« first day (2808 days earlier)      last day (2200 days later) » 

mkt
12:09 PM
I spend a fair bit of time hunting for the proper duplicate to flag questions with, and have started to bookmark a few of the 'canonical' answers for quick access. I wonder if there would be any interest in a meta thread where we organise some of the best answers on different topics?
I realise one can get some of the same information by viewing the top-voted answers under each tag, but having a single page listing a bunch of them could be useful.
I'm ambivalent because it seems a bit against the spirit of CV, so I'd like your opinions. Or if you have better strategies for keeping track of the 'best of CV', please let me know
One option is to star questions for this; at present I use starring to keep track of a wide range of interesting threads.
 
 
1 hour later…
1:16 PM
@mkt It's a great idea. Having embarked on such projects before, I just want to say (a) it's far more work than it looks and (b) it can be exceedingly complex to maintain. If you keep your ambitions low it could be very helpful. Have you considered maintaining just those canonical answers that are difficult to search for?
One example of the latter is that we must have at least 20 good demonstrations of how sums of squares partition, but they're impossible to hunt down because they're almost purely mathematical in content. I would love to have a handy reference to one standard demonstration so I could close each repeat of this question.
 
1:34 PM
@mkt, I think that should be discussed on meta, rather than here. That said, I agree w/ @whuber. Much of this can be done by looking at the threads for a tag sorted by votes or by frequent[ly linked to]. Beyond that, you can favorite the ones that are otherwise hard to find.
 
mkt
Thank you, @whuber and @gung. You're right, this would be a large task and I also don't have the expertise to judge which the best answers are in most areas. So I was thinking instead of a community wiki on meta. I'll start a thread later and see if there's sufficient interest in this.
 

« first day (2808 days earlier)      last day (2200 days later) »