I never paid much attention to the agile movement. I’ve worked with SCRUM and other methodologies and always had some issues with them. They always felt more like a control structure than an improvement by itself. I do remember the first time we used SCRUM. It felt like a breeze. That wasn’t because of SCRUM itself but rather because using any kind of organisational model when we have none, is always an improvement.

But I did feel much constraints using it. I’d say that maybe we never used it properly. But I do feel that SCRUM is great at some things, like managing risk, but not good at others: like productivity. It helps a lot when we need to build tailor made software to a specific client. When we are sharing the risk with the client and the client pays by the sprint.

In this flow, things like demos and sprint deliverables make sure that we’re building the right thing. But my typical experience is different: we usually have a huge backlog of things to do and it’s not so much about the risk of doing the wrong thing, it’s more about the ability to delivery value faster.

Agile in 2018

I admit that I may not have had the best experience with agile methodologies and even that I may not have the best understanding about it. But trying to know more about it and seeing certifications after a couple of days never gave me much assurance that it would bring me additional value (it’s ironic for me saying that, because I always suggest SCRUM certifications to my mentees on their training plans).

The other day I saw a talk from Martin Fowler: Agile in 2018. Even if I don’t follow agile that much, I always try to question my beliefs and try to learn more. So I put that talk on the screen to see what was going on.

And I was amazed. I am completely aligned with everything that Martin Fowler said. The things about agile today being more of a business model and bypassing the agile manifesto agreed with my past experience. And the focus on quality and continuous improvement completely match my work mindset.

I found it interesting that without reading or following Martin’s teachings, I was so aligned with them. That’s when I wondered whether the agile manifesto is not an innovation by itself but rather it’s a side effect of pushing to be better every day as a software developer.

The quest for quality

Quality processes like TDD, RFCs, pair programming, code reviews, having an automatic build pipeline with continuous integration… all these things are sometimes hard to sell. Sometimes I feel that it’s important for a team to feel the pain of bad code before understanding the value of these processes.

It’s hard to buy solutions to problems we never had before.

I also had some resistance to these processes. It took many bad experiences before I started to want something better:

  • Getting a change request that was so hard to incorporate because the code was hard to change
  • Getting bugs all the time, because I though that my code was okay
  • Pushing that quick fix to production that put the application down
  • Doing that fix that was wrong, but the one guy with business context wasn’t aware of it

So many change requests were troublesome. At first I blamed the product people. They should have provided proper specs and if something needs to be changed it’s their responsibility. But after going trough this dance so many times something clicked: I realised that I also needed to be responsible. That it was me that was wasting time and being a bottleneck.

So I started to focus on building code that is easier to change. On how to make code that has high quality and low defects. The technical part comes here. We need experience and training to be better at writing code. We need to decide that from that point on, our goal will be to aim for flawless software.

It’s not just about code

But having code that is easy to change is not enough. Sometimes we have dependencies that we don’t control. Or a change may require modifying database structure that implies data migrations and application downtime. And let’s not forget that simple change that needs to be accounted for on the frontend or by a different team.

Building software is hard. When we realise that the quality of our code is not enough, we try to reach the next level: we need to anticipate what’s coming. It’s not about deep planning. It’s about just knowing what’s ahead and having a clearer picture of the future of what we’re building looks like.

This is why client interaction is so important. We can we iterate daily with out client we can try to understand the end game and prepare our code to be easy to change in that direction. This is another line on the agile manifesto that just comes from trying to be better. We get to some bumps, and we change our way of working to overcome them.

It’s not just about the client

I mean that sometimes we don’t have a client. We may have thousands of them. How do we sit with a client when we are working on a SaaS project that has 10K clients. We can pick some of them and work directly with them, but there’s always the danger of building tailor made software for a subset of our clients.

So we need to adapt. We need to have a way of making informed decisions. We start to gather data and try to get value from it. We need to understand what our customers value and work as a team to deliver value. But we also need to know the direction our company is heading to.

Summary

My point is that maybe if we left several teams that couldn’t communicate with each other, that had a deep focus on continuous improvement and being better, they would reach similar practices to those that are advocated by the agile manifesto. Or maybe they would reach some others. Because each team is unique, because our experience and the type of problems we face force us to see the world from a different point of view.