I don't think limiting groups is the way to go - if groups aren't active and don't thrive, that's fine, let them die off. But creating groups is useful in so many ways - not everyone wants to...
I don't think limiting groups is the way to go - if groups aren't active and don't thrive, that's fine, let them die off. But creating groups is useful in so many ways - not everyone wants to discuss the same things, and if we wait for things to be voted as popular, niche subjects will never get their own corner of the site.
StackExchange has an interesting process for proposing new StackExchange topic sites. It's definitely way beyond what we'd need here, but there might be some interesting things to learn from it:...
StackExchange has an interesting process for proposing new StackExchange topic sites. It's definitely way beyond what we'd need here, but there might be some interesting things to learn from it: https://area51.stackexchange.com/faq
Yeah, I agree with this. Making sure there's interest in a group before creating a group would make sure that there's no dead groups (at least to begin with).
Yeah, I agree with this. Making sure there's interest in a group before creating a group would make sure that there's no dead groups (at least to begin with).
Maybe I'm misunderstanding what you mean, but for example, this post has a "baseball" tag, and if there were a lot of similar posts that could imply that we need a ~sports.baseball.
Maybe I'm misunderstanding what you mean, but for example, this post has a "baseball" tag, and if there were a lot of similar posts that could imply that we need a ~sports.baseball.
That's essentially what is happening already. All the groups that have been added so far were from people expressing an interest in having that group exist and people voted those comments up and...
That's essentially what is happening already. All the groups that have been added so far were from people expressing an interest in having that group exist and people voted those comments up and adding their own suggestions. At some point it may be a more automated process (e.g. Trusted Users being able to vote on subgroup creation, or automatically creating them based on tag usage statistics) but until trust is in place I don't see a straight democratic process as being a good idea... otherwise groups that run contrary to ~ goals (quality, primarily text-based content) will likely pop up over and over again just because they are popular (e.g. meme focused groups, or even hate groups).
No it doesn't. Reddit only requires your account be at least 30 days old and has some positive karma (which is trivial to get) before you can create a subreddit. That is not even remotely the same...
No it doesn't. Reddit only requires your account be at least 30 days old and has some positive karma (which is trivial to get) before you can create a subreddit. That is not even remotely the same thing as what trust/rep will be on ~, see here: https://docs.tildes.net/mechanics-future#trustreputation-system-for-moderation
Will there be any sort of limitation on how many ~ people can moderate? One of the issues I've had with Reddit is many subs on a certain subject share the same exact moderation team. Take Nintendo...
Will there be any sort of limitation on how many ~ people can moderate? One of the issues I've had with Reddit is many subs on a certain subject share the same exact moderation team. Take Nintendo for example. Nearly all of the Nintendo related subs share the same moderators.
Most likely, yes... though to what degree I don't know. I imagine it will be achieved automatically through trust decay (so it's simply impossible to gain and maintain enough trust to get...
Most likely, yes... though to what degree I don't know. I imagine it will be achieved automatically through trust decay (so it's simply impossible to gain and maintain enough trust to get significant power in too many groups) or as a hard cap setting but nothing is really decided yet.
Trust is going to be a complicated system requiring a lot of fine tuning but generally if it's a legitimate problem on reddit it's going to be attempted to be addressed here. And power mods and mod squatters are a recognized problem IMO. Also worth keep in mind is that the ideal here is every active user in a group slowly gains trust and gains access to "mod" like tools so rather than mods making up 0.X% of the community they will likely take up a significant portion of it so there will be significantly more users with powers and with the ability to monitor others actions to hold each other accountable.
You're right, but trust is also per-group. How would you tally trust? Do specific groups count more than another? If someone has a high trust score in ~talk but a low trust score in ~politics,...
You're right, but trust is also per-group. How would you tally trust? Do specific groups count more than another? If someone has a high trust score in ~talk but a low trust score in ~politics, should they be able to create a group?
In general through valuable activity in the community (proper use of "mod"-like actions, valuable contributions, etc), all of which will be judged by other trusted users in order to hold everyone...
How would you tally trust?
In general through valuable activity in the community (proper use of "mod"-like actions, valuable contributions, etc), all of which will be judged by other trusted users in order to hold everyone accountable. The exact mechanics are still be worked on though and at some point will most likely require user feedback to help refine and perfect but that is probably a ways off yet. The site has only been in operation a month and this is an alpha.
Do specific groups count more than another?
I don't really know what you mean by this. Count more, as in?
If someone has a high trust score in ~talk but a low trust score in ~politics, should they be able to create a group?
No one user will likely ever be able to create groups, except admins like @deimos. So a person with high trust in ~talk would probably have a say in subgroup creation for ~talk commensurate with their level of trust. Similarly in ~politics but people with no trust anywhere will likely have little say in automatic, admin-free group creation.
I don't think limiting groups is the way to go - if groups aren't active and don't thrive, that's fine, let them die off. But creating groups is useful in so many ways - not everyone wants to discuss the same things, and if we wait for things to be voted as popular, niche subjects will never get their own corner of the site.
StackExchange has an interesting process for proposing new StackExchange topic sites. It's definitely way beyond what we'd need here, but there might be some interesting things to learn from it: https://area51.stackexchange.com/faq
Yeah, I agree with this. Making sure there's interest in a group before creating a group would make sure that there's no dead groups (at least to begin with).
I think groups should only be creatable after a sufficient number of tagged posts are made which would go there.
But why would someone tag a group that doesn't exist?
Maybe I'm misunderstanding what you mean, but for example, this post has a "baseball" tag, and if there were a lot of similar posts that could imply that we need a ~sports.baseball.
Sure, I meant “pinging” a group, not tagging posts, my fault. I might have misunderstood what @vakieh was saying.
That's essentially what is happening already. All the groups that have been added so far were from people expressing an interest in having that group exist and people voted those comments up and adding their own suggestions. At some point it may be a more automated process (e.g. Trusted Users being able to vote on subgroup creation, or automatically creating them based on tag usage statistics) but until trust is in place I don't see a straight democratic process as being a good idea... otherwise groups that run contrary to ~ goals (quality, primarily text-based content) will likely pop up over and over again just because they are popular (e.g. meme focused groups, or even hate groups).
Reddit requires users to be “trusted” to create a subreddit as well.
No it doesn't. Reddit only requires your account be at least 30 days old and has some positive karma (which is trivial to get) before you can create a subreddit. That is not even remotely the same thing as what trust/rep will be on ~, see here:
https://docs.tildes.net/mechanics-future#trustreputation-system-for-moderation
Will there be any sort of limitation on how many ~ people can moderate? One of the issues I've had with Reddit is many subs on a certain subject share the same exact moderation team. Take Nintendo for example. Nearly all of the Nintendo related subs share the same moderators.
iirc moderation of subs will be given to users who are active in said subs. I don't know if there's a hierarchy or not like on reddit.
Most likely, yes... though to what degree I don't know. I imagine it will be achieved automatically through trust decay (so it's simply impossible to gain and maintain enough trust to get significant power in too many groups) or as a hard cap setting but nothing is really decided yet.
Trust is going to be a complicated system requiring a lot of fine tuning but generally if it's a legitimate problem on reddit it's going to be attempted to be addressed here. And power mods and mod squatters are a recognized problem IMO. Also worth keep in mind is that the ideal here is every active user in a group slowly gains trust and gains access to "mod" like tools so rather than mods making up 0.X% of the community they will likely take up a significant portion of it so there will be significantly more users with powers and with the ability to monitor others actions to hold each other accountable.
You're right, but trust is also per-group. How would you tally trust? Do specific groups count more than another? If someone has a high trust score in ~talk but a low trust score in ~politics, should they be able to create a group?
In general through valuable activity in the community (proper use of "mod"-like actions, valuable contributions, etc), all of which will be judged by other trusted users in order to hold everyone accountable. The exact mechanics are still be worked on though and at some point will most likely require user feedback to help refine and perfect but that is probably a ways off yet. The site has only been in operation a month and this is an alpha.
I don't really know what you mean by this. Count more, as in?
No one user will likely ever be able to create groups, except admins like @deimos. So a person with high trust in ~talk would probably have a say in subgroup creation for ~talk commensurate with their level of trust. Similarly in ~politics but people with no trust anywhere will likely have little say in automatic, admin-free group creation.