Agile Testing Days 2016, Day 1

Never have I ever….Been hungover before a conference even started. Oops, guilty.

I have never started a conference on such a weird note. I guess I have to tell what happened. The short version is: IF you cannot handle a lot of alcoholic beverages, THEN don’t drink a lot of said alcoholic beverages. Seems easy enough, right. Somehow, this little bit of reasoning was too much for me to comprehend.

It all started on Sunday evening, in the bar. This conference is secretly starting on Sunday evening, when some of the people already arrive. It was so nice to see a lot of people again, hugging all over the place! Meike even brought her dog, Hagrid. Such a beautiful animal. We played board games all evening and had a good time.

The next day, Monday, some people already started the tutorials. I didn’t have to give one or participate, so I was free to do my own thing. I spent the morning in Potsdam with my boyfriend, and in the afternoon I went for a nice run in the park nearby. I tinkered with my presentation for a little while as well.

Then, before the speakers dinner started I headed down to the bar and that’s when trouble started. Because I’m taking running pretty seriously now (I train 3 times a week, 10+ km) I barely drink alcohol, and if I do, no more than 2 glasses. You see where this is headed, right. In the bar I had a nice beer on an empty stomach. Strike 1. During the speakers dinner the portions where, how to say this nicely, sized for small children. Strike 2. The wine was nice and flowing freely. Strike 3. And then they had open bar with a special Bowmore whisky??!! Strike 4. Then we were back in the bar and a nice British fellow gave me a Talisker. That proved to be one Talisker too many. I was talking to this guy, feeling quite alright and the next second I suddenly got dizzy, felt like I had to hug a toilet and all that fun stuff. I mumbled that I had to bring my coat upstairs and that I would be back. Right. I woke up at 3:30 am with all my clothes still on and a lot of glitter on my pillow. My head felt like it was about to explode. Silly me, I didn’t take a paracetamol at that point, no, I waited for 2 fucking excruciating hours. I couldn’t sleep anymore. Only at 6 am I finally realized it would be smart to take a paracetamol. At 7 am I still didn’t feel any better. At 9 am, I dragged my sorry ass to the breakfast room and spent 15 minutes trying to down a roll with egg. I went back upstairs and slept again until 11 am. That’s when I finally felt a little better.

But I felt so guilty! I missed the beginning of the conference, my live blogging aspirations in shambles this year…Sorry folks! I only saw a couple of talks today, you’ll have to deal with that…

Keynote after lunch:

#Noestimates – Vasco Duarte.

18 months of work, 500 people, an obscene amount of meetings later, a project failed. All because one program manager stubbornly believed the estimates.

Why do we believe in estimations? We assign a fictional number and then pretend like we now have a certainty about the outcome of a project.

http://j.mp/noestimates-time-limited-promo, free book for 24 hours. So what do you do? You can’t have other people stop using estimates. In this book you can read about how you can follow progress in a more ‘realistic way’, while the others can still use traditional estimate methods.

Estimates are always partly a bet. So why do people in the software industry bet all their money on the success of their project (on time, in budget, etc)? You wouldn’t bet your whole fortune on a football match, that’s insane. But somehow we don’t think it’s insane to estimate when projects are done.

With No Estimates, the focus is more about delivering value to your customer, Vasco says. To me, that sounds like the goal of agile as well. So why did we lose sight of this goal, then? Because I have never been in an agile project that was truly, only focusing on customer value. I too, have spent ridiculous amounts of time in planning meetings. They get me raging internally as well.

Principle 1. “We’re not nervous when things are on time”. If you do not trust your process, do not try to improve it. Build a new process!

Principle 2. Shorten the feedback cycle. This principle seems fairly obvious to me as well, another core thing of the agile manifesto. The idea is very true of course. Fasten the feedback cycle and you will have information.

Every user story is different. Even a story that seems simple from the outside can have ‘accidental complication’. There might be horrible secrets in the code. So, you have given it 2 points, but now it’s….10? 100? “Relative estimation is bullshit!” So, Principle 3: Believe the data, not the estimates.

Now he says something really important. We should test for value! That’s tricky, because you cannot plan for value. You need to put something out there and measure your success. To do that, you need short feedback cycles, so you need small bits of new software. We need to stop focusing on new features, and start focusing on how to add value better.

I’m going to leave it at that, because this last piece is all you need to remember. Question I have is, how are you going to do that? You can strive for this noble goal, but what if the rest of your company is still working the old-fashioned way: feature driven, estimate driven and using KPI’s and all that jazz. One person alone isn’t going to change that? This keynote was fun, but has left me wondering: what can I do to change the way of working at a huge company?!

Break. I took another hour nap again and now I feel sort of whole again.

The lightning talks. I decided to go see some people I like a lot. They have to deliver their message in 7 minutes, so they have to be concise. Since I have about the attention span of a squirrel at this moment in time, that is great.

Gil Zilberfeld – Simple

“It’s quite simple because it talks about complexity”, is how Gil starts. Okay! Every project starts with a simple idea, but over time more features are added, it needs to go faster and that slows the team down: it’s now complex. What do you do then? Embrace change, is what the agile people tell you. So, we try to simplify things in processes (like Waterfall, ewww). The agile process was born. Easy to explain, hard to do. Scrum: easy to explain, hard to do.

How do we shoot ourselves in the foot? We simplify everything (user story format, for example) on the one hand, and we give ourselves a ridiculous amount of options on the other hand (number of programming languages, anyone?).

“Bugs give us meaning in life and time to spend with eachother”. Hahahah, omg I die.

In the end: try things, but keep it as simple as you can.

Great talk by Gil! A lot of info and humor packed in 7 minutes!

Mark Winteringham – The ATDD Paradox

Hungover, but standing strong, he is going to deliver his talk.

(That was actually quite funny. I saw him coming downstairs and just one look at him and I knew we were in the same boat, Boatie McHungoverFace. Only his boat sunk even a little deeper than mine).

One of the reasons we have acceptance test is as a guide, to keep us on the right path. We can do that with tools, of course. We have automated systems to check if the application behaves as we expect it to.

Can you verify business expectations with automated tests? That is the paradox. TDD and ATDD are trying to solve the same problem with a different solution. You have simple organized systems (units) that you can test relatively easily. But on a bigger scale, you have more complex systems (the units come together). We are trying to do the same tactic as TDD on the bigger scale and that doesn’t work as well.

ATDD is not your test automation solution. It’s a guide about what you are building. Other testing is still required: exploratory testing or other types of automation in testing. Orchestrate your testing activities apart from the ATDD thing.

Whew, a LOT of information in 7 minutes. I feel this talk could benefit from more time!

Lalitkumar Bhamare – Thinking about session based programming

Many of us have heard of session based test management, but not about sessions based programming. It is using time boxed sessions for programming. The sessions have different types: analysis session, bug fixing, pair programming, coding for a story session, etc.
It helped his team deliver an estimate to the stakeholders/management. Quite a bold and daring claim after the keynote on No Estimates, me thinks.

I do believe him when he says that using the sessions help them with focus and measuring of some sorts, but with estimation? You don’t know how complex the problem is you’re trying to solve, so if you say ‘I need 3 sessions for this’, is still an estimate.

Alex Schwartz – Forget about redshirts, here come the tribbles

Link to Star Trek! Treat your servers like red shirts: they should be expandable. Here come the tribbles. Tribbles in Star Trek have a calming influence on people, but they reproduce at a terrifyingly fast rate. What is the analogy here? The tribbles stand for micro services.

The red shirts had a name, they were people. The tribbles don’t have a name and they don’t need much. But what are the costs? It’s flexible, but lower level of control.

I’m quite sure if you read this it doesn’t make sense at all. This talk was very hard to blog about, the pictures in his presentation made it clearer!

Melissa Pontes

She is talking about how she includes people with disabilities in her testing activities. The blind and deaf people are often not accounted for in software. They can barely use it, and those people can never be a tester, right? Her team went away from the view that these people have a disability, they just have another view on the world. They tested with deaf people, blind people and people with an intellectual disability. They wanted to include those people so they can also test software. A very noble goal!

Breaktime!

During the break, I had a great conversation with Gitte. She was in need of a massage and we also ended up talking about the different types of people and their needs. Confession time. I’m an introvert and although I love this conference and the people, I need some quiet time. I almost feel guilty about taking this quiet time, because I’m missing out and I feel like I’m the odd one out. Gitte told me that I’m not the first one who talked to her about this and she is part of a movement called ‘inclusive collaboration’. It’s aimed to take advantage of the strengths of people. Introverts have certain strengths, depressive people have strengths, and so on. Instead of labeling those people, we need to work together in a way that everybody feels safe. I was intrigued! It’s on my list to explore this topic more, so this talk was a great demonstration of how serendipity plays a role in our lives. I’m going to ask Gitte for more information and we’ll see how this plays out!

Now onto the last keynote of the day.
Michael “The Wanz” Wansley – From waterfall to agile, the advantage is clear.

This guy can sing, man! The music was failing so he just did it a capella.

He did a lot of different jobs. He’s a QA engineer now, but he also worked in a record store, delivered hair products. He’s gotten into testing because he was working with Windows95 and was trying to figure out why some things work and some things don’t. He even worked for Microsoft for some years after that!

“Software is meant to be transparent. It is supposed to be a tool so the user can accomplish what they want.”

He grew up in a waterfall system, where we wait and we wait and we wait. The process is simple, but the world never works that way. You end up with a bucket of unfixed bugs.

After his career in hip-hop, he’s now back working as a tester. He now works agile! He likes it, releases are faster etcetera. His story telling skills are great. It’s very easy to listen to him, I even forgot to blog. The content of his story, I have to say, is not really ‘new’. I mean, it’s 2016, why are we still talking about how agile is better than waterfall. So, this keynote has very high entertainment and ‘ah, yeah I recognize that feeling’-value, but new stuff? Not so much. And that’s okay. Some people can just mesmerize you with their story telling skills. That alone is something to admire. I hope he inspired people. Even if your message isn’t new, if you find a good way to tell it, people will listen.

And that was the end of the official program of the day. The organization really outdid themselves this year. Outside they built a miniature Christmas market. And, they even had vegetarian bratwurst, for which I am SO GLAD!! I also tried the pancake with applesauce, another brilliant invention. It was really nice to stay near the fire. Met some new people there; had great convo’s.

Then it was time for the dinner! I wasn’t even hungry anymore, but I ate two starters anyway. Great vegetarian options again. We had a nice table with Oana, Göran, Wim, Eric, Bart and some more. I had great fun, but I was also looking forward to the award ceremonies. I have to say, the outcome for both was absolutely perfect. Maaret, you deserve it!!!

And I’m also very happy for the Pan Galactic Gargle Blasters that won the software testing world cup. The look on their faces just before the announcement was priceless; they looked like they were in great pain. Btw, they also had the greatest costumes, scary ghosts! I knew it was them but they closed me in and that was very scary anyhow! Great fun, I was so happy for them. They even played the Netherlands anthem! All the Dutch people in the room stood up to sing it…haha. We might never win a cup in football, at least now we can be proud of our testers.

After dinner, I did the escape room with my team from last year, only this time we failed. Boohoooo. I had nice conversations with people, danced on some songs and now I try to be sensible and go to bed. I didn’t have a drop of alcohol, btw. Tomorrow (Wednesday) is my talk, so don’t expect any blogging in the morning. I will pick it up in the afternoon. Time to get nervous now,

Cheers!