summarized: never-ending micromanagement is not fun.
I can stand behind this 100%. I'd like to see improvements to the features that already exist, allowing them to be actually utilized. This is a thread for negativity, so keep in mind that I'm a huge fan of the Toad and his work while you read my complaints.
I advise skipping the examples unless you read the rest and think I'm exaggerating or over the line. If so, then come back and read them, and see if your opinion has wavered nearer to mine.
There is now a reasonably deep system of genetic heritage and varied genetic traits, however you can't breed a pack of extremely hardy, powerful animals (of any type) because you only get vague descriptions, and unlike a real person your dwarves cannot make relative judgments of size, strength, or health. So you'll never know why that "incredibly muscular" dog is easily outpaced by the dog "belarded with rings of fat". It would probably take about an hour to fix this.
You can build a mighty tower, an epic creation, but designating it in-game will probably take about as long as building a real one would take. Position pointer with many keypresses. 9 more to lengthen wall to maximum length. Move down list to desired material. hit 10 times. Repeat. This is a not-intractable interface problem. A MAJOR holdup to many people.
Another: you can pump magma and water into a giant scaffolding (assuming you get it designated without giving up or having a life) to create a magnificent cast obsidian sculpture, but several bugs with dwarves not wanting to perform their jobs, flows being inconsistent and inefficient, and so on will make it more of a headache than a grand accomplishment. Instead of making it more workable, Toady saw gaps between reality and simulation and handicapped the player with some nerfy changes.
In summary: I see the endless micromanagement issues largely as a result of underdeveloped features. Toady's so quick to put in cool new doodads that he doesn't ever seem to get around to fixing the old features properly. His response (after two years of the same version and same bugs and bugreports) to the plague of the worthless cattle who only served as a cpu drain and annoyance in your forts (no milking, butchering only a stand-in, etc) was "Oh, I think I may or may not have done something about that. On review, realizing my fans/testers are really plagued by it, let me take 2 weeks of programming and take it from a stand-in to a full-blown amazing banquet of features. Now, those features are largely useless due to various small bugs, but I have a really cool new idea so I'm going to do that now."
It's amazing to me to learn that a couple of days are all that is required to flesh out problematic features from annoyances to joys, yet this is never the priority. I understood the mentality of add-it-all-then-fix-it-later before, but now I see the truth: the features to be added are almost limitless, and we're approaching the point where the number of incompletely implemented features will make the game truly unplayable for even the most hard-core fans.
It passed that point already for me. If I can't get a leak stopped because no dwarves will take a job to disassemble a gear, and I can't tell if it is a bug or job problem because idle dwarves held up from work by the leak are constantly trying to throw parties, and I can't keep the fort under control because buggy programming causes children to walk around drowning themselves, and ...
then I have to sit back, take a moment, and think: this isn't a challenge; it's not a game; it's a debacle. It's more frustrating than work, which I actually enjoy. I used to enjoy DF, and I'd like to again. It's such a wonderful compilation of imagination and potential.
Anyway, if I ran DF it wouldn't be anything on the grand scale of Toady's imagination. Then, again, it would be highly playable. So I think the scale has just tipped a little too far the wrong way. I'd like Toady to take a little bit to catch up with himself, get the current features in order, then proceed with adding his new ideas. But we all know what gets him excited about the game, and it's implementing new ideas (not fixing the old ones!)