If this is the plan -- and you bring up lots of good points -- there should be a date or criteria for elections announced up front. For example, something like "we have elections 3 months from the date we hit X number of subscribers."
I'm honestly still not sure when that would be. I'm still trying to process out the long term effects.
Look at wikipedia as an example of how organised, committed groups with nation-state backing have successfully created cabals of editors that liberalise all the content or engage in editing and revision wars in order to press the political goals.
It is an incredibly dangerous path and it fundamentally cedes power to those with money and time to commit to mass manipulation of the service. Now, obviously that's unlikely in the short term but if you're planning out the future path of this organisation you need to be looking at examples of past mistakes for hot NOT to do things. You also need to not be naive about the level of power and commitment they have, look at how reddit struggles to oppose manipulation, twitter bots, and the like. This is all in the sites future too, we can't naively believe that the userbase will just magically defend against such a thing through literally no means that they have to do so.
It needs considerable thought and we need people to write theory for it. I'm not joking.
Recall elections (maybe with a high threshold required to remove a mod) might be a viable solution. Say you have:
An annual "airing of grievances" thread where -- if the community has a problem with a mod -- they can petition for a recall election for that individual mod. You need to get some percentage of the community to upvote a petition for each mod they want to vote on.
If a petition successfully gains the requisite number of upvotes, you automatically have a recall election for that mod in 1 month.
It takes a significantly higher threshold of upvotes to actually remove the mod via recall.
You could limit voting to users who have, say, at least 6-month-old accounts and 50+ comments. For whoever is nominated to replace a recalled mod you could require a more extensive history.
There's no foolproof way to prevent bad-faith actors from taking over, but you can make it a lot harder without dreaming up something too complicated.
If this is the plan -- and you bring up lots of good points -- there should be a date or criteria for elections announced up front. For example, something like "we have elections 3 months from the date we hit X number of subscribers."
I'm honestly still not sure when that would be. I'm still trying to process out the long term effects.
Look at wikipedia as an example of how organised, committed groups with nation-state backing have successfully created cabals of editors that liberalise all the content or engage in editing and revision wars in order to press the political goals.
It is an incredibly dangerous path and it fundamentally cedes power to those with money and time to commit to mass manipulation of the service. Now, obviously that's unlikely in the short term but if you're planning out the future path of this organisation you need to be looking at examples of past mistakes for hot NOT to do things. You also need to not be naive about the level of power and commitment they have, look at how reddit struggles to oppose manipulation, twitter bots, and the like. This is all in the sites future too, we can't naively believe that the userbase will just magically defend against such a thing through literally no means that they have to do so.
It needs considerable thought and we need people to write theory for it. I'm not joking.
Recall elections (maybe with a high threshold required to remove a mod) might be a viable solution. Say you have:
You could limit voting to users who have, say, at least 6-month-old accounts and 50+ comments. For whoever is nominated to replace a recalled mod you could require a more extensive history.
There's no foolproof way to prevent bad-faith actors from taking over, but you can make it a lot harder without dreaming up something too complicated.