May 22-26, 2017
Cologne, Pullman Hotel


Andrea Goulet

Andrea Goulet


Makers and Menders: Putting the right developers on the right projects

Andrea Goulet is the CEO of Corgibytes, a software development shop dedicated to maintaining and modernizing software applications. She’s the founder of LegacyCode.Rocks and hosts a podcast dedicated to changing the way we think about legacy code. Andrea has a knack for solving complex business problems with simple solutions and is passionate about user experience.

It's not often you find someone who "loves" working on legacy code bases. Why did you decide to pick that as your service niche?

Andrea: My background is in business, communications, and marketing. So when my business partner, Scott, approached me about joining Corgibytes as the CEO, I was seeking an opportunity that fit my marketing criteria: something that everyone needs and no one wants to do. One day, Scott and I were watching a home remodeling show, and Scott looked at me longingly and said that's what he loves doing with code. Some of his favorite projects were refactoring incredibly complex codebases, and helping teams get more use out of codebases that have been around for a while. The marketing person in me loved the idea, so we went all in, and over the years I've learned that there's a lot of fun in legacy work if you approach it in the right way.

How do you approach a legacy project? In what ways is it fun for you?

Andrea: The first thing we do is recognize that everyone who touched the project did their best. No one goes to work and intentionally tries to write poor code. Compassion goes a long way with a legacy project. Next, we do what we call a Code Inspection where we take a deep dive into the code and provide a written report that the executive team can understand. One of our core values is "Communication Is Just As Important As Code". We've observed that most codebases that are in a legacy state are that way because of poor communication. We start by getting buy-in from the very top of the organization, so they can understand the value of making their codebase more efficient. My career has always been as a technical communicator. I love watching someone's face light up when they "get it." It's incredibly gratifying for me personally to help people without technical backgrounds understand their own technology better.

At XP2017, you'll be talking about two types of developers: Makers and Menders. Can you tell us a little bit about them and how they're different?

Andrea: Most developers are makers — about 90% from our research. We see this prominently in the hacker culture. You've got Make magazine, Makerfests, Hackathons, Startup Weekends. All of these are designed for folks who love taking an idea and building a proof of concept or minimum viable product. Over the years, I observed that the developers who thrive in this environment get really annoyed when it comes to the polishing stage. They'd rather build an 80% solution and then move on to the next project. Menders are different. They love taking the 80% solution and making it stable. From what we can tell, there's about 5-10% of developers out there who would prefer bug fixes and refactoring to feature development. Our industry treats these developers as second class, but that shouldn't be the case. It's like introverts and extroverts, one isn't better. You just need to know how to motivate them and which projects will set them up so they can shine.

You've launched a project called LegacyCode.Rocks. What's that about?

Andrea: We were at Agile2015 when, for the first time, we felt surrounded by other menders. It was glorious! We had some amazing, engaging conversations that we didn't want to end. So I launched a quick website and slack channel so that we could keep in touch after the event. Over the past two years, that's grown into a community of about 250 people who all love mending. It's a place where we can geek out over TDD and refactoring and get really technical about solutions and ideas. We've started a podcast and have weekly mastermind groups, too.

What are you most looking forward to at XP2017?

Andrea: I have to say, I'm super proud that Corgibytes is able to be the child care sponsor. Since I own my business with my spouse (Scott and I got married after being business partners for two years), we hate choosing between who's going to go to a big conference like this, especially when one of us is speaking. Having the option to bring our children and know they'll be taken care of during the day is a huge step forward. I'm looking forward to connecting with other parents and chatting about how they bring Agile and XP concepts into parenting, too. It's surprising how relevant a family retrospective meeting can be!


Claes Wohlin

Claes Wohlin

Blekinge Institute of Technology

Evidence-driven Change in Software Development, is it Feasible?

Claes Wohlin is a professor in software engineering and dean of the Faculty of Computing at Blekinge Institute of Technology. In 2011, he was elected member of the Royal Swedish Academy of Engineering Sciences. His main research interest include agile and lean software development, and evidence-based software engineering. The research is mostly conducted in close collaboration with industry.

We're very happy to have you as speaker for the opening keynote at XP 2017. Can you say a few words about your relation to industry?

Claes: Although being in academia, I have close ties to industry. I have worked five years in industry including being involved in the first release of GSM when working as a consultant for Ericsson. The industrial background has provided a very good basis for conducting research with industry.

You are a professor at the Blekinge Institute of Technology. What are your main areas of research?

Claes: I have over the years been quite flexible when it comes to specific areas in software engineering. The common denominator is empirical research conducted in close collaboration with industry. The latter means listening to industry needs rather than running my own research agenda.

Your keynote will be about Evidence-Driven Change in Software Development. Without giving away too much, how well do you believe is Agile as commonly applied doing in this regard?

Claes: My perception is that most changes, including Agile, are not driven by evidence. The key question is really how we can make better use of evidence, which includes both generating valid and timely evidence, and actually using the evidence being available.

What do you think is the biggest obstacle to an evidence-based approach to advancing our profession?

Claes: There are some inherent issues that are hard to address, including for example the pace of change, and the time and effort needed to generate valid evidence. However, there is also a gap between practice and research, and this needs to be overcome if we believe in having evidence for changes in software development.

Copyright © 2016-2017 · All Rights Reserved · codecentric AG