In the previous posts of this series various benefits and obstacles of agile software development for distributed (offshored) teams were shown and solutions for these problems were provided. On basis of the Eclipse development process an implementation of this variation of the original agile principles was provided and it could be seen that such an approach can be a success.
In the final post the original defined twelve principles will be evaluated and it will be determined if each of them is still applicable in such a distributed environment.
You might also be interested in the other posts of this series:
Part 1: Motivation and Challenges
Part 2: Best Practices
Part 3: Case-Study on Eclipse
Part 4: Evaluation and Summary
Evaluation of the Agile Principles within Distributed Teams
To further demonstrate the challenges and to show drawbacks of agile distributed software development the twelve agile principles[BBvB+01] will be evaluated in the following section and it will be determined if these are still applicable within distributed teams.
Our highest priority is to satisfy the customer
through early and continuous delivery
of valuable software.[SW07, 11]
This principle can be achieved because it might be even possible to allocate more resources (manpower) in a distributed environment than with only one site [Bat09] and classic agile models which are often limited to very small teams (e.g. crystal clear 1-6 members [Coc04, 150]). Furthermore, as previously stated, the level of excellence of each team member might be increased which also results in a higher level of quality of the software (resulting in a more valuable software).
Welcome changing requirements, even late in
development. Agile processes harness change for
the customer’s competitive advantage.[SW07, 11]
The reduction of costs in an off shored or outsourced environment might lead to an increased flexibility concerning changed requirements [Bat09] but additional communication is required (e.g. visits of coordinators from different teams) to ensure that the changes are understood in the right manner. This communication overhead resulting from the dispersed teams might lead to an increased delay when reacting to changes compared to a single site. Thus this principle is feasible but it with the limitation that the changes might not be processed as fast as within a classic agile software development environment.
Deliver working software frequently, from a
couple of weeks to a couple of months, with a
preference to the shorter timescale.[SW07, 11]
As described at the first principle more resources might be available because of the distributed teams. But as a result of the hampered direct communication between distributed teams the release cycles might be increased compared to a co-located team.
Business people and developers must work
together daily throughout the project.[SW07, 11]
The communication between the developers of sites which are not located close to the origin of the client might be hindered or even be impossible. This can be caused by two factors. The first might be the geographic distances. If these distances are too big the clients representative might not be willing to travel – and rotate – all the way to the other sites. The second can be a result of the different time zones. If the expert from the client is not willing to travel to the different sites and wants to communicate via e.g. telephone it might be hard to find a convenient time for both sides where this communication should take place. [Fow06,Bat09] Because of these factors it is hard to apply this agile principle to distributed teams but as Mikko Korkala et al. showed in their paper it is possible within globally distributed teams to establish frequent customer communication. [KPC10] Furthermore, they pointed out the importance of this principle in an agile globally distributed context:
[…] when adopting an agile approach in the globally
distributed context […] the company should assure that […] at
least some representatives from the customer group is always
involved in the agile meetings.
Build projects around motivated individuals.
Give them the environment and support they need,
and trust them to get the job done.[SW07, 11]
As a result of the increased pool of potential employees the motivation of each team member might be increased in distributed teams. But on the other hand it might be hard to control the motivation on the different sites. [Bat09] Additionally, because of the geographically dispersed sites it might be harder to support the team members and provide the required environment which they need to be fully productive. But with some controlling overhead it is possible to apply this principle to distributed teams.
The most efficient and effective method of
conveying information to and within a development
team is face-to-face conversation.[SW07, 11]
It’s impossible to apply this principle to geographically distributed teams. The only solution is a trade-off by using direct communication tools such as instant messenger, telephone or a video conference. But even with this tools it might still be very hard or impossible to establish a direct communication with a different site because of the different time zones and indirect communication mechanisms must be used e.g. a Wiki. [Fow06,Bat09]
Working software is the primary measure of progress.[SW07, 11]
As previously shown with the example of Continuous Integration additional attention must be put on the test and build system so that this principle can be applied. But in general this is feasible. [Fow06,Bat09]
Agile processes promote sustainable development.
The sponsors, developers, and users should be able
to maintain a constant pace indefinitely.[SW07, 11]
Overall this can also be applied to distributed teams (e.g. 40 hour weeks). But due to communication with sites located in a different time zone additional exhaustion might occur which has to be taken into account by the team leaders. [Bat09]
Continuous attention to technical excellence
and good design enhances agility.[SW07, 11]
Because of the increased level of excellence within the teams resulting from the larger pool of potential software developers this principle might even be more practicable than with a single team.
Simplicity-the art of maximizing the amount
of work not done-is essential.[SW07, 11]
This principle is feasible [Bat09] but it requires more attention because the communication might not be as direct as it would be with only one team which could lead to duplicated work.
The best architectures, requirements, and designs
emerge from self-organizing teams.[SW07, 11]
The lack of face-to-face communication, the different cultural backgrounds of the teams and different holidays within the countries can result in a hindered self organization of the teams. To ensure a proper self organization additional measures have to be taken such as regularly exchanging developers between the teams and continuous awareness by the team members of the different cultural aspects. [Fow06,Bat09]
At regular intervals, the team reflects on how
to become more effective, then tunes and adjusts
its behavior accordingly.[SW07, 11]
Besides the already mentioned communication complications the application of this principle can also be hindered by unawareness of the cultural differences. If this challenge is not faced with the proper commitment activities like the retrospective where the whole team contributes to a process improvement will be hard to be realized successful [SW07,Fow06]. So to make this principle feasible within distributed teams the members with a different cultural background have to be introduced into these new aspects of the agile approach and then be encouraged to apply them.
It should be clear by now that there are several agile principles which are not applicable or limited when developing in distributed teams. These missing or restricted principles lead to many different challenges – where the main challenge lies within communication hindrances resulting from the lack of face-to-face communication and the physical dispersed teams – which makes it hard to apply agile principles to distributed teams. But when additional practices are introduced into the agile software development process it is possible to extend the classic co-located agile development to distributed teams. These additional practices are no replacement for the original practices but when followed with commitment and discipline a sufficient trade-off can be achieved. Especially the establishment and maintenance of communication channels is one key aspect of success in a distributed agile development process.
When a company introduces agile software development methologies to distributed teams they will face risks associated with this approach as Ade Miller concludes from the experiences of his team:
the decision maker(s) must understand that in doing so [distributing the project] they reduce the project’s likelihood of success, increase the delivery time, and reduce the team’s performance and increase its dysfunction. [Mil08, 18]
But the company can also profit from various advantages like the possibility to step into new markets or to increase the level of excellence within their employees.
Whether or not to use the agile methologies for distributed teams must thus be carefully balanced out for every project.
- [Bat09]: Dinesh Batra. Modified agile practices for outsourced software projects. Commun. ACM, 52:143-148, September 2009.
- [BBvB+01]: Kent Beck, Mike Beedle, Arie van Bennekum, Alistair Cockburn, Ward Cunningham, Martin Fowler, James Grenning, Jim Highsmith, Andrew Hunt, Ron Jeffries, Jon Kern, Brian Marick, Robert C. Martin, Steve Mellor, Ken Schwaber, Jeff Sutherland, and Dave Thomas. Manifesto for agile software development. Website, 2001. http://agilemanifesto.org/. Retrieved 28. May 2011.
- [Coc04]: Alistair Cockburn. Crystal Clear: A Human-Powered Methodology for Small Teams. Addison-Wesley Professional, 2004.
- [Fow06]: Martin Fowler. Using an agile software process with offshore development. Website, 2006. http://martinfowler.com/articles/agileOffshore.html. Retrieved 9. February 2011.
- [KPC10]: Mikko Korkala, Minna Pikkarainen, and Kieran Conboy. A case study of customer communication in globally distributed software product development. In Proceedings of the 11th International Conference on Product Focused Software, PROFES ’10, pages 43-46, New York, NY, USA, 2010. ACM.
- [Mil08]: Ade Miller. Distributed agile development at microsoft patterns and practices. Website, Article, 2008. http://www.microsoft.com/download/en/details.aspx?=amp&id=14916
- [SW07]: James Shore and Shane Warden. The Art of Agile Development. O’Reilly Media, 2007.