Hacker News new | ask | show | jobs
by zozbot234 1587 days ago
GP was not banned for arguing, they were banned for placing a throwaway arguing/flamebait comment - "using php is stupid" - in the proper context, by providing some nuance. The fact that they were even banned for that, when at most they should have been warned had they been actually arguing, tells you everything there is to know about how well that 'subreddit' is being managed (hint: not very well).
1 comments

I've reread what GP was posted and he was banned for reason: personal attack, because he did say, that comment is stupid. For exactly the same reason you can get banned, if you use nuanced sarcasm or actually if you are touching anything that is toxic.

Warnings can not be sustainable, when such high volume of warned is involved - you have to keep records of those warnings. Since it is easy to create account in reddit, outright ban is a warning - to not to do the same with new account.

If you are a dick in RL and do snarly comments in RL, you will be banned from communication quicker than in reddit - in some cases even beaten up. So, the only approach is to use reddit and NH comments with attitude, that it is not real conversation.

> get banned, if you use nuanced sarcasm or actually if you are touching anything that is toxic.

I agree that this is a thing, of course. But it makes for very low quality discussion, since non-toxic users will just let the toxicity fester instead of engaging to mitigate it.

> you have to keep records of those warnings.

'Reddit Enhancement Suite' has that as a feature, and it could be added as such on the official site. Or it could be done unofficially in the "modmail".

> Since it is easy to create account in reddit, outright ban is a warning

Returning with a new account after being permbanned from a 'sub' counts as ban evasion, and will trigger a global site-wide ban for both accounts. Reddit has gotten very aggressive about this as of late, quite possibly using AI bots that will try to detect when such "ban evasion" is happening, and trigger global bans.

>>I agree that this is a thing, of course. But it makes for very low quality discussion, since non-toxic users will just let the toxicity fester instead of engaging to mitigate it.

I do not make rules of reddit(I feel that the discussion somehow has took that direction, where... I am defending reddit?) - I just speak from my experience.

There is a reason, why humans can not have on their mind more than one enemy - their brains can not keep track of all the enemies as keeping track of all of them is a nightmare scenario, where brainpower is used just to keep track of all of them. As long as mods are humans, there are no good solutions to keep track of all warnings - maybe AI can do that, but keeping all that information in head is too much garbage - if I have to manage any such forums, banning is the best option for warning, as my brains are not unlimited. If a person convinces, that he has learned a lesson, then the ban can be lifted. Some bans can even expire, so it does not really matter.

When I think of warning, it also has additional information: how long it has to stay, until it is forgotten(because it is not a ban), what that warning was about, because you can't really make warning to kid not to touch hot surface and scald for the same reason, if he touches electric power cable - they are completelly different cases, so I am looking from that viewpoint, that there does not exist proper warning system and 'Reddit Enhancement Suite' is not it as well - from the description it still sounds, that it is meant for banning - not warnings, where warnings are meant to educate and protect people. Warning system means a lot of involvement and without AI, that is not viable, especially when no one is going to pay you for that kind of job.

>>Returning with a new account after being permbanned from a 'sub' counts as ban evasion

Not, if you return with different email, as reddit can ban email, but not a person, so this rule can not be implemented to a person and can be read as guideline. Unless you are a complete idiot and post your id data and after being banned from sub, come back and state that you have returned, then that rule can be applied in that case. And I perceive that rule exactly how it is technically done - that they might ban account and maybe email, that was attached to it - rest takes too much effort and is not really possible to do by reddit(at this stage).

> I do not make rules of reddit(I feel that the discussion somehow has took that direction, where... I am defending reddit?) - I just speak from my experience.

I don't think you're "defending reddit", FWIW. I just wanted to push back lightly on some things you said. And I've also seen a system based on escalating "warnings" work very well at multiple places, including arguably here at HN!

> Not, if you return with different email ...

That's not what multiple users have said in the current discussion. Their AI bot watchdogs are a lot more trigger-happy than that these days. (Which is quite fine and well per se, since plenty of bans of spammers, grossly offensive abusers etc. are quite legitimate.)