Apologies for reviving this old thread.
I have been guilty of reviving several old threads recently. My reasoning was basically that the top search hits for a particular question leads to stale/deprecated answers on discourse and my intention/hope was that the answers would be updated and made useful again.
However, not everyone is happy about this.
Having stale answers that are a top search hit is not great IMHO: it takes one click and a fair amount of scrolling, throwing code into the REPL, attempting to understand the error messages, attempting to fix it, before realizing it was a “false positive” and moving on.
I could certainly start a new question, but there are downsides to that too.
The new question would not be competing fairly for the search engines’ attention. By the time it reaches the top five or ten hits, it might have become stale again. Meanwhile the stale threads will still attract attention.
So I start that new question and the answer is that the method has been renamed foo instead of Foo or simply that it has now been moved to the Bar package. I say thank you and feel a little foolish. In retrospect would such an answer not have been better at the end of the older thread? 
This problem does not arise with Python, R, C++, Javascript, because pretty much every question an inexperienced user may have has some updated answer on stackoverflow and is a top search hit. There is no need to visit the Python or R “discourse” for simple, common questions.
I understand discourse is not stackoverflow, but then it does have a “solution” check feature. It’s a great way for the OP to thank a particular answerer, but is it not also intended to serve as a guide for future visitors of the thread? Unfortunately having stale/useless answers defeats the purpose.
Would it violate the community’s etiquette to flag a question with a “answer incorrect/outdated, suggest to close it”?
Should I always start new questions and let stale threads be stale? 