[kwlug-disc] La-la land, how to recognize?

[Mikalai Birukou]

> At the other end of the spectrum, you have hacked together a solution that barely works, without any clear idea (or architecture) for what you are doing. Further progress increasingly becomes a technical nightmare as you layer kludge on top of kludge, using an unplanned software architecture that doesn't match the problem you are trying to solve.
>
> Somewhere in the middle, between this pole and la-la land is good clean code with a good clean architecture that is easy to maintain and extend. At the extremes you are battling different kinds of complexity. The la-la land complexity is a tower of unnecessary layers and abstraction.
>
> I do know that design methodologies can help to produce better solutions than you would find by programming blind, as long as you understand the strengths and limitations of the methodology and exercise good judgement. Just because a certain methodology requires an investment in time in order to learn and master it, doesn't always mean it is bullshit. Some methodologies are better than others. It's definitely true that design methodologies can be abused to create overly complex la-la land horrors. Which I recognize as involving unnecessarily layering, complexity and abstraction that make it harder to understand and maintain the code. I recognize there is a problem by looking at the resulting code, not by a-priori judging the methodology.

Wise: judge architecture by how convenient/useful it is in development, 
i.e. how good is actual dev experience. 👍

> I have worked on REST architectures (a long time ago), but I have no experience with HATEOAS and therefore I can't meaningfully comment on it.
>> How do you recognize when you've drifted into architects' la-la land?
>>
>> Let's put focus emphasis with capitalization: Architects' La-La land. 😢