Why is pair programming worth trying?

WHAT IS ACTUALLY PAIR PROGRAMMING?

The term refers to the act of developing an IT project or its fragment in pairs. One programmer writes the code while the other follows it on the screen in search of potential errors. The roles of a ‘driver’ and a ‘navigator’ interchange, which means the developers regularly swap after fixed periods of time.

WHERE DID IT COME FROM?

Pair Programming stems from the methodology called Extreme Programming. It is based on a few simple principles, every single one of whom is of strategic importance to a project:

• simple code - means writing only the code that conforms to plans and needs. That leaves no room for chance or the driver’s fancy, which in turn results in succinct and easy to understand code
• effective communication - the members of the team where roles interchange can communicate with each other in a clear and concise way
• mutual motivation - a programmer in the team knows roughly as much as their partner. That means not only do they complement each other, but also expect the coworker to act thoroughly and effectively

HOW DOES IT WORK IN PRACTICE?

The experience of two people working on the same task makes it easier to eliminate erroneous assumptions as early as while designing the code. Logical errors and typos can be quickly detected and fixed. What is more, both roles involve an explicitly stated range of duties, which means the driver, who is in charge of writing the code, focuses exclusively on the implementation, while the navigator makes sure their partner follows the predetermined direction of the process. Much like a rally driver, who handles the car, and the navigator, who provides all the necessary information about the road ahead.

PAIR PROGRAMMING IS NOT A NOVELTY!

The history of this solution dates back to 1953, but it was not until 1992 that the term ‘Dynamic Duo’ was coined. It referred to the act of designing code by one person under the supervision of the other. The following years brought further publications on Pair Programming being employed in multifarious technologies. That means this strategy has been successfully applied in countless various projects for several dozen years now.

WHAT ARE THE ADVANTAGES OF PAIR PROGRAMMING?

There is no doubt about the great increase in code quality. Detecting errors as early as while designing the code optimises the subsequent costs of its maintenance. As a result, there is no need for countless hours of tests and alterations to the code whilst searching for potential errors.

The distribution of knowledge about the system between two people instead of one prevents its dissipation and makes it easier to share when new pairs are formed for further tasks.

Another considerable finding of the research was the increase in efficiency in proportion to the amount of time the tasks took to complete. The same tasks took longer when performed by two programmers separately than when completed in pairs.

So productivity increases and the time it takes to complete a project is shorter, which means the code is ready to submit much sooner.

The team is involved in developing the project and actively contributes to the efficient elimination of any possible threats and errors. That guarantees high quality of the created software.

Nowadays agile software development necessitates flexibility when it comes to change. Pair Programming cuts the inherent effort of communication and coordination in half.

Aside from the obvious advantages to a project, the solution offers benefits to an individual as well:

• the developer can replace their partner, e.g. in case of leave or sickness, with no general loss to a project
• when it comes to recruitment, an experienced programmer is able to assess the potential employee’s qualifications using Pair Programming
• since the project is the result of common effort, the entire team is motivated to introduce every new programmer to the unfamiliar environment

WHAT DO THE DEVELOPERS SAY?

Workplace comfort is one of programmers’ main principles. Ergonomics is as important as problem-free workflow. Over the years countless studies, interviews, and surveys have been carried out in the programming community. The findings make it clear that what employees find most satisfying is the certainty of the flawlessness of the project and the distributed code ownership - that is, the responsibility not being restricted to a single individual.

ARE THERE ANY DISADVANTAGES TO PAIR PROGRAMMING?

With so many benefits to IT projects and their subsequent development and maintenance there is a certain issue that can be considered a great disadvantage by many. The cost of such undertaking amounts to the cost of employing two people in lieu of one.

Besides, Pair Programming renders ineffective as soon as any personal problems arise in the team. When two team members are not too keen on each other, the productivity of their pair is considerably reduced. In such a case there is no ‘pair’ and no ‘programming’, let alone ‘code quality’. This kind of problem needs to be taken care of above all!

There is no doubt, however, that in the long term higher code quality reduces the cost of project maintenance and ultimately guarantees profit despite the initial additional investment.

CONCLUSION

All in all, Pair Programming is a much more efficient and profitable solution than employing a single developer.