There's a kind of progression, ideally.

• Best case scenario it can be achieved through a data-driven workflow, and designers can do it themselves with tools.
• Next up, the desired effect needs some plumbing to make it work, or you don't have a data-driven workflow, and so you need a programmer to be involved -- this is one of the jobs of the gameplay programmer. In theory programmer/gameplay programmers are really doing everything that makes the game a game -- everything that exists between the low-level rendering, input, system abstraction and other systems, and the designers.
• Finally, its possible the engine API makes it difficult or impossible, so you need to pull in engine-level programmers to add support inside the engine.

The idealized model would be strictly tiered, and no tier would ever do the work of another, but just enable the other tier to do what they need by supporting their needs, but there tends to be some role-overlap and plenty of schedule-driven or pragmatic compromises in reality.

Meh... usually designers are the ones tweaking values, not programmers. At least, that's how it works if you have reasonable tools instead of a crap workflow :-P

It was a stage, fortunately.

Edit: To elaborate, I'm now doing engine programming instead. I appreciate it because there are objective measures on how good my work is. If I reduce CPU usage on the MMO server I'm developing for a certain amount of players by 50% without making the code unreadable or introducing weird bugs it's an undeniable improvement and I will not be asked to revert it. When I was doing gameplay programming for a multiplayer online game it would frequently be like "this character needs shoes, please change the Unity prefab object and scripts to accomodate the new model and attachable objects". After being done a few hours later: "revert the change, it didn't look as good as we thought".

Doing gameplay programming strictly as a hobby again.

You were doing it wrong, that is not the job of a programmer thats a character artist and a scripters/designers job to make that change happen. If it needs a programmer your pipeline needs some refactoring.

This is way I like systems work you will still be in the trenches but you get to play arround with the hardware and new API's as well :)

I mean, it should be like you say. Unfortunately the organization overall was inadequate. Not that I blame them, it was their first game, and managed by people without direct game experience. What could possibly go wrong with multi-national multi-continent cooperation between a tech company and an education company. ^^ I'm not angry, those were simply unfortunate circumstances that I accepted at the time of sign up. Because they/we didn't manage to hire a single competent designer all programmers were pulled into direct scripting, level development, and such in addition to engine development. Quite the experience.

Still, I strongly suspect the issue I described in the post you quoted is still more prevalent in the gameplay programmer profession than the engine programmer profession. Not only from my limited own experiences, but also from what I've heard from other programmers.

It depends I have done all kinds of work when I was dedicated Front End programmer, for which the work at the time ranged from pipeline work, back end systems (atlas texturing, render to texture tech), to implementing functional sides of UI screens. However the only time a programmer was interfering with art work would be at the structural level, as in how a scene has to be put together, this was a drawback of the tech. And when a scene contained multiple child anims max only allows one anim line whilst our tech allowed for 4, also drawback in tech, that would end up in not to complicated xmls but complicated enough that everyone forgets who it works until you have to create one again.

We never really got involved with the look or real feel of it unless it would break a usability thing or a tech constraint.

Just wondering but what do you class as engine programming?