The fundamental issue with all of the above is that unlike other games, for example Dual Universe, EVE was only initially designed to fly spaceships, blow them up, make a little isk - rinse and repeat. Once people figured out how to be awesome at and with that, and how teamwork hooked right into enabling more fun, people started to make stories, even mythology, and CCP began to cater to that.
The result of this was that it wasn’t the mechanics of it all that was driving things, it was the stories being made. Not only did this have fundamental impact on shaping the game itself, it also provided the triggers to grow the game. When stories are made, people want to tell them.
In other words: what made EVE work and (!) what made EVE grow was behaviour. That was the driver of everything, including the development of features, mechanisms and later on even game niches.
Note how different this is from CCP’s focus on model changes after the Summer of Rage. But that as an aside, but do not underestimate the impact on foundation elements of EVE.
Anyway 
Since both the container and the mechanisms and the niches are rooted in behaviour, EVE is not a game suitable for a programmable approach, no matter how vulnerable it is to this (especially so). Anything that compromises those foundation elements immediately instigates perception problems (first) and stimulus problems (second).
Every time EVE had a bit of a stumble, you can backtrack the pattern to exactly that. Here’s the kicker, when dealing with a programmable environment perception problems matter little. When dealing with a behavioural environment they determine survivability of said environment. You could replace the entire population with blank slates and you would still face that challenge.
This presents issues, obviously. It puts CCP in the equivalent of the eternal race between the bullet and the armor (and not just in relation to programmable actions / environment interaction). Invent new armor, the botmaker comes up with a new bullet. And so forth. As they said, they cannot win it. They can only prevail facing it constantly.
Letting go of those root / foundation concepts is pretty much impossible. In spite of CCP’s own attempts, EVE’s way too complex, too vested, and unless they kill off every pilot and seed the universe with new babies they can indoctrinate for 5 years there’s no changing this.
Yet at some point, and I hate to say this, CCP will have to consider options (beyond the F2P / historic trauma related focus of strict technical approaches). Unless, and this is the real benchmark, CCP find a way to not just get more accounts, but to get more warm bodies.
Since F2P a character has become a lot less than an identity. It’s become a tool. Interexchangeable, customisable, disposable. This diminishes the foundation of behavioural psychology related to creating and sharing stories. One of the biggest noticeable and measureable identifiers of this is the direction and cohesion of segment / subgroup communicaties in a population. It’s become inward bound, and fragmented. CCP has made this worse, after the Summer of Rage, through policy. Smart policy at the time, necessary even, but it has gone way too far.
As identity diminishes, so does association, so does participation. Sure, it provides brilliant hooks for gratification triggers (shortcuts, F2P), but that is walking a very very fine line in regards to something we touched upon earlier, in spite of content additions everything on a behavioural level has been done a million times over again and has been warded against. EVE has become conservative, conformist, in terms of social psychology. Funny bit, both former CCP’s Zulu and Seagul warned against this respectively.
So what you have is a virtual universe founded on and standing or falling by behavioural routines where perception problems are the prime cause of retention/acquisition/resource allocation problems. Inserting a programmable game design focus into that is lethal.
It’s where Dual Universe has an advantage, beacuse they get to learn from the best and worst practices of others, including CCP, as well as the evolution of technology and behavioural interaction with it. But it will result in something which is not EVE. It won’t be life, it will be a game. As CCP probably knows, as they do have their eyes on the inside
I get what you are saying, honestly I do. You do touch on issues which need tackling. But the technical model approach of game design is inherently flawed as it encompasses less than half of the full picture. Opening the door under such circumstances for what constitutes options of / towards a programmable environment (means, mechanisms, not “environment” as we know it since EVE doesn’t really have that, see the earlier discussion on entropy / cataclysm / environment) is extremely dangerous.
Yes, I agree that the current situation, and the currently present patterns, provide a set of perception and actual problems which are slowly but certainly impacting foundation elements of the dynamic. But the angle of approach to that challenge, should not be automation.
To be blunt: CCP needs to find ways to stimulate non-conformist behaviour, to reaffirm process of identity and association, to unify communication pathways, to insert behavioural triggers for that to become outward bound once again - ideally at minimum chaos actors / options for all of the things individuals and organisations have warded against in their by now conservative behaviour.
To a degree CCP is doing this. But in a very limited manner and scope. It is merely sufficient to provide a technical monitoring output, not a behavioural. NPC Miners, Pirate Strongholds etc. The F2P focuses is dominant in every bit of prioritising and resource allocation. There’s a bit of irony here too, players mirror CCP, CCP mirrors players. Conservative, conformist.
Someone’s got to be, what in my field is known as, the chaos actor with vision and presence. Human dynamics thrive on it. Surges in the real follow the upheaval, but this is the virtual. Behavioural or not, this can be planned and plotted for