Hammock thoughts

After trying to implement the whole system in just my had, I decided to run it by my team. This was a very humbling experience... But it did remind me the importance of surrounding myself with good people. After digging through my thoughts and ideas, they slowly began to crumble away. Although it felt like weeks of my life being torn away slowly, I was happy :)

One of my teammates asked at a certain point - why do we need this part?

Obviously, if we want all the teams (we are currently 6 in the RND) to easily share this part then...

Then I started to hear myself...

I decided to try to find what is the MVP of my architecture plan, and build from there. I can't possibly design something i don't yet understand. So I decided to try and think what would bring the most benefit at the moment.

Although I'd love to explain the MVP and how we continued development from there, but the important lesson here is get out of your head. Designing an architecture is meant to help and serve the team and everyone working on it. So it would seem odd to design it without them. Ask your team, consult with them, and although we heard this mantra way too much - create an MVP, and iterate from there... As much as I hate to admit, it works :)

Due to the odd state of the project - one huge experiment - I thought that consolidating the logic and try to bring it as close as possible to were it is used. The main objective was to separate everything into modules. I have an issue with repositories that are build around file types. Files such as reducers, components, routes, etc. feel odd to me. In most cases, when I fix something, I'll have to edit the action, reducer, selector and possibly the view. Why then should I travel through the whole project for such a simple task? If I had a module, which includes its own action, reducer, components, and anything else it needs - enhancing it would much simpler.