I don't know that either of those are actually problems in my mind.
- If during that payout cycle, the 'historical' content is getting more traffic/votes from users, then why would it matter? The community itself should be deciding where the allocation of the rewards pool is driven.
- I actually think we should go one step further and let user's vote on a single post multiple times, but only during different payout cycles. If I upvoted a blog post a month ago that somehow makes it to the front page of trending again, maybe I want to vote on it again, and give the author another little boost. I may also want to flag it in it's second cycle, which this would also allow me to do.
Both of these situations can already occur in the current system by simply reposting the same content in a new post. These wouldn't be new problems, it would just prevent bloat in the blockchain by encouraging repeat behavior on the same content, as opposed to reposting :)
Another reason to vote again on the same content is that it may be updated, expanded or otherwise edited. A good part of the the idea of allowing edits forever is to encourage maintaining/improving existing and reducing the need to repost it.
[nested]
That isn't possible. The decision to "activate" a post would be a consensus decision so the voting that led to that happening would have to be part of the consensus state.
Interesting view.
Another part that I've heard though (don't know that it still applies) is that there is a cost to keep all of the active posts active as part of the voting consensus.
@timcliff Yeah I've heard that as well, though I imagine posts wouldn't be considered "active" unless they received enough votes to "activate" the post itself. I hope with some of the recent optimizations (moving out of ram to disk, the payout to rewards balances, etc) that a hit like this wouldn't be a reason to deny the change.
There are definitely some potential hurdles/problems, but I think the pros would outweigh the cons and shift the overall dynamic in a positive direction. The life cycle of a blog post is much different than what our payout cycles currently allow and we should try to do something to encourage long term content.
Infinite payouts causes bloat in the memory size of
steemd. RAM is several orders of magnitude more expensive than disk, as you know.