Sunday, January 31, 2010

Scala Programming Language – An Overview

The Scala programming language belongs to a class of programming languages known as ‘Functional Programming Languages’. Before we proceed further, let us have a quick re-cap of some the core concepts of functional programming languages, which we covered in an earlier blog post titled ‘Functional Programming – An Introduction’.

In Mathematics, ‘functions’ express the connection between parameters (inputs, in the case of computers) and the result (the output, in the case of computers) of certain processes. In each computation, the result depends on the parameters in a particular way and hence a ‘function’ is a good way of specifying a computation. This is the basis of ‘Functional Programming’.

The above notion is also more close to the ‘human world’ than to the world of a computer where in the initial days of computing, programs consisted of instructions to modify the memory, executed by the central processing unit. Thus, functional programming languages match the mathematical idea of functions.

A function is fundamentally a transformation. It transforms one or more inputs into exactly one output.

An important property of functions is that they yield no side effects – this means that the same inputs will always yield the same outputs, and that the inputs will not be changed as a result of the function. Every symbol in functional programming language is immutable.

Functional programming treats computations – running a program, solving a numeric calculation – as the evaluation of functions.

Having covered the key concepts of functional programming, let us move on to the industry scenario that led to the evolution of Scala programming language. Moore’s Law states that CPU speeds will double every 18 months. However, these days the focus is to create CPUs with multiple cores – meaning multiple CPUs within a single chip. This means that the multithreaded environment is executing on more than one CPU simultaneously as opposed to the standard ‘round-robin’ cycle executing on a single CPU. Multithreading on multiple CPUs requires that the code must be highly thread-safe. 

Attempts to resolve this problem of having highly thread-safe code has resulted in many new programming languages that addresses the concurrency problem, each language with its own virtual machine or interpreter. This obviously means that a transition to a new platform is required, similar to what happened when organizations moved from C++ to Java, about a decade ago. Such a transition is a non-trivial task and most companies consider another transition risk prone. This sets the stage for the arrival of Scala programming language.

Scala is a statically typed, object-oriented programming language. In addition to being object oriented, Scala is also a functional programming language and blends the best approaches to object-oriented programming and functional programming. Scala is designed and developed to run on the Java Virtual Machine (JVM) and Scala’s operational characteristics are same as Java’s. In fact, the Scala compiler generates byte codes that are nearly similar to that generated by Java compiler. This compatibility ensures that Scala language can utilize existing Java code, which in turn means that Scala has access to the existing ecosystem of Java code, including open-source code.

In Italian language, ‘Scala’ means stairway or steps. The name ‘Scala’ was selected to imply that Scala programming language allows programmers to ‘step-up’ to a programming environment that incorporates the latest in programming language design and at the same time letting programmers use all existing Java code. Scala also means ‘scalable language’, which means the language is designed to grow with the demands of its users.

Scala has been generating significant interest in the software industry and companies are announcing their move to Scala. Twitter, in April 2009, announced that they have switched a large portion of their backend to Scala and intend to convert the rest. Wattzon has mentioned that their entire platform has been written from the ground up in Scala.

Professor Martin Odersky is the creator of the Scala language. As a professor at EPFL in Lausanne, Switzerland, he is working on programming languages, more specifically languages for object-oriented and functional programming. His research thesis is that the two paradigms are two sides of the same coin, to be identified as much as possible. To prove this, he has experimented with a number of language designs, from Pizza to GJ to Functional Nets. He has also influenced the development of Java as a co-designer of Java generics and as the original author of the current javac reference compiler. Since 2001, Prof. Odersky has concentrated on designing, implementing, and refining the Scala programming language.

Before we conclude this discussion, I would like to quote a reference to Scala from a previous blog post, titled ‘Technology Choices for 2009 and Beyond...’ posted on 24 September 2008.
Another relatively new [first public release in 2003] language, Scala, designed and built by the team led by Prof. Martin Odersky (EPFL, Switzerland) [Prof. Odersky has also influenced the development of Java as a co-designer of Java generics and as the original author of the current javac reference compiler] also seems to be promising. On a related note, in the article titled "Java EE meets Web 2.0" written by Constantine Plotnikov, Artem Papkov and Jim Smith in developerWorks, November 2007), the authors identifies principles of the Java EE platform that are incompatible with Web 2.0 and introduces technologies, including Scala, that close the gap.

This concludes our discussion on Scala programming language, which is expected to transform software engineering, the way Java programming language did about a decade ago. 

~ Sunish

No comments: