NB: This blog post is adapted from one I wrote for an internal blog post at work, so it has been sanitised for content.

When learning new things, ideas, skills or exploring new perspectives, the image above reflects how I am feeling when I am trying to assimilate and process all the activities I take part in within the testing community. It’s that beautiful and terrifying moment when you are flying and the sun is ablaze on the horizon. It’s knowing you’ve learned a lot, but there is so much left to learn. Perhaps a move one way or another will lead to failure, but as long as you are quick to learn from those mistakes you can be on a well-lit path again.

Over the last month or so I’ve been attending a few events, including ITAKE Unconference in Bucharest, Romania, Let’s Test in Stockholm, Sweden and Nordic Testing Days in Tallinn, Estonia. Each of these events offered something new and different for myself as a contributor to those events. However, it is the other aspects of what they provide that are important.

ITAKE Unconference

ITAKE Unconference was my first time giving a Keynote talk at a conference. This was a huge honour not just to be asked to give a talk, but also the fact that it wasn’t a testing conference. ITAKE is a developer conference. Every attendee is highly technical, lives and breathes the code they write and the tools they use. It made me realise that whilst I have spent a lot of time learning about security, there is so much else to learn. Especially about how developers learn and work, and how applications are crafted.

I spent hours talking about how to build good environments for testing using tools like Docker and Heroku, or exploring how developers think about testing. A lot of it is about unit testing, some of it is about automation. But there are a lot of developers who understand the value of good testing and want to work with testers to make it happen. There is a lot we can do better to support them in this endeavour. These are things we should be doing at Medidata…testing cannot happen or exist in a vacuum.

Yes, there were those that think the role of testing or testers is now defunct, where a technical person can achieve all things they need to on a project. It was interesting to be able to discuss and challenge some of that thinking, where a tester can be a specialist or advocate for testing on their project; rather than someone who executes tests, gathers test results and creates endless meaningless reports. I’m not saying reporting is bad, just the doing the wrong reporting is bad, and unhelpful. It doesn’t add value, nor does it explain to those who don’t test what the value of the testing is.

ITAKE is a hotbed of software craftspeople. People who want to build and develop great software for their customers and clients. The best talk I went to while I was there was one of the other keynotes. Felienne Hermans, of the Netherlands, gave a talk called:

What is science? On craftsmanship for children

This reflected on her approaches to teaching coding to children. Children learn predominately through play, exploring their environment, and asking questions. It’s something that adults have largely forgotten how to do, or if we haven’t has become more formalised. We’ve turned play and learning into work instead. We can make our learning far more creative through events such as hackathons. We should review, model and landscape our applications inside the environment we are working in. Children do this far more naturally than (some) adults.

 

She also talks about introducing children to the scientific method, how we observe behaviour, theorise about why the behaviour occurs and make a hypothesis, and then on to experiment in order to prove/disprove that hypothesis. This can be applied to coding as much as any branch of science or other learning.

Let’s Test Sweden

This was a bittersweet event for me. This was my third visit to Let’s Test in as many years, but sadly it will be the last ever Let’s Test in Sweden. The organisers have decided to call it a day. This edition of Let’s Test had a distinctly technical focus, with each session being a three-hour workshop, held over two days.

I ran a workshop called Web Application Security, a Hands On Testing Challenge. We have Security Awareness Training in-house, which covers many of the techniques and tools of security testing, so this follows a similar path. Given the time and space in the office, we should be able to make this learning much more hands on. I try and provide a safe space for the attendees to find problems and ask questions about the application under test so that they can also talk with confidence about security in their own environments.

I attended a number of other workshops, including Alan Richardson’s Evil Tester’s Testing Games of Evil Testing. In this workshop, Alan introduced us to how we can use simple tools such as the browser developer tools to interact with simple JavaScript, API clients and HTTP requests to give us a competitive advantage in debugging web based games.

Whilst the topic might sound trivial, the application and usage of these tools are crucial for testing and debugging modern web applications. Browser based development tools will allow us to do so many useful things:

• Viewing the source, both HTML and JavaScript
• Debugging JavaScript, via the Console
• Tampering HTTP requests
• Testing REST services from the browser
• Custom headers
• Throttling application performance
• Emulating other browsers, devices and screen resolutions

During the workshop, we played the games to understand them and their gameplay. We found bugs and fixed them, we wrote code and created cheats to make it easier to win, get massively high scores or infinite lives.

One of the other workshops I attended was Aare Nurm’s Pedal to the Metal. Aare hails from Estonia. His workshop was a fantastic, practical exploration of how software and hardware combine to make products, solving problems and exploring how stakeholder demands can cause issues and constraints with your testing, and how your testing value is perceived and acted upon.

We were given the exercise to test a keyfob for a new kind of car that was coming on the market. We didn’t have access to the vehicle, but only to a prototype to the keyfob. We needed to utilise all our testing savvy to come up with test ideas, find problems, analyse logs and even fix the issues ourselves.

Essentially the keyfob was a Raspberry Pi Zero, with a pin board and wires which could be configured to give different settings. We also had a set of LED’s which would flash according to the function the keyfob was supposed to be executing, including:

• Unlock the car
• Lock the car
• Remote closing the windows
• Activating the headlights
• Remote boot/trunk opening
• Start the engine

We initially observed the behaviour of the keyfob in order to determine it’s function. We were given minimal, but rapidly changing requirements not only for the product but also the business. Essentially if the car failed in the market, the company would go out of business. Hardware and software are so entwined, that even if the product is solid, well made and easy to use if the underlying software architecture is poorly implemented, this can result in a poor customer experience.

Here with my workshop partner Phil, you can observe our testing and learning. This was honestly one of the best learning experiences of my life. Check out the videos here: