Is it okay to majorly re-write a question that already has answers?

Question

I've posted a question here that has spurred much debate and made me wish it could be worded differently for clarity. It's already quite long so if I keep adding to it, reader fatigue will set in. Unfortunately so many answers take direct quotes from the question itself that even a small re-write would make whole answers completely invalid, despite their already huge amount of votes. It doesn't seem right to just abandon the question and ask a new one on the same topic, but it also doesn't seem right to "ninja" the validity away from those already posted answers.

I also don't feel that the question has been answered to the standards I set for it, and the re-write would allow me to define things in a way that may be more accessible and understandable. However, much of the debate that has happened on it has progressed the topic and it would be a shame to lose the history of that conversation by simply dropping it for a new question, or heavily editing it so those conversations become out-of-context.

What's the best course of action here?

Answer 1

Invalidating existing, good answers is a pretty big no-no on Stack Exchange. As noted by others, it is disrespectful of others' efforts and could turn them off from the site. What is a "good answer" in this context? I'm not sure if there's official guidance, but another site treats any upvoted answer as being worthy of this kind of consideration, taking its inspiration from the fact that you can't delete a question if it has an upvoted answer.

This is an idea I want to float, not moderator policy: If you can't fix your question "in place" without either damaging existing answers or forcing their authors to do more work, how about closing the existing question, asking the new one, and notifying all the answerers with a link so they can adapt their answers to the new question if they like? If they don't then ok, you don't get an updated answer -- but at least you haven't taken anything away from the people who wrote them. If people do adapt their answers and end up getting votes on both, so what? In a sense they answered two different questions, so that seems kosher to me.

Answer 2

Make as few changes as possible and be respectful of other's time. If you're just editing for clarity and not changing any of the essential parameters of the question then I think you can get by with lots of changes. However, if essential parameters are changing then make as few as possible.

The fundamental issue is respect for the people who have answered. If you essentially state "Screw you, I don't care about your time investment into this question" by changing essential parameters without telling them about it, that's just inconsiderate. But if you communicate with that that a few essential things have changed about the question and they might need to adapt their answer, then that's okay.

I'd be angry if someone made a huge change to a question I had spent a lot of time on and didn't tell me about the change. Especially so if someone who came later got upvoted more than I did just because they answered later. I don't want to look like an out of touch fool because the OP changed and I didn't know about it.

Answer 3

The issue you mention has happened before, at least once that I can think of. Specifically with this question. The OP kept changing the question so much a diamond moderator closed it. When the question was finalized, the OP notified everyone who had already provided an answer.

I don't see any reason you can't do the same: make your changes and notify everyone who has already provided an answer that the question has changed. Each individual can then decide whether to update or delete their answer, or simply write a new one.