Spotlight on Microsoft Research: Opening a Door to Big Data - Try F#

Posted by Kerry Godes
Senior Manager, Worldwide Marketing and Operations

F# is an open source, cross-platform programming language for writing simple code to solve complex problems. From a business perspective, the primary role of F# is to reduce the time-to-deployment for analytical software components in the modern enterprise.

Its interoperability with other languages and libraries and its ability to tackle the complexity of components such as calculation engines and data-rich analytical services offer compelling business value. F# is both open source under the 2.0 license and a first class language in Visual Studio, as well as it can be used on Mac OS X, Linux, Android, HTML5 and other platforms.

Our colleagues at Microsoft Research (MSR) just released their newest iteration of Try F#, a set of tools designed to make it easy for anyone—not just programmers—to learn F# and take advantage of its big data cross-platform capabilities.

We chatted with Evelyne Viegas, Director of Semantic Computing at MSR, to find out more about how you can use “Try F# to seamlessly discover, access, analyze and visualize big and broad data.”

Q: What is Try F# and what advantages does it offer the user?
A: Try F# allows users to program in F# 3.0 directly in their browsers, with no download of Visual Studio required, and enables users to share code with the community of developers, both experienced and those just starting out. We look at Try F# as a huge door to big data, in other words, it provides the ability to take large quantities of data, and metadata, and manipulate it all right within the browser. You’re not looking at many kinds of programming environments. It’s something that’s comfortable on your desktop that works with all your desktop applications and tools. This presents huge opportunities in today’s data-driven world, and my biggest hope is that Try F# will help more people be able to work with big and broad data. That’s the really exciting part for me.

Q: Who is using Try F#?
A: Students at University College London (UCL) are using Try F#’s to learn the latest updates to F# (version 3.0) through tutorials that focus on solving real-world problems, including analytical programming quandaries of the sort that are encountered in finance. In the financial world, F# is used extensively, so the tutorials provided by Try F# means the students at UCL can get their hands on real-world examples and see how they are applied, using a web-based and user-friendly environment.
From the data scientist perspective, undergraduate and graduate students at the Rensselaer Polytechnic Institute are using Try F# tutorials which incorporate many domains, to help them understand F#’s new powerful “type providers” for data and service programming. For example, there is a type provider for the World Bank that helps answer questions like ‘Is U.S. healthcare cost-effective?’ by offering the researcher quick and easy access to schematized and strongly-typed datasets. They’re able to share libraries, to publish things that other people can use, making them a part of the community.

Q: What has the reaction been like since the launch of Try F#?
A: We unveiled the new release of Try F# at the Symposium on Principles of Programming Languages (POPL) in Rome, Italy, an annual gathering where global thought leaders discuss all aspects of programming languages and systems. We generated a lot of excitement among this hardcore crowd of theoretical programming language computer scientists and got them really interested in Try-ing F#! I chaired the Data Driven Functional Programming Workshop, which examined data-centric programming in today’s era of big data. We invited attendees to tell us more about how they’re using F# and what they’re developing. The experience was designed to help bridge the gap between programmers and data scientists and bring those communities together through Try F#, and the response was great as we received a lot of questions on Try F#’s type providers, among other things.

Q: What else should our readers know about F# and Try F#?
A: Despite the language’s ease of use, it’s also functional and practical—it can benefit everyone from those in finance to those in science and research. It’s open, it’s cross-platform, it goes across domains and we are looking to the community to further extend it. We’d like developers and data scientists to use Try F# to seamlessly discover, access, analyze and visualize big and broad data.

To hear more from Evelyne Viegas and others on how Try F# is being used in education and finance, watch the below video or check out the Microsoft Research blog “Try F# - Data Console to Big and Broad Data.”