Box box cylinder. Just to point out you're just talking box cylinder box. The cylinder is turned on its side is all. If we ignore the many levels of pain that is error handling in EDA, or non functional/cross cutting concerns or it not being very useful for organisations used to request/response or just the general messiness/lack of any sort of tooling to help manage it. And there you find the EDA proponents sitting atop a different ball of "mud" defending it. Service oriented works for some, event for other, bulk/batch for other. No one size fits all.
Well, every architectural pattern has defenders; you're clearly a defender of SOA -- which is not, in itself bad. Pointing out that EDA proponents (like myself!) are passionate defenders of it isn't really an argument, just a fact. Every design pattern has pro's and con's. The fact that I have seen is that technical complexity is what kills tech orgs. The ability to move chunks of functionality to different places, to easily refactor code by simply changing who listens for events and how they react, the decoupling benefits -- these all, *if used correctly*, can be tremendous slayers of the dragon "technical complexity." The biggest resistance I've seen to EDA is one you mention, orgs "not being used to" EDA's. Everybody is used to the "big ball of mud" -- we're comfortable with it, and never stop to question the exponential growth of tech debt and complexity that seems inevitable as a codebase grows. What EDA proponents are saying, and which after writing software for 40 years I'm coming to believe is: it doesn't have to be that way. If killing complexity requires a bit of cognitive discomfort during the transition, I personally think it's worth it.
![Chicken P - People's Favorite (Remix) [Feat. 42 Dugg] (Official Video)](http://i.ytimg.com/vi/a6Bn8FEtxmY/mqdefault.jpg)
"There's no problem that can't be solved by adding another layer of indirection - except for the problem of having too many layers of indirection."
Here's mine: "There's no problem that a well-trained software engineer can't cause."
@@-taz- hahaha
You have a problem. So you pick microservices. Now you have N^2 problems.
:D
Nice video
Box box cylinder.
Just to point out you're just talking box cylinder box.
The cylinder is turned on its side is all.
If we ignore the many levels of pain that is error handling in EDA, or non functional/cross cutting concerns or it not being very useful for organisations used to request/response or just the general messiness/lack of any sort of tooling to help manage it.
And there you find the EDA proponents sitting atop a different ball of "mud" defending it.
Service oriented works for some, event for other, bulk/batch for other. No one size fits all.
Well, every architectural pattern has defenders; you're clearly a defender of SOA -- which is not, in itself bad. Pointing out that EDA proponents (like myself!) are passionate defenders of it isn't really an argument, just a fact.
Every design pattern has pro's and con's. The fact that I have seen is that technical complexity is what kills tech orgs. The ability to move chunks of functionality to different places, to easily refactor code by simply changing who listens for events and how they react, the decoupling benefits -- these all, *if used correctly*, can be tremendous slayers of the dragon "technical complexity."
The biggest resistance I've seen to EDA is one you mention, orgs "not being used to" EDA's. Everybody is used to the "big ball of mud" -- we're comfortable with it, and never stop to question the exponential growth of tech debt and complexity that seems inevitable as a codebase grows. What EDA proponents are saying, and which after writing software for 40 years I'm coming to believe is: it doesn't have to be that way. If killing complexity requires a bit of cognitive discomfort during the transition, I personally think it's worth it.
You just coined "Big Ball of Dung"
awsome