the usual programming language features of conditional statements, loops and recursive functions.
An important aspect of R is that it was created in an Open Source project. Partly because of that and partly because it has been well-thought-out, it has gained considerable traction among data analysts. There are now estimated to be more than two million users of the language. As such it has become a de facto standard.
Revolution Analytics, Riding the R Wave
Revolution Analytics is the Open Source-oriented company, founded in 2007, to support the commercial us of R. Since then it has developed what can best be thought of as an R-based stack of software to enable the efﬁcient use of R by multiple users – both end users who might use BI reports built in R and the data analysts themselves.
In other words Revolution Analytics is, like all Open Source companies, able to provide support and consultancy in its area, but also like many Open Source companies it has enhanced the basic Open Source offering with software of its own. In effect Revolution Analytics provides a software stack which spans the distance between the various data sources (Hadoop, traditional data warehouses, DBMSs, etc.) and BI applications.
We illustrate this in the diagram on the following page. At the heart of the stack are the Revolution R Enterprise Development and Production Environments. Data analysts and data scientists use the development environment for their statistical analysis work but can transfer capabilities – the results of their activities – to the production environment when necessary.