Chef is a popular Infrastructure Automation framework based in Ruby. It comes with a host of testing tools bundled with it like ChefSpec for unit testing, ServerSpec for system testing and TestKitchen for integration testing. This session is a demo of how to use these frameworks to test drive cookbook development.

Is it easy to get a crash course in football by Pele or Maradona for a week and produce a world cup winning football team?

Answer is NO. Then why do lots of people in the corporate world think of hiring scrum trainers & expert developers to train their team for a week and then expect their team to undergo a transformation at a magical scale?

German football team made it a point to transform their team and it took them a lots of years before they were able to reach the pinnacle. A quick side by side comparison of what is causing agile transformation to fail.

Vision

Football: Someone was there owning this entire transformation, the German football association spent a lot of time identifying talent in their teens and groomed them.

Office: In the corporate world switching jobs every few years have become common, but there is no passing on of the context, resulting in the new person taking charge, starting from all over again, as well as frustrating existing good performers who have to rebuild the perception.

Infrastructure

Football: Players expecting state of the art training facilities, fitness programs and new shoes are not a luxury, it is a necessity.

Office: Computers have become so cheap compared to the salaries, yet the policy of providing the best tools and good work environment are archaic.

Coach vs Management

Football: Coaches are given their due powers to help the team achieve the goal. It is very easy for anyone to comment on how professionals should play their game, there would be no use adhering to the metrics if the team cannot win. Winning is the only measure for management.

Office: Often coaches are seen as part of the management or management tries to heavily influence coaching which results in a team which will work either for metrics or to please higher ups without the actual result that it had aimed for.

Training

Football: Just the ability to kick the ball does not make a footballer. Training will be introduced to increase physical strength & stamina, better mental wellness, injury prevention, tactics and strategy.  A heavy investment is made in the training facilities.

Office: In the software industry a generation is about 2-3 years. Computer science degrees are nowhere near what is state of the art in the industry. At many places the on boarding process is either very shallow or not up to date with the recent developments, leaving people to learn most of the things hard way. Given too many things to learn and the information overload, this results in inefficient learning and application of knowledge on the job. We need to prepare people to find answers that are not available on Google.

Team composition

Football: Rookies don’t learn by watching greats from the bench, instead they play along with the veterans. Every sports team makes sure to have the right composition with a mix of rookies, emerging players and veterans. That is how they sustain a winning team.

Office: Architects and Leads often do not code or are not part of the team every day, this means that most of the time the team just looks up to for advice or waits for reviews. There should be a good mix of people at all experience level so that there are enough people to try new things, enough people who have mastered few things and enough people who challenge change.

Above all – Persistence

Traditionally, in software development, user experience (UX) wasn't valued as much as developing of the software itself. But this has changed rather radically. However, creating an enriching user experience in an agile fashion is still challenging. Most of the agile engineering practices in use are around building software but seldom address UX. When building a product in an agile fashion, UX in an incremental fashion becomes important.

In this talk, we will present our experience of creating UX in an incremental fashion for a virtual wallet. We will also talk about the different challenges we faced such as, educating various stakeholders on the value of incremental UX, building collaboration between developers and experience designers and abstracting design components, along with the solutions we devised to tackle these challenges.

We've come far in our journey of Agile as a software development methodology. From stand-ups to showcases to sprint planning meetings to burn-ups (or downs), we've got it down pat when it comes to processes to follow to be considered Agile. However this heads-down, process defined agile, often hinders real agility required to meet business needs. Is doing a three hour sprint planning meeting every week the most important thing to do when you have to get a minimal-viable-product out in the market? How much of automated functional testing should you do when you know that your product's beta version is only going to validate assumptions of your business idea? Should you write tests at all? There is no formulaic answer.

In this talk, KK and Krishnan will talk about their experience of how much Agile is too much Agile. We look at how to find the right balance between following agile practices and being responsive to your business. How much agile is too much and how less is too less?

We will do this by looking at:

• A couple of successful agile adoption stories
• Look at why agile was successful in the contexts above
• Discuss why this success will limit us if we are not careful
• Talk about a start-up and how the things that led to success in the first 2 stories limited us in the start-up context
• Look at approaches to understand what agile practices/processes to follow to be business agile
• Close by summarizing the challenges facing agile (as we see it) and how success in process agility will limit us in business agility

Developing Infrastructure as Code as a practice is increasingly adopted by organizations that want to have Continuous Delivery. Infrastructure As Code requires treating infrastructure as code, applying the same development techniques that one uses to build application software. This talk discusses the need for having infrastructure as code, the several frameworks available in this space (like Puppet, Chef, Ansible), what they bring to the table and the philosophical differences between them and the testing tools and frameworks available to test infrastructure as code. The concepts would be supported by snippets of infrastructure code written for an open source application.

An Ideal Test Pyramid is an invaluable technique to succeed with agile. However, it is challenging to have this ideal pyramid in practical projects for several conditions in the projects, like legacy code bases. To have a good test pyramid, several best practices of software development need to be followed. This talk discusses the practices that help in implementing test pyramid and the challenges faced by many teams in doing that and suggestions on how to overcome them.