Julian aims to find and apply ways to make technology useful and adapt to the needs of users, rather than expecting people to cope with poorly designed software or technology. He also focuses on improving the effectiveness and efficiency of software testing, particularly blending automated and interactive testing. The work includes web automation e.g. as part of the WebDriver Selenium team https://code.google.com/p/selenium/ and mobile phone apps. Julian finds ways to help others to work more effectively where they are fulfilled in their work and enjoy what they do. To know more about him, visit his LinkedIn profile here, read his blog.bettersoftwaretesting.com or follow him on twitter @julianharty.
Q - Hi, I am sitting here with Julian Harty. I have been trying to get Julian’s thoughts for a long time, and finally I got him sitting down in one place, not talking to anyone else, hoping he can share some experiences with us. So Julian, can you tell a few things about yourself?
A - I am mostly human, come from England, have been associated with S/w testing for 15-20 years, its only part of delivering software and I have been to lots of places in the world, currently I am in Pune, India.
Q - You are working on various fronts in the Educational field, to help those in the remote / rural areas learn and gain knowledge. Here challenges are very different - like cost, electricity, sufficient number of teachers, network, to name a few. To still make it work, you experiment with various (relatively) low cost technology devices (example - Raspberry PIs, Portable battery packs, etc.) and help create software / applications in localised languages to counter such challenges. What has prompted you to work on this different path?
A - Software affects people’s lives, running all the devices they are using. Everyone depends on software whether they know it or not, it is used for systems that monitor them, do their taxes, flights, mobile phones, they are all in software. Ultimately they should be able to have definable trust in technology, and shouldn’t talk about it, but should just work for them. Sadly - that’s not where we are in the industry yet, so that is my mission to try and get s/w to work for human beings and help them live better lives. And testing is a means to an end. It’s a necessary part of the process. if testing wasn’t necessary, I wouldn’t do it.
Q - Is there a trigger point that made you get onto this path, or was it always something that you had thought of?
A - I recognised it about this about 10 years back - when I was at Google. I realised that we spend too much time fussing about technologies and the individual traits (dev / testers), and not enough on what we can do with this, apart from keeping ourselves in employment, which is sweet and wonderful, but you have to make it useful for people.
Q - Talking about testing, what are certain things that are going well, or not, for testing as a profession, or tester as a role?
A - No certifications! I have been involved in certain certifications. that is a good idea gone bad. I don’t think there is any value in certification, except for interest of learners, who cannot learn without certifications, and perhaps also helps learn sooner.
What else is good in testing community? People - meeting a couple of people in EuroStar in Dublin, travellers from around the world. It is great to catch up with people.
Testing as a practice is very important - I find most people in the industry are little bit too myopic. We should have people who think broadly in terms of what they are doing and how that work helps and spend less.
There are Testers who tie themselves to their job title. They are polite to people above them and they are rude to people below them (in the organisation hierarchy). I think, by itself, it is hard for someone who is closed confined to software testing to thrive in software industry. A Tester needs to have more broad testing ideas, hypotheses, finding ways to provide information about software ... this is a tremendous role to play. Then take it forward and apply it to the business idea, and how it ties to affecting humans. Testing can be a mechanism to help achieve this.
Q - Talking about the educational aspect of software testing, I too agree that certification should be more of a learning experience than a certification degree. We do not have a standard or any curriculum especially in any of the universities in India which teaches about software testing. Is that a good thing in a way that you are not bound to a certain set of rules or is that a limitation in some way?
A - A program is not bound to a certain set of rules and ... development curricula. There are a few people involved in doing Software Testing at University level like Cem Kaner. I am not sure if he has a strong curriculum but it certainly has a lot of materials available online. That's a good start. ... A curriculum could be some what useful but like education in general, we often do much learning outside of the curriculum as within the curriculum. As there are different flavors of whether people should be instructed in education or they should be allowed to learn freely. Deeply differing viewpoints! I am not sure there is much evidence to show that one dominates the other. I am not sure what exactly could be on the curriculum. Things like CBOK and ISTQB can be a starting point.
Q - How important is passion for work compared to capability and skills for individuals in the software industry? Is one sufficient without the other?
A - I think both are orthogonal. If they are very passionate, they are useless because they make more disruption by being passionate. Passion combined with something else becomes useful. Some of them who are uninterested in work are kind of a loss because they will be unlikely to contribute much. Capabilities are more broad while skills are more narrowly defined. Capabilities mean the ability to do something. I feel, passion has to be combined with capabilities.
Q - Has Open Source development begun to make the ecosystem of technology too fragmented? Has the probability of reuse and quality been made to compromise?
A - I am not aware of any clones to test automation frameworks. Any test frameworks that exist like calabash which are on github, the nice thing about them is that, you can use them in your projects and look for any updates to the github repo / take your ideas back into the core project. I see many commercial tools are often embedding open source testing frameworks like Selenium and since the source code is available they can take the implementation. I hope they go by the license agreement, some do, some don't unfortunately, they go down by GPL. I am all for commercial competition for open source and free software. Open source does not mean it is free. It is not a prerequisite. It just tends to be free. Either way, about fragmentation, there are too many tools. They know which tool is not solving their problem. They try out on their own, they experiment. This is a common disease of programmers who want to solve the problem.
Q - Most of the events and conferences that I have been to, I have not seen as many women participating or coming to these events. I have observed similarly skewed ratio of men Vs women in far too many offices as well. Has that been your experience too?
A - Yes, pretty much! Excepting the Star Conferences (East / West) where they tend to have an equal ratio. I don't see many testing leaders coming from women. I can name them because there are only few of them. The conference that I had been to this week, its 95% of men, which is a pity, because the diversity can help to improve the quality of software. Software is used equally by women, we should make sure that they are involved in the development process. Perhaps it is the industry that has to re-think how to do this. You work for Thoughtworks, what is Thoughtworks doing in this area? How do the teams do it? Are they being more flexible especially in the case of women with kids? In some cultures some men take care of kids at home while women go to work. Women should contribute equally.
Disclaimer: The statements and opinions expressed in this article are those of the author(s) and do not necessarily reflect the positions of Thoughtworks.