> algorithms based on your personal taste,
only for recommending subs to try
> geolocation
only for recommending subs to try
> forced email
no it does not
> shadow banning
extremely rare, and per-sub, not reddit-wide
> banning of politically incorrect subreddits
extremely rare; most of why Reddit is publicly despised is because it allows politically incorrect subreddits.
from your list, it's the only thing it's not correct:
there's a site wide shadow-ban that is only done by admins. that means if you post something, nobody will see but for you everything will show up as normal. subreddit moderators can approve your comment and it will show up. this started as a measure against spammers but it seems like normal users get shadow banned too.
now there's normal ban, where you can be banned from a subreddit (you can't post but you can browse) and there's a reddit-wide ban (your account gets basically deleted).
and there's also AutoModerator "shadow-ban" which makes use of reddit's auto moderator to delete a user's comment as soon as it get posted.
In an age where any opposing opinions are consider "toxic" by some, I don't think trying to "detoxify" should be a goal. If you don't like the opinions of a subreddit, don't go to that one.
The other wave is the circle-jerk - a relentless and all-encompassing embrace of confirmation bias, institutionalized to keep a fundamentalist purity of thought. This is beyond a normal echo chamber, these are high powered ideology amplifiers.
We've never had the concept of unlimited freedom of speech, and most of the time that its been curtailed, it has been to protect against those who use it in bad faith (fire in a crowded theater, etc). That is the toxicity - IMO Reddit is going to learn this lesson far too late, and allow a group that doesn't want to be there anyway take down the whole site when they leave.
When you change your searching and indexing algorithms to suppress fans of a specific politician but not fans of other politicians, that's not really good faith.
The old reddit, allowed actual pedophiles to describe their issues in ask reddit threads.
Theres even a dark ama where a relation which would make people ill, was discussed.
----
Unfortunately what social media is exposing, is weaknesses in how human beings agglomerate.
If someone makes a forum for dead baby jokes - distasteful to many, but not harmful in private - it eventually attracts people who think that dead baby jokes are the norm.
Or in worse terms - a sub for making offcolor jokes soon gets overrun by people who think off color jokes are not a risque deviation from normal behavior, but is actually normal behavior.
----
On top of this, the internet and social communities are extremely low on contextual information - which is critical for most people to understand a conversation in real life.
This means that the only way someone can know what you mean when you are being vague, is if you make an constant effort for tonal and idea accuracy in your comments.
Obviously this is challenging, leading to a cascade of misunderstandings, which only serve to polarize groups more.
----
Theres a lot of research being done on how people behave online, and its just a grim picture.
How Community Feedback Shapes User Behavior (https://arxiv.org/abs/1405.1429)
The spreading of misinformation online( http://www.pnas.org/content/113/3/554.full)
The effect of the ban on hate subs on reddit (http://comp.social.gatech.edu/papers/cscw18-chand-hate.pdf)
There's no money to be made with that.
HackerNews is different cause it's just a funnel and an online play ground for folks at YC. The day they decide that the signal to noise ratio has decreased past their threshold, this place will be gone faster than the Arc server can shut down.
Reddit generates enough to continue being Reddit through the gold program.
What happened was Reddit got sold, and that large amount of money raised expectations about "making money", in particular, equating massive traffic to massive potential dollars.
Reddit would have been just fine, and would have seen few real financial issues had it not been burdened with providing a return for a rather large sale price.
Just because it can be sold, may not mean it should be.
There is an inherent conflict between users goals and how they derive value from a site like Reddit, and the builders who may or may not align with what attracted users in the first place.
It's a difficult problem, and one that Aaron spoke to a few times. The idea of sustainable community discussion was solid, and clearly resonant with a lot of people and probably the broadest demographics around.
Reddit appeals to everyone from teens on up. Kind of amazing.
It's current owners paid a price high enough to remove the sustainable discussion idea from the table, favoring a more monetized model.
That has had ripple effects. Not all good, though some are.
Today, Reddit has factions. The vast majority use Reddit casually, often on mobile. The basic app works, but not well.
There is a much harder core set, a good third or more, and I'm one of those, who just uses the old school browser interface, which works just fine on modern mobile.
This third contains many of the community minded people. There is growing angst over changes and a steady divergence in directions, leaving many who understand Reddit, and it's users well, in an increasingly rough place.
I don't see it ending well.
Frankly, someone is going to do the sustainable discussion thing again, and maybe get it right. Making gazillions won't be in the cards, but a very nice living will be.
Should that happen, whoever does it will own discussion big time.
The strength in users funding user discussion comes from the absence of that inherent conflict that always manifests when money gets too big and expectations fail to align with what users see as the real value.
Maybe people being the customers, in the sense of one another, not so much to deliver ADS, is worth exploring more. It won't be as sexy as something that sells, or that will IPO in a way that sets up founders for life, but it may well be the perfect place to deliver answers to some of the harder questions related to large group discussion dynamics.
I have been experimenting with the idea of not making a bunch of rules that define how shitty people can be to one another.
What is stated is being a good human is no risk. The more shitty someone is, the greater their costs and risks get.
Risks include bans, time outs and such.
Costs include various forms of enforced compliance being required to post.
Remedies for being a good human are basically a return to normal peer status.
In particular, a gradual escalation of cost to contribute long before a time out or ban happens, seems to signal to people. A lot more is recoverable.
I see others modeling these things, demonstrating deescalation, and in general trying to recover things before any admin type action is taken. I also see them mentoring newbies toward the safe positive. Her important as that kind of distributed person to person action does not require tools, and cannot be bought.
This very strongly suggests community norms can be used instead of explicit rules to manage things toward value and away from noise.
Ever notice how alphas looking to start shit have a keen interest in the rules? Of course! They want to know what they can get away with.
Take that away and their behavior does change.
A secondary part of all this involves evaluating risk reward for trolls and others making noise.
I'll not put it here, as my thoughts are incomplete and some work is in process, but I will say indirect means, including tools, automation, filters, rules are all indirect means, and do not address the risk reward inherent in causing grief.
I will also say we can't do much about the low risk inherent in just making an account to chat, but we can definitely impact the reward.
There are more ways to get at this than we see out there right now.
The problem is, rules are also a way for someone who was wronged to appeal. While it's a line that troublemakers want to thread to push boundaries, when rules are unwritten, communities often become a reflection of the will of the moderator rather than a community of diverse opinion. Codified rules are one of the steps we have to encourage diversity of thought, but of course there needs to be rule enforcement...
The appeals are rooted in norms, and the process is a dialog, kept recoverable, not a trial.
Explain to me how a set of rules isn't a formalized expression if the will of those who created them?
There is more to all of that, in particular, security and agency.
But, those details aside, norms operate much like rules, and are far more resonant, and community owned than rules alone are.
Finally, the organic and well distributed nature of norms tends to check varied and manipulative enforcement of rules. A primary example might be a flare up, vs nefarious intent to make noise and or cause grief.
Think family vs Roberts rules of order.
I'm not entirely convinced rules are indicated for all discussion forms and communities.
It's a human problem, and using humans and human ways has advantages.
I've been in communities that were very pleasant and open about a lot of things where the only rule was, quote, "don't be a dick". I've also been in communities, that despite extensive rules that would make make fellows from my law course cry in joy and fully oriented towards left/liberal, were cesspools of hatred.
I think in both cases it's more a matter of what the moderation will do with what they have. "You reap what you sow" probably applies in some way.
There must simply be a commitment to health and purpose of community.
On the other hand, with written rules it's also harder for moderation to abuse their rights.
Either way, there will always be this commitment, the community usually figures out what they need by themselves when the decision becomes necessary.
The best example for reddit is a national park, imo.
As for rules -
Rules are not the first line of your defense.
The first line of defense is the price of admission into your community.
I have seen a large number of forums now, and frankly most of the heavy lifting is done by the type of people and the topic.
For example - Boring, technical/professional forums / Intellectual or otherwise taxing topics / Result in a self selection process that reduces noise.
This is also supported by research on how different types of information are spread through networks - http://www.pnas.org/content/113/3/554.full
A classic example of how barriers to entry make an impact - take a look at the dwarf fortress forums around 2008-2010.
The game is arcane, and to grok the game you need to be willing to put in an unusual effort at the time.
The resulting main boards are pretty good, signal vs noise ratios are healthy.
----
The next major rule is that - general topics are bad. It allows people to farm tangential credibility, and anyone with an opinion can speak.
The worst offender are topics on religion(/identity) and politics.
If you allow pol on your forum, you are fighting a losing battle. Pol has the lowest barrier to entry, but actual Policy and Politics are managed by complicated facts and hidden information.
Politics affects everyone, and is designed to be associated with hot button topics.
This single handedly will poison and polarize your community.
----
Between these 2 rules, you can get enough positive starting runway to create a healthy community for a while. Eventually the community will grow old and degrade, but you will avoid a whole host of other problems.
You are on solid ground otherwise. :D
Norms, established members, and the concepts of owning our end of a discussion, as well as weighing what others say can do wonders for general fuckery.
Assuming a more general discussion, and or one where growth and or participation is desired, having strong norms in place immunizes the community against nefarious players.
When someone new shows up, the idea is to give them some rope. Let them self identify, and self select too.
When most of the community gets this, then the idea of raising both their cost and reducing their reward both pack a big punch.
You aren't wrong. What you put here does work, but it has a lot of friction, which may be undesirable.
See, a nefarious actor has a low risk and generally low cost to entry.
Why enter?
They also have a high reward potential, and it's that favorable ratio that consistently attracts them. Above someone mentions $10 as a barrier to entry.
$10 is cheap ass compared to the fun one can have! No joke. In my study of this, I've met many who would pay that easily.
If that friction, however high it needs to be, is desirable, fine. Raise the cost enough and the outcome will be a low noise, but also low churn, low growth community, generally speaking. A very compelling, but specific topic can alter that too.
Otherwise, the community, it's founders, moderators, others of status, influence, however that all is structured, only has control over the reward part, and they can have some control over the cost too.
See, the topic of discussion here is more general chatter. We have more focused things well on the way to being acceptable signal to noise. What we do t have is general community, say politics, or other broad topics in that same state.
It's that which I have focused on for a very long time. Personal interest.
Norms, and various discussion devices can deny nefarious actors joy, and can frankly make them regret their intent. At the same time, consistent calls to join the community, based on their own arguments framed in positive ways can work wonders.
One discussion norm is weighting of what one gets. When a clown calls you an ass, it's as laughable as it could be a basis for righteous indignation.
The vast majority of the time, righteous indignation is the response. "How dare you!" From there, the reward gets pretty good for the clown.
What happens when the response is laughable? One actually laughs, or does something basic like rate the shit?
I can tell you the reward on all that is much lower. When they realize that is a norm? Low joy, high pain potential, particularly when they figure out they really are seen as a clown, or ass.
All these things, and I'm leaving a lot out, as it would be a short book to model, get at reward.
Ask yourself, what does a reward look like?
Say it's attention, as another example.
Return that with endearment, and most often they will leave, or join and become a member with some genuine basis. They leave because the moment they are familiar, their behavior normalized, understood, it's boring. No fun. No joy.
They join, because they find out there really is a common basis and they didn't realize it. Often, an ask to join will actually work.
A few will just spew forth, and so raise their cost makes great sense.
Simple things, like say not allowing them 4 letter words, carry a very high, often funny cost that can always be removed on good behavior, expression of and demonstration of better intent.
The best is the community can see that, act accordingly and it's all public and largely transparent. Often, some members will step in to help. Often that works, amazingly, and when the basic norms behind it all are in place and solid.
Many other simple, subtle, always recoverable things can be used in tandem with strong community norms to inhibit noise, while promoting signal.
The best part about this kind of approach is it's well distributed. Model it to a few, they apply, model for a few more, and soon most active participants are largely immune.
An "infection" won't spread, and where it does, the cost tools are used to marginalize impact.
A lot really depends on community intent and whether friction to entry is desirable, or not.
The bigger problems cited here are most painful on general discussion, and most discussion is actually general.
Having more specifics out there is good, but not ultimately a solution for the general need and value humans find when they interact.
This is tribal. People seek tribes. A good tribe stands strong because it's members understand how to do that and why it matters.
An investment in those things can be very effective, and that was my point to make. The things you mention can also do that.
You aren't wrong. There are just cases where other ways may well deliver better results, or be better aligned to the community needs and goals.
Say those goals aren't to grow old and decline.
Avoiding those doesn't get us answers to hard questions. It does simply push the problems out of scope.
In some contexts, better understanding one another does a whole lot of good.
That is also a focus of mine personally.
My comments on should be taken with that idea as context.
Just because I have an honesty box for payments doesn't mean you're not obligated to pay for whatever goods/services.