After many failed attempts to get him on the podcast, Craig finally catches up with Dan North at YOW! Conference on his way out the door to the airport and in a quick chat they cover:
BDD – developing an application by looking at its behaviour from the perspective of its stakeholders (people who’s live you touch)
Given When Then – “given” is setting up the world in a well known way, “when” is me interacting with the application as a stakeholder and “then” is what I expect to happen
Craig is at YOW! Conference and catches up with Anne-Marie Charrett who is well known in the testing community as a trainer, coach and consultant but also for her support of the community:
Speak Easy – Speak Easy is a voluntary program designed to increase diversity in tech conferences through dedicated conference spots, mentoring and events
Testing challenges include microservices (the risk of bounded context and breaking things down and missing the whole) and working together as developers and testers
Craig and Tony are at YOW! Conference and are privileged to spend some time with Don Reinertsen, who is considered one of the leading thinkers in the field of lean product development and author of numerous books including “Principles of Product Development Flow”
Taiichi Ohno, the father of the Toyota Production System, hated math and thus preferred to sit on the factory floor and tweak processes, hence it was not a theory driven approach but rather empirically driven
Need to understand why things work so you can transfer it to other domains, a big shortcoming in lean manufacturing is that they don’t have much of a mathematical view on what they are doing
You can use magic in manufacturing because it is highly repetitive
People understand iterations are good to do but do not understand why
“Antifragility is beyond resilience or robustness. The resilient resists shocks and stays the same; the antifragile gets better” (Nassim Nicholas Taleb)
Agile software people are doing a better job at lean product development because software people have already crossed the chasm of inspect and adapt
There are many sources of variability other than just people, such as the Internet and the fact we are constantly doing things people have not done before
To get management to listen about cost of delay you need to benchmark what you are doing today
Agile eliminated the economic gene, hence it works well bottom-up
Easiest way to introduce quantitive based decision making is to find a project manager who wants an economic model (as they will be fighting for resources and the guy with the numbers will end up winning because they can communicate their needs)
Lifecycle pretax profit is far more useful than ROI
Start with Chapter 1 in the book – describes what is wrong with what we are doing today, then look for the tree that is ready to be pushed over in your organisation as there is no one way of approaching this
The low hanging fruit is: visual control boards, economic model, batch size reduction and WIP constraints
The first knob to turn is batch size reduction
It is 175 principles in small little batches that add value, it is not the ten commandments!
Craig is at YOW! Conference and catches up with Aino Vonge Corry who is one of our very few repeat guests on the Agile Revolution. She describes herself as someone who puts speakers on stage, makes developers communicate and messes with the heads of students!
Part of the YOW! conference organising committee
Important to find examples that relate to all of the students in the class (not just a subset)
Microservice lectures – no more than 15 minutes lecture and then a learning activity
If there is interactivity then there is a reason to turn up to a live lecture
Need to respect and acknowledge that other people take in knowledge at different paces, this is important in activities that we give people time to think
People need to relate ideas to the things they are doing now to take new ideas in
Working memory takes in new information and as well as decoding for long term memory
Research says that we can think about 7 +/- 2 things at a time, but newest research says we can only think about 4 things at a time!
Chunk new content or information and then allow people time to process and think
Multitasking is a huge misunderstanding, if you are doing two things at once you are only doing them at 40% rather than one thing at 100%, this is a huge problem for people working in computer science
Try to figure out your learning preferences, realise you can’t chunk a whole lot of new information at once and ensure you sleep because without sleep you cannot learn effectively
Craig and Tony sit down for a conversation at YOW! Conference with Betty Enyonam Kumahor (stands for good for me, on the way there) who is a technology leader in Africa:
Tony and Enyo are mutual members of the Alistair Cockburn fan club
Software engineering uptake in Africa is very low, need more technologists because it is is not an industry it is an enabler
Lots of diversity challenges in Africa – lees than 1% of the South African IT industry is women, but also diversity in languages, education and belief systems
Diversity is a multi-pronged issue, need to be patient but not complacent to move the needle forward, give girls the confidence to be competent and to push the boundaries
Frugal innovation in Africa – building technology in a space of constraints such as inadequate power, everything happens by mobile feature phones, needs to be built fast and cheap
Agile in Africa – need to make communities more aware of Agile practices, share what developed world has learnt but also what needs to adjust for the context of the continent
Growth of techpreneurs, expensive to do business in Africa, focus on local market rather than off-shoring to Europe
Andela program – train to be a software engineer, become a fellow and work for offshore clients
Successful conversions from tech hubs to startups is below 5%
Biggest issue is lack of access to expertise in Agile / Lean practices as well as lack of people to adapt it for the continent
Business and architecture isomorphism – if you look at your architecture you should be able to see your business represented in it and vice-versa
Disruption is causing organisations to think about organisational design as well as architectural design
Microservices is a style that is applicable for certain circumstances, it is not one size fits all – follow the 16th rule of Unix programming “distrust all claims for one true way”
For microservices, Amazon and AWS was the game-changer
If you are not building software using the Agile practices these days, you have probably gone down “the wrong trouser leg of history”
Lean Enterprise is an evolution and description of current thinking
Agile methods need to focus on flow rather than scaling and structure
ThoughtWorks Technology Radar – point in time snapshot on what is going on in current projects, throw systematic darts at the wall, vote on over 300 items to whittle down to 100 items,
Agile as a word has become meaningless, don’t follow the off-the-shelf processes, apply small corrections to move forward
Story of Stone Soup is like Agile consultancies, the hard work is done by the companies
Scrum is a good starting point due to its simplicity
Raccoon is a noun, so not a good replacement name for Agile, because you can buy a pound of it
1,000 working on one thing can never be Agile, you have to make enterprises agile before you can run an agile project
The values in the Agile Manifesto hold up well, would have been nice to have had more diversity, had no expectation they were going to create something so significant
The Agile Manifesto was a reaction to the problems in development at the time, maybe something new is required, it would be a tragic mistake to create Agile Manifesto 2.0, we need to ask what is more relevant today to express our frustrations
Agile is a fundamental way of thinking about doing stuff, that’s why it’s important to understand why we are doing it
The Pragmatic Bookshelf was accidental by saying the dreaded words “how hard could this be”, the strength is knowing nothing about publishing, everything was automated unlike traditional publishers and still runs with 2 main employees, now storyboard books like a movie as the reader is on a learning journey
Ruby has a future, but it needs to distinguish itself as a fantastic general purpose programming language, the community is still very friendly and innovative
The emphasis and dogma around testing is off-putting, the amount of effort around many tests are not moving people forward
After many failed attempts to get him on the podcast, Craig finally catches up with Dan North at YOW! Conference on his way out the door to the airport and in a quick chat they cover:
